diff --git a/.mailmap b/.mailmap index c8a21d72241f6878ff9731097d154da49ae48599..b77cd34cf8525b18a6f1777673f67bd75261aa27 100644 --- a/.mailmap +++ b/.mailmap @@ -426,6 +426,9 @@ Krzysztof Wilczyński Krzysztof Wilczyński Kshitiz Godara Kuninori Morimoto +Kuniyuki Iwashima +Kuniyuki Iwashima +Kuniyuki Iwashima Kuogee Hsieh Lee Jones Lee Jones @@ -719,6 +722,7 @@ Srinivas Ramana Sriram R Sriram Yagnaraman Stanislav Fomichev +Stanislav Fomichev Stefan Wahren Stéphane Witzmann Stephen Hemminger diff --git a/Documentation/ABI/testing/sysfs-driver-intel-xe-hwmon b/Documentation/ABI/testing/sysfs-driver-intel-xe-hwmon index 5a91dcccd3ac3fc663c1ebe621b7ed26fe1dc188..d9e2b17c6872058152766c3629149d0e429b0644 100644 --- a/Documentation/ABI/testing/sysfs-driver-intel-xe-hwmon +++ b/Documentation/ABI/testing/sysfs-driver-intel-xe-hwmon @@ -148,3 +148,51 @@ Contact: intel-xe@lists.freedesktop.org Description: RO. Fan 3 speed in RPM. Only supported for particular Intel Xe graphics platforms. + +What: /sys/bus/pci/drivers/xe/.../hwmon/hwmon/power1_cap +Date: May 2025 +KernelVersion: 6.15 +Contact: intel-xe@lists.freedesktop.org +Description: RW. Card burst (PL2) power limit in microwatts. + + The power controller will throttle the operating frequency + if the power averaged over a window (typically milli seconds) + exceeds this limit. A read value of 0 means that the PL2 + power limit is disabled, writing 0 disables the limit. + PL2 is greater than PL1 and its time window is lesser + compared to PL1. + + Only supported for particular Intel Xe graphics platforms. + +What: /sys/bus/pci/drivers/xe/.../hwmon/hwmon/power2_cap +Date: May 2025 +KernelVersion: 6.15 +Contact: intel-xe@lists.freedesktop.org +Description: RW. Package burst (PL2) power limit in microwatts. + + The power controller will throttle the operating frequency + if the power averaged over a window (typically milli seconds) + exceeds this limit. A read value of 0 means that the PL2 + power limit is disabled, writing 0 disables the limit. + PL2 is greater than PL1 and its time window is lesser + compared to PL1. + + Only supported for particular Intel Xe graphics platforms. + +What: /sys/bus/pci/drivers/xe/.../hwmon/hwmon/power1_cap_interval +Date: May 2025 +KernelVersion: 6.15 +Contact: intel-xe@lists.freedesktop.org +Description: RW. Card burst power limit interval (Tau in PL2/Tau) in + milliseconds over which sustained power is averaged. + + Only supported for particular Intel Xe graphics platforms. + +What: /sys/bus/pci/drivers/xe/.../hwmon/hwmon/power2_cap_interval +Date: May 2025 +KernelVersion: 6.15 +Contact: intel-xe@lists.freedesktop.org +Description: RW. Package burst power limit interval (Tau in PL2/Tau) in + milliseconds over which sustained power is averaged. + + Only supported for particular Intel Xe graphics platforms. diff --git a/Documentation/admin-guide/cifs/usage.rst b/Documentation/admin-guide/cifs/usage.rst index c09674a75a9e3c712339b3dd3b830397dadf4c84..d989ae5778ba317afefe95bfce6044ec38653463 100644 --- a/Documentation/admin-guide/cifs/usage.rst +++ b/Documentation/admin-guide/cifs/usage.rst @@ -270,6 +270,8 @@ configured for Unix Extensions (and the client has not disabled illegal Windows/NTFS/SMB characters to a remap range (this mount parameter is the default for SMB3). This remap (``mapposix``) range is also compatible with Mac (and "Services for Mac" on some older Windows). +When POSIX Extensions for SMB 3.1.1 are negotiated, remapping is automatically +disabled. CIFS VFS Mount Options ====================== diff --git a/Documentation/block/ublk.rst b/Documentation/block/ublk.rst index c368e1081b4111c581567058f87ecb52db08758b..8c4030bcabb6347a0012c66b019f065c7db61fc6 100644 --- a/Documentation/block/ublk.rst +++ b/Documentation/block/ublk.rst @@ -352,6 +352,83 @@ For reaching best IO performance, ublk server should align its segment parameter of `struct ublk_param_segment` with backend for avoiding unnecessary IO split, which usually hurts io_uring performance. +Auto Buffer Registration +------------------------ + +The ``UBLK_F_AUTO_BUF_REG`` feature automatically handles buffer registration +and unregistration for I/O requests, which simplifies the buffer management +process and reduces overhead in the ublk server implementation. + +This is another feature flag for using zero copy, and it is compatible with +``UBLK_F_SUPPORT_ZERO_COPY``. + +Feature Overview +~~~~~~~~~~~~~~~~ + +This feature automatically registers request buffers to the io_uring context +before delivering I/O commands to the ublk server and unregisters them when +completing I/O commands. This eliminates the need for manual buffer +registration/unregistration via ``UBLK_IO_REGISTER_IO_BUF`` and +``UBLK_IO_UNREGISTER_IO_BUF`` commands, then IO handling in ublk server +can avoid dependency on the two uring_cmd operations. + +IOs can't be issued concurrently to io_uring if there is any dependency +among these IOs. So this way not only simplifies ublk server implementation, +but also makes concurrent IO handling becomes possible by removing the +dependency on buffer registration & unregistration commands. + +Usage Requirements +~~~~~~~~~~~~~~~~~~ + +1. The ublk server must create a sparse buffer table on the same ``io_ring_ctx`` + used for ``UBLK_IO_FETCH_REQ`` and ``UBLK_IO_COMMIT_AND_FETCH_REQ``. If + uring_cmd is issued on a different ``io_ring_ctx``, manual buffer + unregistration is required. + +2. Buffer registration data must be passed via uring_cmd's ``sqe->addr`` with the + following structure:: + + struct ublk_auto_buf_reg { + __u16 index; /* Buffer index for registration */ + __u8 flags; /* Registration flags */ + __u8 reserved0; /* Reserved for future use */ + __u32 reserved1; /* Reserved for future use */ + }; + + ublk_auto_buf_reg_to_sqe_addr() is for converting the above structure into + ``sqe->addr``. + +3. All reserved fields in ``ublk_auto_buf_reg`` must be zeroed. + +4. Optional flags can be passed via ``ublk_auto_buf_reg.flags``. + +Fallback Behavior +~~~~~~~~~~~~~~~~~ + +If auto buffer registration fails: + +1. When ``UBLK_AUTO_BUF_REG_FALLBACK`` is enabled: + + - The uring_cmd is completed + - ``UBLK_IO_F_NEED_REG_BUF`` is set in ``ublksrv_io_desc.op_flags`` + - The ublk server must manually deal with the failure, such as, register + the buffer manually, or using user copy feature for retrieving the data + for handling ublk IO + +2. If fallback is not enabled: + + - The ublk I/O request fails silently + - The uring_cmd won't be completed + +Limitations +~~~~~~~~~~~ + +- Requires same ``io_ring_ctx`` for all operations +- May require manual buffer management in fallback cases +- io_ring_ctx buffer table has a max size of 16K, which may not be enough + in case that too many ublk devices are handled by this single io_ring_ctx + and each one has very large queue depth + References ========== diff --git a/Documentation/devicetree/bindings/display/msm/dp-controller.yaml b/Documentation/devicetree/bindings/display/msm/dp-controller.yaml index 246bbb509bea18bed32e3a442d0926a24498c960..9923b065323bbab99de5079b674a0317f3074373 100644 --- a/Documentation/devicetree/bindings/display/msm/dp-controller.yaml +++ b/Documentation/devicetree/bindings/display/msm/dp-controller.yaml @@ -38,6 +38,10 @@ properties: - qcom,sm8450-dp - qcom,sm8550-dp - const: qcom,sm8350-dp + - items: + - enum: + - qcom,sm8750-dp + - const: qcom,sm8650-dp reg: minItems: 4 diff --git a/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml b/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml index 82fe95a6d9599b5799549356451278564dc070de..d4bb65c660af8ce8a6bda129a8275c579a705871 100644 --- a/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml +++ b/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml @@ -42,6 +42,7 @@ properties: - qcom,sm8450-dsi-ctrl - qcom,sm8550-dsi-ctrl - qcom,sm8650-dsi-ctrl + - qcom,sm8750-dsi-ctrl - const: qcom,mdss-dsi-ctrl - enum: - qcom,dsi-ctrl-6g-qcm2290 @@ -70,11 +71,11 @@ properties: - mnoc:: MNOC clock - pixel:: Display pixel clock. minItems: 3 - maxItems: 9 + maxItems: 12 clock-names: minItems: 3 - maxItems: 9 + maxItems: 12 phys: maxItems: 1 @@ -109,7 +110,8 @@ properties: minItems: 2 maxItems: 4 description: | - Parents of "byte" and "pixel" for the given platform. + For DSI on SM8650 and older: parents of "byte" and "pixel" for the given + platform. For DSIv2 platforms this should contain "byte", "esc", "src" and "pixel_src" clocks. @@ -218,8 +220,6 @@ required: - clocks - clock-names - phys - - assigned-clocks - - assigned-clock-parents - ports allOf: @@ -244,6 +244,9 @@ allOf: - const: byte - const: pixel - const: core + required: + - assigned-clocks + - assigned-clock-parents - if: properties: @@ -266,6 +269,9 @@ allOf: - const: byte - const: pixel - const: core + required: + - assigned-clocks + - assigned-clock-parents - if: properties: @@ -288,6 +294,9 @@ allOf: - const: pixel - const: core - const: core_mmss + required: + - assigned-clocks + - assigned-clock-parents - if: properties: @@ -309,6 +318,9 @@ allOf: - const: core_mmss - const: pixel - const: core + required: + - assigned-clocks + - assigned-clock-parents - if: properties: @@ -346,6 +358,35 @@ allOf: - const: core - const: iface - const: bus + required: + - assigned-clocks + - assigned-clock-parents + + - if: + properties: + compatible: + contains: + enum: + - qcom,sm8750-dsi-ctrl + then: + properties: + clocks: + minItems: 12 + maxItems: 12 + clock-names: + items: + - const: byte + - const: byte_intf + - const: pixel + - const: core + - const: iface + - const: bus + - const: dsi_pll_pixel + - const: dsi_pll_byte + - const: esync + - const: osc + - const: byte_src + - const: pixel_src - if: properties: @@ -369,6 +410,9 @@ allOf: - const: core_mmss - const: pixel - const: core + required: + - assigned-clocks + - assigned-clock-parents unevaluatedProperties: false diff --git a/Documentation/devicetree/bindings/display/msm/dsi-phy-7nm.yaml b/Documentation/devicetree/bindings/display/msm/dsi-phy-7nm.yaml index 3c75ff42999a59183d5c6f9ad164023d6361ac07..1ca820a500b725233e161f53cbbbd59406326876 100644 --- a/Documentation/devicetree/bindings/display/msm/dsi-phy-7nm.yaml +++ b/Documentation/devicetree/bindings/display/msm/dsi-phy-7nm.yaml @@ -25,6 +25,7 @@ properties: - qcom,sm8450-dsi-phy-5nm - qcom,sm8550-dsi-phy-4nm - qcom,sm8650-dsi-phy-4nm + - qcom,sm8750-dsi-phy-3nm reg: items: diff --git a/Documentation/devicetree/bindings/display/msm/qcom,sm8650-dpu.yaml b/Documentation/devicetree/bindings/display/msm/qcom,sm8650-dpu.yaml index 01cf79bd754b491349c52c5aef49ba06e835d0bf..0a46120dd8680371ed031f7773859716f49c3aa1 100644 --- a/Documentation/devicetree/bindings/display/msm/qcom,sm8650-dpu.yaml +++ b/Documentation/devicetree/bindings/display/msm/qcom,sm8650-dpu.yaml @@ -16,6 +16,7 @@ properties: enum: - qcom,sa8775p-dpu - qcom,sm8650-dpu + - qcom,sm8750-dpu - qcom,x1e80100-dpu reg: diff --git a/Documentation/devicetree/bindings/display/msm/qcom,sm8750-mdss.yaml b/Documentation/devicetree/bindings/display/msm/qcom,sm8750-mdss.yaml new file mode 100644 index 0000000000000000000000000000000000000000..72c70edc1fb01c61f8aad24fdb58bfb4f62a6e34 --- /dev/null +++ b/Documentation/devicetree/bindings/display/msm/qcom,sm8750-mdss.yaml @@ -0,0 +1,470 @@ +# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/display/msm/qcom,sm8750-mdss.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Qualcomm SM8750 Display MDSS + +maintainers: + - Krzysztof Kozlowski + +description: + SM8650 MSM Mobile Display Subsystem(MDSS), which encapsulates sub-blocks like + DPU display controller, DSI and DP interfaces etc. + +$ref: /schemas/display/msm/mdss-common.yaml# + +properties: + compatible: + const: qcom,sm8750-mdss + + clocks: + items: + - description: Display AHB + - description: Display hf AXI + - description: Display core + + iommus: + maxItems: 1 + + interconnects: + items: + - description: Interconnect path from mdp0 port to the data bus + - description: Interconnect path from CPU to the reg bus + + interconnect-names: + items: + - const: mdp0-mem + - const: cpu-cfg + +patternProperties: + "^display-controller@[0-9a-f]+$": + type: object + additionalProperties: true + properties: + compatible: + const: qcom,sm8750-dpu + + "^displayport-controller@[0-9a-f]+$": + type: object + additionalProperties: true + properties: + compatible: + contains: + const: qcom,sm8750-dp + + "^dsi@[0-9a-f]+$": + type: object + additionalProperties: true + properties: + compatible: + contains: + const: qcom,sm8750-dsi-ctrl + + "^phy@[0-9a-f]+$": + type: object + additionalProperties: true + properties: + compatible: + const: qcom,sm8750-dsi-phy-3nm + +required: + - compatible + +unevaluatedProperties: false + +examples: + - | + #include + #include + #include + #include + #include + #include + + display-subsystem@ae00000 { + compatible = "qcom,sm8750-mdss"; + reg = <0x0ae00000 0x1000>; + reg-names = "mdss"; + + interrupts = ; + + clocks = <&disp_cc_mdss_ahb_clk>, + <&gcc_disp_hf_axi_clk>, + <&disp_cc_mdss_mdp_clk>; + + interconnects = <&mmss_noc MASTER_MDP QCOM_ICC_TAG_ALWAYS + &mc_virt SLAVE_EBI1 QCOM_ICC_TAG_ALWAYS>, + <&gem_noc MASTER_APPSS_PROC QCOM_ICC_TAG_ACTIVE_ONLY + &config_noc SLAVE_DISPLAY_CFG QCOM_ICC_TAG_ACTIVE_ONLY>; + interconnect-names = "mdp0-mem", + "cpu-cfg"; + + resets = <&disp_cc_mdss_core_bcr>; + + power-domains = <&mdss_gdsc>; + + iommus = <&apps_smmu 0x800 0x2>; + + interrupt-controller; + #interrupt-cells = <1>; + + #address-cells = <1>; + #size-cells = <1>; + ranges; + + display-controller@ae01000 { + compatible = "qcom,sm8750-dpu"; + reg = <0x0ae01000 0x93000>, + <0x0aeb0000 0x2008>; + reg-names = "mdp", + "vbif"; + + interrupts-extended = <&mdss 0>; + + clocks = <&gcc_disp_hf_axi_clk>, + <&disp_cc_mdss_ahb_clk>, + <&disp_cc_mdss_mdp_lut_clk>, + <&disp_cc_mdss_mdp_clk>, + <&disp_cc_mdss_vsync_clk>; + clock-names = "nrt_bus", + "iface", + "lut", + "core", + "vsync"; + + assigned-clocks = <&disp_cc_mdss_vsync_clk>; + assigned-clock-rates = <19200000>; + + operating-points-v2 = <&mdp_opp_table>; + + power-domains = <&rpmhpd RPMHPD_MMCX>; + + ports { + #address-cells = <1>; + #size-cells = <0>; + + port@0 { + reg = <0>; + + dpu_intf1_out: endpoint { + remote-endpoint = <&mdss_dsi0_in>; + }; + }; + + port@1 { + reg = <1>; + + dpu_intf2_out: endpoint { + remote-endpoint = <&mdss_dsi1_in>; + }; + }; + + port@2 { + reg = <2>; + + dpu_intf0_out: endpoint { + remote-endpoint = <&mdss_dp0_in>; + }; + }; + }; + + mdp_opp_table: opp-table { + compatible = "operating-points-v2"; + + opp-207000000 { + opp-hz = /bits/ 64 <207000000>; + required-opps = <&rpmhpd_opp_low_svs>; + }; + + opp-337000000 { + opp-hz = /bits/ 64 <337000000>; + required-opps = <&rpmhpd_opp_svs>; + }; + + opp-417000000 { + opp-hz = /bits/ 64 <417000000>; + required-opps = <&rpmhpd_opp_svs_l1>; + }; + + opp-532000000 { + opp-hz = /bits/ 64 <532000000>; + required-opps = <&rpmhpd_opp_nom>; + }; + + opp-575000000 { + opp-hz = /bits/ 64 <575000000>; + required-opps = <&rpmhpd_opp_nom_l1>; + }; + }; + }; + + dsi@ae94000 { + compatible = "qcom,sm8750-dsi-ctrl", "qcom,mdss-dsi-ctrl"; + reg = <0x0ae94000 0x400>; + reg-names = "dsi_ctrl"; + + interrupts-extended = <&mdss 4>; + + clocks = <&disp_cc_mdss_byte0_clk>, + <&disp_cc_mdss_byte0_intf_clk>, + <&disp_cc_mdss_pclk0_clk>, + <&disp_cc_mdss_esc0_clk>, + <&disp_cc_mdss_ahb_clk>, + <&gcc_disp_hf_axi_clk>, + <&mdss_dsi0_phy 1>, + <&mdss_dsi0_phy 0>, + <&disp_cc_esync0_clk>, + <&disp_cc_osc_clk>, + <&disp_cc_mdss_byte0_clk_src>, + <&disp_cc_mdss_pclk0_clk_src>; + clock-names = "byte", + "byte_intf", + "pixel", + "core", + "iface", + "bus", + "dsi_pll_pixel", + "dsi_pll_byte", + "esync", + "osc", + "byte_src", + "pixel_src"; + + operating-points-v2 = <&mdss_dsi_opp_table>; + + power-domains = <&rpmhpd RPMHPD_MMCX>; + + phys = <&mdss_dsi0_phy>; + phy-names = "dsi"; + + vdda-supply = <&vreg_l3g_1p2>; + + #address-cells = <1>; + #size-cells = <0>; + + ports { + #address-cells = <1>; + #size-cells = <0>; + + port@0 { + reg = <0>; + + mdss_dsi0_in: endpoint { + remote-endpoint = <&dpu_intf1_out>; + }; + }; + + port@1 { + reg = <1>; + + mdss_dsi0_out: endpoint { + remote-endpoint = <&panel0_in>; + data-lanes = <0 1 2 3>; + }; + }; + }; + + mdss_dsi_opp_table: opp-table { + compatible = "operating-points-v2"; + + opp-187500000 { + opp-hz = /bits/ 64 <187500000>; + required-opps = <&rpmhpd_opp_low_svs>; + }; + + opp-300000000 { + opp-hz = /bits/ 64 <300000000>; + required-opps = <&rpmhpd_opp_svs>; + }; + + opp-358000000 { + opp-hz = /bits/ 64 <358000000>; + required-opps = <&rpmhpd_opp_svs_l1>; + }; + }; + }; + + mdss_dsi0_phy: phy@ae95000 { + compatible = "qcom,sm8750-dsi-phy-3nm"; + reg = <0x0ae95000 0x200>, + <0x0ae95200 0x280>, + <0x0ae95500 0x400>; + reg-names = "dsi_phy", + "dsi_phy_lane", + "dsi_pll"; + + clocks = <&disp_cc_mdss_ahb_clk>, + <&rpmhcc RPMH_CXO_CLK>; + clock-names = "iface", + "ref"; + + vdds-supply = <&vreg_l3i_0p88>; + + #clock-cells = <1>; + #phy-cells = <0>; + }; + + dsi@ae96000 { + compatible = "qcom,sm8750-dsi-ctrl", "qcom,mdss-dsi-ctrl"; + reg = <0x0ae96000 0x400>; + reg-names = "dsi_ctrl"; + + interrupts-extended = <&mdss 5>; + + clocks = <&disp_cc_mdss_byte1_clk>, + <&disp_cc_mdss_byte1_intf_clk>, + <&disp_cc_mdss_pclk1_clk>, + <&disp_cc_mdss_esc1_clk>, + <&disp_cc_mdss_ahb_clk>, + <&gcc_disp_hf_axi_clk>, + <&mdss_dsi1_phy 1>, + <&mdss_dsi1_phy 0>, + <&disp_cc_esync1_clk>, + <&disp_cc_osc_clk>, + <&disp_cc_mdss_byte1_clk_src>, + <&disp_cc_mdss_pclk1_clk_src>; + clock-names = "byte", + "byte_intf", + "pixel", + "core", + "iface", + "bus", + "dsi_pll_pixel", + "dsi_pll_byte", + "esync", + "osc", + "byte_src", + "pixel_src"; + + operating-points-v2 = <&mdss_dsi_opp_table>; + + power-domains = <&rpmhpd RPMHPD_MMCX>; + + phys = <&mdss_dsi1_phy>; + phy-names = "dsi"; + + #address-cells = <1>; + #size-cells = <0>; + + ports { + #address-cells = <1>; + #size-cells = <0>; + + port@0 { + reg = <0>; + + mdss_dsi1_in: endpoint { + remote-endpoint = <&dpu_intf2_out>; + }; + }; + + port@1 { + reg = <1>; + + mdss_dsi1_out: endpoint { + }; + }; + }; + }; + + mdss_dsi1_phy: phy@ae97000 { + compatible = "qcom,sm8750-dsi-phy-3nm"; + reg = <0x0ae97000 0x200>, + <0x0ae97200 0x280>, + <0x0ae97500 0x400>; + reg-names = "dsi_phy", + "dsi_phy_lane", + "dsi_pll"; + + clocks = <&disp_cc_mdss_ahb_clk>, + <&rpmhcc RPMH_CXO_CLK>; + clock-names = "iface", + "ref"; + + #clock-cells = <1>; + #phy-cells = <0>; + }; + + displayport-controller@af54000 { + compatible = "qcom,sm8750-dp", "qcom,sm8650-dp"; + reg = <0xaf54000 0x104>, + <0xaf54200 0xc0>, + <0xaf55000 0x770>, + <0xaf56000 0x9c>, + <0xaf57000 0x9c>; + + interrupts-extended = <&mdss 12>; + + clocks = <&disp_cc_mdss_ahb_clk>, + <&disp_cc_mdss_dptx0_aux_clk>, + <&disp_cc_mdss_dptx0_link_clk>, + <&disp_cc_mdss_dptx0_link_intf_clk>, + <&disp_cc_mdss_dptx0_pixel0_clk>; + clock-names = "core_iface", + "core_aux", + "ctrl_link", + "ctrl_link_iface", + "stream_pixel"; + + assigned-clocks = <&disp_cc_mdss_dptx0_link_clk_src>, + <&disp_cc_mdss_dptx0_pixel0_clk_src>; + assigned-clock-parents = <&usb_dp_qmpphy QMP_USB43DP_DP_LINK_CLK>, + <&usb_dp_qmpphy QMP_USB43DP_DP_VCO_DIV_CLK>; + + operating-points-v2 = <&dp_opp_table>; + + power-domains = <&rpmhpd RPMHPD_MMCX>; + + phys = <&usb_dp_qmpphy QMP_USB43DP_DP_PHY>; + phy-names = "dp"; + + #sound-dai-cells = <0>; + + dp_opp_table: opp-table { + compatible = "operating-points-v2"; + + opp-192000000 { + opp-hz = /bits/ 64 <192000000>; + required-opps = <&rpmhpd_opp_low_svs_d1>; + }; + + opp-270000000 { + opp-hz = /bits/ 64 <270000000>; + required-opps = <&rpmhpd_opp_low_svs>; + }; + + opp-540000000 { + opp-hz = /bits/ 64 <540000000>; + required-opps = <&rpmhpd_opp_svs_l1>; + }; + + opp-810000000 { + opp-hz = /bits/ 64 <810000000>; + required-opps = <&rpmhpd_opp_nom>; + }; + }; + + ports { + #address-cells = <1>; + #size-cells = <0>; + + port@0 { + reg = <0>; + + mdss_dp0_in: endpoint { + remote-endpoint = <&dpu_intf0_out>; + }; + }; + + port@1 { + reg = <1>; + + mdss_dp0_out: endpoint { + remote-endpoint = <&usb_dp_qmpphy_dp_in>; + }; + }; + }; + }; + }; diff --git a/Documentation/devicetree/bindings/opp/opp-v2-qcom-adreno.yaml b/Documentation/devicetree/bindings/opp/opp-v2-qcom-adreno.yaml index a27ba7b663d456f964628a91a661b51a684de1be..0bd7d6b69755f5f53a045ba7b5e1d08030d980e6 100644 --- a/Documentation/devicetree/bindings/opp/opp-v2-qcom-adreno.yaml +++ b/Documentation/devicetree/bindings/opp/opp-v2-qcom-adreno.yaml @@ -23,7 +23,7 @@ properties: const: operating-points-v2-adreno patternProperties: - '^opp-[0-9]+$': + '^opp(-[0-9]+){1,2}$': type: object additionalProperties: false diff --git a/Documentation/devicetree/bindings/pinctrl/starfive,jh7110-aon-pinctrl.yaml b/Documentation/devicetree/bindings/pinctrl/starfive,jh7110-aon-pinctrl.yaml index b470901f5f562215e06aeee9bf99e74cd2b32056..4dbef86bd9581135e8ef75539c29bd999eb06cd2 100644 --- a/Documentation/devicetree/bindings/pinctrl/starfive,jh7110-aon-pinctrl.yaml +++ b/Documentation/devicetree/bindings/pinctrl/starfive,jh7110-aon-pinctrl.yaml @@ -15,7 +15,7 @@ description: | Some peripherals such as PWM have their I/O go through the 4 "GPIOs". maintainers: - - Jianlong Huang + - Hal Feng properties: compatible: diff --git a/Documentation/devicetree/bindings/pinctrl/starfive,jh7110-sys-pinctrl.yaml b/Documentation/devicetree/bindings/pinctrl/starfive,jh7110-sys-pinctrl.yaml index 222b9e240f8af9adb507834cde2dabd33029435f..e2a25a20f6a6e6cb86c4069bdedc6f6fe395f6f0 100644 --- a/Documentation/devicetree/bindings/pinctrl/starfive,jh7110-sys-pinctrl.yaml +++ b/Documentation/devicetree/bindings/pinctrl/starfive,jh7110-sys-pinctrl.yaml @@ -18,7 +18,7 @@ description: | any GPIO can be set up to be controlled by any of the peripherals. maintainers: - - Jianlong Huang + - Hal Feng properties: compatible: diff --git a/Documentation/filesystems/proc.rst b/Documentation/filesystems/proc.rst index 2a17865dfe39b5902383bc791c56d1be317db101..5236cb52e357dcd00496b26be8578e1dec0a345e 100644 --- a/Documentation/filesystems/proc.rst +++ b/Documentation/filesystems/proc.rst @@ -584,7 +584,6 @@ encoded manner. The codes are the following: ms may share gd stack segment growns down pf pure PFN range - dw disabled write to the mapped file lo pages are locked in memory io memory mapped I/O area sr sequential read advise provided @@ -607,8 +606,11 @@ encoded manner. The codes are the following: mt arm64 MTE allocation tags are enabled um userfaultfd missing tracking uw userfaultfd wr-protect tracking + ui userfaultfd minor fault ss shadow/guarded control stack page sl sealed + lf lock on fault pages + dp always lazily freeable mapping == ======================================= Note that there is no guarantee that every flag and associated mnemonic will diff --git a/Documentation/gpu/amdgpu/debugging.rst b/Documentation/gpu/amdgpu/debugging.rst index 7cbfea0606e15fe4b19f764b0ef0436105981170..ac914d5247410d5ebe39ff9cd93c92bdbbce9821 100644 --- a/Documentation/gpu/amdgpu/debugging.rst +++ b/Documentation/gpu/amdgpu/debugging.rst @@ -85,3 +85,21 @@ UMR GPU debugging and diagnostics tool. Please see the umr `documentation `_ for more information about its capabilities. + +Debugging backlight brightness +============================== +Default backlight brightness is intended to be set via the policy advertised +by the firmware. Firmware will often provide different defaults for AC or DC. +Furthermore, some userspace software will save backlight brightness during +the previous boot and attempt to restore it. + +Some firmware also has support for a feature called "Custom Backlight Curves" +where an input value for brightness is mapped along a linearly interpolated +curve of brightness values that better match display characteristics. + +In the event of problems happening with backlight, there is a trace event +that can be enabled at bootup to log every brightness change request. +This can help isolate where the problem is. To enable the trace event add +the following to the kernel command line: + + tp_printk trace_event=amdgpu_dm:amdgpu_dm_brightness:mod:amdgpu trace_buf_size=1M diff --git a/Documentation/gpu/i915.rst b/Documentation/gpu/i915.rst index 7a469df675d82b8f2dae7349bfd56ad5efb491bb..72932fa31b8d405c71910df02898d313e0f67231 100644 --- a/Documentation/gpu/i915.rst +++ b/Documentation/gpu/i915.rst @@ -112,10 +112,10 @@ panel self refresh. Atomic Plane Helpers -------------------- -.. kernel-doc:: drivers/gpu/drm/i915/display/intel_atomic_plane.c +.. kernel-doc:: drivers/gpu/drm/i915/display/intel_plane.c :doc: atomic plane helpers -.. kernel-doc:: drivers/gpu/drm/i915/display/intel_atomic_plane.c +.. kernel-doc:: drivers/gpu/drm/i915/display/intel_plane.c :internal: Asynchronous Page Flip @@ -204,6 +204,12 @@ DMC Firmware Support .. kernel-doc:: drivers/gpu/drm/i915/display/intel_dmc.c :internal: +DMC Flip Queue +-------------------- + +.. kernel-doc:: drivers/gpu/drm/i915/display/intel_flipq.c + :doc: DMC Flip Queue + DMC wakelock support -------------------- diff --git a/Documentation/gpu/xe/xe_configfs.rst b/Documentation/gpu/xe/xe_configfs.rst index 9b9d941eb20ed620a9513372886b2e1e2b1c2f69..7f8ec39dc6dd5fbf14db07528aa3c8e1fc6912ed 100644 --- a/Documentation/gpu/xe/xe_configfs.rst +++ b/Documentation/gpu/xe/xe_configfs.rst @@ -2,9 +2,15 @@ .. _xe_configfs: -============ +=========== Xe Configfs -============ +=========== .. kernel-doc:: drivers/gpu/drm/xe/xe_configfs.c :doc: Xe Configfs + +Internal API +============ + +.. kernel-doc:: drivers/gpu/drm/xe/xe_configfs.c + :internal: diff --git a/MAINTAINERS b/MAINTAINERS index f69a86b9610aa0708031855b3289cca4dc3a157f..ee2ef9d9db2ab3a2b462aaa1a7a473018c70596d 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -4555,6 +4555,7 @@ BPF [NETWORKING] (tcx & tc BPF, sock_addr) M: Martin KaFai Lau M: Daniel Borkmann R: John Fastabend +R: Stanislav Fomichev L: bpf@vger.kernel.org L: netdev@vger.kernel.org S: Maintained @@ -6254,6 +6255,7 @@ F: include/linux/cpuhotplug.h F: include/linux/smpboot.h F: kernel/cpu.c F: kernel/smpboot.* +F: rust/helper/cpu.c F: rust/kernel/cpu.rs CPU IDLE TIME MANAGEMENT FRAMEWORK @@ -15958,6 +15960,7 @@ R: Liam R. Howlett R: Nico Pache R: Ryan Roberts R: Dev Jain +R: Barry Song L: linux-mm@kvack.org S: Maintained W: http://www.linux-mm.org @@ -17532,7 +17535,7 @@ F: tools/testing/selftests/net/srv6* NETWORKING [TCP] M: Eric Dumazet M: Neal Cardwell -R: Kuniyuki Iwashima +R: Kuniyuki Iwashima L: netdev@vger.kernel.org S: Maintained F: Documentation/networking/net_cachelines/tcp_sock.rst @@ -17562,7 +17565,7 @@ F: net/tls/* NETWORKING [SOCKETS] M: Eric Dumazet -M: Kuniyuki Iwashima +M: Kuniyuki Iwashima M: Paolo Abeni M: Willem de Bruijn S: Maintained @@ -17577,7 +17580,7 @@ F: net/core/scm.c F: net/socket.c NETWORKING [UNIX SOCKETS] -M: Kuniyuki Iwashima +M: Kuniyuki Iwashima S: Maintained F: include/net/af_unix.h F: include/net/netns/unix.h @@ -23701,7 +23704,6 @@ F: include/dt-bindings/clock/starfive?jh71*.h STARFIVE JH71X0 PINCTRL DRIVERS M: Emil Renner Berthing -M: Jianlong Huang M: Hal Feng L: linux-gpio@vger.kernel.org S: Maintained @@ -27007,6 +27009,7 @@ M: David S. Miller M: Jakub Kicinski M: Jesper Dangaard Brouer M: John Fastabend +R: Stanislav Fomichev L: netdev@vger.kernel.org L: bpf@vger.kernel.org S: Supported @@ -27028,6 +27031,7 @@ M: Björn Töpel M: Magnus Karlsson M: Maciej Fijalkowski R: Jonathan Lemon +R: Stanislav Fomichev L: netdev@vger.kernel.org L: bpf@vger.kernel.org S: Maintained diff --git a/Makefile b/Makefile index 35e6e5240c61a8a329011929fcd0352b881dccdc..ba0827a1fccd9dd1cc375d0eecb7e2cfd9d48b83 100644 --- a/Makefile +++ b/Makefile @@ -2,7 +2,7 @@ VERSION = 6 PATCHLEVEL = 16 SUBLEVEL = 0 -EXTRAVERSION = -rc1 +EXTRAVERSION = -rc2 NAME = Baby Opossum Posse # *DOCUMENTATION* @@ -1832,12 +1832,9 @@ rustfmtcheck: rustfmt # Misc # --------------------------------------------------------------------------- -# Run misc checks when ${KBUILD_EXTRA_WARN} contains 1 PHONY += misc-check -ifneq ($(findstring 1,$(KBUILD_EXTRA_WARN)),) misc-check: $(Q)$(srctree)/scripts/misc-check -endif all: misc-check diff --git a/arch/alpha/include/asm/pgtable.h b/arch/alpha/include/asm/pgtable.h index 2676017f42f1759483608e3a5c46beb8625763d0..44e7aedac6e87e9b390fd711ff985d1a235d93d8 100644 --- a/arch/alpha/include/asm/pgtable.h +++ b/arch/alpha/include/asm/pgtable.h @@ -327,7 +327,7 @@ extern inline pte_t mk_swap_pte(unsigned long type, unsigned long offset) #define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) }) #define __swp_entry_to_pte(x) ((pte_t) { (x).val }) -static inline int pte_swp_exclusive(pte_t pte) +static inline bool pte_swp_exclusive(pte_t pte) { return pte_val(pte) & _PAGE_SWP_EXCLUSIVE; } diff --git a/arch/arc/include/asm/arcregs.h b/arch/arc/include/asm/arcregs.h index 005d9e4d187a0440246b131c1ee0852110336dc4..a31bbf5c8bbc8a18d1d90782d89460ce1cf575aa 100644 --- a/arch/arc/include/asm/arcregs.h +++ b/arch/arc/include/asm/arcregs.h @@ -144,7 +144,7 @@ #define ARC_AUX_AGU_MOD2 0x5E2 #define ARC_AUX_AGU_MOD3 0x5E3 -#ifndef __ASSEMBLY__ +#ifndef __ASSEMBLER__ #include diff --git a/arch/arc/include/asm/atomic.h b/arch/arc/include/asm/atomic.h index 592d7fffc223c923f4d63ccf27ad01fb9dc56e49..e615c42b93babd9496dbdb375d413cb90130d530 100644 --- a/arch/arc/include/asm/atomic.h +++ b/arch/arc/include/asm/atomic.h @@ -6,7 +6,7 @@ #ifndef _ASM_ARC_ATOMIC_H #define _ASM_ARC_ATOMIC_H -#ifndef __ASSEMBLY__ +#ifndef __ASSEMBLER__ #include #include @@ -31,6 +31,6 @@ #include #endif -#endif /* !__ASSEMBLY__ */ +#endif /* !__ASSEMBLER__ */ #endif diff --git a/arch/arc/include/asm/atomic64-arcv2.h b/arch/arc/include/asm/atomic64-arcv2.h index 9b5791b8547133b1d7ec4c53ab4a2f611857d165..73080a664369b447d8e67dc973e1fa9822e5af60 100644 --- a/arch/arc/include/asm/atomic64-arcv2.h +++ b/arch/arc/include/asm/atomic64-arcv2.h @@ -137,12 +137,9 @@ ATOMIC64_OPS(xor, xor, xor) #undef ATOMIC64_OP_RETURN #undef ATOMIC64_OP -static inline s64 -arch_atomic64_cmpxchg(atomic64_t *ptr, s64 expected, s64 new) +static inline u64 __arch_cmpxchg64_relaxed(volatile void *ptr, u64 old, u64 new) { - s64 prev; - - smp_mb(); + u64 prev; __asm__ __volatile__( "1: llockd %0, [%1] \n" @@ -152,14 +149,12 @@ arch_atomic64_cmpxchg(atomic64_t *ptr, s64 expected, s64 new) " bnz 1b \n" "2: \n" : "=&r"(prev) - : "r"(ptr), "ir"(expected), "r"(new) - : "cc"); /* memory clobber comes from smp_mb() */ - - smp_mb(); + : "r"(ptr), "ir"(old), "r"(new) + : "memory", "cc"); return prev; } -#define arch_atomic64_cmpxchg arch_atomic64_cmpxchg +#define arch_cmpxchg64_relaxed __arch_cmpxchg64_relaxed static inline s64 arch_atomic64_xchg(atomic64_t *ptr, s64 new) { diff --git a/arch/arc/include/asm/bitops.h b/arch/arc/include/asm/bitops.h index f5a936496f06007b94ee08f4b88426f532604796..5340c287139276a4baead9b2aacbcc6a0272ec57 100644 --- a/arch/arc/include/asm/bitops.h +++ b/arch/arc/include/asm/bitops.h @@ -10,7 +10,7 @@ #error only can be included directly #endif -#ifndef __ASSEMBLY__ +#ifndef __ASSEMBLER__ #include #include @@ -192,6 +192,6 @@ static inline __attribute__ ((const)) unsigned long __ffs(unsigned long x) #include #include -#endif /* !__ASSEMBLY__ */ +#endif /* !__ASSEMBLER__ */ #endif diff --git a/arch/arc/include/asm/bug.h b/arch/arc/include/asm/bug.h index 4c453ba96c51956a0314eaa87d0a92959c725e9a..171c16021f709663b8f78f8728b65d3ad0cdbe42 100644 --- a/arch/arc/include/asm/bug.h +++ b/arch/arc/include/asm/bug.h @@ -6,7 +6,7 @@ #ifndef _ASM_ARC_BUG_H #define _ASM_ARC_BUG_H -#ifndef __ASSEMBLY__ +#ifndef __ASSEMBLER__ #include @@ -29,6 +29,6 @@ void die(const char *str, struct pt_regs *regs, unsigned long address); #include -#endif /* !__ASSEMBLY__ */ +#endif /* !__ASSEMBLER__ */ #endif diff --git a/arch/arc/include/asm/cache.h b/arch/arc/include/asm/cache.h index f0f1fc5d62b662956a2dc42e70dae07f3cac17a3..040a97f4dd829dde90879ea8b9ad57f38cbcd767 100644 --- a/arch/arc/include/asm/cache.h +++ b/arch/arc/include/asm/cache.h @@ -23,7 +23,7 @@ */ #define ARC_UNCACHED_ADDR_SPACE 0xc0000000 -#ifndef __ASSEMBLY__ +#ifndef __ASSEMBLER__ #include @@ -65,7 +65,7 @@ extern int ioc_enable; extern unsigned long perip_base, perip_end; -#endif /* !__ASSEMBLY__ */ +#endif /* !__ASSEMBLER__ */ /* Instruction cache related Auxiliary registers */ #define ARC_REG_IC_BCR 0x77 /* Build Config reg */ diff --git a/arch/arc/include/asm/current.h b/arch/arc/include/asm/current.h index 06be89f6f2f052eb65fb76398d84e6fd77cd92c8..03ffd005f3fa6e10cb9ddfc16518f95a4fc61efa 100644 --- a/arch/arc/include/asm/current.h +++ b/arch/arc/include/asm/current.h @@ -9,7 +9,7 @@ #ifndef _ASM_ARC_CURRENT_H #define _ASM_ARC_CURRENT_H -#ifndef __ASSEMBLY__ +#ifndef __ASSEMBLER__ #ifdef CONFIG_ARC_CURR_IN_REG @@ -20,6 +20,6 @@ register struct task_struct *curr_arc asm("gp"); #include #endif /* ! CONFIG_ARC_CURR_IN_REG */ -#endif /* ! __ASSEMBLY__ */ +#endif /* ! __ASSEMBLER__ */ #endif /* _ASM_ARC_CURRENT_H */ diff --git a/arch/arc/include/asm/dsp-impl.h b/arch/arc/include/asm/dsp-impl.h index cd5636dfeb6f4119013ace57fa5662e129fe7947..fd5fdaad90c169819b93aa35358aced9feab2d93 100644 --- a/arch/arc/include/asm/dsp-impl.h +++ b/arch/arc/include/asm/dsp-impl.h @@ -11,7 +11,7 @@ #define DSP_CTRL_DISABLED_ALL 0 -#ifdef __ASSEMBLY__ +#ifdef __ASSEMBLER__ /* clobbers r5 register */ .macro DSP_EARLY_INIT diff --git a/arch/arc/include/asm/dsp.h b/arch/arc/include/asm/dsp.h index f496dbc4640b2461a7ca7696c21a19dcb6d2d6be..eeaaf4e4eabd31860cea96567f99cc121bb16613 100644 --- a/arch/arc/include/asm/dsp.h +++ b/arch/arc/include/asm/dsp.h @@ -7,7 +7,7 @@ #ifndef __ASM_ARC_DSP_H #define __ASM_ARC_DSP_H -#ifndef __ASSEMBLY__ +#ifndef __ASSEMBLER__ /* * DSP-related saved registers - need to be saved only when you are @@ -24,6 +24,6 @@ struct dsp_callee_regs { #endif }; -#endif /* !__ASSEMBLY__ */ +#endif /* !__ASSEMBLER__ */ #endif /* __ASM_ARC_DSP_H */ diff --git a/arch/arc/include/asm/dwarf.h b/arch/arc/include/asm/dwarf.h index a0d5ebe1bc3f5f37280d9ff37ad3df4827af89fb..1524c5cf8b59e6aea8bda958c7e28650a5828d5a 100644 --- a/arch/arc/include/asm/dwarf.h +++ b/arch/arc/include/asm/dwarf.h @@ -6,7 +6,7 @@ #ifndef _ASM_ARC_DWARF_H #define _ASM_ARC_DWARF_H -#ifdef __ASSEMBLY__ +#ifdef __ASSEMBLER__ #ifdef ARC_DW2_UNWIND_AS_CFI @@ -38,6 +38,6 @@ #endif /* !ARC_DW2_UNWIND_AS_CFI */ -#endif /* __ASSEMBLY__ */ +#endif /* __ASSEMBLER__ */ #endif /* _ASM_ARC_DWARF_H */ diff --git a/arch/arc/include/asm/entry.h b/arch/arc/include/asm/entry.h index 38c35722cebf0312899bf8ac63562550a779dd4e..f453af251a1a1324cb1fc4e2db23d0bea9c29c53 100644 --- a/arch/arc/include/asm/entry.h +++ b/arch/arc/include/asm/entry.h @@ -13,7 +13,7 @@ #include /* For VMALLOC_START */ #include -#ifdef __ASSEMBLY__ +#ifdef __ASSEMBLER__ #ifdef CONFIG_ISA_ARCOMPACT #include /* ISA specific bits */ @@ -146,7 +146,7 @@ #endif /* CONFIG_ARC_CURR_IN_REG */ -#else /* !__ASSEMBLY__ */ +#else /* !__ASSEMBLER__ */ extern void do_signal(struct pt_regs *); extern void do_notify_resume(struct pt_regs *); diff --git a/arch/arc/include/asm/irqflags-arcv2.h b/arch/arc/include/asm/irqflags-arcv2.h index fb3c21f1a2383d718857dfdd66da4fd3a8ccd2bf..30aea562f8aa6f3154548408ac486debcd336b91 100644 --- a/arch/arc/include/asm/irqflags-arcv2.h +++ b/arch/arc/include/asm/irqflags-arcv2.h @@ -50,7 +50,7 @@ #define ISA_INIT_STATUS_BITS (STATUS_IE_MASK | __AD_ENB | \ (ARCV2_IRQ_DEF_PRIO << 1)) -#ifndef __ASSEMBLY__ +#ifndef __ASSEMBLER__ /* * Save IRQ state and disable IRQs @@ -170,6 +170,6 @@ static inline void arc_softirq_clear(int irq) seti .endm -#endif /* __ASSEMBLY__ */ +#endif /* __ASSEMBLER__ */ #endif diff --git a/arch/arc/include/asm/irqflags-compact.h b/arch/arc/include/asm/irqflags-compact.h index 936a2f21f315ed881ff9a972ac8e68d6d93170e2..85c2f6bcde0c1e7fa5d21f5f4eeeb129beff640e 100644 --- a/arch/arc/include/asm/irqflags-compact.h +++ b/arch/arc/include/asm/irqflags-compact.h @@ -40,7 +40,7 @@ #define ISA_INIT_STATUS_BITS STATUS_IE_MASK -#ifndef __ASSEMBLY__ +#ifndef __ASSEMBLER__ /****************************************************************** * IRQ Control Macros @@ -196,6 +196,6 @@ static inline int arch_irqs_disabled(void) flag \scratch .endm -#endif /* __ASSEMBLY__ */ +#endif /* __ASSEMBLER__ */ #endif diff --git a/arch/arc/include/asm/jump_label.h b/arch/arc/include/asm/jump_label.h index a339223d9e052b35ea678d6a3e60faf6e5673671..66ead75784d973ba57cc39f6dc0b81a9ca16af1e 100644 --- a/arch/arc/include/asm/jump_label.h +++ b/arch/arc/include/asm/jump_label.h @@ -2,7 +2,7 @@ #ifndef _ASM_ARC_JUMP_LABEL_H #define _ASM_ARC_JUMP_LABEL_H -#ifndef __ASSEMBLY__ +#ifndef __ASSEMBLER__ #include #include @@ -68,5 +68,5 @@ struct jump_entry { jump_label_t key; }; -#endif /* __ASSEMBLY__ */ +#endif /* __ASSEMBLER__ */ #endif diff --git a/arch/arc/include/asm/linkage.h b/arch/arc/include/asm/linkage.h index 8a3fb71e9cfad2a31c92218611b6a7256e67e2f5..ba3cb65b5eaa44f63261099c4ed64d405cfd255b 100644 --- a/arch/arc/include/asm/linkage.h +++ b/arch/arc/include/asm/linkage.h @@ -12,7 +12,7 @@ #define __ALIGN .align 4 #define __ALIGN_STR __stringify(__ALIGN) -#ifdef __ASSEMBLY__ +#ifdef __ASSEMBLER__ .macro ST2 e, o, off #ifdef CONFIG_ARC_HAS_LL64 @@ -61,7 +61,7 @@ CFI_ENDPROC ASM_NL \ .size name, .-name -#else /* !__ASSEMBLY__ */ +#else /* !__ASSEMBLER__ */ #ifdef CONFIG_ARC_HAS_ICCM #define __arcfp_code __section(".text.arcfp") @@ -75,6 +75,6 @@ #define __arcfp_data __section(".data") #endif -#endif /* __ASSEMBLY__ */ +#endif /* __ASSEMBLER__ */ #endif diff --git a/arch/arc/include/asm/mmu-arcv2.h b/arch/arc/include/asm/mmu-arcv2.h index 41412642f27963f704b601896096737e689e27d5..5e5482026ac9007b2f6de080b7288de5d0092e04 100644 --- a/arch/arc/include/asm/mmu-arcv2.h +++ b/arch/arc/include/asm/mmu-arcv2.h @@ -69,7 +69,7 @@ #define PTE_BITS_NON_RWX_IN_PD1 (PAGE_MASK_PHYS | _PAGE_CACHEABLE) -#ifndef __ASSEMBLY__ +#ifndef __ASSEMBLER__ struct mm_struct; extern int pae40_exist_but_not_enab(void); @@ -100,6 +100,6 @@ static inline void mmu_setup_pgd(struct mm_struct *mm, void *pgd) sr \reg, [ARC_REG_PID] .endm -#endif /* !__ASSEMBLY__ */ +#endif /* !__ASSEMBLER__ */ #endif diff --git a/arch/arc/include/asm/mmu.h b/arch/arc/include/asm/mmu.h index 4ae2db59d494cb7932f129a349aff16ca967dc3d..e3b35ceab582b626b6a13c2a393e84dd316c61d8 100644 --- a/arch/arc/include/asm/mmu.h +++ b/arch/arc/include/asm/mmu.h @@ -6,7 +6,7 @@ #ifndef _ASM_ARC_MMU_H #define _ASM_ARC_MMU_H -#ifndef __ASSEMBLY__ +#ifndef __ASSEMBLER__ #include /* NR_CPUS */ diff --git a/arch/arc/include/asm/page.h b/arch/arc/include/asm/page.h index def0dfb95b4369710fa1aac67aa920893fcd92c8..9720fe6b2c24577ef1a12aab1184b957c72d8a20 100644 --- a/arch/arc/include/asm/page.h +++ b/arch/arc/include/asm/page.h @@ -19,7 +19,7 @@ #endif /* CONFIG_ARC_HAS_PAE40 */ -#ifndef __ASSEMBLY__ +#ifndef __ASSEMBLER__ #define clear_page(paddr) memset((paddr), 0, PAGE_SIZE) #define copy_user_page(to, from, vaddr, pg) copy_page(to, from) @@ -136,6 +136,6 @@ static inline unsigned long virt_to_pfn(const void *kaddr) #include /* page_to_pfn, pfn_to_page */ #include -#endif /* !__ASSEMBLY__ */ +#endif /* !__ASSEMBLER__ */ #endif diff --git a/arch/arc/include/asm/pgtable-bits-arcv2.h b/arch/arc/include/asm/pgtable-bits-arcv2.h index 8ebec1b21d246e61aa688c152dba9b5e2e564104..4630c5acca05d4a758e380f55c495330807ee90e 100644 --- a/arch/arc/include/asm/pgtable-bits-arcv2.h +++ b/arch/arc/include/asm/pgtable-bits-arcv2.h @@ -75,7 +75,7 @@ * This is to enable COW mechanism */ /* xwr */ -#ifndef __ASSEMBLY__ +#ifndef __ASSEMBLER__ #define pte_write(pte) (pte_val(pte) & _PAGE_WRITE) #define pte_dirty(pte) (pte_val(pte) & _PAGE_DIRTY) @@ -130,7 +130,7 @@ void update_mmu_cache_range(struct vm_fault *vmf, struct vm_area_struct *vma, #define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) }) #define __swp_entry_to_pte(x) ((pte_t) { (x).val }) -static inline int pte_swp_exclusive(pte_t pte) +static inline bool pte_swp_exclusive(pte_t pte) { return pte_val(pte) & _PAGE_SWP_EXCLUSIVE; } @@ -142,6 +142,6 @@ PTE_BIT_FUNC(swp_clear_exclusive, &= ~(_PAGE_SWP_EXCLUSIVE)); #include #endif -#endif /* __ASSEMBLY__ */ +#endif /* __ASSEMBLER__ */ #endif diff --git a/arch/arc/include/asm/pgtable-levels.h b/arch/arc/include/asm/pgtable-levels.h index d1ce4b0f1071b69d794ba1375945ff4691434d7c..c8f9273372c073a2e42517ac0a62b5ea868b5da3 100644 --- a/arch/arc/include/asm/pgtable-levels.h +++ b/arch/arc/include/asm/pgtable-levels.h @@ -85,7 +85,7 @@ #define PTRS_PER_PTE BIT(PMD_SHIFT - PAGE_SHIFT) -#ifndef __ASSEMBLY__ +#ifndef __ASSEMBLER__ #if CONFIG_PGTABLE_LEVELS > 3 #include @@ -181,6 +181,6 @@ #define pmd_leaf(x) (pmd_val(x) & _PAGE_HW_SZ) #endif -#endif /* !__ASSEMBLY__ */ +#endif /* !__ASSEMBLER__ */ #endif diff --git a/arch/arc/include/asm/pgtable.h b/arch/arc/include/asm/pgtable.h index 4cf45a99fd792a5e1b528c96a610cfdd9120f14e..bd580e2b62d7c073ef82a9ed1a93755996b67643 100644 --- a/arch/arc/include/asm/pgtable.h +++ b/arch/arc/include/asm/pgtable.h @@ -19,7 +19,7 @@ */ #define USER_PTRS_PER_PGD (TASK_SIZE / PGDIR_SIZE) -#ifndef __ASSEMBLY__ +#ifndef __ASSEMBLER__ extern char empty_zero_page[PAGE_SIZE]; #define ZERO_PAGE(vaddr) (virt_to_page(empty_zero_page)) @@ -29,6 +29,6 @@ extern pgd_t swapper_pg_dir[] __aligned(PAGE_SIZE); /* to cope with aliasing VIPT cache */ #define HAVE_ARCH_UNMAPPED_AREA -#endif /* __ASSEMBLY__ */ +#endif /* __ASSEMBLER__ */ #endif diff --git a/arch/arc/include/asm/processor.h b/arch/arc/include/asm/processor.h index d606658e2fe73149ef3afd35cd2e50b08f1d5edd..7f7901ac6643c395b08cd2d275753abd7f0d71f7 100644 --- a/arch/arc/include/asm/processor.h +++ b/arch/arc/include/asm/processor.h @@ -11,7 +11,7 @@ #ifndef __ASM_ARC_PROCESSOR_H #define __ASM_ARC_PROCESSOR_H -#ifndef __ASSEMBLY__ +#ifndef __ASSEMBLER__ #include #include @@ -66,7 +66,7 @@ extern void start_thread(struct pt_regs * regs, unsigned long pc, extern unsigned int __get_wchan(struct task_struct *p); -#endif /* !__ASSEMBLY__ */ +#endif /* !__ASSEMBLER__ */ /* * Default System Memory Map on ARC diff --git a/arch/arc/include/asm/ptrace.h b/arch/arc/include/asm/ptrace.h index cf79df0b257053494fcc20ba48c6105828bd4318..f6c052af8f4d308000a1ede319be057f89731619 100644 --- a/arch/arc/include/asm/ptrace.h +++ b/arch/arc/include/asm/ptrace.h @@ -10,7 +10,7 @@ #include #include -#ifndef __ASSEMBLY__ +#ifndef __ASSEMBLER__ typedef union { struct { @@ -172,6 +172,6 @@ static inline unsigned long regs_get_register(struct pt_regs *regs, extern int syscall_trace_enter(struct pt_regs *); extern void syscall_trace_exit(struct pt_regs *); -#endif /* !__ASSEMBLY__ */ +#endif /* !__ASSEMBLER__ */ #endif /* __ASM_PTRACE_H */ diff --git a/arch/arc/include/asm/switch_to.h b/arch/arc/include/asm/switch_to.h index 1f85de8288b17ba72d3f1f88fb9848ded86e8d0f..5806106a65f9016b270727f3ef8c121eab2a66fd 100644 --- a/arch/arc/include/asm/switch_to.h +++ b/arch/arc/include/asm/switch_to.h @@ -6,7 +6,7 @@ #ifndef _ASM_ARC_SWITCH_TO_H #define _ASM_ARC_SWITCH_TO_H -#ifndef __ASSEMBLY__ +#ifndef __ASSEMBLER__ #include #include diff --git a/arch/arc/include/asm/thread_info.h b/arch/arc/include/asm/thread_info.h index 12daaf3a61eaf613dcbb259643107558092db1c2..255d2c7742190075c0207c1ba6ef7883bff666c6 100644 --- a/arch/arc/include/asm/thread_info.h +++ b/arch/arc/include/asm/thread_info.h @@ -24,7 +24,7 @@ #define THREAD_SIZE (PAGE_SIZE << THREAD_SIZE_ORDER) #define THREAD_SHIFT (PAGE_SHIFT << THREAD_SIZE_ORDER) -#ifndef __ASSEMBLY__ +#ifndef __ASSEMBLER__ #include @@ -62,7 +62,7 @@ static inline __attribute_const__ struct thread_info *current_thread_info(void) return (struct thread_info *)(sp & ~(THREAD_SIZE - 1)); } -#endif /* !__ASSEMBLY__ */ +#endif /* !__ASSEMBLER__ */ /* * thread information flags diff --git a/arch/arc/include/uapi/asm/ptrace.h b/arch/arc/include/uapi/asm/ptrace.h index 2a6eff57f6dd814c9fa4797c366e228d7a397502..3ae832db278cd4bc66d6d9d2f022477c2c5d454e 100644 --- a/arch/arc/include/uapi/asm/ptrace.h +++ b/arch/arc/include/uapi/asm/ptrace.h @@ -14,7 +14,7 @@ #define PTRACE_GET_THREAD_AREA 25 -#ifndef __ASSEMBLY__ +#ifndef __ASSEMBLER__ /* * Userspace ABI: Register state needed by * -ptrace (gdbserver) @@ -53,6 +53,6 @@ struct user_regs_arcv2 { unsigned long r30, r58, r59; }; -#endif /* !__ASSEMBLY__ */ +#endif /* !__ASSEMBLER__ */ #endif /* _UAPI__ASM_ARC_PTRACE_H */ diff --git a/arch/arc/kernel/unwind.c b/arch/arc/kernel/unwind.c index d8969dab12d42d515a3bf53d763d268ca1a16134..789cfb9ea14ed8607b5bb5e31f21a3cc47314fc5 100644 --- a/arch/arc/kernel/unwind.c +++ b/arch/arc/kernel/unwind.c @@ -241,15 +241,6 @@ static int cmp_eh_frame_hdr_table_entries(const void *p1, const void *p2) return (e1->start > e2->start) - (e1->start < e2->start); } -static void swap_eh_frame_hdr_table_entries(void *p1, void *p2, int size) -{ - struct eh_frame_hdr_table_entry *e1 = p1; - struct eh_frame_hdr_table_entry *e2 = p2; - - swap(e1->start, e2->start); - swap(e1->fde, e2->fde); -} - static void init_unwind_hdr(struct unwind_table *table, void *(*alloc) (unsigned long)) { @@ -345,7 +336,7 @@ static void init_unwind_hdr(struct unwind_table *table, sort(header->table, n, sizeof(*header->table), - cmp_eh_frame_hdr_table_entries, swap_eh_frame_hdr_table_entries); + cmp_eh_frame_hdr_table_entries, NULL); table->hdrsz = hdrSize; smp_wmb(); diff --git a/arch/arm/include/asm/pgtable.h b/arch/arm/include/asm/pgtable.h index 7f1c3b4e3e0449cc26d41062c4e59c7a42376c41..86378eec77573bf840b0e8aed6c3d7c9d18ada23 100644 --- a/arch/arm/include/asm/pgtable.h +++ b/arch/arm/include/asm/pgtable.h @@ -301,7 +301,7 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot) #define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) }) #define __swp_entry_to_pte(swp) __pte((swp).val) -static inline int pte_swp_exclusive(pte_t pte) +static inline bool pte_swp_exclusive(pte_t pte) { return pte_isset(pte, L_PTE_SWP_EXCLUSIVE); } diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h index 6ce2c51734820dbdb4110ef13b1c48d36d89abb6..5ccca509dff1a5e96c8836c439053e21c192bde4 100644 --- a/arch/arm64/include/asm/kvm_host.h +++ b/arch/arm64/include/asm/kvm_host.h @@ -1107,14 +1107,36 @@ static inline u64 *___ctxt_sys_reg(const struct kvm_cpu_context *ctxt, int r) #define ctxt_sys_reg(c,r) (*__ctxt_sys_reg(c,r)) u64 kvm_vcpu_apply_reg_masks(const struct kvm_vcpu *, enum vcpu_sysreg, u64); + +#define __vcpu_assign_sys_reg(v, r, val) \ + do { \ + const struct kvm_cpu_context *ctxt = &(v)->arch.ctxt; \ + u64 __v = (val); \ + if (vcpu_has_nv((v)) && (r) >= __SANITISED_REG_START__) \ + __v = kvm_vcpu_apply_reg_masks((v), (r), __v); \ + \ + ctxt_sys_reg(ctxt, (r)) = __v; \ + } while (0) + +#define __vcpu_rmw_sys_reg(v, r, op, val) \ + do { \ + const struct kvm_cpu_context *ctxt = &(v)->arch.ctxt; \ + u64 __v = ctxt_sys_reg(ctxt, (r)); \ + __v op (val); \ + if (vcpu_has_nv((v)) && (r) >= __SANITISED_REG_START__) \ + __v = kvm_vcpu_apply_reg_masks((v), (r), __v); \ + \ + ctxt_sys_reg(ctxt, (r)) = __v; \ + } while (0) + #define __vcpu_sys_reg(v,r) \ - (*({ \ + ({ \ const struct kvm_cpu_context *ctxt = &(v)->arch.ctxt; \ - u64 *__r = __ctxt_sys_reg(ctxt, (r)); \ + u64 __v = ctxt_sys_reg(ctxt, (r)); \ if (vcpu_has_nv((v)) && (r) >= __SANITISED_REG_START__) \ - *__r = kvm_vcpu_apply_reg_masks((v), (r), *__r);\ - __r; \ - })) + __v = kvm_vcpu_apply_reg_masks((v), (r), __v); \ + __v; \ + }) u64 vcpu_read_sys_reg(const struct kvm_vcpu *vcpu, int reg); void vcpu_write_sys_reg(struct kvm_vcpu *vcpu, u64 val, int reg); diff --git a/arch/arm64/include/asm/pgtable.h b/arch/arm64/include/asm/pgtable.h index 88db8a0c0b3716ae8d4287a5fc6348c14330983a..192d86e1cc76ea3459f9851bc316cb039e5b8734 100644 --- a/arch/arm64/include/asm/pgtable.h +++ b/arch/arm64/include/asm/pgtable.h @@ -563,7 +563,7 @@ static inline pte_t pte_swp_mkexclusive(pte_t pte) return set_pte_bit(pte, __pgprot(PTE_SWP_EXCLUSIVE)); } -static inline int pte_swp_exclusive(pte_t pte) +static inline bool pte_swp_exclusive(pte_t pte) { return pte_val(pte) & PTE_SWP_EXCLUSIVE; } diff --git a/arch/arm64/kvm/arch_timer.c b/arch/arm64/kvm/arch_timer.c index fdbc8beec930bd3a1ad033a06102ff54a97fb89e..701ea10a63f1eff98fa4b4295e14163e8a65da89 100644 --- a/arch/arm64/kvm/arch_timer.c +++ b/arch/arm64/kvm/arch_timer.c @@ -108,16 +108,16 @@ static void timer_set_ctl(struct arch_timer_context *ctxt, u32 ctl) switch(arch_timer_ctx_index(ctxt)) { case TIMER_VTIMER: - __vcpu_sys_reg(vcpu, CNTV_CTL_EL0) = ctl; + __vcpu_assign_sys_reg(vcpu, CNTV_CTL_EL0, ctl); break; case TIMER_PTIMER: - __vcpu_sys_reg(vcpu, CNTP_CTL_EL0) = ctl; + __vcpu_assign_sys_reg(vcpu, CNTP_CTL_EL0, ctl); break; case TIMER_HVTIMER: - __vcpu_sys_reg(vcpu, CNTHV_CTL_EL2) = ctl; + __vcpu_assign_sys_reg(vcpu, CNTHV_CTL_EL2, ctl); break; case TIMER_HPTIMER: - __vcpu_sys_reg(vcpu, CNTHP_CTL_EL2) = ctl; + __vcpu_assign_sys_reg(vcpu, CNTHP_CTL_EL2, ctl); break; default: WARN_ON(1); @@ -130,16 +130,16 @@ static void timer_set_cval(struct arch_timer_context *ctxt, u64 cval) switch(arch_timer_ctx_index(ctxt)) { case TIMER_VTIMER: - __vcpu_sys_reg(vcpu, CNTV_CVAL_EL0) = cval; + __vcpu_assign_sys_reg(vcpu, CNTV_CVAL_EL0, cval); break; case TIMER_PTIMER: - __vcpu_sys_reg(vcpu, CNTP_CVAL_EL0) = cval; + __vcpu_assign_sys_reg(vcpu, CNTP_CVAL_EL0, cval); break; case TIMER_HVTIMER: - __vcpu_sys_reg(vcpu, CNTHV_CVAL_EL2) = cval; + __vcpu_assign_sys_reg(vcpu, CNTHV_CVAL_EL2, cval); break; case TIMER_HPTIMER: - __vcpu_sys_reg(vcpu, CNTHP_CVAL_EL2) = cval; + __vcpu_assign_sys_reg(vcpu, CNTHP_CVAL_EL2, cval); break; default: WARN_ON(1); @@ -1036,7 +1036,7 @@ void kvm_timer_vcpu_reset(struct kvm_vcpu *vcpu) if (vcpu_has_nv(vcpu)) { struct arch_timer_offset *offs = &vcpu_vtimer(vcpu)->offset; - offs->vcpu_offset = &__vcpu_sys_reg(vcpu, CNTVOFF_EL2); + offs->vcpu_offset = __ctxt_sys_reg(&vcpu->arch.ctxt, CNTVOFF_EL2); offs->vm_offset = &vcpu->kvm->arch.timer_data.poffset; } diff --git a/arch/arm64/kvm/debug.c b/arch/arm64/kvm/debug.c index 0e4c805e7e89191de89b0af573213b2148661303..1a7dab333f5575fc3785c407912d858df2fd1c8d 100644 --- a/arch/arm64/kvm/debug.c +++ b/arch/arm64/kvm/debug.c @@ -216,9 +216,9 @@ void kvm_debug_set_guest_ownership(struct kvm_vcpu *vcpu) void kvm_debug_handle_oslar(struct kvm_vcpu *vcpu, u64 val) { if (val & OSLAR_EL1_OSLK) - __vcpu_sys_reg(vcpu, OSLSR_EL1) |= OSLSR_EL1_OSLK; + __vcpu_rmw_sys_reg(vcpu, OSLSR_EL1, |=, OSLSR_EL1_OSLK); else - __vcpu_sys_reg(vcpu, OSLSR_EL1) &= ~OSLSR_EL1_OSLK; + __vcpu_rmw_sys_reg(vcpu, OSLSR_EL1, &=, ~OSLSR_EL1_OSLK); preempt_disable(); kvm_arch_vcpu_put(vcpu); diff --git a/arch/arm64/kvm/fpsimd.c b/arch/arm64/kvm/fpsimd.c index 7f6e43d256915f507061c7c0fefd5b24febd6043..8f6c8f57c6b9c15041d170fcb394db371c07aeec 100644 --- a/arch/arm64/kvm/fpsimd.c +++ b/arch/arm64/kvm/fpsimd.c @@ -103,8 +103,8 @@ void kvm_arch_vcpu_ctxsync_fp(struct kvm_vcpu *vcpu) fp_state.sve_state = vcpu->arch.sve_state; fp_state.sve_vl = vcpu->arch.sve_max_vl; fp_state.sme_state = NULL; - fp_state.svcr = &__vcpu_sys_reg(vcpu, SVCR); - fp_state.fpmr = &__vcpu_sys_reg(vcpu, FPMR); + fp_state.svcr = __ctxt_sys_reg(&vcpu->arch.ctxt, SVCR); + fp_state.fpmr = __ctxt_sys_reg(&vcpu->arch.ctxt, FPMR); fp_state.fp_type = &vcpu->arch.fp_type; if (vcpu_has_sve(vcpu)) diff --git a/arch/arm64/kvm/hyp/exception.c b/arch/arm64/kvm/hyp/exception.c index 424a5107cddb5e1cdd75ef3581adef03aaadabb7..6a2a899a344e6437609edc138b4ede9425ec3263 100644 --- a/arch/arm64/kvm/hyp/exception.c +++ b/arch/arm64/kvm/hyp/exception.c @@ -37,7 +37,7 @@ static inline void __vcpu_write_sys_reg(struct kvm_vcpu *vcpu, u64 val, int reg) if (unlikely(vcpu_has_nv(vcpu))) vcpu_write_sys_reg(vcpu, val, reg); else if (!__vcpu_write_sys_reg_to_cpu(val, reg)) - __vcpu_sys_reg(vcpu, reg) = val; + __vcpu_assign_sys_reg(vcpu, reg, val); } static void __vcpu_write_spsr(struct kvm_vcpu *vcpu, unsigned long target_mode, @@ -51,7 +51,7 @@ static void __vcpu_write_spsr(struct kvm_vcpu *vcpu, unsigned long target_mode, } else if (has_vhe()) { write_sysreg_el1(val, SYS_SPSR); } else { - __vcpu_sys_reg(vcpu, SPSR_EL1) = val; + __vcpu_assign_sys_reg(vcpu, SPSR_EL1, val); } } diff --git a/arch/arm64/kvm/hyp/include/hyp/switch.h b/arch/arm64/kvm/hyp/include/hyp/switch.h index bb9f2eecfb670a9e181a68b3967fd139324ed1be..76dfda116e56347de50282b3b6eca96d5a34cc12 100644 --- a/arch/arm64/kvm/hyp/include/hyp/switch.h +++ b/arch/arm64/kvm/hyp/include/hyp/switch.h @@ -45,7 +45,7 @@ static inline void __fpsimd_save_fpexc32(struct kvm_vcpu *vcpu) if (!vcpu_el1_is_32bit(vcpu)) return; - __vcpu_sys_reg(vcpu, FPEXC32_EL2) = read_sysreg(fpexc32_el2); + __vcpu_assign_sys_reg(vcpu, FPEXC32_EL2, read_sysreg(fpexc32_el2)); } static inline void __activate_traps_fpsimd32(struct kvm_vcpu *vcpu) @@ -456,7 +456,7 @@ static inline void fpsimd_lazy_switch_to_host(struct kvm_vcpu *vcpu) */ if (vcpu_has_sve(vcpu)) { zcr_el1 = read_sysreg_el1(SYS_ZCR); - __vcpu_sys_reg(vcpu, vcpu_sve_zcr_elx(vcpu)) = zcr_el1; + __vcpu_assign_sys_reg(vcpu, vcpu_sve_zcr_elx(vcpu), zcr_el1); /* * The guest's state is always saved using the guest's max VL. diff --git a/arch/arm64/kvm/hyp/include/hyp/sysreg-sr.h b/arch/arm64/kvm/hyp/include/hyp/sysreg-sr.h index b9cff893bbe0b54ee77b20478f11e70fdbc58952..4d0dbea4c56f7890a3a46623479224c93af43311 100644 --- a/arch/arm64/kvm/hyp/include/hyp/sysreg-sr.h +++ b/arch/arm64/kvm/hyp/include/hyp/sysreg-sr.h @@ -307,11 +307,11 @@ static inline void __sysreg32_save_state(struct kvm_vcpu *vcpu) vcpu->arch.ctxt.spsr_irq = read_sysreg(spsr_irq); vcpu->arch.ctxt.spsr_fiq = read_sysreg(spsr_fiq); - __vcpu_sys_reg(vcpu, DACR32_EL2) = read_sysreg(dacr32_el2); - __vcpu_sys_reg(vcpu, IFSR32_EL2) = read_sysreg(ifsr32_el2); + __vcpu_assign_sys_reg(vcpu, DACR32_EL2, read_sysreg(dacr32_el2)); + __vcpu_assign_sys_reg(vcpu, IFSR32_EL2, read_sysreg(ifsr32_el2)); if (has_vhe() || kvm_debug_regs_in_use(vcpu)) - __vcpu_sys_reg(vcpu, DBGVCR32_EL2) = read_sysreg(dbgvcr32_el2); + __vcpu_assign_sys_reg(vcpu, DBGVCR32_EL2, read_sysreg(dbgvcr32_el2)); } static inline void __sysreg32_restore_state(struct kvm_vcpu *vcpu) diff --git a/arch/arm64/kvm/hyp/nvhe/hyp-main.c b/arch/arm64/kvm/hyp/nvhe/hyp-main.c index 8e8848de4d4707cfb3260d7f013fdd2ef57db327..e9198e56e784b5ca49e242839899eda81c47dae8 100644 --- a/arch/arm64/kvm/hyp/nvhe/hyp-main.c +++ b/arch/arm64/kvm/hyp/nvhe/hyp-main.c @@ -26,7 +26,7 @@ void __kvm_hyp_host_forward_smc(struct kvm_cpu_context *host_ctxt); static void __hyp_sve_save_guest(struct kvm_vcpu *vcpu) { - __vcpu_sys_reg(vcpu, ZCR_EL1) = read_sysreg_el1(SYS_ZCR); + __vcpu_assign_sys_reg(vcpu, ZCR_EL1, read_sysreg_el1(SYS_ZCR)); /* * On saving/restoring guest sve state, always use the maximum VL for * the guest. The layout of the data when saving the sve state depends @@ -79,7 +79,7 @@ static void fpsimd_sve_sync(struct kvm_vcpu *vcpu) has_fpmr = kvm_has_fpmr(kern_hyp_va(vcpu->kvm)); if (has_fpmr) - __vcpu_sys_reg(vcpu, FPMR) = read_sysreg_s(SYS_FPMR); + __vcpu_assign_sys_reg(vcpu, FPMR, read_sysreg_s(SYS_FPMR)); if (system_supports_sve()) __hyp_sve_restore_host(); diff --git a/arch/arm64/kvm/hyp/vhe/switch.c b/arch/arm64/kvm/hyp/vhe/switch.c index c9b330dc2066976741fcbb23fa2cb43a52f87a3d..09df2b42bc1b7c3764223572aa7c5636f962fd36 100644 --- a/arch/arm64/kvm/hyp/vhe/switch.c +++ b/arch/arm64/kvm/hyp/vhe/switch.c @@ -223,9 +223,9 @@ static void __deactivate_traps(struct kvm_vcpu *vcpu) */ val = read_sysreg_el0(SYS_CNTP_CVAL); if (map.direct_ptimer == vcpu_ptimer(vcpu)) - __vcpu_sys_reg(vcpu, CNTP_CVAL_EL0) = val; + __vcpu_assign_sys_reg(vcpu, CNTP_CVAL_EL0, val); if (map.direct_ptimer == vcpu_hptimer(vcpu)) - __vcpu_sys_reg(vcpu, CNTHP_CVAL_EL2) = val; + __vcpu_assign_sys_reg(vcpu, CNTHP_CVAL_EL2, val); offset = read_sysreg_s(SYS_CNTPOFF_EL2); diff --git a/arch/arm64/kvm/hyp/vhe/sysreg-sr.c b/arch/arm64/kvm/hyp/vhe/sysreg-sr.c index 3814b0b2c937f5214edcc7c6d37ecf04b96b2688..73e4bc7fde9e4c25a898436cb6f5e74e7d7d42e7 100644 --- a/arch/arm64/kvm/hyp/vhe/sysreg-sr.c +++ b/arch/arm64/kvm/hyp/vhe/sysreg-sr.c @@ -18,17 +18,17 @@ static void __sysreg_save_vel2_state(struct kvm_vcpu *vcpu) { /* These registers are common with EL1 */ - __vcpu_sys_reg(vcpu, PAR_EL1) = read_sysreg(par_el1); - __vcpu_sys_reg(vcpu, TPIDR_EL1) = read_sysreg(tpidr_el1); - - __vcpu_sys_reg(vcpu, ESR_EL2) = read_sysreg_el1(SYS_ESR); - __vcpu_sys_reg(vcpu, AFSR0_EL2) = read_sysreg_el1(SYS_AFSR0); - __vcpu_sys_reg(vcpu, AFSR1_EL2) = read_sysreg_el1(SYS_AFSR1); - __vcpu_sys_reg(vcpu, FAR_EL2) = read_sysreg_el1(SYS_FAR); - __vcpu_sys_reg(vcpu, MAIR_EL2) = read_sysreg_el1(SYS_MAIR); - __vcpu_sys_reg(vcpu, VBAR_EL2) = read_sysreg_el1(SYS_VBAR); - __vcpu_sys_reg(vcpu, CONTEXTIDR_EL2) = read_sysreg_el1(SYS_CONTEXTIDR); - __vcpu_sys_reg(vcpu, AMAIR_EL2) = read_sysreg_el1(SYS_AMAIR); + __vcpu_assign_sys_reg(vcpu, PAR_EL1, read_sysreg(par_el1)); + __vcpu_assign_sys_reg(vcpu, TPIDR_EL1, read_sysreg(tpidr_el1)); + + __vcpu_assign_sys_reg(vcpu, ESR_EL2, read_sysreg_el1(SYS_ESR)); + __vcpu_assign_sys_reg(vcpu, AFSR0_EL2, read_sysreg_el1(SYS_AFSR0)); + __vcpu_assign_sys_reg(vcpu, AFSR1_EL2, read_sysreg_el1(SYS_AFSR1)); + __vcpu_assign_sys_reg(vcpu, FAR_EL2, read_sysreg_el1(SYS_FAR)); + __vcpu_assign_sys_reg(vcpu, MAIR_EL2, read_sysreg_el1(SYS_MAIR)); + __vcpu_assign_sys_reg(vcpu, VBAR_EL2, read_sysreg_el1(SYS_VBAR)); + __vcpu_assign_sys_reg(vcpu, CONTEXTIDR_EL2, read_sysreg_el1(SYS_CONTEXTIDR)); + __vcpu_assign_sys_reg(vcpu, AMAIR_EL2, read_sysreg_el1(SYS_AMAIR)); /* * In VHE mode those registers are compatible between EL1 and EL2, @@ -46,21 +46,21 @@ static void __sysreg_save_vel2_state(struct kvm_vcpu *vcpu) * are always trapped, ensuring that the in-memory * copy is always up-to-date. A small blessing... */ - __vcpu_sys_reg(vcpu, SCTLR_EL2) = read_sysreg_el1(SYS_SCTLR); - __vcpu_sys_reg(vcpu, TTBR0_EL2) = read_sysreg_el1(SYS_TTBR0); - __vcpu_sys_reg(vcpu, TTBR1_EL2) = read_sysreg_el1(SYS_TTBR1); - __vcpu_sys_reg(vcpu, TCR_EL2) = read_sysreg_el1(SYS_TCR); + __vcpu_assign_sys_reg(vcpu, SCTLR_EL2, read_sysreg_el1(SYS_SCTLR)); + __vcpu_assign_sys_reg(vcpu, TTBR0_EL2, read_sysreg_el1(SYS_TTBR0)); + __vcpu_assign_sys_reg(vcpu, TTBR1_EL2, read_sysreg_el1(SYS_TTBR1)); + __vcpu_assign_sys_reg(vcpu, TCR_EL2, read_sysreg_el1(SYS_TCR)); if (ctxt_has_tcrx(&vcpu->arch.ctxt)) { - __vcpu_sys_reg(vcpu, TCR2_EL2) = read_sysreg_el1(SYS_TCR2); + __vcpu_assign_sys_reg(vcpu, TCR2_EL2, read_sysreg_el1(SYS_TCR2)); if (ctxt_has_s1pie(&vcpu->arch.ctxt)) { - __vcpu_sys_reg(vcpu, PIRE0_EL2) = read_sysreg_el1(SYS_PIRE0); - __vcpu_sys_reg(vcpu, PIR_EL2) = read_sysreg_el1(SYS_PIR); + __vcpu_assign_sys_reg(vcpu, PIRE0_EL2, read_sysreg_el1(SYS_PIRE0)); + __vcpu_assign_sys_reg(vcpu, PIR_EL2, read_sysreg_el1(SYS_PIR)); } if (ctxt_has_s1poe(&vcpu->arch.ctxt)) - __vcpu_sys_reg(vcpu, POR_EL2) = read_sysreg_el1(SYS_POR); + __vcpu_assign_sys_reg(vcpu, POR_EL2, read_sysreg_el1(SYS_POR)); } /* @@ -70,13 +70,13 @@ static void __sysreg_save_vel2_state(struct kvm_vcpu *vcpu) */ val = read_sysreg_el1(SYS_CNTKCTL); val &= CNTKCTL_VALID_BITS; - __vcpu_sys_reg(vcpu, CNTHCTL_EL2) &= ~CNTKCTL_VALID_BITS; - __vcpu_sys_reg(vcpu, CNTHCTL_EL2) |= val; + __vcpu_rmw_sys_reg(vcpu, CNTHCTL_EL2, &=, ~CNTKCTL_VALID_BITS); + __vcpu_rmw_sys_reg(vcpu, CNTHCTL_EL2, |=, val); } - __vcpu_sys_reg(vcpu, SP_EL2) = read_sysreg(sp_el1); - __vcpu_sys_reg(vcpu, ELR_EL2) = read_sysreg_el1(SYS_ELR); - __vcpu_sys_reg(vcpu, SPSR_EL2) = read_sysreg_el1(SYS_SPSR); + __vcpu_assign_sys_reg(vcpu, SP_EL2, read_sysreg(sp_el1)); + __vcpu_assign_sys_reg(vcpu, ELR_EL2, read_sysreg_el1(SYS_ELR)); + __vcpu_assign_sys_reg(vcpu, SPSR_EL2, read_sysreg_el1(SYS_SPSR)); } static void __sysreg_restore_vel2_state(struct kvm_vcpu *vcpu) diff --git a/arch/arm64/kvm/nested.c b/arch/arm64/kvm/nested.c index 4a53e4147fb012813fad8abb965feed4161197e1..5b191f4dc5668c89682082d27409327044ff512a 100644 --- a/arch/arm64/kvm/nested.c +++ b/arch/arm64/kvm/nested.c @@ -1757,7 +1757,7 @@ int kvm_init_nv_sysregs(struct kvm_vcpu *vcpu) out: for (enum vcpu_sysreg sr = __SANITISED_REG_START__; sr < NR_SYS_REGS; sr++) - (void)__vcpu_sys_reg(vcpu, sr); + __vcpu_rmw_sys_reg(vcpu, sr, |=, 0); return 0; } diff --git a/arch/arm64/kvm/pmu-emul.c b/arch/arm64/kvm/pmu-emul.c index 25c29107f13fda842e85d8accab77557fd7b1b48..b03dbda7f1ab995894c49b5cd37892c73dae819d 100644 --- a/arch/arm64/kvm/pmu-emul.c +++ b/arch/arm64/kvm/pmu-emul.c @@ -178,7 +178,7 @@ static void kvm_pmu_set_pmc_value(struct kvm_pmc *pmc, u64 val, bool force) val |= lower_32_bits(val); } - __vcpu_sys_reg(vcpu, reg) = val; + __vcpu_assign_sys_reg(vcpu, reg, val); /* Recreate the perf event to reflect the updated sample_period */ kvm_pmu_create_perf_event(pmc); @@ -204,7 +204,7 @@ void kvm_pmu_set_counter_value(struct kvm_vcpu *vcpu, u64 select_idx, u64 val) void kvm_pmu_set_counter_value_user(struct kvm_vcpu *vcpu, u64 select_idx, u64 val) { kvm_pmu_release_perf_event(kvm_vcpu_idx_to_pmc(vcpu, select_idx)); - __vcpu_sys_reg(vcpu, counter_index_to_reg(select_idx)) = val; + __vcpu_assign_sys_reg(vcpu, counter_index_to_reg(select_idx), val); kvm_make_request(KVM_REQ_RELOAD_PMU, vcpu); } @@ -239,7 +239,7 @@ static void kvm_pmu_stop_counter(struct kvm_pmc *pmc) reg = counter_index_to_reg(pmc->idx); - __vcpu_sys_reg(vcpu, reg) = val; + __vcpu_assign_sys_reg(vcpu, reg, val); kvm_pmu_release_perf_event(pmc); } @@ -503,14 +503,14 @@ static void kvm_pmu_counter_increment(struct kvm_vcpu *vcpu, reg = __vcpu_sys_reg(vcpu, counter_index_to_reg(i)) + 1; if (!kvm_pmc_is_64bit(pmc)) reg = lower_32_bits(reg); - __vcpu_sys_reg(vcpu, counter_index_to_reg(i)) = reg; + __vcpu_assign_sys_reg(vcpu, counter_index_to_reg(i), reg); /* No overflow? move on */ if (kvm_pmc_has_64bit_overflow(pmc) ? reg : lower_32_bits(reg)) continue; /* Mark overflow */ - __vcpu_sys_reg(vcpu, PMOVSSET_EL0) |= BIT(i); + __vcpu_rmw_sys_reg(vcpu, PMOVSSET_EL0, |=, BIT(i)); if (kvm_pmu_counter_can_chain(pmc)) kvm_pmu_counter_increment(vcpu, BIT(i + 1), @@ -556,7 +556,7 @@ static void kvm_pmu_perf_overflow(struct perf_event *perf_event, perf_event->attr.sample_period = period; perf_event->hw.sample_period = period; - __vcpu_sys_reg(vcpu, PMOVSSET_EL0) |= BIT(idx); + __vcpu_rmw_sys_reg(vcpu, PMOVSSET_EL0, |=, BIT(idx)); if (kvm_pmu_counter_can_chain(pmc)) kvm_pmu_counter_increment(vcpu, BIT(idx + 1), @@ -602,7 +602,7 @@ void kvm_pmu_handle_pmcr(struct kvm_vcpu *vcpu, u64 val) kvm_make_request(KVM_REQ_RELOAD_PMU, vcpu); /* The reset bits don't indicate any state, and shouldn't be saved. */ - __vcpu_sys_reg(vcpu, PMCR_EL0) = val & ~(ARMV8_PMU_PMCR_C | ARMV8_PMU_PMCR_P); + __vcpu_assign_sys_reg(vcpu, PMCR_EL0, (val & ~(ARMV8_PMU_PMCR_C | ARMV8_PMU_PMCR_P))); if (val & ARMV8_PMU_PMCR_C) kvm_pmu_set_counter_value(vcpu, ARMV8_PMU_CYCLE_IDX, 0); @@ -779,7 +779,7 @@ void kvm_pmu_set_counter_event_type(struct kvm_vcpu *vcpu, u64 data, u64 reg; reg = counter_index_to_evtreg(pmc->idx); - __vcpu_sys_reg(vcpu, reg) = data & kvm_pmu_evtyper_mask(vcpu->kvm); + __vcpu_assign_sys_reg(vcpu, reg, (data & kvm_pmu_evtyper_mask(vcpu->kvm))); kvm_pmu_create_perf_event(pmc); } @@ -914,9 +914,9 @@ void kvm_vcpu_reload_pmu(struct kvm_vcpu *vcpu) { u64 mask = kvm_pmu_implemented_counter_mask(vcpu); - __vcpu_sys_reg(vcpu, PMOVSSET_EL0) &= mask; - __vcpu_sys_reg(vcpu, PMINTENSET_EL1) &= mask; - __vcpu_sys_reg(vcpu, PMCNTENSET_EL0) &= mask; + __vcpu_rmw_sys_reg(vcpu, PMOVSSET_EL0, &=, mask); + __vcpu_rmw_sys_reg(vcpu, PMINTENSET_EL1, &=, mask); + __vcpu_rmw_sys_reg(vcpu, PMCNTENSET_EL0, &=, mask); kvm_pmu_reprogram_counter_mask(vcpu, mask); } @@ -1038,7 +1038,7 @@ static void kvm_arm_set_nr_counters(struct kvm *kvm, unsigned int nr) u64 val = __vcpu_sys_reg(vcpu, MDCR_EL2); val &= ~MDCR_EL2_HPMN; val |= FIELD_PREP(MDCR_EL2_HPMN, kvm->arch.nr_pmu_counters); - __vcpu_sys_reg(vcpu, MDCR_EL2) = val; + __vcpu_assign_sys_reg(vcpu, MDCR_EL2, val); } } } diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c index a6cf2888d150307707895f944b06047e371565c2..76c2f0da821f8c992cc4a706555dc5b099438316 100644 --- a/arch/arm64/kvm/sys_regs.c +++ b/arch/arm64/kvm/sys_regs.c @@ -228,7 +228,7 @@ void vcpu_write_sys_reg(struct kvm_vcpu *vcpu, u64 val, int reg) * to reverse-translate virtual EL2 system registers for a * non-VHE guest hypervisor. */ - __vcpu_sys_reg(vcpu, reg) = val; + __vcpu_assign_sys_reg(vcpu, reg, val); switch (reg) { case CNTHCTL_EL2: @@ -263,7 +263,7 @@ void vcpu_write_sys_reg(struct kvm_vcpu *vcpu, u64 val, int reg) return; memory_write: - __vcpu_sys_reg(vcpu, reg) = val; + __vcpu_assign_sys_reg(vcpu, reg, val); } /* CSSELR values; used to index KVM_REG_ARM_DEMUX_ID_CCSIDR */ @@ -605,7 +605,7 @@ static int set_oslsr_el1(struct kvm_vcpu *vcpu, const struct sys_reg_desc *rd, if ((val ^ rd->val) & ~OSLSR_EL1_OSLK) return -EINVAL; - __vcpu_sys_reg(vcpu, rd->reg) = val; + __vcpu_assign_sys_reg(vcpu, rd->reg, val); return 0; } @@ -791,7 +791,7 @@ static u64 reset_pmu_reg(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) mask |= GENMASK(n - 1, 0); reset_unknown(vcpu, r); - __vcpu_sys_reg(vcpu, r->reg) &= mask; + __vcpu_rmw_sys_reg(vcpu, r->reg, &=, mask); return __vcpu_sys_reg(vcpu, r->reg); } @@ -799,7 +799,7 @@ static u64 reset_pmu_reg(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) static u64 reset_pmevcntr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) { reset_unknown(vcpu, r); - __vcpu_sys_reg(vcpu, r->reg) &= GENMASK(31, 0); + __vcpu_rmw_sys_reg(vcpu, r->reg, &=, GENMASK(31, 0)); return __vcpu_sys_reg(vcpu, r->reg); } @@ -811,7 +811,7 @@ static u64 reset_pmevtyper(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) return 0; reset_unknown(vcpu, r); - __vcpu_sys_reg(vcpu, r->reg) &= kvm_pmu_evtyper_mask(vcpu->kvm); + __vcpu_rmw_sys_reg(vcpu, r->reg, &=, kvm_pmu_evtyper_mask(vcpu->kvm)); return __vcpu_sys_reg(vcpu, r->reg); } @@ -819,7 +819,7 @@ static u64 reset_pmevtyper(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) static u64 reset_pmselr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) { reset_unknown(vcpu, r); - __vcpu_sys_reg(vcpu, r->reg) &= PMSELR_EL0_SEL_MASK; + __vcpu_rmw_sys_reg(vcpu, r->reg, &=, PMSELR_EL0_SEL_MASK); return __vcpu_sys_reg(vcpu, r->reg); } @@ -835,7 +835,7 @@ static u64 reset_pmcr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) * The value of PMCR.N field is included when the * vCPU register is read via kvm_vcpu_read_pmcr(). */ - __vcpu_sys_reg(vcpu, r->reg) = pmcr; + __vcpu_assign_sys_reg(vcpu, r->reg, pmcr); return __vcpu_sys_reg(vcpu, r->reg); } @@ -907,7 +907,7 @@ static bool access_pmselr(struct kvm_vcpu *vcpu, struct sys_reg_params *p, return false; if (p->is_write) - __vcpu_sys_reg(vcpu, PMSELR_EL0) = p->regval; + __vcpu_assign_sys_reg(vcpu, PMSELR_EL0, p->regval); else /* return PMSELR.SEL field */ p->regval = __vcpu_sys_reg(vcpu, PMSELR_EL0) @@ -1076,7 +1076,7 @@ static int set_pmreg(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r, u64 va { u64 mask = kvm_pmu_accessible_counter_mask(vcpu); - __vcpu_sys_reg(vcpu, r->reg) = val & mask; + __vcpu_assign_sys_reg(vcpu, r->reg, val & mask); kvm_make_request(KVM_REQ_RELOAD_PMU, vcpu); return 0; @@ -1103,10 +1103,10 @@ static bool access_pmcnten(struct kvm_vcpu *vcpu, struct sys_reg_params *p, val = p->regval & mask; if (r->Op2 & 0x1) /* accessing PMCNTENSET_EL0 */ - __vcpu_sys_reg(vcpu, PMCNTENSET_EL0) |= val; + __vcpu_rmw_sys_reg(vcpu, PMCNTENSET_EL0, |=, val); else /* accessing PMCNTENCLR_EL0 */ - __vcpu_sys_reg(vcpu, PMCNTENSET_EL0) &= ~val; + __vcpu_rmw_sys_reg(vcpu, PMCNTENSET_EL0, &=, ~val); kvm_pmu_reprogram_counter_mask(vcpu, val); } else { @@ -1129,10 +1129,10 @@ static bool access_pminten(struct kvm_vcpu *vcpu, struct sys_reg_params *p, if (r->Op2 & 0x1) /* accessing PMINTENSET_EL1 */ - __vcpu_sys_reg(vcpu, PMINTENSET_EL1) |= val; + __vcpu_rmw_sys_reg(vcpu, PMINTENSET_EL1, |=, val); else /* accessing PMINTENCLR_EL1 */ - __vcpu_sys_reg(vcpu, PMINTENSET_EL1) &= ~val; + __vcpu_rmw_sys_reg(vcpu, PMINTENSET_EL1, &=, ~val); } else { p->regval = __vcpu_sys_reg(vcpu, PMINTENSET_EL1); } @@ -1151,10 +1151,10 @@ static bool access_pmovs(struct kvm_vcpu *vcpu, struct sys_reg_params *p, if (p->is_write) { if (r->CRm & 0x2) /* accessing PMOVSSET_EL0 */ - __vcpu_sys_reg(vcpu, PMOVSSET_EL0) |= (p->regval & mask); + __vcpu_rmw_sys_reg(vcpu, PMOVSSET_EL0, |=, (p->regval & mask)); else /* accessing PMOVSCLR_EL0 */ - __vcpu_sys_reg(vcpu, PMOVSSET_EL0) &= ~(p->regval & mask); + __vcpu_rmw_sys_reg(vcpu, PMOVSSET_EL0, &=, ~(p->regval & mask)); } else { p->regval = __vcpu_sys_reg(vcpu, PMOVSSET_EL0); } @@ -1185,8 +1185,8 @@ static bool access_pmuserenr(struct kvm_vcpu *vcpu, struct sys_reg_params *p, if (!vcpu_mode_priv(vcpu)) return undef_access(vcpu, p, r); - __vcpu_sys_reg(vcpu, PMUSERENR_EL0) = - p->regval & ARMV8_PMU_USERENR_MASK; + __vcpu_assign_sys_reg(vcpu, PMUSERENR_EL0, + (p->regval & ARMV8_PMU_USERENR_MASK)); } else { p->regval = __vcpu_sys_reg(vcpu, PMUSERENR_EL0) & ARMV8_PMU_USERENR_MASK; @@ -1237,7 +1237,7 @@ static int set_pmcr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r, if (!kvm_supports_32bit_el0()) val |= ARMV8_PMU_PMCR_LC; - __vcpu_sys_reg(vcpu, r->reg) = val; + __vcpu_assign_sys_reg(vcpu, r->reg, val); kvm_make_request(KVM_REQ_RELOAD_PMU, vcpu); return 0; @@ -2213,7 +2213,7 @@ static u64 reset_clidr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) if (kvm_has_mte(vcpu->kvm)) clidr |= 2ULL << CLIDR_TTYPE_SHIFT(loc); - __vcpu_sys_reg(vcpu, r->reg) = clidr; + __vcpu_assign_sys_reg(vcpu, r->reg, clidr); return __vcpu_sys_reg(vcpu, r->reg); } @@ -2227,7 +2227,7 @@ static int set_clidr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *rd, if ((val & CLIDR_EL1_RES0) || (!(ctr_el0 & CTR_EL0_IDC) && idc)) return -EINVAL; - __vcpu_sys_reg(vcpu, rd->reg) = val; + __vcpu_assign_sys_reg(vcpu, rd->reg, val); return 0; } @@ -2404,7 +2404,7 @@ static bool access_sp_el1(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) { if (p->is_write) - __vcpu_sys_reg(vcpu, SP_EL1) = p->regval; + __vcpu_assign_sys_reg(vcpu, SP_EL1, p->regval); else p->regval = __vcpu_sys_reg(vcpu, SP_EL1); @@ -2428,7 +2428,7 @@ static bool access_spsr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) { if (p->is_write) - __vcpu_sys_reg(vcpu, SPSR_EL1) = p->regval; + __vcpu_assign_sys_reg(vcpu, SPSR_EL1, p->regval); else p->regval = __vcpu_sys_reg(vcpu, SPSR_EL1); @@ -2440,7 +2440,7 @@ static bool access_cntkctl_el12(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) { if (p->is_write) - __vcpu_sys_reg(vcpu, CNTKCTL_EL1) = p->regval; + __vcpu_assign_sys_reg(vcpu, CNTKCTL_EL1, p->regval); else p->regval = __vcpu_sys_reg(vcpu, CNTKCTL_EL1); @@ -2454,7 +2454,9 @@ static u64 reset_hcr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) if (!cpus_have_final_cap(ARM64_HAS_HCR_NV1)) val |= HCR_E2H; - return __vcpu_sys_reg(vcpu, r->reg) = val; + __vcpu_assign_sys_reg(vcpu, r->reg, val); + + return __vcpu_sys_reg(vcpu, r->reg); } static unsigned int __el2_visibility(const struct kvm_vcpu *vcpu, @@ -2625,7 +2627,7 @@ static bool access_mdcr(struct kvm_vcpu *vcpu, u64_replace_bits(val, hpmn, MDCR_EL2_HPMN); } - __vcpu_sys_reg(vcpu, MDCR_EL2) = val; + __vcpu_assign_sys_reg(vcpu, MDCR_EL2, val); /* * Request a reload of the PMU to enable/disable the counters @@ -2754,7 +2756,7 @@ static int set_imp_id_reg(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r, static u64 reset_mdcr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) { - __vcpu_sys_reg(vcpu, r->reg) = vcpu->kvm->arch.nr_pmu_counters; + __vcpu_assign_sys_reg(vcpu, r->reg, vcpu->kvm->arch.nr_pmu_counters); return vcpu->kvm->arch.nr_pmu_counters; } @@ -4790,7 +4792,7 @@ void kvm_reset_sys_regs(struct kvm_vcpu *vcpu) r->reset(vcpu, r); if (r->reg >= __SANITISED_REG_START__ && r->reg < NR_SYS_REGS) - (void)__vcpu_sys_reg(vcpu, r->reg); + __vcpu_rmw_sys_reg(vcpu, r->reg, |=, 0); } set_bit(KVM_ARCH_FLAG_ID_REGS_INITIALIZED, &kvm->arch.flags); @@ -5012,7 +5014,7 @@ int kvm_sys_reg_set_user(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg, if (r->set_user) { ret = (r->set_user)(vcpu, r, val); } else { - __vcpu_sys_reg(vcpu, r->reg) = val; + __vcpu_assign_sys_reg(vcpu, r->reg, val); ret = 0; } diff --git a/arch/arm64/kvm/sys_regs.h b/arch/arm64/kvm/sys_regs.h index cc6338d3876638a69d432462608c5e7e8eb34ac8..ef97d9fc67cc536c291ff91ab6ccf9d4b2571198 100644 --- a/arch/arm64/kvm/sys_regs.h +++ b/arch/arm64/kvm/sys_regs.h @@ -137,7 +137,7 @@ static inline u64 reset_unknown(struct kvm_vcpu *vcpu, { BUG_ON(!r->reg); BUG_ON(r->reg >= NR_SYS_REGS); - __vcpu_sys_reg(vcpu, r->reg) = 0x1de7ec7edbadc0deULL; + __vcpu_assign_sys_reg(vcpu, r->reg, 0x1de7ec7edbadc0deULL); return __vcpu_sys_reg(vcpu, r->reg); } @@ -145,7 +145,7 @@ static inline u64 reset_val(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) { BUG_ON(!r->reg); BUG_ON(r->reg >= NR_SYS_REGS); - __vcpu_sys_reg(vcpu, r->reg) = r->val; + __vcpu_assign_sys_reg(vcpu, r->reg, r->val); return __vcpu_sys_reg(vcpu, r->reg); } diff --git a/arch/arm64/kvm/vgic/vgic-v3-nested.c b/arch/arm64/kvm/vgic/vgic-v3-nested.c index 4f6954c306747bf6d84756b9e39d6beb022c4f94..d22a8ad7bcc518053643fd2515a0bcc808312d5f 100644 --- a/arch/arm64/kvm/vgic/vgic-v3-nested.c +++ b/arch/arm64/kvm/vgic/vgic-v3-nested.c @@ -356,12 +356,12 @@ void vgic_v3_put_nested(struct kvm_vcpu *vcpu) val = __vcpu_sys_reg(vcpu, ICH_HCR_EL2); val &= ~ICH_HCR_EL2_EOIcount_MASK; val |= (s_cpu_if->vgic_hcr & ICH_HCR_EL2_EOIcount_MASK); - __vcpu_sys_reg(vcpu, ICH_HCR_EL2) = val; - __vcpu_sys_reg(vcpu, ICH_VMCR_EL2) = s_cpu_if->vgic_vmcr; + __vcpu_assign_sys_reg(vcpu, ICH_HCR_EL2, val); + __vcpu_assign_sys_reg(vcpu, ICH_VMCR_EL2, s_cpu_if->vgic_vmcr); for (i = 0; i < 4; i++) { - __vcpu_sys_reg(vcpu, ICH_AP0RN(i)) = s_cpu_if->vgic_ap0r[i]; - __vcpu_sys_reg(vcpu, ICH_AP1RN(i)) = s_cpu_if->vgic_ap1r[i]; + __vcpu_assign_sys_reg(vcpu, ICH_AP0RN(i), s_cpu_if->vgic_ap0r[i]); + __vcpu_assign_sys_reg(vcpu, ICH_AP1RN(i), s_cpu_if->vgic_ap1r[i]); } for_each_set_bit(i, &shadow_if->lr_map, kvm_vgic_global_state.nr_lr) { @@ -370,7 +370,7 @@ void vgic_v3_put_nested(struct kvm_vcpu *vcpu) val &= ~ICH_LR_STATE; val |= s_cpu_if->vgic_lr[i] & ICH_LR_STATE; - __vcpu_sys_reg(vcpu, ICH_LRN(i)) = val; + __vcpu_assign_sys_reg(vcpu, ICH_LRN(i), val); s_cpu_if->vgic_lr[i] = 0; } diff --git a/arch/csky/include/asm/pgtable.h b/arch/csky/include/asm/pgtable.h index b8378431aeff7c5e1705d8fac2981a7123f39c2f..5a394be09c35b6b3117d8b2c26297ec7c7339cf2 100644 --- a/arch/csky/include/asm/pgtable.h +++ b/arch/csky/include/asm/pgtable.h @@ -200,7 +200,7 @@ static inline pte_t pte_mkyoung(pte_t pte) return pte; } -static inline int pte_swp_exclusive(pte_t pte) +static inline bool pte_swp_exclusive(pte_t pte) { return pte_val(pte) & _PAGE_SWP_EXCLUSIVE; } diff --git a/arch/hexagon/include/asm/pgtable.h b/arch/hexagon/include/asm/pgtable.h index 9fbdfdbc539f8539b0377a601b8a1027d6a26212..fbf24d1d1ca6aafc95179cac6af720bf29f10231 100644 --- a/arch/hexagon/include/asm/pgtable.h +++ b/arch/hexagon/include/asm/pgtable.h @@ -387,7 +387,7 @@ static inline unsigned long pmd_page_vaddr(pmd_t pmd) (((type & 0x1f) << 1) | \ ((offset & 0x3ffff8) << 10) | ((offset & 0x7) << 7)) }) -static inline int pte_swp_exclusive(pte_t pte) +static inline bool pte_swp_exclusive(pte_t pte) { return pte_val(pte) & _PAGE_SWP_EXCLUSIVE; } diff --git a/arch/loongarch/include/asm/pgtable.h b/arch/loongarch/include/asm/pgtable.h index a3f17914dbab1c54f7adf7f6d9b503de75aaf314..b30185302c07127504e0e26ba3e6ad2d5c4b39ec 100644 --- a/arch/loongarch/include/asm/pgtable.h +++ b/arch/loongarch/include/asm/pgtable.h @@ -301,7 +301,7 @@ static inline pte_t mk_swap_pte(unsigned long type, unsigned long offset) #define __pmd_to_swp_entry(pmd) ((swp_entry_t) { pmd_val(pmd) }) #define __swp_entry_to_pmd(x) ((pmd_t) { (x).val | _PAGE_HUGE }) -static inline int pte_swp_exclusive(pte_t pte) +static inline bool pte_swp_exclusive(pte_t pte) { return pte_val(pte) & _PAGE_SWP_EXCLUSIVE; } diff --git a/arch/m68k/include/asm/mcf_pgtable.h b/arch/m68k/include/asm/mcf_pgtable.h index f5c596b211d422f8f7c4d9d7387e5cf8203d252d..d79fef609194d6ce92a5a3a8e2ff994a679b10d5 100644 --- a/arch/m68k/include/asm/mcf_pgtable.h +++ b/arch/m68k/include/asm/mcf_pgtable.h @@ -268,7 +268,7 @@ extern pgd_t kernel_pg_dir[PTRS_PER_PGD]; #define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) }) #define __swp_entry_to_pte(x) (__pte((x).val)) -static inline int pte_swp_exclusive(pte_t pte) +static inline bool pte_swp_exclusive(pte_t pte) { return pte_val(pte) & _PAGE_SWP_EXCLUSIVE; } diff --git a/arch/m68k/include/asm/motorola_pgtable.h b/arch/m68k/include/asm/motorola_pgtable.h index 040ac3bad713cd2a5e9d3b930f4c47c321937fdf..14fee64d3e605c979d62851bf3d61c2a514f5e10 100644 --- a/arch/m68k/include/asm/motorola_pgtable.h +++ b/arch/m68k/include/asm/motorola_pgtable.h @@ -185,7 +185,7 @@ extern pgd_t kernel_pg_dir[128]; #define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) }) #define __swp_entry_to_pte(x) ((pte_t) { (x).val }) -static inline int pte_swp_exclusive(pte_t pte) +static inline bool pte_swp_exclusive(pte_t pte) { return pte_val(pte) & _PAGE_SWP_EXCLUSIVE; } diff --git a/arch/m68k/include/asm/sun3_pgtable.h b/arch/m68k/include/asm/sun3_pgtable.h index 73745dc0ec0e0f4ba8f86cce875492273617c3ce..858cbe936f5b3f69d6ec1a970bcc7f9ce03b9dd2 100644 --- a/arch/m68k/include/asm/sun3_pgtable.h +++ b/arch/m68k/include/asm/sun3_pgtable.h @@ -169,7 +169,7 @@ extern pgd_t kernel_pg_dir[PTRS_PER_PGD]; #define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) }) #define __swp_entry_to_pte(x) ((pte_t) { (x).val }) -static inline int pte_swp_exclusive(pte_t pte) +static inline bool pte_swp_exclusive(pte_t pte) { return pte_val(pte) & _PAGE_SWP_EXCLUSIVE; } diff --git a/arch/microblaze/include/asm/pgtable.h b/arch/microblaze/include/asm/pgtable.h index b1bb2c65dd047e0d412a6b81db1447d2b0846060..bae1abfa6f6b863924de9f0cc954ebdb4d09a778 100644 --- a/arch/microblaze/include/asm/pgtable.h +++ b/arch/microblaze/include/asm/pgtable.h @@ -398,7 +398,7 @@ extern pgd_t swapper_pg_dir[PTRS_PER_PGD]; #define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) >> 2 }) #define __swp_entry_to_pte(x) ((pte_t) { (x).val << 2 }) -static inline int pte_swp_exclusive(pte_t pte) +static inline bool pte_swp_exclusive(pte_t pte) { return pte_val(pte) & _PAGE_SWP_EXCLUSIVE; } diff --git a/arch/mips/include/asm/pgtable.h b/arch/mips/include/asm/pgtable.h index 4852b005a72d3689f216b31c61f2f4e5dafacf22..ae73ecf4c41aacf9bd01911e62b451d115692c42 100644 --- a/arch/mips/include/asm/pgtable.h +++ b/arch/mips/include/asm/pgtable.h @@ -534,7 +534,7 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot) #endif #if defined(CONFIG_PHYS_ADDR_T_64BIT) && defined(CONFIG_CPU_MIPS32) -static inline int pte_swp_exclusive(pte_t pte) +static inline bool pte_swp_exclusive(pte_t pte) { return pte.pte_low & _PAGE_SWP_EXCLUSIVE; } @@ -551,7 +551,7 @@ static inline pte_t pte_swp_clear_exclusive(pte_t pte) return pte; } #else -static inline int pte_swp_exclusive(pte_t pte) +static inline bool pte_swp_exclusive(pte_t pte) { return pte_val(pte) & _PAGE_SWP_EXCLUSIVE; } diff --git a/arch/nios2/include/asm/pgtable.h b/arch/nios2/include/asm/pgtable.h index e98578e27e26bcb127a66cfc00d11b2197934a1c..844dce55569f34b8a324470d32a9e43ab9b666b3 100644 --- a/arch/nios2/include/asm/pgtable.h +++ b/arch/nios2/include/asm/pgtable.h @@ -259,7 +259,7 @@ static inline unsigned long pmd_page_vaddr(pmd_t pmd) #define __swp_entry_to_pte(swp) ((pte_t) { (swp).val }) #define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) }) -static inline int pte_swp_exclusive(pte_t pte) +static inline bool pte_swp_exclusive(pte_t pte) { return pte_val(pte) & _PAGE_SWP_EXCLUSIVE; } diff --git a/arch/openrisc/include/asm/pgtable.h b/arch/openrisc/include/asm/pgtable.h index 71bfb8c8c482a3b3c31ea48e4311f45c86558307..5bd6463bd514497c56456713eaed3611bb2d0f01 100644 --- a/arch/openrisc/include/asm/pgtable.h +++ b/arch/openrisc/include/asm/pgtable.h @@ -411,7 +411,7 @@ static inline void update_mmu_cache_range(struct vm_fault *vmf, #define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) }) #define __swp_entry_to_pte(x) ((pte_t) { (x).val }) -static inline int pte_swp_exclusive(pte_t pte) +static inline bool pte_swp_exclusive(pte_t pte) { return pte_val(pte) & _PAGE_SWP_EXCLUSIVE; } diff --git a/arch/parisc/include/asm/pgtable.h b/arch/parisc/include/asm/pgtable.h index 80f5e2a2841314e1ee3110e99dfddc46f5db7eeb..1a86a4370b298af1c6904c2f8b37d327b3438076 100644 --- a/arch/parisc/include/asm/pgtable.h +++ b/arch/parisc/include/asm/pgtable.h @@ -425,7 +425,7 @@ static inline void set_ptes(struct mm_struct *mm, unsigned long addr, #define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) }) #define __swp_entry_to_pte(x) ((pte_t) { (x).val }) -static inline int pte_swp_exclusive(pte_t pte) +static inline bool pte_swp_exclusive(pte_t pte) { return pte_val(pte) & _PAGE_SWP_EXCLUSIVE; } diff --git a/arch/powerpc/include/asm/book3s/32/pgtable.h b/arch/powerpc/include/asm/book3s/32/pgtable.h index 42c3af90d1f0ff8f23ddd6511a98a68f5751c1d7..92d21c6faf1e57665f7078b8ca3f2c267e1ce9d4 100644 --- a/arch/powerpc/include/asm/book3s/32/pgtable.h +++ b/arch/powerpc/include/asm/book3s/32/pgtable.h @@ -365,7 +365,7 @@ static inline void __ptep_set_access_flags(struct vm_area_struct *vma, #define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) >> 3 }) #define __swp_entry_to_pte(x) ((pte_t) { (x).val << 3 }) -static inline int pte_swp_exclusive(pte_t pte) +static inline bool pte_swp_exclusive(pte_t pte) { return pte_val(pte) & _PAGE_SWP_EXCLUSIVE; } diff --git a/arch/powerpc/include/asm/book3s/64/pgtable.h b/arch/powerpc/include/asm/book3s/64/pgtable.h index 6ed93e290c2feda820f7e47d3d58b862295b6c53..a2ddcbb3fcb94f119de7b9c7afa566985443da80 100644 --- a/arch/powerpc/include/asm/book3s/64/pgtable.h +++ b/arch/powerpc/include/asm/book3s/64/pgtable.h @@ -693,7 +693,7 @@ static inline pte_t pte_swp_mkexclusive(pte_t pte) return __pte_raw(pte_raw(pte) | cpu_to_be64(_PAGE_SWP_EXCLUSIVE)); } -static inline int pte_swp_exclusive(pte_t pte) +static inline bool pte_swp_exclusive(pte_t pte) { return !!(pte_raw(pte) & cpu_to_be64(_PAGE_SWP_EXCLUSIVE)); } diff --git a/arch/powerpc/include/asm/nohash/pgtable.h b/arch/powerpc/include/asm/nohash/pgtable.h index 8d1f0b7062eb2f11805b6e0f8c9e1202fd6e6723..7d6b9e5b286ef93f49823ee15c5ff93e10b7f159 100644 --- a/arch/powerpc/include/asm/nohash/pgtable.h +++ b/arch/powerpc/include/asm/nohash/pgtable.h @@ -286,7 +286,7 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot) return __pte((pte_val(pte) & _PAGE_CHG_MASK) | pgprot_val(newprot)); } -static inline int pte_swp_exclusive(pte_t pte) +static inline bool pte_swp_exclusive(pte_t pte) { return pte_val(pte) & _PAGE_SWP_EXCLUSIVE; } diff --git a/arch/powerpc/platforms/book3s/vas-api.c b/arch/powerpc/platforms/book3s/vas-api.c index 0b6365d85d11714a87174dc8ebfee854eb3442c8..dc6f75d3ac6ef76b272fbd3997d0511abb9204df 100644 --- a/arch/powerpc/platforms/book3s/vas-api.c +++ b/arch/powerpc/platforms/book3s/vas-api.c @@ -521,6 +521,15 @@ static int coproc_mmap(struct file *fp, struct vm_area_struct *vma) return -EINVAL; } + /* + * Map complete page to the paste address. So the user + * space should pass 0ULL to the offset parameter. + */ + if (vma->vm_pgoff) { + pr_debug("Page offset unsupported to map paste address\n"); + return -EINVAL; + } + /* Ensure instance has an open send window */ if (!txwin) { pr_err("No send window open?\n"); diff --git a/arch/powerpc/platforms/powernv/memtrace.c b/arch/powerpc/platforms/powernv/memtrace.c index 4ac9808e55a44d13a97de03168ce0d95f60d1bb4..2ea30b34335415be91d0aa44ec4f7bf96c4e6d71 100644 --- a/arch/powerpc/platforms/powernv/memtrace.c +++ b/arch/powerpc/platforms/powernv/memtrace.c @@ -48,11 +48,15 @@ static ssize_t memtrace_read(struct file *filp, char __user *ubuf, static int memtrace_mmap(struct file *filp, struct vm_area_struct *vma) { struct memtrace_entry *ent = filp->private_data; + unsigned long ent_nrpages = ent->size >> PAGE_SHIFT; + unsigned long vma_nrpages = vma_pages(vma); - if (ent->size < vma->vm_end - vma->vm_start) + /* The requested page offset should be within object's page count */ + if (vma->vm_pgoff >= ent_nrpages) return -EINVAL; - if (vma->vm_pgoff << PAGE_SHIFT >= ent->size) + /* The requested mapping range should remain within the bounds */ + if (vma_nrpages > ent_nrpages - vma->vm_pgoff) return -EINVAL; vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot); diff --git a/arch/riscv/include/asm/pgtable.h b/arch/riscv/include/asm/pgtable.h index a11816bbf9e7f95bf07c144c47f6821d6945d904..438ce7df24c39707e4ad5fa1c8b33b0d827ed8f9 100644 --- a/arch/riscv/include/asm/pgtable.h +++ b/arch/riscv/include/asm/pgtable.h @@ -1028,7 +1028,7 @@ static inline pud_t pud_modify(pud_t pud, pgprot_t newprot) #define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) }) #define __swp_entry_to_pte(x) ((pte_t) { (x).val }) -static inline int pte_swp_exclusive(pte_t pte) +static inline bool pte_swp_exclusive(pte_t pte) { return pte_val(pte) & _PAGE_SWP_EXCLUSIVE; } diff --git a/arch/s390/include/asm/pgtable.h b/arch/s390/include/asm/pgtable.h index 1c661ac62ce81b7d57963c6a1f0766b530eec103..6d8bc27a366e4f17abc0e04d01dff5fcf20eb2e7 100644 --- a/arch/s390/include/asm/pgtable.h +++ b/arch/s390/include/asm/pgtable.h @@ -915,7 +915,7 @@ static inline int pmd_protnone(pmd_t pmd) } #endif -static inline int pte_swp_exclusive(pte_t pte) +static inline bool pte_swp_exclusive(pte_t pte) { return pte_val(pte) & _PAGE_SWP_EXCLUSIVE; } diff --git a/arch/sh/include/asm/pgtable_32.h b/arch/sh/include/asm/pgtable_32.h index 71b18741cc11a7759cc780f23046f74e16886592..5f51af18997b57ff531d13aed53f9119882a0f6c 100644 --- a/arch/sh/include/asm/pgtable_32.h +++ b/arch/sh/include/asm/pgtable_32.h @@ -470,7 +470,7 @@ static inline unsigned long pmd_page_vaddr(pmd_t pmd) /* In both cases, we borrow bit 6 to store the exclusive marker in swap PTEs. */ #define _PAGE_SWP_EXCLUSIVE _PAGE_USER -static inline int pte_swp_exclusive(pte_t pte) +static inline bool pte_swp_exclusive(pte_t pte) { return pte.pte_low & _PAGE_SWP_EXCLUSIVE; } diff --git a/arch/sparc/include/asm/pgtable_32.h b/arch/sparc/include/asm/pgtable_32.h index 1454ebe915393e7c574b238c9836edac104db03c..7c199c003ffe3b402644407ea6f63b07352e0c19 100644 --- a/arch/sparc/include/asm/pgtable_32.h +++ b/arch/sparc/include/asm/pgtable_32.h @@ -348,7 +348,7 @@ static inline swp_entry_t __swp_entry(unsigned long type, unsigned long offset) #define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) }) #define __swp_entry_to_pte(x) ((pte_t) { (x).val }) -static inline int pte_swp_exclusive(pte_t pte) +static inline bool pte_swp_exclusive(pte_t pte) { return pte_val(pte) & SRMMU_SWP_EXCLUSIVE; } diff --git a/arch/sparc/include/asm/pgtable_64.h b/arch/sparc/include/asm/pgtable_64.h index 4af03e3c161b4bdacda6e075bb0c68dc6793f6f3..669cd02469a1d94dd1b0f5941af17afbc690da48 100644 --- a/arch/sparc/include/asm/pgtable_64.h +++ b/arch/sparc/include/asm/pgtable_64.h @@ -1023,7 +1023,7 @@ pgtable_t pgtable_trans_huge_withdraw(struct mm_struct *mm, pmd_t *pmdp); #define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) }) #define __swp_entry_to_pte(x) ((pte_t) { (x).val }) -static inline int pte_swp_exclusive(pte_t pte) +static inline bool pte_swp_exclusive(pte_t pte) { return pte_val(pte) & _PAGE_SWP_EXCLUSIVE; } diff --git a/arch/um/include/asm/pgtable.h b/arch/um/include/asm/pgtable.h index ca2a519d53ab1d0fd6b537384c56cd451286bac1..24fdea6f88c37f4672876f76e9d5067610a87c2f 100644 --- a/arch/um/include/asm/pgtable.h +++ b/arch/um/include/asm/pgtable.h @@ -314,7 +314,7 @@ extern pte_t *virt_to_pte(struct mm_struct *mm, unsigned long addr); ((swp_entry_t) { pte_val(pte_mkuptodate(pte)) }) #define __swp_entry_to_pte(x) ((pte_t) { (x).val }) -static inline int pte_swp_exclusive(pte_t pte) +static inline bool pte_swp_exclusive(pte_t pte) { return pte_get_bits(pte, _PAGE_SWP_EXCLUSIVE); } diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h index 774430c3abff21074865f47b6b608cb81ed3264b..97954c936c5480f2cc7cb22ea62c6fe8a8f772dd 100644 --- a/arch/x86/include/asm/pgtable.h +++ b/arch/x86/include/asm/pgtable.h @@ -1561,7 +1561,7 @@ static inline pte_t pte_swp_mkexclusive(pte_t pte) return pte_set_flags(pte, _PAGE_SWP_EXCLUSIVE); } -static inline int pte_swp_exclusive(pte_t pte) +static inline bool pte_swp_exclusive(pte_t pte) { return pte_flags(pte) & _PAGE_SWP_EXCLUSIVE; } diff --git a/arch/x86/kernel/smp.c b/arch/x86/kernel/smp.c index 18266cc3d98c182a3822ebf6c025924442994edb..b014e6d229f9519add6f3fb315f1a2f375675a50 100644 --- a/arch/x86/kernel/smp.c +++ b/arch/x86/kernel/smp.c @@ -299,3 +299,27 @@ struct smp_ops smp_ops = { .send_call_func_single_ipi = native_send_call_func_single_ipi, }; EXPORT_SYMBOL_GPL(smp_ops); + +int arch_cpu_rescan_dead_smt_siblings(void) +{ + enum cpuhp_smt_control old = cpu_smt_control; + int ret; + + /* + * If SMT has been disabled and SMT siblings are in HLT, bring them back + * online and offline them again so that they end up in MWAIT proper. + * + * Called with hotplug enabled. + */ + if (old != CPU_SMT_DISABLED && old != CPU_SMT_FORCE_DISABLED) + return 0; + + ret = cpuhp_smt_enable(); + if (ret) + return ret; + + ret = cpuhp_smt_disable(old); + + return ret; +} +EXPORT_SYMBOL_GPL(arch_cpu_rescan_dead_smt_siblings); diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index fc78c2325fd2986a40ab46d7b59b237b0adc40a8..58ede3fa6a75ba53019d9cde5f6a4964ec68f020 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -1244,6 +1244,10 @@ void play_dead_common(void) local_irq_disable(); } +/* + * We need to flush the caches before going to sleep, lest we have + * dirty data in our caches when we come back up. + */ void __noreturn mwait_play_dead(unsigned int eax_hint) { struct mwait_cpu_dead *md = this_cpu_ptr(&mwait_cpu_dead); @@ -1289,50 +1293,6 @@ void __noreturn mwait_play_dead(unsigned int eax_hint) } } -/* - * We need to flush the caches before going to sleep, lest we have - * dirty data in our caches when we come back up. - */ -static inline void mwait_play_dead_cpuid_hint(void) -{ - unsigned int eax, ebx, ecx, edx; - unsigned int highest_cstate = 0; - unsigned int highest_subcstate = 0; - int i; - - if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD || - boot_cpu_data.x86_vendor == X86_VENDOR_HYGON) - return; - if (!this_cpu_has(X86_FEATURE_MWAIT)) - return; - if (!this_cpu_has(X86_FEATURE_CLFLUSH)) - return; - - eax = CPUID_LEAF_MWAIT; - ecx = 0; - native_cpuid(&eax, &ebx, &ecx, &edx); - - /* - * eax will be 0 if EDX enumeration is not valid. - * Initialized below to cstate, sub_cstate value when EDX is valid. - */ - if (!(ecx & CPUID5_ECX_EXTENSIONS_SUPPORTED)) { - eax = 0; - } else { - edx >>= MWAIT_SUBSTATE_SIZE; - for (i = 0; i < 7 && edx; i++, edx >>= MWAIT_SUBSTATE_SIZE) { - if (edx & MWAIT_SUBSTATE_MASK) { - highest_cstate = i; - highest_subcstate = edx & MWAIT_SUBSTATE_MASK; - } - } - eax = (highest_cstate << MWAIT_SUBSTATE_SIZE) | - (highest_subcstate - 1); - } - - mwait_play_dead(eax); -} - /* * Kick all "offline" CPUs out of mwait on kexec(). See comment in * mwait_play_dead(). @@ -1383,9 +1343,9 @@ void native_play_dead(void) play_dead_common(); tboot_shutdown(TB_SHUTDOWN_WFS); - mwait_play_dead_cpuid_hint(); - if (cpuidle_play_dead()) - hlt_play_dead(); + /* Below returns only on error. */ + cpuidle_play_dead(); + hlt_play_dead(); } #else /* ... !CONFIG_HOTPLUG_CPU */ diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c index cbc84c6abc2e3f8b1f18ed8d1c9629fd7f827466..4e06e2e89a8fa7e8710e9683346f8527182327d2 100644 --- a/arch/x86/kvm/mmu/mmu.c +++ b/arch/x86/kvm/mmu/mmu.c @@ -4896,12 +4896,16 @@ long kvm_arch_vcpu_pre_fault_memory(struct kvm_vcpu *vcpu, { u64 error_code = PFERR_GUEST_FINAL_MASK; u8 level = PG_LEVEL_4K; + u64 direct_bits; u64 end; int r; if (!vcpu->kvm->arch.pre_fault_allowed) return -EOPNOTSUPP; + if (kvm_is_gfn_alias(vcpu->kvm, gpa_to_gfn(range->gpa))) + return -EINVAL; + /* * reload is efficient when called repeatedly, so we can do it on * every iteration. @@ -4910,15 +4914,18 @@ long kvm_arch_vcpu_pre_fault_memory(struct kvm_vcpu *vcpu, if (r) return r; + direct_bits = 0; if (kvm_arch_has_private_mem(vcpu->kvm) && kvm_mem_is_private(vcpu->kvm, gpa_to_gfn(range->gpa))) error_code |= PFERR_PRIVATE_ACCESS; + else + direct_bits = gfn_to_gpa(kvm_gfn_direct_bits(vcpu->kvm)); /* * Shadow paging uses GVA for kvm page fault, so restrict to * two-dimensional paging. */ - r = kvm_tdp_map_page(vcpu, range->gpa, error_code, &level); + r = kvm_tdp_map_page(vcpu, range->gpa | direct_bits, error_code, &level); if (r < 0) return r; diff --git a/arch/x86/kvm/svm/sev.c b/arch/x86/kvm/svm/sev.c index 5a69b657dae9ac74790bf120f89c05f8ac905745..459c3b791fd4325a9ac7a22efee8e90f495b76ec 100644 --- a/arch/x86/kvm/svm/sev.c +++ b/arch/x86/kvm/svm/sev.c @@ -2871,6 +2871,33 @@ void __init sev_set_cpu_caps(void) } } +static bool is_sev_snp_initialized(void) +{ + struct sev_user_data_snp_status *status; + struct sev_data_snp_addr buf; + bool initialized = false; + int ret, error = 0; + + status = snp_alloc_firmware_page(GFP_KERNEL | __GFP_ZERO); + if (!status) + return false; + + buf.address = __psp_pa(status); + ret = sev_do_cmd(SEV_CMD_SNP_PLATFORM_STATUS, &buf, &error); + if (ret) { + pr_err("SEV: SNP_PLATFORM_STATUS failed ret=%d, fw_error=%d (%#x)\n", + ret, error, error); + goto out; + } + + initialized = !!status->state; + +out: + snp_free_firmware_page(status); + + return initialized; +} + void __init sev_hardware_setup(void) { unsigned int eax, ebx, ecx, edx, sev_asid_count, sev_es_asid_count; @@ -2975,6 +3002,14 @@ void __init sev_hardware_setup(void) sev_snp_supported = sev_snp_enabled && cc_platform_has(CC_ATTR_HOST_SEV_SNP); out: + if (sev_enabled) { + init_args.probe = true; + if (sev_platform_init(&init_args)) + sev_supported = sev_es_supported = sev_snp_supported = false; + else if (sev_snp_supported) + sev_snp_supported = is_sev_snp_initialized(); + } + if (boot_cpu_has(X86_FEATURE_SEV)) pr_info("SEV %s (ASIDs %u - %u)\n", sev_supported ? min_sev_asid <= max_sev_asid ? "enabled" : @@ -3001,15 +3036,6 @@ out: sev_supported_vmsa_features = 0; if (sev_es_debug_swap_enabled) sev_supported_vmsa_features |= SVM_SEV_FEAT_DEBUG_SWAP; - - if (!sev_enabled) - return; - - /* - * Do both SNP and SEV initialization at KVM module load. - */ - init_args.probe = true; - sev_platform_init(&init_args); } void sev_hardware_unsetup(void) diff --git a/arch/x86/power/hibernate.c b/arch/x86/power/hibernate.c index a7c23f2a58c994c0d3de553fa6fe944660995d28..a2294c1649f65e43cdac025a5a315fe6222754fd 100644 --- a/arch/x86/power/hibernate.c +++ b/arch/x86/power/hibernate.c @@ -192,7 +192,8 @@ out: int arch_resume_nosmt(void) { - int ret = 0; + int ret; + /* * We reached this while coming out of hibernation. This means * that SMT siblings are sleeping in hlt, as mwait is not safe @@ -206,18 +207,10 @@ int arch_resume_nosmt(void) * Called with hotplug disabled. */ cpu_hotplug_enable(); - if (cpu_smt_control == CPU_SMT_DISABLED || - cpu_smt_control == CPU_SMT_FORCE_DISABLED) { - enum cpuhp_smt_control old = cpu_smt_control; - - ret = cpuhp_smt_enable(); - if (ret) - goto out; - ret = cpuhp_smt_disable(old); - if (ret) - goto out; - } -out: + + ret = arch_cpu_rescan_dead_smt_siblings(); + cpu_hotplug_disable(); + return ret; } diff --git a/arch/xtensa/include/asm/pgtable.h b/arch/xtensa/include/asm/pgtable.h index cb1725c40e369200656e034c8a493064dd074780..d62aa1c316fcb1d8cef18c4734bbe7748276fbbe 100644 --- a/arch/xtensa/include/asm/pgtable.h +++ b/arch/xtensa/include/asm/pgtable.h @@ -349,7 +349,7 @@ ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr, pte_t *ptep) #define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) }) #define __swp_entry_to_pte(x) ((pte_t) { (x).val }) -static inline int pte_swp_exclusive(pte_t pte) +static inline bool pte_swp_exclusive(pte_t pte) { return pte_val(pte) & _PAGE_SWP_EXCLUSIVE; } diff --git a/block/blk-merge.c b/block/blk-merge.c index 3af1d284add5096e502c89ffebb6e3633f5854d5..70d704615be52046c46518746d15ce584515d7a2 100644 --- a/block/blk-merge.c +++ b/block/blk-merge.c @@ -998,20 +998,20 @@ bool blk_attempt_plug_merge(struct request_queue *q, struct bio *bio, if (!plug || rq_list_empty(&plug->mq_list)) return false; - rq_list_for_each(&plug->mq_list, rq) { - if (rq->q == q) { - if (blk_attempt_bio_merge(q, rq, bio, nr_segs, false) == - BIO_MERGE_OK) - return true; - break; - } + rq = plug->mq_list.tail; + if (rq->q == q) + return blk_attempt_bio_merge(q, rq, bio, nr_segs, false) == + BIO_MERGE_OK; + else if (!plug->multiple_queues) + return false; - /* - * Only keep iterating plug list for merges if we have multiple - * queues - */ - if (!plug->multiple_queues) - break; + rq_list_for_each(&plug->mq_list, rq) { + if (rq->q != q) + continue; + if (blk_attempt_bio_merge(q, rq, bio, nr_segs, false) == + BIO_MERGE_OK) + return true; + break; } return false; } diff --git a/block/blk-zoned.c b/block/blk-zoned.c index 8f15d1aa6eb89a99b7244de5d6583ad88ae72752..351d659280e116c2b6ffa61d4bdda7ce3f253d0e 100644 --- a/block/blk-zoned.c +++ b/block/blk-zoned.c @@ -1225,6 +1225,7 @@ void blk_zone_write_plug_bio_endio(struct bio *bio) if (bio_flagged(bio, BIO_EMULATES_ZONE_APPEND)) { bio->bi_opf &= ~REQ_OP_MASK; bio->bi_opf |= REQ_OP_ZONE_APPEND; + bio_clear_flag(bio, BIO_EMULATES_ZONE_APPEND); } /* @@ -1306,7 +1307,6 @@ again: spin_unlock_irqrestore(&zwplug->lock, flags); bdev = bio->bi_bdev; - submit_bio_noacct_nocheck(bio); /* * blk-mq devices will reuse the extra reference on the request queue @@ -1314,8 +1314,12 @@ again: * path for BIO-based devices will not do that. So drop this extra * reference here. */ - if (bdev_test_flag(bdev, BD_HAS_SUBMIT_BIO)) + if (bdev_test_flag(bdev, BD_HAS_SUBMIT_BIO)) { + bdev->bd_disk->fops->submit_bio(bio); blk_queue_exit(bdev->bd_disk->queue); + } else { + blk_mq_submit_bio(bio); + } put_zwplug: /* Drop the reference we took in disk_zone_wplug_schedule_bio_work(). */ diff --git a/crypto/hkdf.c b/crypto/hkdf.c index f24c2a8d4df99a1f410f3f34f19f1ee0f1e69edd..82d1b32ca6ce474869570f7533c039590b68349d 100644 --- a/crypto/hkdf.c +++ b/crypto/hkdf.c @@ -566,7 +566,7 @@ static int __init crypto_hkdf_module_init(void) static void __exit crypto_hkdf_module_exit(void) {} -module_init(crypto_hkdf_module_init); +late_initcall(crypto_hkdf_module_init); module_exit(crypto_hkdf_module_exit); MODULE_LICENSE("GPL"); diff --git a/drivers/accel/amdxdna/aie2_psp.c b/drivers/accel/amdxdna/aie2_psp.c index dc3a072ce3b6df6d952291fef9620f30a257db5a..f28a060a88109beba6837d5244123558a43aedc1 100644 --- a/drivers/accel/amdxdna/aie2_psp.c +++ b/drivers/accel/amdxdna/aie2_psp.c @@ -126,8 +126,8 @@ struct psp_device *aie2m_psp_create(struct drm_device *ddev, struct psp_config * psp->ddev = ddev; memcpy(psp->psp_regs, conf->psp_regs, sizeof(psp->psp_regs)); - psp->fw_buf_sz = ALIGN(conf->fw_size, PSP_FW_ALIGN) + PSP_FW_ALIGN; - psp->fw_buffer = drmm_kmalloc(ddev, psp->fw_buf_sz, GFP_KERNEL); + psp->fw_buf_sz = ALIGN(conf->fw_size, PSP_FW_ALIGN); + psp->fw_buffer = drmm_kmalloc(ddev, psp->fw_buf_sz + PSP_FW_ALIGN, GFP_KERNEL); if (!psp->fw_buffer) { drm_err(ddev, "no memory for fw buffer"); return NULL; diff --git a/drivers/acpi/acpi_pad.c b/drivers/acpi/acpi_pad.c index 6f8bbe1247a5d4f3fa44ee0792cd834a4203e8a3..c9a0bcaba2e4c148b45c3d8f5da73ccb50022858 100644 --- a/drivers/acpi/acpi_pad.c +++ b/drivers/acpi/acpi_pad.c @@ -33,7 +33,7 @@ static DEFINE_MUTEX(isolated_cpus_lock); static DEFINE_MUTEX(round_robin_lock); -static unsigned long power_saving_mwait_eax; +static unsigned int power_saving_mwait_eax; static unsigned char tsc_detected_unstable; static unsigned char tsc_marked_unstable; diff --git a/drivers/acpi/apei/einj-core.c b/drivers/acpi/apei/einj-core.c index fea11a35eea3e18ada0f87f9840f2b613b2410db..9b041415a9d0186e34a5fbc7975566dd6aa15a80 100644 --- a/drivers/acpi/apei/einj-core.c +++ b/drivers/acpi/apei/einj-core.c @@ -883,19 +883,16 @@ static int __init einj_init(void) } einj_dev = faux_device_create("acpi-einj", NULL, &einj_device_ops); - if (!einj_dev) - return -ENODEV; - einj_initialized = true; + if (einj_dev) + einj_initialized = true; return 0; } static void __exit einj_exit(void) { - if (einj_initialized) - faux_device_destroy(einj_dev); - + faux_device_destroy(einj_dev); } module_init(einj_init); diff --git a/drivers/acpi/cppc_acpi.c b/drivers/acpi/cppc_acpi.c index a9ae2fd628630de691f246d9878e49a56c14f2de..6b649031808f805b2bf4212d2ffca6719013d098 100644 --- a/drivers/acpi/cppc_acpi.c +++ b/drivers/acpi/cppc_acpi.c @@ -476,7 +476,7 @@ bool cppc_allow_fast_switch(void) struct cpc_desc *cpc_ptr; int cpu; - for_each_possible_cpu(cpu) { + for_each_present_cpu(cpu) { cpc_ptr = per_cpu(cpc_desc_ptr, cpu); desired_reg = &cpc_ptr->cpc_regs[DESIRED_PERF]; if (!CPC_IN_SYSTEM_MEMORY(desired_reg) && diff --git a/drivers/acpi/ec.c b/drivers/acpi/ec.c index 6f4203716b53f6320a4ad110c8ce158175b75fb6..75c7db8b156a206a5436189afba9890673cf1694 100644 --- a/drivers/acpi/ec.c +++ b/drivers/acpi/ec.c @@ -23,8 +23,10 @@ #include #include #include +#include #include #include +#include #include #include #include @@ -2031,6 +2033,21 @@ void __init acpi_ec_ecdt_probe(void) goto out; } + if (!strstarts(ecdt_ptr->id, "\\")) { + /* + * The ECDT table on some MSI notebooks contains invalid data, together + * with an empty ID string (""). + * + * Section 5.2.15 of the ACPI specification requires the ID string to be + * a "fully qualified reference to the (...) embedded controller device", + * so this string always has to start with a backslash. + * + * By verifying this we can avoid such faulty ECDT tables in a safe way. + */ + pr_err(FW_BUG "Ignoring ECDT due to invalid ID string \"%s\"\n", ecdt_ptr->id); + goto out; + } + ec = acpi_ec_alloc(); if (!ec) goto out; diff --git a/drivers/acpi/internal.h b/drivers/acpi/internal.h index 00910ccd7eda81de382d4a1c249bf43280e2a777..e2781864fdcebd54cc3a16ae33f3a7ec36fd5754 100644 --- a/drivers/acpi/internal.h +++ b/drivers/acpi/internal.h @@ -175,6 +175,12 @@ bool processor_physically_present(acpi_handle handle); static inline void acpi_early_processor_control_setup(void) {} #endif +#ifdef CONFIG_ACPI_PROCESSOR_CSTATE +void acpi_idle_rescan_dead_smt_siblings(void); +#else +static inline void acpi_idle_rescan_dead_smt_siblings(void) {} +#endif + /* -------------------------------------------------------------------------- Embedded Controller -------------------------------------------------------------------------- */ diff --git a/drivers/acpi/processor_driver.c b/drivers/acpi/processor_driver.c index 3b281bc1e73c3b35f76cbad1e3207e7907dc05c0..65e779be64ffcc6b728fc1b78df14435514c45db 100644 --- a/drivers/acpi/processor_driver.c +++ b/drivers/acpi/processor_driver.c @@ -279,6 +279,9 @@ static int __init acpi_processor_driver_init(void) * after acpi_cppc_processor_probe() has been called for all online CPUs */ acpi_processor_init_invariance_cppc(); + + acpi_idle_rescan_dead_smt_siblings(); + return 0; err: driver_unregister(&acpi_processor_driver); diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c index e2febca2ec13e0d88e879cfd2f63c4844b81eec5..2c2dc559e0f8def8469f3b94f2d7935f2cdb78a4 100644 --- a/drivers/acpi/processor_idle.c +++ b/drivers/acpi/processor_idle.c @@ -24,6 +24,8 @@ #include #include +#include "internal.h" + /* * Include the apic definitions for x86 to have the APIC timer related defines * available also for UP (on SMP it gets magically included via linux/smp.h). @@ -55,6 +57,12 @@ struct cpuidle_driver acpi_idle_driver = { }; #ifdef CONFIG_ACPI_PROCESSOR_CSTATE +void acpi_idle_rescan_dead_smt_siblings(void) +{ + if (cpuidle_get_driver() == &acpi_idle_driver) + arch_cpu_rescan_dead_smt_siblings(); +} + static DEFINE_PER_CPU(struct acpi_processor_cx * [CPUIDLE_STATE_MAX], acpi_cstate); diff --git a/drivers/acpi/resource.c b/drivers/acpi/resource.c index 7d59c6c9185fc1eacc5be5eaa353c5db1e3e248d..b1ab192d7a0809fe967e7a6d1318c5a3e04848ac 100644 --- a/drivers/acpi/resource.c +++ b/drivers/acpi/resource.c @@ -666,6 +666,13 @@ static const struct dmi_system_id irq1_edge_low_force_override[] = { DMI_MATCH(DMI_BOARD_NAME, "GMxHGxx"), }, }, + { + /* MACHENIKE L16P/L16P */ + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "MACHENIKE"), + DMI_MATCH(DMI_BOARD_NAME, "L16P"), + }, + }, { /* * TongFang GM5HG0A in case of the SKIKK Vanaheim relabel the diff --git a/drivers/base/faux.c b/drivers/base/faux.c index 9054d346bd7fe89575b95c1491467850bcd0393a..f5fbda0a9a44bd8e4831ee0ef9fc87841146d2da 100644 --- a/drivers/base/faux.c +++ b/drivers/base/faux.c @@ -86,6 +86,7 @@ static struct device_driver faux_driver = { .name = "faux_driver", .bus = &faux_bus_type, .probe_type = PROBE_FORCE_SYNCHRONOUS, + .suppress_bind_attrs = true, }; static void faux_device_release(struct device *dev) @@ -169,7 +170,7 @@ struct faux_device *faux_device_create_with_groups(const char *name, * successful is almost impossible to determine by the caller. */ if (!dev->driver) { - dev_err(dev, "probe did not succeed, tearing down the device\n"); + dev_dbg(dev, "probe did not succeed, tearing down the device\n"); faux_device_destroy(faux_dev); faux_dev = NULL; } diff --git a/drivers/block/loop.c b/drivers/block/loop.c index f8d136684109aa72ac1a897bb4340fcc3907f198..500840e4a74efc16c213522787ba8c1079f09380 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -1248,12 +1248,6 @@ loop_set_status(struct loop_device *lo, const struct loop_info64 *info) lo->lo_flags &= ~LOOP_SET_STATUS_CLEARABLE_FLAGS; lo->lo_flags |= (info->lo_flags & LOOP_SET_STATUS_SETTABLE_FLAGS); - if (size_changed) { - loff_t new_size = get_size(lo->lo_offset, lo->lo_sizelimit, - lo->lo_backing_file); - loop_set_size(lo, new_size); - } - /* update the direct I/O flag if lo_offset changed */ loop_update_dio(lo); @@ -1261,6 +1255,11 @@ out_unfreeze: blk_mq_unfreeze_queue(lo->lo_queue, memflags); if (partscan) clear_bit(GD_SUPPRESS_PART_SCAN, &lo->lo_disk->state); + if (!err && size_changed) { + loff_t new_size = get_size(lo->lo_offset, lo->lo_sizelimit, + lo->lo_backing_file); + loop_set_size(lo, new_size); + } out_unlock: mutex_unlock(&lo->lo_mutex); if (partscan) diff --git a/drivers/bluetooth/btintel_pcie.c b/drivers/bluetooth/btintel_pcie.c index 50fe17f1e1d182236a9a0e9255cb6961002fd937..563165c5efae33b7840107c7e550f56ee3df01c9 100644 --- a/drivers/bluetooth/btintel_pcie.c +++ b/drivers/bluetooth/btintel_pcie.c @@ -396,8 +396,13 @@ static int btintel_pcie_submit_rx(struct btintel_pcie_data *data) static int btintel_pcie_start_rx(struct btintel_pcie_data *data) { int i, ret; + struct rxq *rxq = &data->rxq; + + /* Post (BTINTEL_PCIE_RX_DESCS_COUNT - 3) buffers to overcome the + * hardware issues leading to race condition at the firmware. + */ - for (i = 0; i < BTINTEL_PCIE_RX_MAX_QUEUE; i++) { + for (i = 0; i < rxq->count - 3; i++) { ret = btintel_pcie_submit_rx(data); if (ret) return ret; @@ -1782,8 +1787,8 @@ static int btintel_pcie_alloc(struct btintel_pcie_data *data) * + size of index * Number of queues(2) * type of index array(4) * + size of context information */ - total = (sizeof(struct tfd) + sizeof(struct urbd0) + sizeof(struct frbd) - + sizeof(struct urbd1)) * BTINTEL_DESCS_COUNT; + total = (sizeof(struct tfd) + sizeof(struct urbd0)) * BTINTEL_PCIE_TX_DESCS_COUNT; + total += (sizeof(struct frbd) + sizeof(struct urbd1)) * BTINTEL_PCIE_RX_DESCS_COUNT; /* Add the sum of size of index array and size of ci struct */ total += (sizeof(u16) * BTINTEL_PCIE_NUM_QUEUES * 4) + sizeof(struct ctx_info); @@ -1808,36 +1813,36 @@ static int btintel_pcie_alloc(struct btintel_pcie_data *data) data->dma_v_addr = v_addr; /* Setup descriptor count */ - data->txq.count = BTINTEL_DESCS_COUNT; - data->rxq.count = BTINTEL_DESCS_COUNT; + data->txq.count = BTINTEL_PCIE_TX_DESCS_COUNT; + data->rxq.count = BTINTEL_PCIE_RX_DESCS_COUNT; /* Setup tfds */ data->txq.tfds_p_addr = p_addr; data->txq.tfds = v_addr; - p_addr += (sizeof(struct tfd) * BTINTEL_DESCS_COUNT); - v_addr += (sizeof(struct tfd) * BTINTEL_DESCS_COUNT); + p_addr += (sizeof(struct tfd) * BTINTEL_PCIE_TX_DESCS_COUNT); + v_addr += (sizeof(struct tfd) * BTINTEL_PCIE_TX_DESCS_COUNT); /* Setup urbd0 */ data->txq.urbd0s_p_addr = p_addr; data->txq.urbd0s = v_addr; - p_addr += (sizeof(struct urbd0) * BTINTEL_DESCS_COUNT); - v_addr += (sizeof(struct urbd0) * BTINTEL_DESCS_COUNT); + p_addr += (sizeof(struct urbd0) * BTINTEL_PCIE_TX_DESCS_COUNT); + v_addr += (sizeof(struct urbd0) * BTINTEL_PCIE_TX_DESCS_COUNT); /* Setup FRBD*/ data->rxq.frbds_p_addr = p_addr; data->rxq.frbds = v_addr; - p_addr += (sizeof(struct frbd) * BTINTEL_DESCS_COUNT); - v_addr += (sizeof(struct frbd) * BTINTEL_DESCS_COUNT); + p_addr += (sizeof(struct frbd) * BTINTEL_PCIE_RX_DESCS_COUNT); + v_addr += (sizeof(struct frbd) * BTINTEL_PCIE_RX_DESCS_COUNT); /* Setup urbd1 */ data->rxq.urbd1s_p_addr = p_addr; data->rxq.urbd1s = v_addr; - p_addr += (sizeof(struct urbd1) * BTINTEL_DESCS_COUNT); - v_addr += (sizeof(struct urbd1) * BTINTEL_DESCS_COUNT); + p_addr += (sizeof(struct urbd1) * BTINTEL_PCIE_RX_DESCS_COUNT); + v_addr += (sizeof(struct urbd1) * BTINTEL_PCIE_RX_DESCS_COUNT); /* Setup data buffers for txq */ err = btintel_pcie_setup_txq_bufs(data, &data->txq); diff --git a/drivers/bluetooth/btintel_pcie.h b/drivers/bluetooth/btintel_pcie.h index 21b964b15c1c924c86a2d639b41f2c0fcf1e96df..7dad4523236c9fcc1fdb0d057ce902c16a5d8029 100644 --- a/drivers/bluetooth/btintel_pcie.h +++ b/drivers/bluetooth/btintel_pcie.h @@ -154,8 +154,11 @@ enum msix_mbox_int_causes { /* Default interrupt timeout in msec */ #define BTINTEL_DEFAULT_INTR_TIMEOUT_MS 3000 -/* The number of descriptors in TX/RX queues */ -#define BTINTEL_DESCS_COUNT 16 +/* The number of descriptors in TX queues */ +#define BTINTEL_PCIE_TX_DESCS_COUNT 32 + +/* The number of descriptors in RX queues */ +#define BTINTEL_PCIE_RX_DESCS_COUNT 64 /* Number of Queue for TX and RX * It indicates the index of the IA(Index Array) @@ -177,9 +180,6 @@ enum { /* Doorbell vector for TFD */ #define BTINTEL_PCIE_TX_DB_VEC 0 -/* Number of pending RX requests for downlink */ -#define BTINTEL_PCIE_RX_MAX_QUEUE 6 - /* Doorbell vector for FRBD */ #define BTINTEL_PCIE_RX_DB_VEC 513 diff --git a/drivers/cpufreq/rcpufreq_dt.rs b/drivers/cpufreq/rcpufreq_dt.rs index 94ed81644fe1c3f8a0240dede2299102a2118e73..43c87d0259b6ad89b7bf222fc991f043fd513957 100644 --- a/drivers/cpufreq/rcpufreq_dt.rs +++ b/drivers/cpufreq/rcpufreq_dt.rs @@ -26,9 +26,9 @@ fn find_supply_name_exact(dev: &Device, name: &str) -> Option { } /// Finds supply name for the CPU from DT. -fn find_supply_names(dev: &Device, cpu: u32) -> Option> { +fn find_supply_names(dev: &Device, cpu: cpu::CpuId) -> Option> { // Try "cpu0" for older DTs, fallback to "cpu". - let name = (cpu == 0) + let name = (cpu.as_u32() == 0) .then(|| find_supply_name_exact(dev, "cpu0")) .flatten() .or_else(|| find_supply_name_exact(dev, "cpu"))?; diff --git a/drivers/dma-buf/dma-buf.c b/drivers/dma-buf/dma-buf.c index 890ecac04dac8619fc64c0bc5e2f5323eb30449d..2bcf9ceca99758fa0ee89a4d281b49925676899e 100644 --- a/drivers/dma-buf/dma-buf.c +++ b/drivers/dma-buf/dma-buf.c @@ -1118,7 +1118,7 @@ struct sg_table *dma_buf_map_attachment(struct dma_buf_attachment *attach, * Catch exporters making buffers inaccessible even when * attachments preventing that exist. */ - WARN_ON_ONCE(ret == EBUSY); + WARN_ON_ONCE(ret == -EBUSY); if (ret) return ERR_PTR(ret); } diff --git a/drivers/dma-buf/udmabuf.c b/drivers/dma-buf/udmabuf.c index af5200e360a6a95a8566ad4762efc112b2396fae..40399c26e6be6283e76ab3d9a8eedca9e4c6f932 100644 --- a/drivers/dma-buf/udmabuf.c +++ b/drivers/dma-buf/udmabuf.c @@ -257,8 +257,7 @@ static int begin_cpu_udmabuf(struct dma_buf *buf, ubuf->sg = NULL; } } else { - dma_sync_sg_for_cpu(dev, ubuf->sg->sgl, ubuf->sg->nents, - direction); + dma_sync_sgtable_for_cpu(dev, ubuf->sg, direction); } return ret; @@ -273,7 +272,7 @@ static int end_cpu_udmabuf(struct dma_buf *buf, if (!ubuf->sg) return -EINVAL; - dma_sync_sg_for_device(dev, ubuf->sg->sgl, ubuf->sg->nents, direction); + dma_sync_sgtable_for_device(dev, ubuf->sg, direction); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile index 87080c06e5fccb7d84f7e715dbf5a077e76db29f..930de203d533c37267affaf10a1a25f666cd07bf 100644 --- a/drivers/gpu/drm/amd/amdgpu/Makefile +++ b/drivers/gpu/drm/amd/amdgpu/Makefile @@ -66,7 +66,7 @@ amdgpu-y += amdgpu_device.o amdgpu_doorbell_mgr.o amdgpu_kms.o \ amdgpu_fw_attestation.o amdgpu_securedisplay.o \ amdgpu_eeprom.o amdgpu_mca.o amdgpu_psp_ta.o amdgpu_lsdma.o \ amdgpu_ring_mux.o amdgpu_xcp.o amdgpu_seq64.o amdgpu_aca.o amdgpu_dev_coredump.o \ - amdgpu_cper.o amdgpu_userq_fence.o amdgpu_eviction_fence.o + amdgpu_cper.o amdgpu_userq_fence.o amdgpu_eviction_fence.o amdgpu_ip.o amdgpu-$(CONFIG_PROC_FS) += amdgpu_fdinfo.o diff --git a/drivers/gpu/drm/amd/amdgpu/aldebaran.c b/drivers/gpu/drm/amd/amdgpu/aldebaran.c index e13fbd97414126ef068bece1b57c61c6767803d9..9569dc16dd3dac6a6a558c923f01ff2a355f69e0 100644 --- a/drivers/gpu/drm/amd/amdgpu/aldebaran.c +++ b/drivers/gpu/drm/amd/amdgpu/aldebaran.c @@ -71,18 +71,29 @@ aldebaran_get_reset_handler(struct amdgpu_reset_control *reset_ctl, return NULL; } +static inline uint32_t aldebaran_get_ip_block_mask(struct amdgpu_device *adev) +{ + uint32_t ip_block_mask = BIT(AMD_IP_BLOCK_TYPE_GFX) | + BIT(AMD_IP_BLOCK_TYPE_SDMA); + + if (adev->aid_mask) + ip_block_mask |= BIT(AMD_IP_BLOCK_TYPE_IH); + + return ip_block_mask; +} + static int aldebaran_mode2_suspend_ip(struct amdgpu_device *adev) { + uint32_t ip_block_mask = aldebaran_get_ip_block_mask(adev); + uint32_t ip_block; int r, i; amdgpu_device_set_pg_state(adev, AMD_PG_STATE_UNGATE); amdgpu_device_set_cg_state(adev, AMD_CG_STATE_UNGATE); for (i = adev->num_ip_blocks - 1; i >= 0; i--) { - if (!(adev->ip_blocks[i].version->type == - AMD_IP_BLOCK_TYPE_GFX || - adev->ip_blocks[i].version->type == - AMD_IP_BLOCK_TYPE_SDMA)) + ip_block = BIT(adev->ip_blocks[i].version->type); + if (!(ip_block_mask & ip_block)) continue; r = amdgpu_ip_block_suspend(&adev->ip_blocks[i]); @@ -200,8 +211,10 @@ aldebaran_mode2_perform_reset(struct amdgpu_reset_control *reset_ctl, static int aldebaran_mode2_restore_ip(struct amdgpu_device *adev) { struct amdgpu_firmware_info *ucode_list[AMDGPU_UCODE_ID_MAXIMUM]; + uint32_t ip_block_mask = aldebaran_get_ip_block_mask(adev); struct amdgpu_firmware_info *ucode; struct amdgpu_ip_block *cmn_block; + struct amdgpu_ip_block *ih_block; int ucode_count = 0; int i, r; @@ -243,6 +256,18 @@ static int aldebaran_mode2_restore_ip(struct amdgpu_device *adev) if (r) return r; + if (ip_block_mask & BIT(AMD_IP_BLOCK_TYPE_IH)) { + ih_block = amdgpu_device_ip_get_ip_block(adev, + AMD_IP_BLOCK_TYPE_IH); + if (unlikely(!ih_block)) { + dev_err(adev->dev, "Failed to get IH handle\n"); + return -EINVAL; + } + r = amdgpu_ip_block_resume(ih_block); + if (r) + return r; + } + /* Reinit GFXHUB */ adev->gfxhub.funcs->init(adev); r = adev->gfxhub.funcs->gart_enable(adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index a5ccd0ada16ab03241f9e61d55fb2a312fd21f31..7edb7ba917869239c4c8883addebf06d708f3b59 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -470,9 +470,6 @@ struct amdgpu_sa_manager { void *cpu_ptr; }; -int amdgpu_fence_slab_init(void); -void amdgpu_fence_slab_fini(void); - /* * IRQS. */ @@ -1282,6 +1279,7 @@ struct amdgpu_device { bool debug_exp_resets; bool debug_disable_gpu_ring_reset; bool debug_vm_userptr; + bool debug_disable_ce_logs; /* Protection for the following isolation structure */ struct mutex enforce_isolation_mutex; @@ -1336,6 +1334,11 @@ static inline struct amdgpu_device *amdgpu_ttm_adev(struct ttm_device *bdev) return container_of(bdev, struct amdgpu_device, mman.bdev); } +static inline bool amdgpu_is_multi_aid(struct amdgpu_device *adev) +{ + return !!adev->aid_mask; +} + int amdgpu_device_init(struct amdgpu_device *adev, uint32_t flags); void amdgpu_device_fini_hw(struct amdgpu_device *adev); @@ -1387,7 +1390,8 @@ void amdgpu_device_indirect_wreg64(struct amdgpu_device *adev, void amdgpu_device_indirect_wreg64_ext(struct amdgpu_device *adev, u64 reg_addr, u64 reg_data); u32 amdgpu_device_get_rev_id(struct amdgpu_device *adev); -bool amdgpu_device_asic_has_dc_support(enum amd_asic_type asic_type); +bool amdgpu_device_asic_has_dc_support(struct pci_dev *pdev, + enum amd_asic_type asic_type); bool amdgpu_device_has_dc_support(struct amdgpu_device *adev); void amdgpu_device_set_sriov_virtual_display(struct amdgpu_device *adev); @@ -1619,6 +1623,7 @@ void amdgpu_driver_release_kms(struct drm_device *dev); int amdgpu_device_ip_suspend(struct amdgpu_device *adev); int amdgpu_device_prepare(struct drm_device *dev); +void amdgpu_device_complete(struct drm_device *dev); int amdgpu_device_suspend(struct drm_device *dev, bool fbcon); int amdgpu_device_resume(struct drm_device *dev, bool fbcon); u32 amdgpu_get_vblank_counter_kms(struct drm_crtc *crtc); @@ -1760,4 +1765,19 @@ extern const struct attribute_group amdgpu_flash_attr_group; void amdgpu_set_init_level(struct amdgpu_device *adev, enum amdgpu_init_lvl_id lvl); + +static inline int amdgpu_device_bus_status_check(struct amdgpu_device *adev) +{ + u32 status; + int r; + + r = pci_read_config_dword(adev->pdev, PCI_COMMAND, &status); + if (r || PCI_POSSIBLE_ERROR(status)) { + dev_err(adev->dev, "device lost from bus!"); + return -ENODEV; + } + + return 0; +} + #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c index 3835f25929142a03a38bbda15bba910be3782e0b..cbc40cad581b4455c34586fbbeff6f7c6162f699 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c @@ -115,6 +115,11 @@ static void aca_smu_bank_dump(struct amdgpu_device *adev, int idx, int total, st u64 event_id = qctx ? qctx->evid.event_id : RAS_EVENT_INVALID_ID; int i; + if (adev->debug_disable_ce_logs && + bank->smu_err_type == ACA_SMU_TYPE_CE && + !ACA_BANK_ERR_IS_DEFFERED(bank)) + return; + RAS_EVENT_LOG(adev, event_id, HW_ERR "Accelerator Check Architecture events logged\n"); /* plus 1 for output format, e.g: ACA[08/08]: xxxx */ for (i = 0; i < ARRAY_SIZE(aca_regs); i++) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c index 55161e5cdc304a1cb531624c77775c41eae89f76..fbe7616555c83f6988910194116dc7df6bdbc75d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c @@ -248,18 +248,34 @@ void amdgpu_amdkfd_interrupt(struct amdgpu_device *adev, kgd2kfd_interrupt(adev->kfd.dev, ih_ring_entry); } -void amdgpu_amdkfd_suspend(struct amdgpu_device *adev, bool run_pm) +void amdgpu_amdkfd_suspend(struct amdgpu_device *adev, bool suspend_proc) { if (adev->kfd.dev) - kgd2kfd_suspend(adev->kfd.dev, run_pm); + kgd2kfd_suspend(adev->kfd.dev, suspend_proc); } -int amdgpu_amdkfd_resume(struct amdgpu_device *adev, bool run_pm) +int amdgpu_amdkfd_resume(struct amdgpu_device *adev, bool resume_proc) { int r = 0; if (adev->kfd.dev) - r = kgd2kfd_resume(adev->kfd.dev, run_pm); + r = kgd2kfd_resume(adev->kfd.dev, resume_proc); + + return r; +} + +void amdgpu_amdkfd_suspend_process(struct amdgpu_device *adev) +{ + if (adev->kfd.dev) + kgd2kfd_suspend_process(adev->kfd.dev); +} + +int amdgpu_amdkfd_resume_process(struct amdgpu_device *adev) +{ + int r = 0; + + if (adev->kfd.dev) + r = kgd2kfd_resume_process(adev->kfd.dev); return r; } @@ -749,12 +765,12 @@ int amdgpu_amdkfd_send_close_event_drain_irq(struct amdgpu_device *adev, int amdgpu_amdkfd_check_and_lock_kfd(struct amdgpu_device *adev) { - return kgd2kfd_check_and_lock_kfd(); + return kgd2kfd_check_and_lock_kfd(adev->kfd.dev); } void amdgpu_amdkfd_unlock_kfd(struct amdgpu_device *adev) { - kgd2kfd_unlock_kfd(); + kgd2kfd_unlock_kfd(adev->kfd.dev); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h index b6ca41859b53676a37ca8aa07a61e4c8c9279eae..33eb4826b58b1a84a410b112b38e777064ccf9af 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h @@ -154,8 +154,10 @@ struct amdkfd_process_info { int amdgpu_amdkfd_init(void); void amdgpu_amdkfd_fini(void); -void amdgpu_amdkfd_suspend(struct amdgpu_device *adev, bool run_pm); -int amdgpu_amdkfd_resume(struct amdgpu_device *adev, bool run_pm); +void amdgpu_amdkfd_suspend(struct amdgpu_device *adev, bool suspend_proc); +int amdgpu_amdkfd_resume(struct amdgpu_device *adev, bool resume_proc); +void amdgpu_amdkfd_suspend_process(struct amdgpu_device *adev); +int amdgpu_amdkfd_resume_process(struct amdgpu_device *adev); void amdgpu_amdkfd_interrupt(struct amdgpu_device *adev, const void *ih_ring_entry); void amdgpu_amdkfd_device_probe(struct amdgpu_device *adev); @@ -411,16 +413,18 @@ struct kfd_dev *kgd2kfd_probe(struct amdgpu_device *adev, bool vf); bool kgd2kfd_device_init(struct kfd_dev *kfd, const struct kgd2kfd_shared_resources *gpu_resources); void kgd2kfd_device_exit(struct kfd_dev *kfd); -void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm); -int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm); +void kgd2kfd_suspend(struct kfd_dev *kfd, bool suspend_proc); +int kgd2kfd_resume(struct kfd_dev *kfd, bool resume_proc); +void kgd2kfd_suspend_process(struct kfd_dev *kfd); +int kgd2kfd_resume_process(struct kfd_dev *kfd); int kgd2kfd_pre_reset(struct kfd_dev *kfd, struct amdgpu_reset_context *reset_context); int kgd2kfd_post_reset(struct kfd_dev *kfd); void kgd2kfd_interrupt(struct kfd_dev *kfd, const void *ih_ring_entry); void kgd2kfd_set_sram_ecc_flag(struct kfd_dev *kfd); void kgd2kfd_smi_event_throttle(struct kfd_dev *kfd, uint64_t throttle_bitmask); -int kgd2kfd_check_and_lock_kfd(void); -void kgd2kfd_unlock_kfd(void); +int kgd2kfd_check_and_lock_kfd(struct kfd_dev *kfd); +void kgd2kfd_unlock_kfd(struct kfd_dev *kfd); int kgd2kfd_start_sched(struct kfd_dev *kfd, uint32_t node_id); int kgd2kfd_stop_sched(struct kfd_dev *kfd, uint32_t node_id); bool kgd2kfd_compute_active(struct kfd_dev *kfd, uint32_t node_id); @@ -454,11 +458,20 @@ static inline void kgd2kfd_device_exit(struct kfd_dev *kfd) { } -static inline void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm) +static inline void kgd2kfd_suspend(struct kfd_dev *kfd, bool suspend_proc) { } -static inline int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm) +static inline int kgd2kfd_resume(struct kfd_dev *kfd, bool resume_proc) +{ + return 0; +} + +static inline void kgd2kfd_suspend_process(struct kfd_dev *kfd) +{ +} + +static inline int kgd2kfd_resume_process(struct kfd_dev *kfd) { return 0; } @@ -489,12 +502,12 @@ void kgd2kfd_smi_event_throttle(struct kfd_dev *kfd, uint64_t throttle_bitmask) { } -static inline int kgd2kfd_check_and_lock_kfd(void) +static inline int kgd2kfd_check_and_lock_kfd(struct kfd_dev *kfd) { return 0; } -static inline void kgd2kfd_unlock_kfd(void) +static inline void kgd2kfd_unlock_kfd(struct kfd_dev *kfd) { } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c index ffbaa8bc5eea9ea2ee23ec595eedc08ea97666ab..1105a09e55dc18309c47d01f652f07e6dda399d2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c @@ -320,7 +320,7 @@ static void set_barrier_auto_waitcnt(struct amdgpu_device *adev, bool enable_wai if (!down_read_trylock(&adev->reset_domain->sem)) return; - amdgpu_amdkfd_suspend(adev, false); + amdgpu_amdkfd_suspend(adev, true); if (suspend_resume_compute_scheduler(adev, true)) goto out; @@ -333,7 +333,7 @@ static void set_barrier_auto_waitcnt(struct amdgpu_device *adev, bool enable_wai out: suspend_resume_compute_scheduler(adev, false); - amdgpu_amdkfd_resume(adev, false); + amdgpu_amdkfd_resume(adev, true); up_read(&adev->reset_domain->sem); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c index ca4a6b82817f5349c8434c9aa0c1c89658853c30..df77558e03ef21fd754897440c47ca484fdb562d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c @@ -561,6 +561,13 @@ static uint32_t read_vmid_from_vmfault_reg(struct amdgpu_device *adev) return REG_GET_FIELD(status, VM_CONTEXT1_PROTECTION_FAULT_STATUS, VMID); } +static uint32_t kgd_hqd_sdma_get_doorbell(struct amdgpu_device *adev, + int engine, int queue) + +{ + return 0; +} + const struct kfd2kgd_calls gfx_v7_kfd2kgd = { .program_sh_mem_settings = kgd_program_sh_mem_settings, .set_pasid_vmid_mapping = kgd_set_pasid_vmid_mapping, @@ -578,4 +585,5 @@ const struct kfd2kgd_calls gfx_v7_kfd2kgd = { .set_scratch_backing_va = set_scratch_backing_va, .set_vm_context_page_table_base = set_vm_context_page_table_base, .read_vmid_from_vmfault_reg = read_vmid_from_vmfault_reg, + .hqd_sdma_get_doorbell = kgd_hqd_sdma_get_doorbell, }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c index 0f3e2944edd7e9d4aca62d48bcfa00ab7c6b0a16..e68c0fa8d7513a3f0ee195563e454f85ff61d024 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c @@ -582,6 +582,13 @@ static void set_vm_context_page_table_base(struct amdgpu_device *adev, lower_32_bits(page_table_base)); } +static uint32_t kgd_hqd_sdma_get_doorbell(struct amdgpu_device *adev, + int engine, int queue) + +{ + return 0; +} + const struct kfd2kgd_calls gfx_v8_kfd2kgd = { .program_sh_mem_settings = kgd_program_sh_mem_settings, .set_pasid_vmid_mapping = kgd_set_pasid_vmid_mapping, @@ -599,4 +606,5 @@ const struct kfd2kgd_calls gfx_v8_kfd2kgd = { get_atc_vmid_pasid_mapping_info, .set_scratch_backing_va = set_scratch_backing_va, .set_vm_context_page_table_base = set_vm_context_page_table_base, + .hqd_sdma_get_doorbell = kgd_hqd_sdma_get_doorbell, }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c index 5a234eadae8b3a26a940784044fd30c7e7efc109..15dde1f5032842420c7cbcfc30fed212fc20b022 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c @@ -212,7 +212,7 @@ int amdgpu_cper_entry_fill_bad_page_threshold_section(struct amdgpu_device *adev NONSTD_SEC_OFFSET(hdr->sec_cnt, idx)); amdgpu_cper_entry_fill_section_desc(adev, section_desc, true, false, - CPER_SEV_NUM, RUNTIME, NONSTD_SEC_LEN, + CPER_SEV_FATAL, RUNTIME, NONSTD_SEC_LEN, NONSTD_SEC_OFFSET(hdr->sec_cnt, idx)); section->hdr.valid_bits.err_info_cnt = 1; @@ -326,7 +326,9 @@ int amdgpu_cper_generate_bp_threshold_record(struct amdgpu_device *adev) return -ENOMEM; } - amdgpu_cper_entry_fill_hdr(adev, bp_threshold, AMDGPU_CPER_TYPE_BP_THRESHOLD, CPER_SEV_NUM); + amdgpu_cper_entry_fill_hdr(adev, bp_threshold, + AMDGPU_CPER_TYPE_BP_THRESHOLD, + CPER_SEV_FATAL); ret = amdgpu_cper_entry_fill_bad_page_threshold_section(adev, bp_threshold, 0); if (ret) return ret; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c index 85567d0d9545cf514a66560f333080ceb36c45d7..f5d5c45ddc0d15b0e1789e7e6209f52e7786551a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c @@ -944,6 +944,7 @@ static void amdgpu_ctx_mgr_entity_fini(struct amdgpu_ctx_mgr *mgr) drm_sched_entity_fini(entity); } } + kref_put(&ctx->refcount, amdgpu_ctx_fini); } } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c index e49890a23ef6ed9c9c8601e83021c7a2f529d57b..2353455d4aff8533854960dd8c869907f84fc61d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c @@ -1902,7 +1902,7 @@ no_preempt: continue; } job = to_amdgpu_job(s_job); - if (preempted && (&job->hw_fence) == fence) + if (preempted && (&job->hw_fence.base) == fence) /* mark the job as preempted */ job->preemption_status |= AMDGPU_IB_PREEMPTED; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index a59f194e3360e5555a4d78097f1a9b30bd7a039f..d282c0753b14b0b4ca890263885baa54ed611e14 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -232,7 +232,7 @@ static int amdgpu_device_attr_sysfs_init(struct amdgpu_device *adev) { int ret = 0; - if (!amdgpu_sriov_vf(adev)) + if (amdgpu_nbio_is_replay_cnt_supported(adev)) ret = sysfs_create_file(&adev->dev->kobj, &dev_attr_pcie_replay_count.attr); @@ -241,7 +241,7 @@ static int amdgpu_device_attr_sysfs_init(struct amdgpu_device *adev) static void amdgpu_device_attr_sysfs_fini(struct amdgpu_device *adev) { - if (!amdgpu_sriov_vf(adev)) + if (amdgpu_nbio_is_replay_cnt_supported(adev)) sysfs_remove_file(&adev->dev->kobj, &dev_attr_pcie_replay_count.attr); } @@ -1288,14 +1288,14 @@ u32 amdgpu_device_get_rev_id(struct amdgpu_device *adev) */ static uint32_t amdgpu_invalid_rreg(struct amdgpu_device *adev, uint32_t reg) { - DRM_ERROR("Invalid callback to read register 0x%04X\n", reg); + dev_err(adev->dev, "Invalid callback to read register 0x%04X\n", reg); BUG(); return 0; } static uint32_t amdgpu_invalid_rreg_ext(struct amdgpu_device *adev, uint64_t reg) { - DRM_ERROR("Invalid callback to read register 0x%llX\n", reg); + dev_err(adev->dev, "Invalid callback to read register 0x%llX\n", reg); BUG(); return 0; } @@ -1312,15 +1312,17 @@ static uint32_t amdgpu_invalid_rreg_ext(struct amdgpu_device *adev, uint64_t reg */ static void amdgpu_invalid_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v) { - DRM_ERROR("Invalid callback to write register 0x%04X with 0x%08X\n", - reg, v); + dev_err(adev->dev, + "Invalid callback to write register 0x%04X with 0x%08X\n", reg, + v); BUG(); } static void amdgpu_invalid_wreg_ext(struct amdgpu_device *adev, uint64_t reg, uint32_t v) { - DRM_ERROR("Invalid callback to write register 0x%llX with 0x%08X\n", - reg, v); + dev_err(adev->dev, + "Invalid callback to write register 0x%llX with 0x%08X\n", reg, + v); BUG(); } @@ -1336,14 +1338,15 @@ static void amdgpu_invalid_wreg_ext(struct amdgpu_device *adev, uint64_t reg, ui */ static uint64_t amdgpu_invalid_rreg64(struct amdgpu_device *adev, uint32_t reg) { - DRM_ERROR("Invalid callback to read 64 bit register 0x%04X\n", reg); + dev_err(adev->dev, "Invalid callback to read 64 bit register 0x%04X\n", + reg); BUG(); return 0; } static uint64_t amdgpu_invalid_rreg64_ext(struct amdgpu_device *adev, uint64_t reg) { - DRM_ERROR("Invalid callback to read register 0x%llX\n", reg); + dev_err(adev->dev, "Invalid callback to read register 0x%llX\n", reg); BUG(); return 0; } @@ -1360,15 +1363,17 @@ static uint64_t amdgpu_invalid_rreg64_ext(struct amdgpu_device *adev, uint64_t r */ static void amdgpu_invalid_wreg64(struct amdgpu_device *adev, uint32_t reg, uint64_t v) { - DRM_ERROR("Invalid callback to write 64 bit register 0x%04X with 0x%08llX\n", - reg, v); + dev_err(adev->dev, + "Invalid callback to write 64 bit register 0x%04X with 0x%08llX\n", + reg, v); BUG(); } static void amdgpu_invalid_wreg64_ext(struct amdgpu_device *adev, uint64_t reg, uint64_t v) { - DRM_ERROR("Invalid callback to write 64 bit register 0x%llX with 0x%08llX\n", - reg, v); + dev_err(adev->dev, + "Invalid callback to write 64 bit register 0x%llX with 0x%08llX\n", + reg, v); BUG(); } @@ -1386,8 +1391,9 @@ static void amdgpu_invalid_wreg64_ext(struct amdgpu_device *adev, uint64_t reg, static uint32_t amdgpu_block_invalid_rreg(struct amdgpu_device *adev, uint32_t block, uint32_t reg) { - DRM_ERROR("Invalid callback to read register 0x%04X in block 0x%04X\n", - reg, block); + dev_err(adev->dev, + "Invalid callback to read register 0x%04X in block 0x%04X\n", + reg, block); BUG(); return 0; } @@ -1407,8 +1413,9 @@ static void amdgpu_block_invalid_wreg(struct amdgpu_device *adev, uint32_t block, uint32_t reg, uint32_t v) { - DRM_ERROR("Invalid block callback to write register 0x%04X in block 0x%04X with 0x%08X\n", - reg, block, v); + dev_err(adev->dev, + "Invalid block callback to write register 0x%04X in block 0x%04X with 0x%08X\n", + reg, block, v); BUG(); } @@ -1694,7 +1701,9 @@ int amdgpu_device_resize_fb_bar(struct amdgpu_device *adev) /* PCI_EXT_CAP_ID_VNDR extended capability is located at 0x100 */ if (!pci_find_ext_capability(adev->pdev, PCI_EXT_CAP_ID_VNDR)) - DRM_WARN("System can't access extended configuration space, please check!!\n"); + dev_warn( + adev->dev, + "System can't access extended configuration space, please check!!\n"); /* skip if the bios has already enabled large BAR */ if (adev->gmc.real_vram_size && @@ -1734,9 +1743,10 @@ int amdgpu_device_resize_fb_bar(struct amdgpu_device *adev) r = pci_resize_resource(adev->pdev, 0, rbar_size); if (r == -ENOSPC) - DRM_INFO("Not enough PCI address space for a large BAR."); + dev_info(adev->dev, + "Not enough PCI address space for a large BAR."); else if (r && r != -ENOTSUPP) - DRM_ERROR("Problem resizing BAR0 (%d).", r); + dev_err(adev->dev, "Problem resizing BAR0 (%d).", r); pci_assign_unassigned_bus_resources(adev->pdev->bus); @@ -1838,8 +1848,8 @@ bool amdgpu_device_seamless_boot_supported(struct amdgpu_device *adev) case 0: return false; default: - DRM_ERROR("Invalid value for amdgpu.seamless: %d\n", - amdgpu_seamless); + dev_err(adev->dev, "Invalid value for amdgpu.seamless: %d\n", + amdgpu_seamless); return false; } @@ -2015,7 +2025,7 @@ static void amdgpu_device_check_smu_prv_buffer_size(struct amdgpu_device *adev) return; if (!is_os_64) { - DRM_WARN("Not 64-bit OS, feature not supported\n"); + dev_warn(adev->dev, "Not 64-bit OS, feature not supported\n"); goto def_value; } si_meminfo(&si); @@ -2030,7 +2040,7 @@ static void amdgpu_device_check_smu_prv_buffer_size(struct amdgpu_device *adev) if (total_memory < dram_size_seven_GB) goto def_value1; } else { - DRM_WARN("Smu memory pool size not supported\n"); + dev_warn(adev->dev, "Smu memory pool size not supported\n"); goto def_value; } adev->pm.smu_prv_buffer_size = amdgpu_smu_memory_pool_size << 28; @@ -2038,7 +2048,7 @@ static void amdgpu_device_check_smu_prv_buffer_size(struct amdgpu_device *adev) return; def_value1: - DRM_WARN("No enough system memory\n"); + dev_warn(adev->dev, "No enough system memory\n"); def_value: adev->pm.smu_prv_buffer_size = 0; } @@ -2202,12 +2212,13 @@ static void amdgpu_switcheroo_set_state(struct pci_dev *pdev, amdgpu_device_load_pci_state(pdev); r = pci_enable_device(pdev); if (r) - DRM_WARN("pci_enable_device failed (%d)\n", r); + dev_warn(&pdev->dev, "pci_enable_device failed (%d)\n", + r); amdgpu_device_resume(dev, true); dev->switch_power_state = DRM_SWITCH_POWER_ON; } else { - pr_info("switched off\n"); + dev_info(&pdev->dev, "switched off\n"); dev->switch_power_state = DRM_SWITCH_POWER_CHANGING; amdgpu_device_prepare(dev); amdgpu_device_suspend(dev, true); @@ -2274,8 +2285,9 @@ int amdgpu_device_ip_set_clockgating_state(void *dev, r = adev->ip_blocks[i].version->funcs->set_clockgating_state( &adev->ip_blocks[i], state); if (r) - DRM_ERROR("set_clockgating_state of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + dev_err(adev->dev, + "set_clockgating_state of IP block <%s> failed %d\n", + adev->ip_blocks[i].version->funcs->name, r); } return r; } @@ -2308,8 +2320,9 @@ int amdgpu_device_ip_set_powergating_state(void *dev, r = adev->ip_blocks[i].version->funcs->set_powergating_state( &adev->ip_blocks[i], state); if (r) - DRM_ERROR("set_powergating_state of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + dev_err(adev->dev, + "set_powergating_state of IP block <%s> failed %d\n", + adev->ip_blocks[i].version->funcs->name, r); } return r; } @@ -2525,9 +2538,11 @@ static void amdgpu_device_enable_virtual_display(struct amdgpu_device *adev) } } - DRM_INFO("virtual display string:%s, %s:virtual_display:%d, num_crtc:%d\n", - amdgpu_virtual_display, pci_address_name, - adev->enable_virtual_display, adev->mode_info.num_crtc); + dev_info( + adev->dev, + "virtual display string:%s, %s:virtual_display:%d, num_crtc:%d\n", + amdgpu_virtual_display, pci_address_name, + adev->enable_virtual_display, adev->mode_info.num_crtc); kfree(pciaddstr); } @@ -2538,8 +2553,9 @@ void amdgpu_device_set_sriov_virtual_display(struct amdgpu_device *adev) if (amdgpu_sriov_vf(adev) && !adev->enable_virtual_display) { adev->mode_info.num_crtc = 1; adev->enable_virtual_display = true; - DRM_INFO("virtual_display:%d, num_crtc:%d\n", - adev->enable_virtual_display, adev->mode_info.num_crtc); + dev_info(adev->dev, "virtual_display:%d, num_crtc:%d\n", + adev->enable_virtual_display, + adev->mode_info.num_crtc); } } @@ -2773,21 +2789,29 @@ static int amdgpu_device_ip_early_init(struct amdgpu_device *adev) if (!amdgpu_device_pcie_dynamic_switching_supported(adev)) adev->pm.pp_feature &= ~PP_PCIE_DPM_MASK; + adev->virt.is_xgmi_node_migrate_enabled = false; + if (amdgpu_sriov_vf(adev)) { + adev->virt.is_xgmi_node_migrate_enabled = + amdgpu_ip_version((adev), GC_HWIP, 0) == IP_VERSION(9, 4, 4); + } + total = true; for (i = 0; i < adev->num_ip_blocks; i++) { ip_block = &adev->ip_blocks[i]; if ((amdgpu_ip_block_mask & (1 << i)) == 0) { - DRM_WARN("disabled ip block: %d <%s>\n", - i, adev->ip_blocks[i].version->funcs->name); + dev_warn(adev->dev, "disabled ip block: %d <%s>\n", i, + adev->ip_blocks[i].version->funcs->name); adev->ip_blocks[i].status.valid = false; } else if (ip_block->version->funcs->early_init) { r = ip_block->version->funcs->early_init(ip_block); if (r == -ENOENT) { adev->ip_blocks[i].status.valid = false; } else if (r) { - DRM_ERROR("early_init of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + dev_err(adev->dev, + "early_init of IP block <%s> failed %d\n", + adev->ip_blocks[i].version->funcs->name, + r); total = false; } else { adev->ip_blocks[i].status.valid = true; @@ -2868,8 +2892,10 @@ static int amdgpu_device_ip_hw_init_phase1(struct amdgpu_device *adev) adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_IH) { r = adev->ip_blocks[i].version->funcs->hw_init(&adev->ip_blocks[i]); if (r) { - DRM_ERROR("hw_init of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + dev_err(adev->dev, + "hw_init of IP block <%s> failed %d\n", + adev->ip_blocks[i].version->funcs->name, + r); return r; } adev->ip_blocks[i].status.hw = true; @@ -2893,8 +2919,9 @@ static int amdgpu_device_ip_hw_init_phase2(struct amdgpu_device *adev) continue; r = adev->ip_blocks[i].version->funcs->hw_init(&adev->ip_blocks[i]); if (r) { - DRM_ERROR("hw_init of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + dev_err(adev->dev, + "hw_init of IP block <%s> failed %d\n", + adev->ip_blocks[i].version->funcs->name, r); return r; } adev->ip_blocks[i].status.hw = true; @@ -2932,8 +2959,11 @@ static int amdgpu_device_fw_loading(struct amdgpu_device *adev) } else { r = adev->ip_blocks[i].version->funcs->hw_init(&adev->ip_blocks[i]); if (r) { - DRM_ERROR("hw_init of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + dev_err(adev->dev, + "hw_init of IP block <%s> failed %d\n", + adev->ip_blocks[i] + .version->funcs->name, + r); return r; } adev->ip_blocks[i].status.hw = true; @@ -2988,25 +3018,29 @@ static int amdgpu_device_init_schedulers(struct amdgpu_device *adev) r = drm_sched_init(&ring->sched, &args); if (r) { - DRM_ERROR("Failed to create scheduler on ring %s.\n", - ring->name); + dev_err(adev->dev, + "Failed to create scheduler on ring %s.\n", + ring->name); return r; } r = amdgpu_uvd_entity_init(adev, ring); if (r) { - DRM_ERROR("Failed to create UVD scheduling entity on ring %s.\n", - ring->name); + dev_err(adev->dev, + "Failed to create UVD scheduling entity on ring %s.\n", + ring->name); return r; } r = amdgpu_vce_entity_init(adev, ring); if (r) { - DRM_ERROR("Failed to create VCE scheduling entity on ring %s.\n", - ring->name); + dev_err(adev->dev, + "Failed to create VCE scheduling entity on ring %s.\n", + ring->name); return r; } } - amdgpu_xcp_update_partition_sched_list(adev); + if (adev->xcp_mgr) + amdgpu_xcp_update_partition_sched_list(adev); return 0; } @@ -3038,8 +3072,10 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev) if (adev->ip_blocks[i].version->funcs->sw_init) { r = adev->ip_blocks[i].version->funcs->sw_init(&adev->ip_blocks[i]); if (r) { - DRM_ERROR("sw_init of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + dev_err(adev->dev, + "sw_init of IP block <%s> failed %d\n", + adev->ip_blocks[i].version->funcs->name, + r); goto init_failed; } } @@ -3053,7 +3089,8 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev) /* need to do common hw init early so everything is set up for gmc */ r = adev->ip_blocks[i].version->funcs->hw_init(&adev->ip_blocks[i]); if (r) { - DRM_ERROR("hw_init %d failed %d\n", i, r); + dev_err(adev->dev, "hw_init %d failed %d\n", i, + r); goto init_failed; } adev->ip_blocks[i].status.hw = true; @@ -3065,17 +3102,21 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev) r = amdgpu_device_mem_scratch_init(adev); if (r) { - DRM_ERROR("amdgpu_mem_scratch_init failed %d\n", r); + dev_err(adev->dev, + "amdgpu_mem_scratch_init failed %d\n", + r); goto init_failed; } r = adev->ip_blocks[i].version->funcs->hw_init(&adev->ip_blocks[i]); if (r) { - DRM_ERROR("hw_init %d failed %d\n", i, r); + dev_err(adev->dev, "hw_init %d failed %d\n", i, + r); goto init_failed; } r = amdgpu_device_wb_init(adev); if (r) { - DRM_ERROR("amdgpu_device_wb_init failed %d\n", r); + dev_err(adev->dev, + "amdgpu_device_wb_init failed %d\n", r); goto init_failed; } adev->ip_blocks[i].status.hw = true; @@ -3087,14 +3128,16 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev) AMDGPU_GEM_DOMAIN_GTT, AMDGPU_CSA_SIZE); if (r) { - DRM_ERROR("allocate CSA failed %d\n", r); + dev_err(adev->dev, + "allocate CSA failed %d\n", r); goto init_failed; } } r = amdgpu_seq64_init(adev); if (r) { - DRM_ERROR("allocate seq64 failed %d\n", r); + dev_err(adev->dev, "allocate seq64 failed %d\n", + r); goto init_failed; } } @@ -3284,8 +3327,10 @@ int amdgpu_device_set_cg_state(struct amdgpu_device *adev, r = adev->ip_blocks[i].version->funcs->set_clockgating_state(&adev->ip_blocks[i], state); if (r) { - DRM_ERROR("set_clockgating_state(gate) of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + dev_err(adev->dev, + "set_clockgating_state(gate) of IP block <%s> failed %d\n", + adev->ip_blocks[i].version->funcs->name, + r); return r; } } @@ -3321,8 +3366,10 @@ int amdgpu_device_set_pg_state(struct amdgpu_device *adev, r = adev->ip_blocks[i].version->funcs->set_powergating_state(&adev->ip_blocks[i], state); if (r) { - DRM_ERROR("set_powergating_state(gate) of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + dev_err(adev->dev, + "set_powergating_state(gate) of IP block <%s> failed %d\n", + adev->ip_blocks[i].version->funcs->name, + r); return r; } } @@ -3388,8 +3435,10 @@ static int amdgpu_device_ip_late_init(struct amdgpu_device *adev) if (adev->ip_blocks[i].version->funcs->late_init) { r = adev->ip_blocks[i].version->funcs->late_init(&adev->ip_blocks[i]); if (r) { - DRM_ERROR("late_init of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + dev_err(adev->dev, + "late_init of IP block <%s> failed %d\n", + adev->ip_blocks[i].version->funcs->name, + r); return r; } } @@ -3398,7 +3447,7 @@ static int amdgpu_device_ip_late_init(struct amdgpu_device *adev) r = amdgpu_ras_late_init(adev); if (r) { - DRM_ERROR("amdgpu_ras_late_init failed %d", r); + dev_err(adev->dev, "amdgpu_ras_late_init failed %d", r); return r; } @@ -3412,7 +3461,7 @@ static int amdgpu_device_ip_late_init(struct amdgpu_device *adev) r = amdgpu_device_enable_mgpu_fan_boost(); if (r) - DRM_ERROR("enable mgpu fan boost failed (%d).\n", r); + dev_err(adev->dev, "enable mgpu fan boost failed (%d).\n", r); /* For passthrough configuration on arcturus and aldebaran, enable special handling SBR */ if (amdgpu_passthrough(adev) && @@ -3445,7 +3494,9 @@ static int amdgpu_device_ip_late_init(struct amdgpu_device *adev) r = amdgpu_xgmi_set_pstate(gpu_instance->adev, AMDGPU_XGMI_PSTATE_MIN); if (r) { - DRM_ERROR("pstate setting failed (%d).\n", r); + dev_err(adev->dev, + "pstate setting failed (%d).\n", + r); break; } } @@ -3459,17 +3510,19 @@ static int amdgpu_device_ip_late_init(struct amdgpu_device *adev) static void amdgpu_ip_block_hw_fini(struct amdgpu_ip_block *ip_block) { + struct amdgpu_device *adev = ip_block->adev; int r; if (!ip_block->version->funcs->hw_fini) { - DRM_ERROR("hw_fini of IP block <%s> not defined\n", - ip_block->version->funcs->name); + dev_err(adev->dev, "hw_fini of IP block <%s> not defined\n", + ip_block->version->funcs->name); } else { r = ip_block->version->funcs->hw_fini(ip_block); /* XXX handle errors */ if (r) { - DRM_DEBUG("hw_fini of IP block <%s> failed %d\n", - ip_block->version->funcs->name, r); + dev_dbg(adev->dev, + "hw_fini of IP block <%s> failed %d\n", + ip_block->version->funcs->name, r); } } @@ -3510,15 +3563,16 @@ static int amdgpu_device_ip_fini_early(struct amdgpu_device *adev) r = adev->ip_blocks[i].version->funcs->early_fini(&adev->ip_blocks[i]); if (r) { - DRM_DEBUG("early_fini of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + dev_dbg(adev->dev, + "early_fini of IP block <%s> failed %d\n", + adev->ip_blocks[i].version->funcs->name, r); } } amdgpu_device_set_pg_state(adev, AMD_PG_STATE_UNGATE); amdgpu_device_set_cg_state(adev, AMD_CG_STATE_UNGATE); - amdgpu_amdkfd_suspend(adev, false); + amdgpu_amdkfd_suspend(adev, true); amdgpu_userq_suspend(adev); /* Workaround for ASICs need to disable SMC first */ @@ -3533,7 +3587,8 @@ static int amdgpu_device_ip_fini_early(struct amdgpu_device *adev) if (amdgpu_sriov_vf(adev)) { if (amdgpu_virt_release_full_gpu(adev, false)) - DRM_ERROR("failed to release exclusive mode on fini\n"); + dev_err(adev->dev, + "failed to release exclusive mode on fini\n"); } return 0; @@ -3581,8 +3636,10 @@ static int amdgpu_device_ip_fini(struct amdgpu_device *adev) r = adev->ip_blocks[i].version->funcs->sw_fini(&adev->ip_blocks[i]); /* XXX handle errors */ if (r) { - DRM_DEBUG("sw_fini of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + dev_dbg(adev->dev, + "sw_fini of IP block <%s> failed %d\n", + adev->ip_blocks[i].version->funcs->name, + r); } } adev->ip_blocks[i].status.sw = false; @@ -3615,7 +3672,7 @@ static void amdgpu_device_delayed_init_work_handler(struct work_struct *work) r = amdgpu_ib_ring_tests(adev); if (r) - DRM_ERROR("ib ring test failed (%d).\n", r); + dev_err(adev->dev, "ib ring test failed (%d).\n", r); } static void amdgpu_device_delay_enable_gfx_off(struct work_struct *work) @@ -3756,8 +3813,9 @@ static int amdgpu_device_ip_suspend_phase2(struct amdgpu_device *adev) if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_SMC) { r = amdgpu_dpm_set_mp1_state(adev, adev->mp1_state); if (r) { - DRM_ERROR("SMC failed to set mp1 state %d, %d\n", - adev->mp1_state, r); + dev_err(adev->dev, + "SMC failed to set mp1 state %d, %d\n", + adev->mp1_state, r); return r; } } @@ -4041,12 +4099,14 @@ static void amdgpu_device_detect_sriov_bios(struct amdgpu_device *adev) /** * amdgpu_device_asic_has_dc_support - determine if DC supports the asic * + * @pdev : pci device context * @asic_type: AMD asic type * * Check if there is DC (new modesetting infrastructre) support for an asic. * returns true if DC has support, false if not. */ -bool amdgpu_device_asic_has_dc_support(enum amd_asic_type asic_type) +bool amdgpu_device_asic_has_dc_support(struct pci_dev *pdev, + enum amd_asic_type asic_type) { switch (asic_type) { #ifdef CONFIG_DRM_AMDGPU_SI @@ -4089,7 +4149,9 @@ bool amdgpu_device_asic_has_dc_support(enum amd_asic_type asic_type) #else default: if (amdgpu_dc > 0) - DRM_INFO_ONCE("Display Core has been requested via kernel parameter but isn't supported by ASIC, ignoring\n"); + dev_info_once( + &pdev->dev, + "Display Core has been requested via kernel parameter but isn't supported by ASIC, ignoring\n"); return false; #endif } @@ -4108,7 +4170,7 @@ bool amdgpu_device_has_dc_support(struct amdgpu_device *adev) (adev->harvest_ip_mask & AMD_HARVEST_IP_DMU_MASK)) return false; - return amdgpu_device_asic_has_dc_support(adev->asic_type); + return amdgpu_device_asic_has_dc_support(adev->pdev, adev->asic_type); } static void amdgpu_device_xgmi_reset_func(struct work_struct *__work) @@ -4150,7 +4212,8 @@ static void amdgpu_device_xgmi_reset_func(struct work_struct *__work) fail: if (adev->asic_reset_res) - DRM_WARN("ASIC reset failed with error, %d for drm dev, %s", + dev_warn(adev->dev, + "ASIC reset failed with error, %d for drm dev, %s", adev->asic_reset_res, adev_to_drm(adev)->unique); amdgpu_put_xgmi_hive(hive); } @@ -4274,7 +4337,7 @@ static void amdgpu_device_set_mcbp(struct amdgpu_device *adev) adev->gfx.mcbp = true; if (adev->gfx.mcbp) - DRM_INFO("MCBP is enabled\n"); + dev_info(adev->dev, "MCBP is enabled\n"); } /** @@ -4342,9 +4405,11 @@ int amdgpu_device_init(struct amdgpu_device *adev, adev->audio_endpt_rreg = &amdgpu_block_invalid_rreg; adev->audio_endpt_wreg = &amdgpu_block_invalid_wreg; - DRM_INFO("initializing kernel modesetting (%s 0x%04X:0x%04X 0x%04X:0x%04X 0x%02X).\n", - amdgpu_asic_name[adev->asic_type], pdev->vendor, pdev->device, - pdev->subsystem_vendor, pdev->subsystem_device, pdev->revision); + dev_info( + adev->dev, + "initializing kernel modesetting (%s 0x%04X:0x%04X 0x%04X:0x%04X 0x%02X).\n", + amdgpu_asic_name[adev->asic_type], pdev->vendor, pdev->device, + pdev->subsystem_vendor, pdev->subsystem_device, pdev->revision); /* mutex initialization are all done here so we * can recall function without having locking issues @@ -4461,8 +4526,10 @@ int amdgpu_device_init(struct amdgpu_device *adev, if (!adev->rmmio) return -ENOMEM; - DRM_INFO("register mmio base: 0x%08X\n", (uint32_t)adev->rmmio_base); - DRM_INFO("register mmio size: %u\n", (unsigned int)adev->rmmio_size); + dev_info(adev->dev, "register mmio base: 0x%08X\n", + (uint32_t)adev->rmmio_base); + dev_info(adev->dev, "register mmio size: %u\n", + (unsigned int)adev->rmmio_size); /* * Reset domain needs to be present early, before XGMI hive discovered @@ -4599,7 +4666,7 @@ int amdgpu_device_init(struct amdgpu_device *adev, r = -EINVAL; goto failed; } - DRM_INFO("GPU posting now...\n"); + dev_info(adev->dev, "GPU posting now...\n"); r = amdgpu_device_asic_init(adev); if (r) { dev_err(adev->dev, "gpu post error!\n"); @@ -4709,12 +4776,12 @@ fence_driver_init: r = amdgpu_pm_sysfs_init(adev); if (r) - DRM_ERROR("registering pm sysfs failed (%d).\n", r); + dev_err(adev->dev, "registering pm sysfs failed (%d).\n", r); r = amdgpu_ucode_sysfs_init(adev); if (r) { adev->ucode_sysfs_en = false; - DRM_ERROR("Creating firmware sysfs failed (%d).\n", r); + dev_err(adev->dev, "Creating firmware sysfs failed (%d).\n", r); } else adev->ucode_sysfs_en = true; @@ -4963,7 +5030,7 @@ static int amdgpu_device_evict_resources(struct amdgpu_device *adev) ret = amdgpu_ttm_evict_resources(adev, TTM_PL_VRAM); if (ret) - DRM_WARN("evicting device resources failed\n"); + dev_warn(adev->dev, "evicting device resources failed\n"); return ret; } @@ -5034,6 +5101,28 @@ int amdgpu_device_prepare(struct drm_device *dev) return 0; } +/** + * amdgpu_device_complete - complete power state transition + * + * @dev: drm dev pointer + * + * Undo the changes from amdgpu_device_prepare. This will be + * called on all resume transitions, including those that failed. + */ +void amdgpu_device_complete(struct drm_device *dev) +{ + struct amdgpu_device *adev = drm_to_adev(dev); + int i; + + for (i = 0; i < adev->num_ip_blocks; i++) { + if (!adev->ip_blocks[i].status.valid) + continue; + if (!adev->ip_blocks[i].version->funcs->complete) + continue; + adev->ip_blocks[i].version->funcs->complete(&adev->ip_blocks[i]); + } +} + /** * amdgpu_device_suspend - initiate device suspend * @@ -5055,6 +5144,8 @@ int amdgpu_device_suspend(struct drm_device *dev, bool notify_clients) adev->in_suspend = true; if (amdgpu_sriov_vf(adev)) { + if (!adev->in_s0ix && !adev->in_runpm) + amdgpu_amdkfd_suspend_process(adev); amdgpu_virt_fini_data_exchange(adev); r = amdgpu_virt_request_full_gpu(adev, false); if (r) @@ -5062,7 +5153,7 @@ int amdgpu_device_suspend(struct drm_device *dev, bool notify_clients) } if (amdgpu_acpi_smart_shift_update(dev, AMDGPU_SS_DEV_D3)) - DRM_WARN("smart shift update failed\n"); + dev_warn(adev->dev, "smart shift update failed\n"); if (notify_clients) drm_client_dev_suspend(adev_to_drm(adev), false); @@ -5074,7 +5165,7 @@ int amdgpu_device_suspend(struct drm_device *dev, bool notify_clients) amdgpu_device_ip_suspend_phase1(adev); if (!adev->in_s0ix) { - amdgpu_amdkfd_suspend(adev, adev->in_runpm); + amdgpu_amdkfd_suspend(adev, !amdgpu_sriov_vf(adev) && !adev->in_runpm); amdgpu_userq_suspend(adev); } @@ -5098,6 +5189,32 @@ int amdgpu_device_suspend(struct drm_device *dev, bool notify_clients) return 0; } +static inline int amdgpu_virt_resume(struct amdgpu_device *adev) +{ + int r; + unsigned int prev_physical_node_id = adev->gmc.xgmi.physical_node_id; + + /* During VM resume, QEMU programming of VF MSIX table (register GFXMSIX_VECT0_ADDR_LO) + * may not work. The access could be blocked by nBIF protection as VF isn't in + * exclusive access mode. Exclusive access is enabled now, disable/enable MSIX + * so that QEMU reprograms MSIX table. + */ + amdgpu_restore_msix(adev); + + r = adev->gfxhub.funcs->get_xgmi_info(adev); + if (r) + return r; + + dev_info(adev->dev, "xgmi node, old id %d, new id %d\n", + prev_physical_node_id, adev->gmc.xgmi.physical_node_id); + + adev->vm_manager.vram_base_offset = adev->gfxhub.funcs->get_mc_fb_offset(adev); + adev->vm_manager.vram_base_offset += + adev->gmc.xgmi.physical_node_id * adev->gmc.xgmi.node_segment_size; + + return 0; +} + /** * amdgpu_device_resume - initiate device resume * @@ -5119,6 +5236,12 @@ int amdgpu_device_resume(struct drm_device *dev, bool notify_clients) return r; } + if (amdgpu_virt_xgmi_migrate_enabled(adev)) { + r = amdgpu_virt_resume(adev); + if (r) + goto exit; + } + if (dev->switch_power_state == DRM_SWITCH_POWER_OFF) return 0; @@ -5140,7 +5263,7 @@ int amdgpu_device_resume(struct drm_device *dev, bool notify_clients) } if (!adev->in_s0ix) { - r = amdgpu_amdkfd_resume(adev, adev->in_runpm); + r = amdgpu_amdkfd_resume(adev, !amdgpu_sriov_vf(adev) && !adev->in_runpm); if (r) goto exit; @@ -5159,6 +5282,9 @@ exit: if (amdgpu_sriov_vf(adev)) { amdgpu_virt_init_data_exchange(adev); amdgpu_virt_release_full_gpu(adev, true); + + if (!adev->in_s0ix && !r && !adev->in_runpm) + r = amdgpu_amdkfd_resume_process(adev); } if (r) @@ -5196,7 +5322,7 @@ exit: adev->in_suspend = false; if (amdgpu_acpi_smart_shift_update(dev, AMDGPU_SS_DEV_D0)) - DRM_WARN("smart shift update failed\n"); + dev_warn(adev->dev, "smart shift update failed\n"); return 0; } @@ -5727,7 +5853,9 @@ int amdgpu_device_reinit_after_reset(struct amdgpu_reset_context *reset_context) amdgpu_coredump(tmp_adev, false, vram_lost, reset_context->job); if (vram_lost) { - DRM_INFO("VRAM is lost due to GPU reset!\n"); + dev_info( + tmp_adev->dev, + "VRAM is lost due to GPU reset!\n"); amdgpu_inc_vram_lost(tmp_adev); } @@ -6006,29 +6134,20 @@ static int amdgpu_device_health_check(struct list_head *device_list_handle) { struct amdgpu_device *tmp_adev; int ret = 0; - u32 status; list_for_each_entry(tmp_adev, device_list_handle, reset_list) { - pci_read_config_dword(tmp_adev->pdev, PCI_COMMAND, &status); - if (PCI_POSSIBLE_ERROR(status)) { - dev_err(tmp_adev->dev, "device lost from bus!"); - ret = -ENODEV; - } + ret |= amdgpu_device_bus_status_check(tmp_adev); } return ret; } -static int amdgpu_device_halt_activities(struct amdgpu_device *adev, - struct amdgpu_job *job, - struct amdgpu_reset_context *reset_context, - struct list_head *device_list, - struct amdgpu_hive_info *hive, - bool need_emergency_restart) +static int amdgpu_device_recovery_prepare(struct amdgpu_device *adev, + struct list_head *device_list, + struct amdgpu_hive_info *hive) { - struct list_head *device_list_handle = NULL; struct amdgpu_device *tmp_adev = NULL; - int i, r = 0; + int r; /* * Build list of devices to reset. @@ -6045,26 +6164,55 @@ static int amdgpu_device_halt_activities(struct amdgpu_device *adev, } if (!list_is_first(&adev->reset_list, device_list)) list_rotate_to_front(&adev->reset_list, device_list); - device_list_handle = device_list; } else { list_add_tail(&adev->reset_list, device_list); - device_list_handle = device_list; } if (!amdgpu_sriov_vf(adev) && (!adev->pcie_reset_ctx.occurs_dpc)) { - r = amdgpu_device_health_check(device_list_handle); + r = amdgpu_device_health_check(device_list); if (r) return r; } - /* We need to lock reset domain only once both for XGMI and single device */ - tmp_adev = list_first_entry(device_list_handle, struct amdgpu_device, - reset_list); + return 0; +} + +static void amdgpu_device_recovery_get_reset_lock(struct amdgpu_device *adev, + struct list_head *device_list) +{ + struct amdgpu_device *tmp_adev = NULL; + + if (list_empty(device_list)) + return; + tmp_adev = + list_first_entry(device_list, struct amdgpu_device, reset_list); amdgpu_device_lock_reset_domain(tmp_adev->reset_domain); +} - /* block all schedulers and reset given job's ring */ - list_for_each_entry(tmp_adev, device_list_handle, reset_list) { +static void amdgpu_device_recovery_put_reset_lock(struct amdgpu_device *adev, + struct list_head *device_list) +{ + struct amdgpu_device *tmp_adev = NULL; + + if (list_empty(device_list)) + return; + tmp_adev = + list_first_entry(device_list, struct amdgpu_device, reset_list); + amdgpu_device_unlock_reset_domain(tmp_adev->reset_domain); +} +static void amdgpu_device_halt_activities(struct amdgpu_device *adev, + struct amdgpu_job *job, + struct amdgpu_reset_context *reset_context, + struct list_head *device_list, + struct amdgpu_hive_info *hive, + bool need_emergency_restart) +{ + struct amdgpu_device *tmp_adev = NULL; + int i; + + /* block all schedulers and reset given job's ring */ + list_for_each_entry(tmp_adev, device_list, reset_list) { amdgpu_device_set_mp1_state(tmp_adev); /* @@ -6113,8 +6261,6 @@ static int amdgpu_device_halt_activities(struct amdgpu_device *adev, } atomic_inc(&tmp_adev->gpu_reset_counter); } - - return r; } static int amdgpu_device_asic_reset(struct amdgpu_device *adev, @@ -6220,7 +6366,8 @@ static int amdgpu_device_sched_resume(struct list_head *device_list, } else { dev_info(tmp_adev->dev, "GPU reset(%d) succeeded!\n", atomic_read(&tmp_adev->gpu_reset_counter)); if (amdgpu_acpi_smart_shift_update(adev_to_drm(tmp_adev), AMDGPU_SS_DEV_D0)) - DRM_WARN("smart shift update failed\n"); + dev_warn(tmp_adev->dev, + "smart shift update failed\n"); } } @@ -6252,11 +6399,6 @@ static void amdgpu_device_gpu_resume(struct amdgpu_device *adev, amdgpu_ras_set_error_query_ready(tmp_adev, true); } - - tmp_adev = list_first_entry(device_list, struct amdgpu_device, - reset_list); - amdgpu_device_unlock_reset_domain(tmp_adev->reset_domain); - } @@ -6306,7 +6448,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, */ if (need_emergency_restart && amdgpu_ras_get_context(adev) && amdgpu_ras_get_context(adev)->reboot) { - DRM_WARN("Emergency reboot."); + dev_warn(adev->dev, "Emergency reboot."); ksys_sync_helper(); emergency_restart(); @@ -6324,11 +6466,14 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, reset_context->hive = hive; INIT_LIST_HEAD(&device_list); - r = amdgpu_device_halt_activities(adev, job, reset_context, &device_list, - hive, need_emergency_restart); - if (r) + if (amdgpu_device_recovery_prepare(adev, &device_list, hive)) goto end_reset; + /* We need to lock reset domain only once both for XGMI and single device */ + amdgpu_device_recovery_get_reset_lock(adev, &device_list); + + amdgpu_device_halt_activities(adev, job, reset_context, &device_list, + hive, need_emergency_restart); if (need_emergency_restart) goto skip_sched_resume; /* @@ -6337,7 +6482,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, * * job->base holds a reference to parent fence */ - if (job && dma_fence_is_signaled(&job->hw_fence)) { + if (job && dma_fence_is_signaled(&job->hw_fence.base)) { job_signaled = true; dev_info(adev->dev, "Guilty job already signaled, skipping HW reset"); goto skip_hw_reset; @@ -6345,13 +6490,15 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, r = amdgpu_device_asic_reset(adev, &device_list, reset_context); if (r) - goto end_reset; + goto reset_unlock; skip_hw_reset: r = amdgpu_device_sched_resume(&device_list, reset_context, job_signaled); if (r) - goto end_reset; + goto reset_unlock; skip_sched_resume: amdgpu_device_gpu_resume(adev, &device_list, need_emergency_restart); +reset_unlock: + amdgpu_device_recovery_put_reset_lock(adev, &device_list); end_reset: if (hive) { mutex_unlock(&hive->hive_lock); @@ -6747,7 +6894,6 @@ pci_ers_result_t amdgpu_pci_error_detected(struct pci_dev *pdev, pci_channel_sta struct amdgpu_hive_info *hive = amdgpu_get_xgmi_hive(adev); struct amdgpu_reset_context reset_context; struct list_head device_list; - int r = 0; dev_info(adev->dev, "PCI error: detected callback!!\n"); @@ -6772,14 +6918,14 @@ pci_ers_result_t amdgpu_pci_error_detected(struct pci_dev *pdev, pci_channel_sta memset(&reset_context, 0, sizeof(reset_context)); INIT_LIST_HEAD(&device_list); - r = amdgpu_device_halt_activities(adev, NULL, &reset_context, &device_list, - hive, false); + amdgpu_device_recovery_prepare(adev, &device_list, hive); + amdgpu_device_recovery_get_reset_lock(adev, &device_list); + amdgpu_device_halt_activities(adev, NULL, &reset_context, &device_list, + hive, false); if (hive) { mutex_unlock(&hive->hive_lock); amdgpu_put_xgmi_hive(hive); } - if (r) - return PCI_ERS_RESULT_DISCONNECT; return PCI_ERS_RESULT_NEED_RESET; case pci_channel_io_perm_failure: /* Permanent error, prepare for device removal */ @@ -6889,8 +7035,8 @@ out: if (hive) { list_for_each_entry(tmp_adev, &device_list, reset_list) amdgpu_device_unset_mp1_state(tmp_adev); - amdgpu_device_unlock_reset_domain(adev->reset_domain); } + amdgpu_device_recovery_put_reset_lock(adev, &device_list); } if (hive) { @@ -6936,6 +7082,7 @@ void amdgpu_pci_resume(struct pci_dev *pdev) amdgpu_device_sched_resume(&device_list, NULL, NULL); amdgpu_device_gpu_resume(adev, &device_list, false); + amdgpu_device_recovery_put_reset_lock(adev, &device_list); adev->pcie_reset_ctx.occurs_dpc = false; if (hive) { @@ -6960,11 +7107,11 @@ bool amdgpu_device_cache_pci_state(struct pci_dev *pdev) adev->pci_state = pci_store_saved_state(pdev); if (!adev->pci_state) { - DRM_ERROR("Failed to store PCI saved state"); + dev_err(adev->dev, "Failed to store PCI saved state"); return false; } } else { - DRM_WARN("Failed to save PCI state, err:%d\n", r); + dev_warn(adev->dev, "Failed to save PCI state, err:%d\n", r); return false; } @@ -6985,7 +7132,7 @@ bool amdgpu_device_load_pci_state(struct pci_dev *pdev) if (!r) { pci_restore_state(pdev); } else { - DRM_WARN("Failed to load PCI state, err:%d\n", r); + dev_warn(adev->dev, "Failed to load PCI state, err:%d\n", r); return false; } @@ -7231,7 +7378,7 @@ struct dma_fence *amdgpu_device_enforce_isolation(struct amdgpu_device *adev, dep = amdgpu_sync_peek_fence(&isolation->prev, ring); r = amdgpu_sync_fence(&isolation->active, &f->finished, GFP_NOWAIT); if (r) - DRM_WARN("OOM tracking isolation\n"); + dev_warn(adev->dev, "OOM tracking isolation\n"); out_grab_ref: dma_fence_get(dep); @@ -7299,9 +7446,11 @@ uint32_t amdgpu_device_wait_on_rreg(struct amdgpu_device *adev, tmp_ = RREG32(reg_addr); loop--; if (!loop) { - DRM_WARN("Register(%d) [%s] failed to reach value 0x%08x != 0x%08xn", - inst, reg_name, (uint32_t)expected_value, - (uint32_t)(tmp_ & (mask))); + dev_warn( + adev->dev, + "Register(%d) [%s] failed to reach value 0x%08x != 0x%08xn", + inst, reg_name, (uint32_t)expected_value, + (uint32_t)(tmp_ & (mask))); ret = -ETIMEDOUT; break; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c index a0e9bf9b27108c372623e10292dcdcabea69f1de..81b3443c8d7f48392ccf14d4c9cf3961a6fdd239 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c @@ -321,10 +321,12 @@ static int amdgpu_discovery_read_binary_from_file(struct amdgpu_device *adev, const struct firmware *fw; int r; - r = request_firmware(&fw, fw_name, adev->dev); + r = firmware_request_nowarn(&fw, fw_name, adev->dev); if (r) { - dev_err(adev->dev, "can't load firmware \"%s\"\n", - fw_name); + if (amdgpu_discovery == 2) + dev_err(adev->dev, "can't load firmware \"%s\"\n", fw_name); + else + drm_info(&adev->ddev, "Optional firmware \"%s\" was not found\n", fw_name); return r; } @@ -459,16 +461,12 @@ static int amdgpu_discovery_init(struct amdgpu_device *adev) /* Read from file if it is the preferred option */ fw_name = amdgpu_discovery_get_fw_name(adev); if (fw_name != NULL) { - dev_info(adev->dev, "use ip discovery information from file"); + drm_dbg(&adev->ddev, "use ip discovery information from file"); r = amdgpu_discovery_read_binary_from_file(adev, adev->mman.discovery_bin, fw_name); - - if (r) { - dev_err(adev->dev, "failed to read ip discovery binary from file\n"); - r = -EINVAL; + if (r) goto out; - } - } else { + drm_dbg(&adev->ddev, "use ip discovery information from memory"); r = amdgpu_discovery_read_binary_from_mem( adev, adev->mman.discovery_bin); if (r) @@ -1338,10 +1336,8 @@ static int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev) int r; r = amdgpu_discovery_init(adev); - if (r) { - DRM_ERROR("amdgpu_discovery_init failed\n"); + if (r) return r; - } wafl_ver = 0; adev->gfx.xcc_mask = 0; @@ -2579,8 +2575,10 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev) break; default: r = amdgpu_discovery_reg_base_init(adev); - if (r) - return -EINVAL; + if (r) { + drm_err(&adev->ddev, "discovery failed: %d\n", r); + return r; + } amdgpu_discovery_harvest_ip(adev); amdgpu_discovery_get_gfx_info(adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c index 35c778426a7c704f37dfdc8533dc61d79e3bf9e3..9e463d3ee92779a8ece4836ee682b44b4a0d312e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c @@ -1317,7 +1317,7 @@ amdgpu_display_user_framebuffer_create(struct drm_device *dev, /* Handle is imported dma-buf, so cannot be migrated to VRAM for scanout */ bo = gem_to_amdgpu_bo(obj); domains = amdgpu_display_supported_domains(drm_to_adev(dev), bo->flags); - if (obj->import_attach && !(domains & AMDGPU_GEM_DOMAIN_GTT)) { + if (drm_gem_is_imported(obj) && !(domains & AMDGPU_GEM_DOMAIN_GTT)) { drm_dbg_kms(dev, "Cannot create framebuffer from imported dma_buf\n"); drm_gem_object_put(obj); return ERR_PTR(-EINVAL); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c index 44e120f9f76497dd6045e6dea0cda8a03e6ec087..ff98c87b2e0b91665fb6a9b9c0045df4f9ada1fd 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c @@ -513,8 +513,8 @@ bool amdgpu_dmabuf_is_xgmi_accessible(struct amdgpu_device *adev, if (!adev) return false; - if (obj->import_attach) { - struct dma_buf *dma_buf = obj->import_attach->dmabuf; + if (drm_gem_is_imported(obj)) { + struct dma_buf *dma_buf = obj->dma_buf; if (dma_buf->ops != &amdgpu_dmabuf_ops) /* No XGMI with non AMD GPUs */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell_mgr.c index 3f3662e8b871032c21e1c02e308b78515cfc05c3..3040437d99c2390806dd025cb0119d459da2ae1e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell_mgr.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell_mgr.c @@ -41,7 +41,8 @@ u32 amdgpu_mm_rdoorbell(struct amdgpu_device *adev, u32 index) if (index < adev->doorbell.num_kernel_doorbells) return readl(adev->doorbell.cpu_addr + index); - DRM_ERROR("reading beyond doorbell aperture: 0x%08x!\n", index); + dev_err(adev->dev, "reading beyond doorbell aperture: 0x%08x!\n", + index); return 0; } @@ -63,7 +64,8 @@ void amdgpu_mm_wdoorbell(struct amdgpu_device *adev, u32 index, u32 v) if (index < adev->doorbell.num_kernel_doorbells) writel(v, adev->doorbell.cpu_addr + index); else - DRM_ERROR("writing beyond doorbell aperture: 0x%08x!\n", index); + dev_err(adev->dev, + "writing beyond doorbell aperture: 0x%08x!\n", index); } /** @@ -83,7 +85,8 @@ u64 amdgpu_mm_rdoorbell64(struct amdgpu_device *adev, u32 index) if (index < adev->doorbell.num_kernel_doorbells) return atomic64_read((atomic64_t *)(adev->doorbell.cpu_addr + index)); - DRM_ERROR("reading beyond doorbell aperture: 0x%08x!\n", index); + dev_err(adev->dev, "reading beyond doorbell aperture: 0x%08x!\n", + index); return 0; } @@ -105,7 +108,8 @@ void amdgpu_mm_wdoorbell64(struct amdgpu_device *adev, u32 index, u64 v) if (index < adev->doorbell.num_kernel_doorbells) atomic64_set((atomic64_t *)(adev->doorbell.cpu_addr + index), v); else - DRM_ERROR("writing beyond doorbell aperture: 0x%08x!\n", index); + dev_err(adev->dev, + "writing beyond doorbell aperture: 0x%08x!\n", index); } /** @@ -166,7 +170,8 @@ int amdgpu_doorbell_create_kernel_doorbells(struct amdgpu_device *adev) NULL, (void **)&adev->doorbell.cpu_addr); if (r) { - DRM_ERROR("Failed to allocate kernel doorbells, err=%d\n", r); + dev_err(adev->dev, + "Failed to allocate kernel doorbells, err=%d\n", r); return r; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 4db92e0a60da7b37ad0be30fbc4c6f0d30836153..b299e15bb5e5134a006d89fe8018357aa7e73ae7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -144,6 +144,7 @@ enum AMDGPU_DEBUG_MASK { AMDGPU_DEBUG_DISABLE_GPU_RING_RESET = BIT(6), AMDGPU_DEBUG_SMU_POOL = BIT(7), AMDGPU_DEBUG_VM_USERPTR = BIT(8), + AMDGPU_DEBUG_DISABLE_RAS_CE_LOG = BIT(9) }; unsigned int amdgpu_vram_limit = UINT_MAX; @@ -2278,6 +2279,11 @@ static void amdgpu_init_debug_options(struct amdgpu_device *adev) pr_info("debug: VM mode debug for userptr is enabled\n"); adev->debug_vm_userptr = true; } + + if (amdgpu_debug_mask & AMDGPU_DEBUG_DISABLE_RAS_CE_LOG) { + pr_info("debug: disable kernel logs of correctable errors\n"); + adev->debug_disable_ce_logs = true; + } } static unsigned long amdgpu_fix_asic_type(struct pci_dev *pdev, unsigned long flags) @@ -2321,7 +2327,7 @@ static int amdgpu_pci_probe(struct pci_dev *pdev, amdgpu_aspm = 0; if (amdgpu_virtual_display || - amdgpu_device_asic_has_dc_support(flags & AMD_ASIC_MASK)) + amdgpu_device_asic_has_dc_support(pdev, flags & AMD_ASIC_MASK)) supports_atomic = true; if ((flags & AMD_EXP_HW_SUPPORT) && !amdgpu_exp_hw_support) { @@ -2570,7 +2576,7 @@ static int amdgpu_pmops_prepare(struct device *dev) static void amdgpu_pmops_complete(struct device *dev) { - /* nothing to do */ + amdgpu_device_complete(dev_get_drvdata(dev)); } static int amdgpu_pmops_suspend(struct device *dev) @@ -2906,20 +2912,6 @@ done: return ret; } -static int amdgpu_drm_release(struct inode *inode, struct file *filp) -{ - struct drm_file *file_priv = filp->private_data; - struct amdgpu_fpriv *fpriv = file_priv->driver_priv; - - if (fpriv) { - fpriv->evf_mgr.fd_closing = true; - amdgpu_eviction_fence_destroy(&fpriv->evf_mgr); - amdgpu_userq_mgr_fini(&fpriv->userq_mgr); - } - - return drm_release(inode, filp); -} - long amdgpu_drm_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { @@ -2971,7 +2963,7 @@ static const struct file_operations amdgpu_driver_kms_fops = { .owner = THIS_MODULE, .open = drm_open, .flush = amdgpu_flush, - .release = amdgpu_drm_release, + .release = drm_release, .unlocked_ioctl = amdgpu_drm_ioctl, .mmap = drm_gem_mmap, .poll = drm_poll, @@ -3107,10 +3099,6 @@ static int __init amdgpu_init(void) if (r) goto error_sync; - r = amdgpu_fence_slab_init(); - if (r) - goto error_fence; - r = amdgpu_userq_fence_slab_init(); if (r) goto error_fence; @@ -3145,7 +3133,6 @@ static void __exit amdgpu_exit(void) amdgpu_unregister_atpx_handler(); amdgpu_acpi_release(); amdgpu_sync_fini(); - amdgpu_fence_slab_fini(); amdgpu_userq_fence_slab_fini(); mmu_notifier_synchronize(); amdgpu_xcp_drv_release(); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c index 8cecf25996edc19146fa851c9abae14ac9965011..2c3547f4cea420c94dffc60d6931fd80888bae35 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c @@ -41,37 +41,6 @@ #include "amdgpu_trace.h" #include "amdgpu_reset.h" -/* - * Fences mark an event in the GPUs pipeline and are used - * for GPU/CPU synchronization. When the fence is written, - * it is expected that all buffers associated with that fence - * are no longer in use by the associated ring on the GPU and - * that the relevant GPU caches have been flushed. - */ - -struct amdgpu_fence { - struct dma_fence base; - - /* RB, DMA, etc. */ - struct amdgpu_ring *ring; - ktime_t start_timestamp; -}; - -static struct kmem_cache *amdgpu_fence_slab; - -int amdgpu_fence_slab_init(void) -{ - amdgpu_fence_slab = KMEM_CACHE(amdgpu_fence, SLAB_HWCACHE_ALIGN); - if (!amdgpu_fence_slab) - return -ENOMEM; - return 0; -} - -void amdgpu_fence_slab_fini(void) -{ - rcu_barrier(); - kmem_cache_destroy(amdgpu_fence_slab); -} /* * Cast helper */ @@ -130,14 +99,14 @@ static u32 amdgpu_fence_read(struct amdgpu_ring *ring) * * @ring: ring the fence is associated with * @f: resulting fence object - * @job: job the fence is embedded in + * @af: amdgpu fence input * @flags: flags to pass into the subordinate .emit_fence() call * * Emits a fence command on the requested ring (all asics). * Returns 0 on success, -ENOMEM on failure. */ -int amdgpu_fence_emit(struct amdgpu_ring *ring, struct dma_fence **f, struct amdgpu_job *job, - unsigned int flags) +int amdgpu_fence_emit(struct amdgpu_ring *ring, struct dma_fence **f, + struct amdgpu_fence *af, unsigned int flags) { struct amdgpu_device *adev = ring->adev; struct dma_fence *fence; @@ -146,36 +115,28 @@ int amdgpu_fence_emit(struct amdgpu_ring *ring, struct dma_fence **f, struct amd uint32_t seq; int r; - if (job == NULL) { - /* create a sperate hw fence */ - am_fence = kmem_cache_alloc(amdgpu_fence_slab, GFP_ATOMIC); - if (am_fence == NULL) + if (!af) { + /* create a separate hw fence */ + am_fence = kzalloc(sizeof(*am_fence), GFP_KERNEL); + if (!am_fence) return -ENOMEM; - fence = &am_fence->base; - am_fence->ring = ring; } else { - /* take use of job-embedded fence */ - fence = &job->hw_fence; + am_fence = af; } + fence = &am_fence->base; + am_fence->ring = ring; seq = ++ring->fence_drv.sync_seq; - if (job && job->job_run_counter) { - /* reinit seq for resubmitted jobs */ - fence->seqno = seq; - /* TO be inline with external fence creation and other drivers */ + if (af) { + dma_fence_init(fence, &amdgpu_job_fence_ops, + &ring->fence_drv.lock, + adev->fence_context + ring->idx, seq); + /* Against remove in amdgpu_job_{free, free_cb} */ dma_fence_get(fence); } else { - if (job) { - dma_fence_init(fence, &amdgpu_job_fence_ops, - &ring->fence_drv.lock, - adev->fence_context + ring->idx, seq); - /* Against remove in amdgpu_job_{free, free_cb} */ - dma_fence_get(fence); - } else { - dma_fence_init(fence, &amdgpu_fence_ops, - &ring->fence_drv.lock, - adev->fence_context + ring->idx, seq); - } + dma_fence_init(fence, &amdgpu_fence_ops, + &ring->fence_drv.lock, + adev->fence_context + ring->idx, seq); } amdgpu_ring_emit_fence(ring, ring->fence_drv.gpu_addr, @@ -326,7 +287,9 @@ static void amdgpu_fence_fallback(struct timer_list *t) fence_drv.fallback_timer); if (amdgpu_fence_process(ring)) - DRM_WARN("Fence fallback timer expired on ring %s\n", ring->name); + dev_warn(ring->adev->dev, + "Fence fallback timer expired on ring %s\n", + ring->name); } /** @@ -718,7 +681,7 @@ void amdgpu_fence_driver_clear_job_fences(struct amdgpu_ring *ring) * it right here or we won't be able to track them in fence_drv * and they will remain unsignaled during sa_bo free. */ - job = container_of(old, struct amdgpu_job, hw_fence); + job = container_of(old, struct amdgpu_job, hw_fence.base); if (!job->base.s_fence && !dma_fence_is_signaled(old)) dma_fence_signal(old); RCU_INIT_POINTER(*ptr, NULL); @@ -780,7 +743,7 @@ static const char *amdgpu_fence_get_timeline_name(struct dma_fence *f) static const char *amdgpu_job_fence_get_timeline_name(struct dma_fence *f) { - struct amdgpu_job *job = container_of(f, struct amdgpu_job, hw_fence); + struct amdgpu_job *job = container_of(f, struct amdgpu_job, hw_fence.base); return (const char *)to_amdgpu_ring(job->base.sched)->name; } @@ -810,7 +773,7 @@ static bool amdgpu_fence_enable_signaling(struct dma_fence *f) */ static bool amdgpu_job_fence_enable_signaling(struct dma_fence *f) { - struct amdgpu_job *job = container_of(f, struct amdgpu_job, hw_fence); + struct amdgpu_job *job = container_of(f, struct amdgpu_job, hw_fence.base); if (!timer_pending(&to_amdgpu_ring(job->base.sched)->fence_drv.fallback_timer)) amdgpu_fence_schedule_fallback(to_amdgpu_ring(job->base.sched)); @@ -830,7 +793,7 @@ static void amdgpu_fence_free(struct rcu_head *rcu) struct dma_fence *f = container_of(rcu, struct dma_fence, rcu); /* free fence_slab if it's separated fence*/ - kmem_cache_free(amdgpu_fence_slab, to_amdgpu_fence(f)); + kfree(to_amdgpu_fence(f)); } /** @@ -845,7 +808,7 @@ static void amdgpu_job_fence_free(struct rcu_head *rcu) struct dma_fence *f = container_of(rcu, struct dma_fence, rcu); /* free job if fence has a parent job */ - kfree(container_of(f, struct amdgpu_job, hw_fence)); + kfree(container_of(f, struct amdgpu_job, hw_fence.base)); } /** diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c index 1ae88c459da513dd0e49a3c141302f48039530dc..b0082aa7f3c618bcf20784ae431c692fae9b0899 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c @@ -144,7 +144,8 @@ int amdgpu_fru_get_product_info(struct amdgpu_device *adev) /* If algo exists, it means that the i2c_adapter's initialized */ if (!adev->pm.fru_eeprom_i2c_bus || !adev->pm.fru_eeprom_i2c_bus->algo) { - DRM_WARN("Cannot access FRU, EEPROM accessor not initialized"); + dev_warn(adev->dev, + "Cannot access FRU, EEPROM accessor not initialized"); return -ENODEV; } @@ -152,19 +153,22 @@ int amdgpu_fru_get_product_info(struct amdgpu_device *adev) len = amdgpu_eeprom_read(adev->pm.fru_eeprom_i2c_bus, fru_addr, buf, sizeof(buf)); if (len != 8) { - DRM_ERROR("Couldn't read the IPMI Common Header: %d", len); + dev_err(adev->dev, "Couldn't read the IPMI Common Header: %d", + len); return len < 0 ? len : -EIO; } if (buf[0] != 1) { - DRM_ERROR("Bad IPMI Common Header version: 0x%02x", buf[0]); + dev_err(adev->dev, "Bad IPMI Common Header version: 0x%02x", + buf[0]); return -EIO; } for (csum = 0; len > 0; len--) csum += buf[len - 1]; if (csum) { - DRM_ERROR("Bad IPMI Common Header checksum: 0x%02x", csum); + dev_err(adev->dev, "Bad IPMI Common Header checksum: 0x%02x", + csum); return -EIO; } @@ -179,12 +183,14 @@ int amdgpu_fru_get_product_info(struct amdgpu_device *adev) /* Read the header of the PIA. */ len = amdgpu_eeprom_read(adev->pm.fru_eeprom_i2c_bus, addr, buf, 3); if (len != 3) { - DRM_ERROR("Couldn't read the Product Info Area header: %d", len); + dev_err(adev->dev, + "Couldn't read the Product Info Area header: %d", len); return len < 0 ? len : -EIO; } if (buf[0] != 1) { - DRM_ERROR("Bad IPMI Product Info Area version: 0x%02x", buf[0]); + dev_err(adev->dev, "Bad IPMI Product Info Area version: 0x%02x", + buf[0]); return -EIO; } @@ -197,14 +203,16 @@ int amdgpu_fru_get_product_info(struct amdgpu_device *adev) len = amdgpu_eeprom_read(adev->pm.fru_eeprom_i2c_bus, addr, pia, size); if (len != size) { kfree(pia); - DRM_ERROR("Couldn't read the Product Info Area: %d", len); + dev_err(adev->dev, "Couldn't read the Product Info Area: %d", + len); return len < 0 ? len : -EIO; } for (csum = 0; size > 0; size--) csum += pia[size - 1]; if (csum) { - DRM_ERROR("Bad Product Info Area checksum: 0x%02x", csum); + dev_err(adev->dev, "Bad Product Info Area checksum: 0x%02x", + csum); kfree(pia); return -EIO; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index e5e33a68d935abe5fce6f5f866478cf9043b7228..6626a6e64ff5f8b67d99f30bfaee4aafee7e50fb 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -317,8 +317,7 @@ static int amdgpu_gem_object_open(struct drm_gem_object *obj, */ if (!vm->is_compute_context || !vm->process_info) return 0; - if (!obj->import_attach || - !dma_buf_is_dynamic(obj->import_attach->dmabuf)) + if (!drm_gem_is_imported(obj) || !dma_buf_is_dynamic(obj->dma_buf)) return 0; mutex_lock_nested(&vm->process_info->lock, 1); if (!WARN_ON(!vm->process_info->eviction_fence)) { @@ -1024,7 +1023,7 @@ int amdgpu_gem_op_ioctl(struct drm_device *dev, void *data, break; } case AMDGPU_GEM_OP_SET_PLACEMENT: - if (robj->tbo.base.import_attach && + if (drm_gem_is_imported(&robj->tbo.base) && args->value & AMDGPU_GEM_DOMAIN_VRAM) { r = -EINVAL; amdgpu_bo_unreserve(robj); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c index c5646af055abcbce1db22aa61295d9a0f5e06165..c80c8f54353211d96222936588929c809f58015b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c @@ -149,7 +149,7 @@ static bool amdgpu_gfx_is_graphics_multipipe_capable(struct amdgpu_device *adev) static bool amdgpu_gfx_is_compute_multipipe_capable(struct amdgpu_device *adev) { if (amdgpu_compute_multipipe != -1) { - DRM_INFO("amdgpu: forcing compute pipe policy %d\n", + dev_info(adev->dev, "amdgpu: forcing compute pipe policy %d\n", amdgpu_compute_multipipe); return amdgpu_compute_multipipe == 1; } @@ -674,7 +674,7 @@ int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev, int xcc_id) * generation exposes more than 64 queues. If so, the * definition of queue_mask needs updating */ if (WARN_ON(i > (sizeof(queue_mask)*8))) { - DRM_ERROR("Invalid KCQ enabled: %d\n", i); + dev_err(adev->dev, "Invalid KCQ enabled: %d\n", i); break; } @@ -683,15 +683,15 @@ int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev, int xcc_id) amdgpu_device_flush_hdp(adev, NULL); - DRM_INFO("kiq ring mec %d pipe %d q %d\n", kiq_ring->me, kiq_ring->pipe, - kiq_ring->queue); + dev_info(adev->dev, "kiq ring mec %d pipe %d q %d\n", kiq_ring->me, + kiq_ring->pipe, kiq_ring->queue); spin_lock(&kiq->ring_lock); r = amdgpu_ring_alloc(kiq_ring, kiq->pmf->map_queues_size * adev->gfx.num_compute_rings + kiq->pmf->set_resources_size); if (r) { - DRM_ERROR("Failed to lock KIQ (%d).\n", r); + dev_err(adev->dev, "Failed to lock KIQ (%d).\n", r); spin_unlock(&kiq->ring_lock); return r; } @@ -712,7 +712,7 @@ int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev, int xcc_id) r = amdgpu_ring_test_helper(kiq_ring); spin_unlock(&kiq->ring_lock); if (r) - DRM_ERROR("KCQ enable failed\n"); + dev_err(adev->dev, "KCQ enable failed\n"); return r; } @@ -734,7 +734,7 @@ int amdgpu_gfx_enable_kgq(struct amdgpu_device *adev, int xcc_id) r = amdgpu_mes_map_legacy_queue(adev, &adev->gfx.gfx_ring[j]); if (r) { - DRM_ERROR("failed to map gfx queue\n"); + dev_err(adev->dev, "failed to map gfx queue\n"); return r; } } @@ -748,7 +748,7 @@ int amdgpu_gfx_enable_kgq(struct amdgpu_device *adev, int xcc_id) r = amdgpu_ring_alloc(kiq_ring, kiq->pmf->map_queues_size * adev->gfx.num_gfx_rings); if (r) { - DRM_ERROR("Failed to lock KIQ (%d).\n", r); + dev_err(adev->dev, "Failed to lock KIQ (%d).\n", r); spin_unlock(&kiq->ring_lock); return r; } @@ -769,7 +769,7 @@ int amdgpu_gfx_enable_kgq(struct amdgpu_device *adev, int xcc_id) r = amdgpu_ring_test_helper(kiq_ring); spin_unlock(&kiq->ring_lock); if (r) - DRM_ERROR("KGQ enable failed\n"); + dev_err(adev->dev, "KGQ enable failed\n"); return r; } @@ -1030,7 +1030,7 @@ int amdgpu_gfx_cp_ecc_error_irq(struct amdgpu_device *adev, ih_data.head = *ras_if; - DRM_ERROR("CP ECC ERROR IRQ\n"); + dev_err(adev->dev, "CP ECC ERROR IRQ\n"); amdgpu_ras_interrupt_dispatch(adev, &ih_data); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c index 6b0fbbb91e5795db95b3c7dcbbb22db8a8b89e13..97b562a79ea8ee04c3b5321ef93117a311160a75 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c @@ -38,6 +38,13 @@ #include #include +static const u64 four_gb = 0x100000000ULL; + +bool amdgpu_gmc_is_pdb0_enabled(struct amdgpu_device *adev) +{ + return adev->gmc.xgmi.connected_to_cpu || amdgpu_virt_xgmi_migrate_enabled(adev); +} + /** * amdgpu_gmc_pdb0_alloc - allocate vram for pdb0 * @@ -251,10 +258,20 @@ void amdgpu_gmc_sysvm_location(struct amdgpu_device *adev, struct amdgpu_gmc *mc u64 hive_vram_end = mc->xgmi.node_segment_size * mc->xgmi.num_physical_nodes - 1; mc->vram_start = mc->xgmi.node_segment_size * mc->xgmi.physical_node_id; mc->vram_end = mc->vram_start + mc->xgmi.node_segment_size - 1; - mc->gart_start = hive_vram_end + 1; + /* node_segment_size may not 4GB aligned on SRIOV, align up is needed. */ + mc->gart_start = ALIGN(hive_vram_end + 1, four_gb); mc->gart_end = mc->gart_start + mc->gart_size - 1; - mc->fb_start = hive_vram_start; - mc->fb_end = hive_vram_end; + if (amdgpu_virt_xgmi_migrate_enabled(adev)) { + /* set mc->vram_start to 0 to switch the returned GPU address of + * amdgpu_bo_create_reserved() from FB aperture to GART aperture. + */ + mc->vram_start = 0; + mc->vram_end = mc->vram_start + mc->mc_vram_size - 1; + mc->visible_vram_size = min(mc->visible_vram_size, mc->real_vram_size); + } else { + mc->fb_start = hive_vram_start; + mc->fb_end = hive_vram_end; + } dev_info(adev->dev, "VRAM: %lluM 0x%016llX - 0x%016llX (%lluM used)\n", mc->mc_vram_size >> 20, mc->vram_start, mc->vram_end, mc->real_vram_size >> 20); @@ -276,7 +293,6 @@ void amdgpu_gmc_sysvm_location(struct amdgpu_device *adev, struct amdgpu_gmc *mc void amdgpu_gmc_gart_location(struct amdgpu_device *adev, struct amdgpu_gmc *mc, enum amdgpu_gart_placement gart_placement) { - const uint64_t four_gb = 0x100000000ULL; u64 size_af, size_bf; /*To avoid the hole, limit the max mc address to AMDGPU_GMC_HOLE_START*/ u64 max_mc_address = min(adev->gmc.mc_mask, AMDGPU_GMC_HOLE_START - 1); @@ -1041,9 +1057,7 @@ void amdgpu_gmc_init_pdb0(struct amdgpu_device *adev) */ u64 vram_size = adev->gmc.xgmi.node_segment_size * adev->gmc.xgmi.num_physical_nodes; u64 pde0_page_size = (1ULL<gmc.vmid0_page_table_block_size)<<21; - u64 vram_addr = adev->vm_manager.vram_base_offset - - adev->gmc.xgmi.physical_node_id * adev->gmc.xgmi.node_segment_size; - u64 vram_end = vram_addr + vram_size; + u64 vram_addr, vram_end; u64 gart_ptb_gpu_pa = amdgpu_gmc_vram_pa(adev, adev->gart.bo); int idx; @@ -1056,6 +1070,11 @@ void amdgpu_gmc_init_pdb0(struct amdgpu_device *adev) flags |= AMDGPU_PTE_FRAG((adev->gmc.vmid0_page_table_block_size + 9*1)); flags |= AMDGPU_PDE_PTE_FLAG(adev); + vram_addr = adev->vm_manager.vram_base_offset; + if (!amdgpu_virt_xgmi_migrate_enabled(adev)) + vram_addr -= adev->gmc.xgmi.physical_node_id * adev->gmc.xgmi.node_segment_size; + vram_end = vram_addr + vram_size; + /* The first n PDE0 entries are used as PTE, * pointing to vram */ @@ -1429,3 +1448,232 @@ bool amdgpu_gmc_need_reset_on_init(struct amdgpu_device *adev) return false; } + +enum amdgpu_memory_partition +amdgpu_gmc_get_vf_memory_partition(struct amdgpu_device *adev) +{ + switch (adev->gmc.num_mem_partitions) { + case 0: + return UNKNOWN_MEMORY_PARTITION_MODE; + case 1: + return AMDGPU_NPS1_PARTITION_MODE; + case 2: + return AMDGPU_NPS2_PARTITION_MODE; + case 4: + return AMDGPU_NPS4_PARTITION_MODE; + case 8: + return AMDGPU_NPS8_PARTITION_MODE; + default: + return AMDGPU_NPS1_PARTITION_MODE; + } +} + +enum amdgpu_memory_partition +amdgpu_gmc_get_memory_partition(struct amdgpu_device *adev, u32 *supp_modes) +{ + enum amdgpu_memory_partition mode = UNKNOWN_MEMORY_PARTITION_MODE; + + if (adev->nbio.funcs && + adev->nbio.funcs->get_memory_partition_mode) + mode = adev->nbio.funcs->get_memory_partition_mode(adev, + supp_modes); + else + dev_warn(adev->dev, "memory partition mode query is not supported\n"); + + return mode; +} + +enum amdgpu_memory_partition +amdgpu_gmc_query_memory_partition(struct amdgpu_device *adev) +{ + if (amdgpu_sriov_vf(adev)) + return amdgpu_gmc_get_vf_memory_partition(adev); + else + return amdgpu_gmc_get_memory_partition(adev, NULL); +} + +static bool amdgpu_gmc_validate_partition_info(struct amdgpu_device *adev) +{ + enum amdgpu_memory_partition mode; + u32 supp_modes; + bool valid; + + mode = amdgpu_gmc_get_memory_partition(adev, &supp_modes); + + /* Mode detected by hardware not present in supported modes */ + if ((mode != UNKNOWN_MEMORY_PARTITION_MODE) && + !(BIT(mode - 1) & supp_modes)) + return false; + + switch (mode) { + case UNKNOWN_MEMORY_PARTITION_MODE: + case AMDGPU_NPS1_PARTITION_MODE: + valid = (adev->gmc.num_mem_partitions == 1); + break; + case AMDGPU_NPS2_PARTITION_MODE: + valid = (adev->gmc.num_mem_partitions == 2); + break; + case AMDGPU_NPS4_PARTITION_MODE: + valid = (adev->gmc.num_mem_partitions == 3 || + adev->gmc.num_mem_partitions == 4); + break; + case AMDGPU_NPS8_PARTITION_MODE: + valid = (adev->gmc.num_mem_partitions == 8); + break; + default: + valid = false; + } + + return valid; +} + +static bool amdgpu_gmc_is_node_present(int *node_ids, int num_ids, int nid) +{ + int i; + + /* Check if node with id 'nid' is present in 'node_ids' array */ + for (i = 0; i < num_ids; ++i) + if (node_ids[i] == nid) + return true; + + return false; +} + +static void +amdgpu_gmc_init_acpi_mem_ranges(struct amdgpu_device *adev, + struct amdgpu_mem_partition_info *mem_ranges) +{ + struct amdgpu_numa_info numa_info; + int node_ids[AMDGPU_MAX_MEM_RANGES]; + int num_ranges = 0, ret; + int num_xcc, xcc_id; + uint32_t xcc_mask; + + num_xcc = NUM_XCC(adev->gfx.xcc_mask); + xcc_mask = (1U << num_xcc) - 1; + + for_each_inst(xcc_id, xcc_mask) { + ret = amdgpu_acpi_get_mem_info(adev, xcc_id, &numa_info); + if (ret) + continue; + + if (numa_info.nid == NUMA_NO_NODE) { + mem_ranges[0].size = numa_info.size; + mem_ranges[0].numa.node = numa_info.nid; + num_ranges = 1; + break; + } + + if (amdgpu_gmc_is_node_present(node_ids, num_ranges, + numa_info.nid)) + continue; + + node_ids[num_ranges] = numa_info.nid; + mem_ranges[num_ranges].numa.node = numa_info.nid; + mem_ranges[num_ranges].size = numa_info.size; + ++num_ranges; + } + + adev->gmc.num_mem_partitions = num_ranges; +} + +void amdgpu_gmc_init_sw_mem_ranges(struct amdgpu_device *adev, + struct amdgpu_mem_partition_info *mem_ranges) +{ + enum amdgpu_memory_partition mode; + u32 start_addr = 0, size; + int i, r, l; + + mode = amdgpu_gmc_query_memory_partition(adev); + + switch (mode) { + case UNKNOWN_MEMORY_PARTITION_MODE: + adev->gmc.num_mem_partitions = 0; + break; + case AMDGPU_NPS1_PARTITION_MODE: + adev->gmc.num_mem_partitions = 1; + break; + case AMDGPU_NPS2_PARTITION_MODE: + adev->gmc.num_mem_partitions = 2; + break; + case AMDGPU_NPS4_PARTITION_MODE: + if (adev->flags & AMD_IS_APU) + adev->gmc.num_mem_partitions = 3; + else + adev->gmc.num_mem_partitions = 4; + break; + case AMDGPU_NPS8_PARTITION_MODE: + adev->gmc.num_mem_partitions = 8; + break; + default: + adev->gmc.num_mem_partitions = 1; + break; + } + + /* Use NPS range info, if populated */ + r = amdgpu_gmc_get_nps_memranges(adev, mem_ranges, + &adev->gmc.num_mem_partitions); + if (!r) { + l = 0; + for (i = 1; i < adev->gmc.num_mem_partitions; ++i) { + if (mem_ranges[i].range.lpfn > + mem_ranges[i - 1].range.lpfn) + l = i; + } + + } else { + if (!adev->gmc.num_mem_partitions) { + dev_warn(adev->dev, + "Not able to detect NPS mode, fall back to NPS1\n"); + adev->gmc.num_mem_partitions = 1; + } + /* Fallback to sw based calculation */ + size = (adev->gmc.real_vram_size + SZ_16M) >> AMDGPU_GPU_PAGE_SHIFT; + size /= adev->gmc.num_mem_partitions; + + for (i = 0; i < adev->gmc.num_mem_partitions; ++i) { + mem_ranges[i].range.fpfn = start_addr; + mem_ranges[i].size = + ((u64)size << AMDGPU_GPU_PAGE_SHIFT); + mem_ranges[i].range.lpfn = start_addr + size - 1; + start_addr += size; + } + + l = adev->gmc.num_mem_partitions - 1; + } + + /* Adjust the last one */ + mem_ranges[l].range.lpfn = + (adev->gmc.real_vram_size >> AMDGPU_GPU_PAGE_SHIFT) - 1; + mem_ranges[l].size = + adev->gmc.real_vram_size - + ((u64)mem_ranges[l].range.fpfn << AMDGPU_GPU_PAGE_SHIFT); +} + +int amdgpu_gmc_init_mem_ranges(struct amdgpu_device *adev) +{ + bool valid; + + adev->gmc.mem_partitions = kcalloc(AMDGPU_MAX_MEM_RANGES, + sizeof(struct amdgpu_mem_partition_info), + GFP_KERNEL); + if (!adev->gmc.mem_partitions) + return -ENOMEM; + + if (adev->gmc.is_app_apu) + amdgpu_gmc_init_acpi_mem_ranges(adev, adev->gmc.mem_partitions); + else + amdgpu_gmc_init_sw_mem_ranges(adev, adev->gmc.mem_partitions); + + if (amdgpu_sriov_vf(adev)) + valid = true; + else + valid = amdgpu_gmc_validate_partition_info(adev); + if (!valid) { + /* TODO: handle invalid case */ + dev_warn(adev->dev, + "Mem ranges not matching with hardware config\n"); + } + + return 0; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h index 80fa29c26e9eeefca623668da7671582dbf3cde0..397c6ccdb9033bac752e5aa5ee8f566d428bdc0f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h @@ -84,6 +84,8 @@ enum amdgpu_memory_partition { #define AMDGPU_GMC_INIT_RESET_NPS BIT(0) +#define AMDGPU_MAX_MEM_RANGES 8 + /* * GMC page fault information */ @@ -394,6 +396,7 @@ static inline uint64_t amdgpu_gmc_sign_extend(uint64_t addr) return addr; } +bool amdgpu_gmc_is_pdb0_enabled(struct amdgpu_device *adev); int amdgpu_gmc_pdb0_alloc(struct amdgpu_device *adev); void amdgpu_gmc_get_pde_for_bo(struct amdgpu_bo *bo, int level, uint64_t *addr, uint64_t *flags); @@ -455,5 +458,13 @@ int amdgpu_gmc_request_memory_partition(struct amdgpu_device *adev, int nps_mode); void amdgpu_gmc_prepare_nps_mode_change(struct amdgpu_device *adev); bool amdgpu_gmc_need_reset_on_init(struct amdgpu_device *adev); - +enum amdgpu_memory_partition +amdgpu_gmc_get_vf_memory_partition(struct amdgpu_device *adev); +enum amdgpu_memory_partition +amdgpu_gmc_get_memory_partition(struct amdgpu_device *adev, u32 *supp_modes); +enum amdgpu_memory_partition +amdgpu_gmc_query_memory_partition(struct amdgpu_device *adev); +int amdgpu_gmc_init_mem_ranges(struct amdgpu_device *adev); +void amdgpu_gmc_init_sw_mem_ranges(struct amdgpu_device *adev, + struct amdgpu_mem_partition_info *mem_ranges); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.c index 8179d0814db99a8da8b8816b79b523a1a28a1f13..57101d24422f2bb90e800dd3af6457a1fb8e963f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.c @@ -24,7 +24,6 @@ * Alex Deucher */ -#include #include #include diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c index 802743efa3b39aa64f084341e0d23ae1df30ccc8..206b70acb29a0af88e96cc06e6a463c59935c188 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c @@ -128,6 +128,7 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned int num_ibs, struct amdgpu_device *adev = ring->adev; struct amdgpu_ib *ib = &ibs[0]; struct dma_fence *tmp = NULL; + struct amdgpu_fence *af; bool need_ctx_switch; struct amdgpu_vm *vm; uint64_t fence_ctx; @@ -154,6 +155,7 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned int num_ibs, csa_va = job->csa_va; gds_va = job->gds_va; init_shadow = job->init_shadow; + af = &job->hw_fence; } else { vm = NULL; fence_ctx = 0; @@ -161,6 +163,7 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned int num_ibs, csa_va = 0; gds_va = 0; init_shadow = false; + af = NULL; } if (!ring->sched.ready) { @@ -282,7 +285,7 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned int num_ibs, amdgpu_ring_init_cond_exec(ring, ring->cond_exe_gpu_addr); } - r = amdgpu_fence_emit(ring, f, job, fence_flags); + r = amdgpu_fence_emit(ring, f, af, fence_flags); if (r) { dev_err(adev->dev, "failed to emit fence (%d)\n", r); if (job && job->vmid) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c index 30f16968b5788691ee403c93dd6c3c696cef1985..a6419246e9c24716b58fdd27e494b0a35dade09c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c @@ -218,7 +218,7 @@ int amdgpu_ih_process(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih) restart_ih: count = AMDGPU_IH_MAX_NUM_IVS; - DRM_DEBUG("%s: rptr %d, wptr %d\n", __func__, ih->rptr, wptr); + dev_dbg(adev->dev, "%s: rptr %d, wptr %d\n", __func__, ih->rptr, wptr); /* Order reading of wptr vs. reading of IH ring data */ rmb(); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ip.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ip.c new file mode 100644 index 0000000000000000000000000000000000000000..99e1cf4fc955eacb996645d11bb70a10e0de6996 --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ip.c @@ -0,0 +1,96 @@ +/* + * Copyright 2025 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ + +#include "amdgpu.h" +#include "amdgpu_ip.h" + +static int8_t amdgpu_logical_to_dev_inst(struct amdgpu_device *adev, + enum amd_hw_ip_block_type block, + int8_t inst) +{ + int8_t dev_inst; + + switch (block) { + case GC_HWIP: + case SDMA0_HWIP: + /* Both JPEG and VCN as JPEG is only alias of VCN */ + case VCN_HWIP: + dev_inst = adev->ip_map.dev_inst[block][inst]; + break; + default: + /* For rest of the IPs, no look up required. + * Assume 'logical instance == physical instance' for all configs. */ + dev_inst = inst; + break; + } + + return dev_inst; +} + +static uint32_t amdgpu_logical_to_dev_mask(struct amdgpu_device *adev, + enum amd_hw_ip_block_type block, + uint32_t mask) +{ + uint32_t dev_mask = 0; + int8_t log_inst, dev_inst; + + while (mask) { + log_inst = ffs(mask) - 1; + dev_inst = amdgpu_logical_to_dev_inst(adev, block, log_inst); + dev_mask |= (1 << dev_inst); + mask &= ~(1 << log_inst); + } + + return dev_mask; +} + +static void amdgpu_populate_ip_map(struct amdgpu_device *adev, + enum amd_hw_ip_block_type ip_block, + uint32_t inst_mask) +{ + int l = 0, i; + + while (inst_mask) { + i = ffs(inst_mask) - 1; + adev->ip_map.dev_inst[ip_block][l++] = i; + inst_mask &= ~(1 << i); + } + for (; l < HWIP_MAX_INSTANCE; l++) + adev->ip_map.dev_inst[ip_block][l] = -1; +} + +void amdgpu_ip_map_init(struct amdgpu_device *adev) +{ + u32 ip_map[][2] = { + { GC_HWIP, adev->gfx.xcc_mask }, + { SDMA0_HWIP, adev->sdma.sdma_mask }, + { VCN_HWIP, adev->vcn.inst_mask }, + }; + int i; + + for (i = 0; i < ARRAY_SIZE(ip_map); ++i) + amdgpu_populate_ip_map(adev, ip_map[i][0], ip_map[i][1]); + + adev->ip_map.logical_to_dev_inst = amdgpu_logical_to_dev_inst; + adev->ip_map.logical_to_dev_mask = amdgpu_logical_to_dev_mask; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ip.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ip.h new file mode 100644 index 0000000000000000000000000000000000000000..2490fd322aecea01b28c2485ebece4ba2270597b --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ip.h @@ -0,0 +1,29 @@ +/* + * Copyright 2025 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ + +#ifndef __AMDGPU_IP_H__ +#define __AMDGPU_IP_H__ + +void amdgpu_ip_map_init(struct amdgpu_device *adev); + +#endif /* __AMDGPU_IP_H__ */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c index 13c60cac4261bacb8f9da5bcb18b139b6fd8eee9..183fa33c243479426f932c42c27f677f02c6e4bd 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c @@ -142,8 +142,9 @@ void amdgpu_irq_disable_all(struct amdgpu_device *adev) r = src->funcs->set(adev, src, k, AMDGPU_IRQ_STATE_DISABLE); if (r) - DRM_ERROR("error disabling interrupt (%d)\n", - r); + dev_err(adev->dev, + "error disabling interrupt (%d)\n", + r); } } } @@ -242,7 +243,7 @@ static bool amdgpu_msi_ok(struct amdgpu_device *adev) return true; } -static void amdgpu_restore_msix(struct amdgpu_device *adev) +void amdgpu_restore_msix(struct amdgpu_device *adev) { u16 ctrl; @@ -315,7 +316,7 @@ int amdgpu_irq_init(struct amdgpu_device *adev) adev->irq.irq = irq; adev_to_drm(adev)->max_vblank_count = 0x00ffffff; - DRM_DEBUG("amdgpu: irq initialized.\n"); + dev_dbg(adev->dev, "amdgpu: irq initialized.\n"); return 0; free_vectors: @@ -461,10 +462,10 @@ void amdgpu_irq_dispatch(struct amdgpu_device *adev, src_id = entry.src_id; if (client_id >= AMDGPU_IRQ_CLIENTID_MAX) { - DRM_DEBUG("Invalid client_id in IV: %d\n", client_id); + dev_dbg(adev->dev, "Invalid client_id in IV: %d\n", client_id); } else if (src_id >= AMDGPU_MAX_IRQ_SRC_ID) { - DRM_DEBUG("Invalid src_id in IV: %d\n", src_id); + dev_dbg(adev->dev, "Invalid src_id in IV: %d\n", src_id); } else if (((client_id == AMDGPU_IRQ_CLIENTID_LEGACY) || (client_id == SOC15_IH_CLIENTID_ISP)) && @@ -472,18 +473,21 @@ void amdgpu_irq_dispatch(struct amdgpu_device *adev, generic_handle_domain_irq(adev->irq.domain, src_id); } else if (!adev->irq.client[client_id].sources) { - DRM_DEBUG("Unregistered interrupt client_id: %d src_id: %d\n", - client_id, src_id); + dev_dbg(adev->dev, + "Unregistered interrupt client_id: %d src_id: %d\n", + client_id, src_id); } else if ((src = adev->irq.client[client_id].sources[src_id])) { r = src->funcs->process(adev, src, &entry); if (r < 0) - DRM_ERROR("error processing interrupt (%d)\n", r); + dev_err(adev->dev, "error processing interrupt (%d)\n", + r); else if (r) handled = true; } else { - DRM_DEBUG("Unregistered interrupt src_id: %d of client_id:%d\n", + dev_dbg(adev->dev, + "Unregistered interrupt src_id: %d of client_id:%d\n", src_id, client_id); } @@ -732,7 +736,7 @@ int amdgpu_irq_add_domain(struct amdgpu_device *adev) adev->irq.domain = irq_domain_create_linear(NULL, AMDGPU_MAX_IRQ_SRC_ID, &amdgpu_hw_irqdomain_ops, adev); if (!adev->irq.domain) { - DRM_ERROR("GPU irq add domain failed\n"); + dev_err(adev->dev, "GPU irq add domain failed\n"); return -ENODEV; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h index 04c0b4fa17a4e7bf9628c52f9d03cbcd443f81ee..9f0417456abda16cd562c330425a768f58ad2925 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.h @@ -146,5 +146,6 @@ void amdgpu_irq_gpu_reset_resume_helper(struct amdgpu_device *adev); int amdgpu_irq_add_domain(struct amdgpu_device *adev); void amdgpu_irq_remove_domain(struct amdgpu_device *adev); unsigned amdgpu_irq_create_mapping(struct amdgpu_device *adev, unsigned src_id); +void amdgpu_restore_msix(struct amdgpu_device *adev); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_isp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_isp.h index 4f3b7b5d9c1fec0dc0d4f277c6ef453a1aa3ac11..1d1c4b1ec7e74e2d5865971710d4eec6ac106da0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_isp.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_isp.h @@ -28,6 +28,8 @@ #ifndef __AMDGPU_ISP_H__ #define __AMDGPU_ISP_H__ +#include + #define ISP_REGS_OFFSET_END 0x629A4 struct amdgpu_isp; @@ -54,6 +56,7 @@ struct amdgpu_isp { struct isp_platform_data *isp_pdata; unsigned int harvest_config; const struct firmware *fw; + struct generic_pm_domain ispgpd; }; extern const struct amdgpu_ip_block_version isp_v4_1_0_ip_block; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c index 1e24590ae1449f49e4632fbf2b931e04c03af8d5..2b58e353cca154223ee5314f0285cc1f805430f6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c @@ -92,8 +92,7 @@ static enum drm_gpu_sched_stat amdgpu_job_timedout(struct drm_sched_job *s_job) struct drm_wedge_task_info *info = NULL; struct amdgpu_task_info *ti; struct amdgpu_device *adev = ring->adev; - int idx; - int r; + int idx, r; if (!drm_dev_enter(adev_to_drm(adev), &idx)) { dev_info(adev->dev, "%s - device unplugged skipping recovery on scheduler:%s", @@ -133,43 +132,20 @@ static enum drm_gpu_sched_stat amdgpu_job_timedout(struct drm_sched_job *s_job) if (unlikely(adev->debug_disable_gpu_ring_reset)) { dev_err(adev->dev, "Ring reset disabled by debug mask\n"); } else if (amdgpu_gpu_recovery && ring->funcs->reset) { - bool is_guilty; - - dev_err(adev->dev, "Starting %s ring reset\n", s_job->sched->name); - /* stop the scheduler, but don't mess with the - * bad job yet because if ring reset fails - * we'll fall back to full GPU reset. - */ - drm_sched_wqueue_stop(&ring->sched); - - /* for engine resets, we need to reset the engine, - * but individual queues may be unaffected. - * check here to make sure the accounting is correct. - */ - if (ring->funcs->is_guilty) - is_guilty = ring->funcs->is_guilty(ring); - else - is_guilty = true; - - if (is_guilty) - dma_fence_set_error(&s_job->s_fence->finished, -ETIME); - - r = amdgpu_ring_reset(ring, job->vmid); + dev_err(adev->dev, "Starting %s ring reset\n", + s_job->sched->name); + r = amdgpu_ring_reset(ring, job->vmid, NULL); if (!r) { - if (amdgpu_ring_sched_ready(ring)) - drm_sched_stop(&ring->sched, s_job); - if (is_guilty) { - atomic_inc(&ring->adev->gpu_reset_counter); - amdgpu_fence_driver_force_completion(ring); - } - if (amdgpu_ring_sched_ready(ring)) - drm_sched_start(&ring->sched, 0); - dev_err(adev->dev, "Ring %s reset succeeded\n", ring->sched.name); - drm_dev_wedged_event(adev_to_drm(adev), DRM_WEDGE_RECOVERY_NONE, info); + atomic_inc(&ring->adev->gpu_reset_counter); + dev_err(adev->dev, "Ring %s reset succeeded\n", + ring->sched.name); + drm_dev_wedged_event(adev_to_drm(adev), + DRM_WEDGE_RECOVERY_NONE, info); goto exit; } - dev_err(adev->dev, "Ring %s reset failure\n", ring->sched.name); + dev_err(adev->dev, "Ring %s reset failed\n", ring->sched.name); } + dma_fence_set_error(&s_job->s_fence->finished, -ETIME); amdgpu_vm_put_task_info(ti); @@ -275,8 +251,8 @@ void amdgpu_job_free_resources(struct amdgpu_job *job) /* Check if any fences where initialized */ if (job->base.s_fence && job->base.s_fence->finished.ops) f = &job->base.s_fence->finished; - else if (job->hw_fence.ops) - f = &job->hw_fence; + else if (job->hw_fence.base.ops) + f = &job->hw_fence.base; else f = NULL; @@ -293,10 +269,10 @@ static void amdgpu_job_free_cb(struct drm_sched_job *s_job) amdgpu_sync_free(&job->explicit_sync); /* only put the hw fence if has embedded fence */ - if (!job->hw_fence.ops) + if (!job->hw_fence.base.ops) kfree(job); else - dma_fence_put(&job->hw_fence); + dma_fence_put(&job->hw_fence.base); } void amdgpu_job_set_gang_leader(struct amdgpu_job *job, @@ -325,10 +301,10 @@ void amdgpu_job_free(struct amdgpu_job *job) if (job->gang_submit != &job->base.s_fence->scheduled) dma_fence_put(job->gang_submit); - if (!job->hw_fence.ops) + if (!job->hw_fence.base.ops) kfree(job); else - dma_fence_put(&job->hw_fence); + dma_fence_put(&job->hw_fence.base); } struct dma_fence *amdgpu_job_submit(struct amdgpu_job *job) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h index 529045d604129f64fdaeb97b855bb6e4af1fb23a..2f302266662bcb1997ce62a77760a082737189a1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h @@ -48,7 +48,7 @@ struct amdgpu_job { struct drm_sched_job base; struct amdgpu_vm *vm; struct amdgpu_sync explicit_sync; - struct dma_fence hw_fence; + struct amdgpu_fence hw_fence; struct dma_fence *gang_submit; uint32_t preamble_status; uint32_t preemption_status; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c index dda29132dfb2c8164727e54bd438bac7bce1870e..82d58ac7afb011b0a4b72cc3b1fed6b9a8dde1f9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c @@ -463,7 +463,8 @@ int amdgpu_jpeg_reg_dump_init(struct amdgpu_device *adev, adev->jpeg.ip_dump = kcalloc(adev->jpeg.num_jpeg_inst * count, sizeof(uint32_t), GFP_KERNEL); if (!adev->jpeg.ip_dump) { - DRM_ERROR("Failed to allocate memory for JPEG IP Dump\n"); + dev_err(adev->dev, + "Failed to allocate memory for JPEG IP Dump\n"); return -ENOMEM; } adev->jpeg.reg_list = reg; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index d555853c5717137d63c6f1aa108cd143a98e8e8f..4c072451a8a108c5d8171acd65a0038f689f0b94 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -1503,6 +1503,9 @@ void amdgpu_driver_postclose_kms(struct drm_device *dev, amdgpu_vm_bo_del(adev, fpriv->prt_va); amdgpu_bo_unreserve(pd); } + fpriv->evf_mgr.fd_closing = true; + amdgpu_eviction_fence_destroy(&fpriv->evf_mgr); + amdgpu_userq_mgr_fini(&fpriv->userq_mgr); amdgpu_ctx_mgr_fini(&fpriv->ctx_mgr); amdgpu_vm_fini(adev, &fpriv->vm); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c index 6fa9fa11c8f31dfd04bb12a3548e92790db7098d..135598502c8d01c0d65ebf741e5b8c754acc0c75 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c @@ -47,7 +47,7 @@ static int amdgpu_mes_doorbell_init(struct amdgpu_device *adev) /* Bitmap for dynamic allocation of kernel doorbells */ mes->doorbell_bitmap = bitmap_zalloc(PAGE_SIZE / sizeof(u32), GFP_KERNEL); if (!mes->doorbell_bitmap) { - DRM_ERROR("Failed to allocate MES doorbell bitmap\n"); + dev_err(adev->dev, "Failed to allocate MES doorbell bitmap\n"); return -ENOMEM; } @@ -256,7 +256,7 @@ int amdgpu_mes_suspend(struct amdgpu_device *adev) r = adev->mes.funcs->suspend_gang(&adev->mes, &input); amdgpu_mes_unlock(&adev->mes); if (r) - DRM_ERROR("failed to suspend all gangs"); + dev_err(adev->dev, "failed to suspend all gangs"); return r; } @@ -280,7 +280,7 @@ int amdgpu_mes_resume(struct amdgpu_device *adev) r = adev->mes.funcs->resume_gang(&adev->mes, &input); amdgpu_mes_unlock(&adev->mes); if (r) - DRM_ERROR("failed to resume all gangs"); + dev_err(adev->dev, "failed to resume all gangs"); return r; } @@ -304,7 +304,7 @@ int amdgpu_mes_map_legacy_queue(struct amdgpu_device *adev, r = adev->mes.funcs->map_legacy_queue(&adev->mes, &queue_input); amdgpu_mes_unlock(&adev->mes); if (r) - DRM_ERROR("failed to map legacy queue\n"); + dev_err(adev->dev, "failed to map legacy queue\n"); return r; } @@ -329,7 +329,7 @@ int amdgpu_mes_unmap_legacy_queue(struct amdgpu_device *adev, r = adev->mes.funcs->unmap_legacy_queue(&adev->mes, &queue_input); amdgpu_mes_unlock(&adev->mes); if (r) - DRM_ERROR("failed to unmap legacy queue\n"); + dev_err(adev->dev, "failed to unmap legacy queue\n"); return r; } @@ -361,7 +361,7 @@ int amdgpu_mes_reset_legacy_queue(struct amdgpu_device *adev, r = adev->mes.funcs->reset_hw_queue(&adev->mes, &queue_input); amdgpu_mes_unlock(&adev->mes); if (r) - DRM_ERROR("failed to reset legacy queue\n"); + dev_err(adev->dev, "failed to reset legacy queue\n"); return r; } @@ -469,7 +469,8 @@ int amdgpu_mes_set_shader_debugger(struct amdgpu_device *adev, int r; if (!adev->mes.funcs->misc_op) { - DRM_ERROR("mes set shader debugger is not supported!\n"); + dev_err(adev->dev, + "mes set shader debugger is not supported!\n"); return -EINVAL; } @@ -493,7 +494,7 @@ int amdgpu_mes_set_shader_debugger(struct amdgpu_device *adev, r = adev->mes.funcs->misc_op(&adev->mes, &op_input); if (r) - DRM_ERROR("failed to set_shader_debugger\n"); + dev_err(adev->dev, "failed to set_shader_debugger\n"); amdgpu_mes_unlock(&adev->mes); @@ -507,7 +508,8 @@ int amdgpu_mes_flush_shader_debugger(struct amdgpu_device *adev, int r; if (!adev->mes.funcs->misc_op) { - DRM_ERROR("mes flush shader debugger is not supported!\n"); + dev_err(adev->dev, + "mes flush shader debugger is not supported!\n"); return -EINVAL; } @@ -519,7 +521,7 @@ int amdgpu_mes_flush_shader_debugger(struct amdgpu_device *adev, r = adev->mes.funcs->misc_op(&adev->mes, &op_input); if (r) - DRM_ERROR("failed to set_shader_debugger\n"); + dev_err(adev->dev, "failed to set_shader_debugger\n"); amdgpu_mes_unlock(&adev->mes); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c index d085687a47ea16dfb1eb138b69f1a13e1786bc1e..e56ba93a8df6433b12d6f5a48dde66b87591aa77 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c @@ -53,6 +53,15 @@ u64 amdgpu_nbio_get_pcie_replay_count(struct amdgpu_device *adev) return 0; } +bool amdgpu_nbio_is_replay_cnt_supported(struct amdgpu_device *adev) +{ + if (amdgpu_sriov_vf(adev) || !adev->asic_funcs->get_pcie_replay_count || + (!adev->nbio.funcs || !adev->nbio.funcs->get_pcie_replay_count)) + return false; + + return true; +} + int amdgpu_nbio_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *ras_block) { int r; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h index 79c2f807b9fe87eb479fd94c0a3511dbaf23d305..b528de6a01f6d3d57b5151afbbeba1014afb9991 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h @@ -119,4 +119,6 @@ int amdgpu_nbio_ras_sw_init(struct amdgpu_device *adev); int amdgpu_nbio_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *ras_block); u64 amdgpu_nbio_get_pcie_replay_count(struct amdgpu_device *adev); +bool amdgpu_nbio_is_replay_cnt_supported(struct amdgpu_device *adev); + #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c index 73403744331a5af6f67ee8b276576be569e46748..c5fda18967c8d859526aeabd280c97f10c65886d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c @@ -32,6 +32,7 @@ #include #include #include +#include #include #include @@ -62,7 +63,7 @@ static void amdgpu_bo_destroy(struct ttm_buffer_object *tbo) amdgpu_bo_kunmap(bo); - if (bo->tbo.base.import_attach) + if (drm_gem_is_imported(&bo->tbo.base)) drm_prime_gem_destroy(&bo->tbo.base, bo->tbo.sg); drm_gem_object_release(&bo->tbo.base); amdgpu_bo_unref(&bo->parent); @@ -939,7 +940,7 @@ int amdgpu_bo_pin(struct amdgpu_bo *bo, u32 domain) domain = bo->preferred_domains & domain; /* A shared bo cannot be migrated to VRAM */ - if (bo->tbo.base.import_attach) { + if (drm_gem_is_imported(&bo->tbo.base)) { if (domain & AMDGPU_GEM_DOMAIN_GTT) domain = AMDGPU_GEM_DOMAIN_GTT; else @@ -967,7 +968,7 @@ int amdgpu_bo_pin(struct amdgpu_bo *bo, u32 domain) */ domain = amdgpu_bo_get_preferred_domain(adev, domain); - if (bo->tbo.base.import_attach) + if (drm_gem_is_imported(&bo->tbo.base)) dma_buf_pin(bo->tbo.base.import_attach); /* force to pin into visible video ram */ @@ -1018,7 +1019,7 @@ void amdgpu_bo_unpin(struct amdgpu_bo *bo) if (bo->tbo.pin_count) return; - if (bo->tbo.base.import_attach) + if (drm_gem_is_imported(&bo->tbo.base)) dma_buf_unpin(bo->tbo.base.import_attach); if (bo->tbo.resource->mem_type == TTM_PL_VRAM) { @@ -1263,7 +1264,7 @@ void amdgpu_bo_move_notify(struct ttm_buffer_object *bo, amdgpu_bo_kunmap(abo); - if (abo->tbo.base.dma_buf && !abo->tbo.base.import_attach && + if (abo->tbo.base.dma_buf && !drm_gem_is_imported(&abo->tbo.base) && old_mem && old_mem->mem_type != TTM_PL_SYSTEM) dma_buf_move_notify(abo->tbo.base.dma_buf); @@ -1472,6 +1473,26 @@ u64 amdgpu_bo_gpu_offset(struct amdgpu_bo *bo) return amdgpu_bo_gpu_offset_no_check(bo); } +/** + * amdgpu_bo_fb_aper_addr - return FB aperture GPU offset of the VRAM bo + * @bo: amdgpu VRAM buffer object for which we query the offset + * + * Returns: + * current FB aperture GPU offset of the object. + */ +u64 amdgpu_bo_fb_aper_addr(struct amdgpu_bo *bo) +{ + struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev); + uint64_t offset, fb_base; + + WARN_ON_ONCE(bo->tbo.resource->mem_type != TTM_PL_VRAM); + + fb_base = adev->gmc.fb_start; + fb_base += adev->gmc.xgmi.physical_node_id * adev->gmc.xgmi.node_segment_size; + offset = (bo->tbo.resource->start << PAGE_SHIFT) + fb_base; + return amdgpu_gmc_sign_extend(offset); +} + /** * amdgpu_bo_gpu_offset_no_check - return GPU offset of bo * @bo: amdgpu object for which we query the offset diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h index 375448627f7bc1f9515d3e3081b7d2a4ae4ab389..c316920f34509e630deb5fc97ab2346eae2f4aeb 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h @@ -304,6 +304,7 @@ int amdgpu_bo_sync_wait_resv(struct amdgpu_device *adev, struct dma_resv *resv, bool intr); int amdgpu_bo_sync_wait(struct amdgpu_bo *bo, void *owner, bool intr); u64 amdgpu_bo_gpu_offset(struct amdgpu_bo *bo); +u64 amdgpu_bo_fb_aper_addr(struct amdgpu_bo *bo); u64 amdgpu_bo_gpu_offset_no_check(struct amdgpu_bo *bo); uint32_t amdgpu_bo_mem_stats_placement(struct amdgpu_bo *bo); uint32_t amdgpu_bo_get_preferred_domain(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c index e6f0b035e20b22e111a7396f6d78ae9b922cd223..a4bede28df173de8a4868a28f7576b7779e4c4db 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c @@ -252,6 +252,7 @@ static int psp_early_init(struct amdgpu_ip_block *ip_block) break; case IP_VERSION(14, 0, 2): case IP_VERSION(14, 0, 3): + adev->psp.sup_ifwi_up = !amdgpu_sriov_vf(adev); psp_v14_0_set_psp_funcs(psp); break; case IP_VERSION(14, 0, 5): @@ -596,6 +597,10 @@ int psp_wait_for(struct psp_context *psp, uint32_t reg_index, udelay(1); } + dev_err(adev->dev, + "psp reg (0x%x) wait timed out, mask: %x, read: %x exp: %x", + reg_index, mask, val, reg_val); + return -ETIME; } @@ -654,6 +659,10 @@ static const char *psp_gfx_cmd_name(enum psp_gfx_cmd_id cmd_id) return "BOOT_CFG"; case GFX_CMD_ID_CONFIG_SQ_PERFMON: return "CONFIG_SQ_PERFMON"; + case GFX_CMD_ID_FB_FW_RESERV_ADDR: + return "FB_FW_RESERV_ADDR"; + case GFX_CMD_ID_FB_FW_RESERV_EXT_ADDR: + return "FB_FW_RESERV_EXT_ADDR"; default: return "UNKNOWN CMD"; } @@ -871,6 +880,8 @@ static int psp_tmr_init(struct psp_context *psp) &psp->tmr_bo, &psp->tmr_mc_addr, pptr); } + if (amdgpu_virt_xgmi_migrate_enabled(psp->adev) && psp->tmr_bo) + psp->tmr_mc_addr = amdgpu_bo_fb_aper_addr(psp->tmr_bo); return ret; } @@ -984,6 +995,93 @@ int psp_get_fw_attestation_records_addr(struct psp_context *psp, return ret; } +static int psp_get_fw_reservation_info(struct psp_context *psp, + uint32_t cmd_id, + uint64_t *addr, + uint32_t *size) +{ + int ret; + uint32_t status; + struct psp_gfx_cmd_resp *cmd; + + cmd = acquire_psp_cmd_buf(psp); + + cmd->cmd_id = cmd_id; + + ret = psp_cmd_submit_buf(psp, NULL, cmd, + psp->fence_buf_mc_addr); + if (ret) { + release_psp_cmd_buf(psp); + return ret; + } + + status = cmd->resp.status; + if (status == PSP_ERR_UNKNOWN_COMMAND) { + release_psp_cmd_buf(psp); + *addr = 0; + *size = 0; + return 0; + } + + *addr = (uint64_t)cmd->resp.uresp.fw_reserve_info.reserve_base_address_hi << 32 | + cmd->resp.uresp.fw_reserve_info.reserve_base_address_lo; + *size = cmd->resp.uresp.fw_reserve_info.reserve_size; + + release_psp_cmd_buf(psp); + + return 0; +} + +int psp_update_fw_reservation(struct psp_context *psp) +{ + int ret; + uint64_t reserv_addr, reserv_addr_ext; + uint32_t reserv_size, reserv_size_ext; + struct amdgpu_device *adev = psp->adev; + + if (amdgpu_sriov_vf(psp->adev)) + return 0; + + if ((amdgpu_ip_version(adev, MP0_HWIP, 0) != IP_VERSION(14, 0, 2)) && + (amdgpu_ip_version(adev, MP0_HWIP, 0) != IP_VERSION(14, 0, 3))) + return 0; + + ret = psp_get_fw_reservation_info(psp, GFX_CMD_ID_FB_FW_RESERV_ADDR, &reserv_addr, &reserv_size); + if (ret) + return ret; + ret = psp_get_fw_reservation_info(psp, GFX_CMD_ID_FB_FW_RESERV_EXT_ADDR, &reserv_addr_ext, &reserv_size_ext); + if (ret) + return ret; + + if (reserv_addr != adev->gmc.real_vram_size - reserv_size) { + dev_warn(adev->dev, "reserve fw region is not valid!\n"); + return 0; + } + + amdgpu_bo_free_kernel(&adev->mman.fw_reserved_memory, NULL, NULL); + + reserv_size = roundup(reserv_size, SZ_1M); + + ret = amdgpu_bo_create_kernel_at(adev, reserv_addr, reserv_size, &adev->mman.fw_reserved_memory, NULL); + if (ret) { + dev_err(adev->dev, "reserve fw region failed(%d)!\n", ret); + amdgpu_bo_free_kernel(&adev->mman.fw_reserved_memory, NULL, NULL); + return ret; + } + + reserv_size_ext = roundup(reserv_size_ext, SZ_1M); + + ret = amdgpu_bo_create_kernel_at(adev, reserv_addr_ext, reserv_size_ext, + &adev->mman.fw_reserved_memory_extend, NULL); + if (ret) { + dev_err(adev->dev, "reserve extend fw region failed(%d)!\n", ret); + amdgpu_bo_free_kernel(&adev->mman.fw_reserved_memory_extend, NULL, NULL); + return ret; + } + + return 0; +} + static int psp_boot_config_get(struct amdgpu_device *adev, uint32_t *boot_cfg) { struct psp_context *psp = &adev->psp; @@ -1270,6 +1368,11 @@ int psp_ta_load(struct psp_context *psp, struct ta_context *context) psp_copy_fw(psp, context->bin_desc.start_addr, context->bin_desc.size_bytes); + if (amdgpu_virt_xgmi_migrate_enabled(psp->adev) && + context->mem_context.shared_bo) + context->mem_context.shared_mc_addr = + amdgpu_bo_fb_aper_addr(context->mem_context.shared_bo); + psp_prep_ta_load_cmd_buf(cmd, psp->fw_pri_mc_addr, context); ret = psp_cmd_submit_buf(psp, NULL, cmd, @@ -2337,11 +2440,27 @@ bool amdgpu_psp_tos_reload_needed(struct amdgpu_device *adev) return false; } +static void psp_update_gpu_addresses(struct amdgpu_device *adev) +{ + struct psp_context *psp = &adev->psp; + + if (psp->cmd_buf_bo && psp->cmd_buf_mem) { + psp->fw_pri_mc_addr = amdgpu_bo_fb_aper_addr(psp->fw_pri_bo); + psp->fence_buf_mc_addr = amdgpu_bo_fb_aper_addr(psp->fence_buf_bo); + psp->cmd_buf_mc_addr = amdgpu_bo_fb_aper_addr(psp->cmd_buf_bo); + } + if (adev->firmware.rbuf && psp->km_ring.ring_mem) + psp->km_ring.ring_mem_mc_addr = amdgpu_bo_fb_aper_addr(adev->firmware.rbuf); +} + static int psp_hw_start(struct psp_context *psp) { struct amdgpu_device *adev = psp->adev; int ret; + if (amdgpu_virt_xgmi_migrate_enabled(adev)) + psp_update_gpu_addresses(adev); + if (!amdgpu_sriov_vf(adev)) { if ((is_psp_fw_valid(psp->kdb)) && (psp->funcs->bootloader_load_kdb != NULL)) { @@ -2440,6 +2559,14 @@ static int psp_hw_start(struct psp_context *psp) return ret; } + if (!amdgpu_in_reset(adev) && !adev->in_suspend) { + ret = psp_update_fw_reservation(psp); + if (ret) { + dev_err(adev->dev, "update fw reservation failed!\n"); + return ret; + } + } + if (amdgpu_sriov_vf(adev) && amdgpu_in_reset(adev)) goto skip_pin_bo; @@ -3522,8 +3649,12 @@ int psp_init_sos_microcode(struct psp_context *psp, const char *chip_name) uint8_t *ucode_array_start_addr; int err = 0; - err = amdgpu_ucode_request(adev, &adev->psp.sos_fw, AMDGPU_UCODE_REQUIRED, - "amdgpu/%s_sos.bin", chip_name); + if (amdgpu_is_kicker_fw(adev)) + err = amdgpu_ucode_request(adev, &adev->psp.sos_fw, AMDGPU_UCODE_REQUIRED, + "amdgpu/%s_sos_kicker.bin", chip_name); + else + err = amdgpu_ucode_request(adev, &adev->psp.sos_fw, AMDGPU_UCODE_REQUIRED, + "amdgpu/%s_sos.bin", chip_name); if (err) goto out; @@ -3799,8 +3930,12 @@ int psp_init_ta_microcode(struct psp_context *psp, const char *chip_name) struct amdgpu_device *adev = psp->adev; int err; - err = amdgpu_ucode_request(adev, &adev->psp.ta_fw, AMDGPU_UCODE_REQUIRED, - "amdgpu/%s_ta.bin", chip_name); + if (amdgpu_is_kicker_fw(adev)) + err = amdgpu_ucode_request(adev, &adev->psp.ta_fw, AMDGPU_UCODE_REQUIRED, + "amdgpu/%s_ta_kicker.bin", chip_name); + else + err = amdgpu_ucode_request(adev, &adev->psp.ta_fw, AMDGPU_UCODE_REQUIRED, + "amdgpu/%s_ta.bin", chip_name); if (err) return err; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h index 428adc7f741de39ffdbfe76769bf345287de97a4..4bc0ec49d2e9d6b855241209c73d2cd610eea3a8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h @@ -51,6 +51,17 @@ #define C2PMSG_CMD_SPI_GET_ROM_IMAGE_ADDR_HI 0x10 #define C2PMSG_CMD_SPI_GET_FLASH_IMAGE 0x11 +/* Command register bit 31 set to indicate readiness */ +#define MBOX_TOS_READY_FLAG (GFX_FLAG_RESPONSE) +#define MBOX_TOS_READY_MASK (GFX_CMD_RESPONSE_MASK | GFX_CMD_STATUS_MASK) + +/* Values to check for a successful GFX_CMD response wait. Check against + * both status bits and response state - helps to detect a command failure + * or other unexpected cases like a device drop reading all 0xFFs + */ +#define MBOX_TOS_RESP_FLAG (GFX_FLAG_RESPONSE) +#define MBOX_TOS_RESP_MASK (GFX_CMD_RESPONSE_MASK | GFX_CMD_STATUS_MASK) + extern const struct attribute_group amdgpu_flash_attr_group; enum psp_shared_mem_size { @@ -588,7 +599,7 @@ int psp_init_cap_microcode(struct psp_context *psp, const char *chip_name); int psp_get_fw_attestation_records_addr(struct psp_context *psp, uint64_t *output_ptr); - +int psp_update_fw_reservation(struct psp_context *psp); int psp_load_fw_list(struct psp_context *psp, struct amdgpu_firmware_info **ucode_list, int ucode_count); void psp_copy_fw(struct psp_context *psp, uint8_t *start_addr, uint32_t bin_size); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index de0944947eaf1e52c9f544fc3096fd9cc383d7d1..c508697a34122edbe70264384d001c834b7260b2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -1107,6 +1107,9 @@ static void amdgpu_ras_error_print_error_data(struct amdgpu_device *adev, err_info->de_count, blk_name); } } else { + if (adev->debug_disable_ce_logs) + return; + for_each_ras_error(err_node, err_data) { err_info = &err_node->err_info; mcm_info = &err_info->mcm_info; @@ -3003,6 +3006,15 @@ int amdgpu_ras_save_bad_pages(struct amdgpu_device *adev, return 0; } + if (!con->eeprom_control.is_eeprom_valid) { + dev_warn(adev->dev, + "Failed to save EEPROM table data because of EEPROM data corruption!"); + if (new_cnt) + *new_cnt = 0; + + return 0; + } + mutex_lock(&con->recovery_lock); control = &con->eeprom_control; data = con->eh_data; @@ -3294,7 +3306,6 @@ static int amdgpu_ras_poison_creation_handler(struct amdgpu_device *adev, uint64_t de_queried_count; uint32_t new_detect_count, total_detect_count; uint32_t need_query_count = poison_creation_count; - bool query_data_timeout = false; enum ras_event_type type = RAS_EVENT_TYPE_POISON_CREATION; memset(&info, 0, sizeof(info)); @@ -3323,21 +3334,13 @@ static int amdgpu_ras_poison_creation_handler(struct amdgpu_device *adev, timeout = MAX_UMC_POISON_POLLING_TIME_ASYNC; if (timeout) { - if (!--timeout) { - query_data_timeout = true; + if (!--timeout) break; - } msleep(1); } } } while (total_detect_count < need_query_count); - if (query_data_timeout) { - dev_warn(adev->dev, "Can't find deferred error! count: %u\n", - (need_query_count - total_detect_count)); - return -ENOENT; - } - if (total_detect_count) schedule_delayed_work(&ras->page_retirement_dwork, 0); @@ -3488,8 +3491,7 @@ int amdgpu_ras_init_badpage_info(struct amdgpu_device *adev) control = &con->eeprom_control; ret = amdgpu_ras_eeprom_init(control); - if (ret) - return ret; + control->is_eeprom_valid = !ret; if (!adev->umc.ras || !adev->umc.ras->convert_ras_err_addr) control->ras_num_pa_recs = control->ras_num_recs; @@ -3498,10 +3500,12 @@ int amdgpu_ras_init_badpage_info(struct amdgpu_device *adev) adev->umc.ras->get_retire_flip_bits) adev->umc.ras->get_retire_flip_bits(adev); - if (control->ras_num_recs) { + if (control->ras_num_recs && control->is_eeprom_valid) { ret = amdgpu_ras_load_bad_pages(adev); - if (ret) - return ret; + if (ret) { + control->is_eeprom_valid = false; + return 0; + } amdgpu_dpm_send_hbm_bad_pages_num( adev, control->ras_num_bad_pages); @@ -3520,7 +3524,7 @@ int amdgpu_ras_init_badpage_info(struct amdgpu_device *adev) dev_warn(adev->dev, "Failed to format RAS EEPROM data in V3 version!\n"); } - return ret; + return 0; } int amdgpu_ras_recovery_init(struct amdgpu_device *adev, bool init_bp_info) @@ -4414,8 +4418,10 @@ void amdgpu_ras_clear_err_state(struct amdgpu_device *adev) struct amdgpu_ras *ras; ras = amdgpu_ras_get_context(adev); - if (ras) + if (ras) { ras->ras_err_state = 0; + ras->gpu_reset_flags = 0; + } } void amdgpu_ras_set_err_poison(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c index 2c58e09e56f95db6be33a0824f162bdcb83a759e..54838746f97dd317d889a1a683596c33d0edc2a9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c @@ -277,10 +277,11 @@ static int __write_table_header(struct amdgpu_ras_eeprom_control *control) up_read(&adev->reset_domain->sem); if (res < 0) { - DRM_ERROR("Failed to write EEPROM table header:%d", res); + dev_err(adev->dev, "Failed to write EEPROM table header:%d", + res); } else if (res < RAS_TABLE_HEADER_SIZE) { - DRM_ERROR("Short write:%d out of %d\n", - res, RAS_TABLE_HEADER_SIZE); + dev_err(adev->dev, "Short write:%d out of %d\n", res, + RAS_TABLE_HEADER_SIZE); res = -EIO; } else { res = 0; @@ -323,7 +324,8 @@ static int __write_table_ras_info(struct amdgpu_ras_eeprom_control *control) buf = kzalloc(RAS_TABLE_V2_1_INFO_SIZE, GFP_KERNEL); if (!buf) { - DRM_ERROR("Failed to alloc buf to write table ras info\n"); + dev_err(adev->dev, + "Failed to alloc buf to write table ras info\n"); return -ENOMEM; } @@ -338,10 +340,11 @@ static int __write_table_ras_info(struct amdgpu_ras_eeprom_control *control) up_read(&adev->reset_domain->sem); if (res < 0) { - DRM_ERROR("Failed to write EEPROM table ras info:%d", res); + dev_err(adev->dev, "Failed to write EEPROM table ras info:%d", + res); } else if (res < RAS_TABLE_V2_1_INFO_SIZE) { - DRM_ERROR("Short write:%d out of %d\n", - res, RAS_TABLE_V2_1_INFO_SIZE); + dev_err(adev->dev, "Short write:%d out of %d\n", res, + RAS_TABLE_V2_1_INFO_SIZE); res = -EIO; } else { res = 0; @@ -476,6 +479,8 @@ int amdgpu_ras_eeprom_reset_table(struct amdgpu_ras_eeprom_control *control) control->ras_num_recs = 0; control->ras_num_bad_pages = 0; + control->ras_num_mca_recs = 0; + control->ras_num_pa_recs = 0; control->ras_fri = 0; amdgpu_dpm_send_hbm_bad_pages_num(adev, control->ras_num_bad_pages); @@ -607,13 +612,13 @@ static int __amdgpu_ras_eeprom_write(struct amdgpu_ras_eeprom_control *control, buf, buf_size); up_read(&adev->reset_domain->sem); if (res < 0) { - DRM_ERROR("Writing %d EEPROM table records error:%d", - num, res); + dev_err(adev->dev, "Writing %d EEPROM table records error:%d", + num, res); } else if (res < buf_size) { /* Short write, return error. */ - DRM_ERROR("Wrote %d records out of %d", - res / RAS_TABLE_RECORD_SIZE, num); + dev_err(adev->dev, "Wrote %d records out of %d", + res / RAS_TABLE_RECORD_SIZE, num); res = -EIO; } else { res = 0; @@ -761,18 +766,17 @@ amdgpu_ras_eeprom_update_header(struct amdgpu_ras_eeprom_control *control) dev_warn(adev->dev, "Saved bad pages %d reaches threshold value %d\n", control->ras_num_bad_pages, ras->bad_page_cnt_threshold); - control->tbl_hdr.header = RAS_TABLE_HDR_BAD; - if (control->tbl_hdr.version >= RAS_TABLE_VER_V2_1) { - control->tbl_rai.rma_status = GPU_RETIRED__ECC_REACH_THRESHOLD; - control->tbl_rai.health_percent = 0; - } - if ((amdgpu_bad_page_threshold != -1) && - (amdgpu_bad_page_threshold != -2)) + (amdgpu_bad_page_threshold != -2)) { + control->tbl_hdr.header = RAS_TABLE_HDR_BAD; + if (control->tbl_hdr.version >= RAS_TABLE_VER_V2_1) { + control->tbl_rai.rma_status = GPU_RETIRED__ECC_REACH_THRESHOLD; + control->tbl_rai.health_percent = 0; + } ras->is_rma = true; - - /* ignore the -ENOTSUPP return value */ - amdgpu_dpm_send_rma_reason(adev); + /* ignore the -ENOTSUPP return value */ + amdgpu_dpm_send_rma_reason(adev); + } } if (control->tbl_hdr.version >= RAS_TABLE_VER_V2_1) @@ -787,8 +791,9 @@ amdgpu_ras_eeprom_update_header(struct amdgpu_ras_eeprom_control *control) buf_size = control->ras_num_recs * RAS_TABLE_RECORD_SIZE; buf = kcalloc(control->ras_num_recs, RAS_TABLE_RECORD_SIZE, GFP_KERNEL); if (!buf) { - DRM_ERROR("allocating memory for table of size %d bytes failed\n", - control->tbl_hdr.tbl_size); + dev_err(adev->dev, + "allocating memory for table of size %d bytes failed\n", + control->tbl_hdr.tbl_size); res = -ENOMEM; goto Out; } @@ -800,12 +805,11 @@ amdgpu_ras_eeprom_update_header(struct amdgpu_ras_eeprom_control *control) buf, buf_size); up_read(&adev->reset_domain->sem); if (res < 0) { - DRM_ERROR("EEPROM failed reading records:%d\n", - res); + dev_err(adev->dev, "EEPROM failed reading records:%d\n", res); goto Out; } else if (res < buf_size) { - DRM_ERROR("EEPROM read %d out of %d bytes\n", - res, buf_size); + dev_err(adev->dev, "EEPROM read %d out of %d bytes\n", res, + buf_size); res = -EIO; goto Out; } @@ -866,11 +870,12 @@ int amdgpu_ras_eeprom_append(struct amdgpu_ras_eeprom_control *control, return 0; if (num == 0) { - DRM_ERROR("will not append 0 records\n"); + dev_err(adev->dev, "will not append 0 records\n"); return -EINVAL; } else if (num > control->ras_max_record_count) { - DRM_ERROR("cannot append %d records than the size of table %d\n", - num, control->ras_max_record_count); + dev_err(adev->dev, + "cannot append %d records than the size of table %d\n", + num, control->ras_max_record_count); return -EINVAL; } @@ -924,13 +929,13 @@ static int __amdgpu_ras_eeprom_read(struct amdgpu_ras_eeprom_control *control, buf, buf_size); up_read(&adev->reset_domain->sem); if (res < 0) { - DRM_ERROR("Reading %d EEPROM table records error:%d", - num, res); + dev_err(adev->dev, "Reading %d EEPROM table records error:%d", + num, res); } else if (res < buf_size) { /* Short read, return error. */ - DRM_ERROR("Read %d records out of %d", - res / RAS_TABLE_RECORD_SIZE, num); + dev_err(adev->dev, "Read %d records out of %d", + res / RAS_TABLE_RECORD_SIZE, num); res = -EIO; } else { res = 0; @@ -964,11 +969,11 @@ int amdgpu_ras_eeprom_read(struct amdgpu_ras_eeprom_control *control, return 0; if (num == 0) { - DRM_ERROR("will not read 0 records\n"); + dev_err(adev->dev, "will not read 0 records\n"); return -EINVAL; } else if (num > control->ras_num_recs) { - DRM_ERROR("too many records to read:%d available:%d\n", - num, control->ras_num_recs); + dev_err(adev->dev, "too many records to read:%d available:%d\n", + num, control->ras_num_recs); return -EINVAL; } @@ -1300,7 +1305,8 @@ static int __verify_ras_table_checksum(struct amdgpu_ras_eeprom_control *control buf = kzalloc(buf_size, GFP_KERNEL); if (!buf) { - DRM_ERROR("Out of memory checking RAS table checksum.\n"); + dev_err(adev->dev, + "Out of memory checking RAS table checksum.\n"); return -ENOMEM; } @@ -1309,7 +1315,7 @@ static int __verify_ras_table_checksum(struct amdgpu_ras_eeprom_control *control control->ras_header_offset, buf, buf_size); if (res < buf_size) { - DRM_ERROR("Partial read for checksum, res:%d\n", res); + dev_err(adev->dev, "Partial read for checksum, res:%d\n", res); /* On partial reads, return -EIO. */ if (res >= 0) @@ -1334,7 +1340,8 @@ static int __read_table_ras_info(struct amdgpu_ras_eeprom_control *control) buf = kzalloc(RAS_TABLE_V2_1_INFO_SIZE, GFP_KERNEL); if (!buf) { - DRM_ERROR("Failed to alloc buf to read EEPROM table ras info\n"); + dev_err(adev->dev, + "Failed to alloc buf to read EEPROM table ras info\n"); return -ENOMEM; } @@ -1346,7 +1353,8 @@ static int __read_table_ras_info(struct amdgpu_ras_eeprom_control *control) control->i2c_address + control->ras_info_offset, buf, RAS_TABLE_V2_1_INFO_SIZE); if (res < RAS_TABLE_V2_1_INFO_SIZE) { - DRM_ERROR("Failed to read EEPROM table ras info, res:%d", res); + dev_err(adev->dev, + "Failed to read EEPROM table ras info, res:%d", res); res = res >= 0 ? -EIO : res; goto Out; } @@ -1387,7 +1395,8 @@ int amdgpu_ras_eeprom_init(struct amdgpu_ras_eeprom_control *control) control->i2c_address + control->ras_header_offset, buf, RAS_TABLE_HEADER_SIZE); if (res < RAS_TABLE_HEADER_SIZE) { - DRM_ERROR("Failed to read EEPROM table header, res:%d", res); + dev_err(adev->dev, "Failed to read EEPROM table header, res:%d", + res); return res >= 0 ? -EIO : res; } @@ -1452,8 +1461,9 @@ int amdgpu_ras_eeprom_check(struct amdgpu_ras_eeprom_control *control) control->ras_num_mca_recs * adev->umc.retire_unit; if (hdr->header == RAS_TABLE_HDR_VAL) { - DRM_DEBUG_DRIVER("Found existing EEPROM table with %d records", - control->ras_num_bad_pages); + dev_dbg(adev->dev, + "Found existing EEPROM table with %d records", + control->ras_num_bad_pages); if (hdr->version >= RAS_TABLE_VER_V2_1) { res = __read_table_ras_info(control); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h index ec6d7ea37ad071d102e25277e2562f059b808f8e..35c69ac3dbeb5e836c22b326401442cc3b361dac 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.h @@ -114,6 +114,8 @@ struct amdgpu_ras_eeprom_control { /* Record channel info which occurred bad pages */ u32 bad_channel_bitmap; + + bool is_eeprom_valid; }; /* diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h index b95b4711076923d9d52e87d7732c67f3bcb47ecc..784ba2ec354c7e136d5255dfe4501d038fc6f081 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h @@ -127,6 +127,22 @@ struct amdgpu_fence_driver { struct dma_fence **fences; }; +/* + * Fences mark an event in the GPUs pipeline and are used + * for GPU/CPU synchronization. When the fence is written, + * it is expected that all buffers associated with that fence + * are no longer in use by the associated ring on the GPU and + * that the relevant GPU caches have been flushed. + */ + +struct amdgpu_fence { + struct dma_fence base; + + /* RB, DMA, etc. */ + struct amdgpu_ring *ring; + ktime_t start_timestamp; +}; + extern const struct drm_sched_backend_ops amdgpu_sched_ops; void amdgpu_fence_driver_clear_job_fences(struct amdgpu_ring *ring); @@ -141,8 +157,8 @@ void amdgpu_fence_driver_hw_init(struct amdgpu_device *adev); void amdgpu_fence_driver_hw_fini(struct amdgpu_device *adev); int amdgpu_fence_driver_sw_init(struct amdgpu_device *adev); void amdgpu_fence_driver_sw_fini(struct amdgpu_device *adev); -int amdgpu_fence_emit(struct amdgpu_ring *ring, struct dma_fence **fence, struct amdgpu_job *job, - unsigned flags); +int amdgpu_fence_emit(struct amdgpu_ring *ring, struct dma_fence **f, + struct amdgpu_fence *af, unsigned int flags); int amdgpu_fence_emit_polling(struct amdgpu_ring *ring, uint32_t *s, uint32_t timeout); bool amdgpu_fence_process(struct amdgpu_ring *ring); @@ -252,9 +268,9 @@ struct amdgpu_ring_funcs { void (*patch_cntl)(struct amdgpu_ring *ring, unsigned offset); void (*patch_ce)(struct amdgpu_ring *ring, unsigned offset); void (*patch_de)(struct amdgpu_ring *ring, unsigned offset); - int (*reset)(struct amdgpu_ring *ring, unsigned int vmid); + int (*reset)(struct amdgpu_ring *ring, unsigned int vmid, + struct amdgpu_fence *timedout_fence); void (*emit_cleaner_shader)(struct amdgpu_ring *ring); - bool (*is_guilty)(struct amdgpu_ring *ring); }; /** @@ -409,7 +425,7 @@ struct amdgpu_ring { #define amdgpu_ring_patch_cntl(r, o) ((r)->funcs->patch_cntl((r), (o))) #define amdgpu_ring_patch_ce(r, o) ((r)->funcs->patch_ce((r), (o))) #define amdgpu_ring_patch_de(r, o) ((r)->funcs->patch_de((r), (o))) -#define amdgpu_ring_reset(r, v) (r)->funcs->reset((r), (v)) +#define amdgpu_ring_reset(r, v, f) (r)->funcs->reset((r), (v), (f)) unsigned int amdgpu_ring_max_ibs(enum amdgpu_ring_type type); int amdgpu_ring_alloc(struct amdgpu_ring *ring, unsigned ndw); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c index 6716ac281c49863dfd403cd0b45c5de2e92d5c82..7e26a44dcc1fd1306d892e4f9e16efc9c1c1d88c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c @@ -534,41 +534,17 @@ bool amdgpu_sdma_is_shared_inv_eng(struct amdgpu_device *adev, struct amdgpu_rin static int amdgpu_sdma_soft_reset(struct amdgpu_device *adev, u32 instance_id) { struct amdgpu_sdma_instance *sdma_instance = &adev->sdma.instance[instance_id]; - int r = -EOPNOTSUPP; - - switch (amdgpu_ip_version(adev, SDMA0_HWIP, 0)) { - case IP_VERSION(4, 4, 2): - case IP_VERSION(4, 4, 4): - case IP_VERSION(4, 4, 5): - /* For SDMA 4.x, use the existing DPM interface for backward compatibility */ - r = amdgpu_dpm_reset_sdma(adev, 1 << instance_id); - break; - case IP_VERSION(5, 0, 0): - case IP_VERSION(5, 0, 1): - case IP_VERSION(5, 0, 2): - case IP_VERSION(5, 0, 5): - case IP_VERSION(5, 2, 0): - case IP_VERSION(5, 2, 2): - case IP_VERSION(5, 2, 4): - case IP_VERSION(5, 2, 5): - case IP_VERSION(5, 2, 6): - case IP_VERSION(5, 2, 3): - case IP_VERSION(5, 2, 1): - case IP_VERSION(5, 2, 7): - if (sdma_instance->funcs->soft_reset_kernel_queue) - r = sdma_instance->funcs->soft_reset_kernel_queue(adev, instance_id); - break; - default: - break; - } - return r; + if (sdma_instance->funcs->soft_reset_kernel_queue) + return sdma_instance->funcs->soft_reset_kernel_queue(adev, instance_id); + + return -EOPNOTSUPP; } /** * amdgpu_sdma_reset_engine - Reset a specific SDMA engine * @adev: Pointer to the AMDGPU device - * @instance_id: ID of the SDMA engine instance to reset + * @instance_id: Logical ID of the SDMA engine instance to reset * * Returns: 0 on success, or a negative error code on failure. */ @@ -578,35 +554,35 @@ int amdgpu_sdma_reset_engine(struct amdgpu_device *adev, uint32_t instance_id) struct amdgpu_sdma_instance *sdma_instance = &adev->sdma.instance[instance_id]; struct amdgpu_ring *gfx_ring = &sdma_instance->ring; struct amdgpu_ring *page_ring = &sdma_instance->page; - bool gfx_sched_stopped = false, page_sched_stopped = false; mutex_lock(&sdma_instance->engine_reset_mutex); /* Stop the scheduler's work queue for the GFX and page rings if they are running. * This ensures that no new tasks are submitted to the queues while * the reset is in progress. */ - if (!amdgpu_ring_sched_ready(gfx_ring)) { - drm_sched_wqueue_stop(&gfx_ring->sched); - gfx_sched_stopped = true; - } + drm_sched_wqueue_stop(&gfx_ring->sched); - if (adev->sdma.has_page_queue && !amdgpu_ring_sched_ready(page_ring)) { + if (adev->sdma.has_page_queue) drm_sched_wqueue_stop(&page_ring->sched); - page_sched_stopped = true; - } - if (sdma_instance->funcs->stop_kernel_queue) + if (sdma_instance->funcs->stop_kernel_queue) { sdma_instance->funcs->stop_kernel_queue(gfx_ring); + if (adev->sdma.has_page_queue) + sdma_instance->funcs->stop_kernel_queue(page_ring); + } /* Perform the SDMA reset for the specified instance */ ret = amdgpu_sdma_soft_reset(adev, instance_id); if (ret) { - dev_err(adev->dev, "Failed to reset SDMA instance %u\n", instance_id); + dev_err(adev->dev, "Failed to reset SDMA logical instance %u\n", instance_id); goto exit; } - if (sdma_instance->funcs->start_kernel_queue) + if (sdma_instance->funcs->start_kernel_queue) { sdma_instance->funcs->start_kernel_queue(gfx_ring); + if (adev->sdma.has_page_queue) + sdma_instance->funcs->start_kernel_queue(page_ring); + } exit: /* Restart the scheduler's work queue for the GFX and page rings @@ -614,12 +590,9 @@ exit: * to be submitted to the queues after the reset is complete. */ if (!ret) { - if (gfx_sched_stopped && amdgpu_ring_sched_ready(gfx_ring)) { - drm_sched_wqueue_start(&gfx_ring->sched); - } - if (page_sched_stopped && amdgpu_ring_sched_ready(page_ring)) { + drm_sched_wqueue_start(&gfx_ring->sched); + if (adev->sdma.has_page_queue) drm_sched_wqueue_start(&page_ring->sched); - } } mutex_unlock(&sdma_instance->engine_reset_mutex); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index 9c5df35f05b7b86396957756e6ae1c2d319b27b9..27ab4e754b2a9f431d7a5269ebf7f711ed3c06c8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -299,7 +299,8 @@ int amdgpu_ttm_copy_mem_to_mem(struct amdgpu_device *adev, struct amdgpu_bo *abo_src, *abo_dst; if (!adev->mman.buffer_funcs_enabled) { - DRM_ERROR("Trying to move memory with ring turned off.\n"); + dev_err(adev->dev, + "Trying to move memory with ring turned off.\n"); return -EINVAL; } @@ -934,7 +935,7 @@ static int amdgpu_ttm_backend_bind(struct ttm_device *bdev, if (gtt->userptr) { r = amdgpu_ttm_tt_pin_userptr(bdev, ttm); if (r) { - DRM_ERROR("failed to pin userptr\n"); + dev_err(adev->dev, "failed to pin userptr\n"); return r; } } else if (ttm->page_flags & TTM_TT_FLAG_EXTERNAL) { @@ -1060,7 +1061,7 @@ static void amdgpu_ttm_backend_unbind(struct ttm_device *bdev, /* if the pages have userptr pinning then clear that first */ if (gtt->userptr) { amdgpu_ttm_tt_unpin_userptr(bdev, ttm); - } else if (ttm->sg && gtt->gobj->import_attach) { + } else if (ttm->sg && drm_gem_is_imported(gtt->gobj)) { struct dma_buf_attachment *attach; attach = gtt->gobj->import_attach; @@ -1781,7 +1782,7 @@ static int amdgpu_ttm_reserve_tmr(struct amdgpu_device *adev) &ctx->c2p_bo, NULL); if (ret) { - DRM_ERROR("alloc c2p_bo failed(%d)!\n", ret); + dev_err(adev->dev, "alloc c2p_bo failed(%d)!\n", ret); amdgpu_ttm_training_reserve_vram_fini(adev); return ret; } @@ -1793,7 +1794,7 @@ static int amdgpu_ttm_reserve_tmr(struct amdgpu_device *adev) adev, adev->gmc.real_vram_size - reserve_size, reserve_size, &adev->mman.fw_reserved_memory, NULL); if (ret) { - DRM_ERROR("alloc tmr failed(%d)!\n", ret); + dev_err(adev->dev, "alloc tmr failed(%d)!\n", ret); amdgpu_bo_free_kernel(&adev->mman.fw_reserved_memory, NULL, NULL); return ret; @@ -1864,13 +1865,14 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) adev->need_swiotlb, dma_addressing_limited(adev->dev)); if (r) { - DRM_ERROR("failed initializing buffer object driver(%d).\n", r); + dev_err(adev->dev, + "failed initializing buffer object driver(%d).\n", r); return r; } r = amdgpu_ttm_pools_init(adev); if (r) { - DRM_ERROR("failed to init ttm pools(%d).\n", r); + dev_err(adev->dev, "failed to init ttm pools(%d).\n", r); return r; } adev->mman.initialized = true; @@ -1878,7 +1880,7 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) /* Initialize VRAM pool with all of VRAM divided into pages */ r = amdgpu_vram_mgr_init(adev); if (r) { - DRM_ERROR("Failed initializing VRAM heap.\n"); + dev_err(adev->dev, "Failed initializing VRAM heap.\n"); return r; } @@ -1958,7 +1960,7 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) DRM_DEBUG_DRIVER("Skipped stolen memory reservation\n"); } - DRM_INFO("amdgpu: %uM of VRAM memory ready\n", + dev_info(adev->dev, "amdgpu: %uM of VRAM memory ready\n", (unsigned int)(adev->gmc.real_vram_size / (1024 * 1024))); /* Compute GTT size, either based on TTM limit @@ -1981,10 +1983,10 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) /* Initialize GTT memory pool */ r = amdgpu_gtt_mgr_init(adev, gtt_size); if (r) { - DRM_ERROR("Failed initializing GTT heap.\n"); + dev_err(adev->dev, "Failed initializing GTT heap.\n"); return r; } - DRM_INFO("amdgpu: %uM of GTT memory ready.\n", + dev_info(adev->dev, "amdgpu: %uM of GTT memory ready.\n", (unsigned int)(gtt_size / (1024 * 1024))); if (adev->flags & AMD_IS_APU) { @@ -1995,40 +1997,40 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) /* Initialize doorbell pool on PCI BAR */ r = amdgpu_ttm_init_on_chip(adev, AMDGPU_PL_DOORBELL, adev->doorbell.size / PAGE_SIZE); if (r) { - DRM_ERROR("Failed initializing doorbell heap.\n"); + dev_err(adev->dev, "Failed initializing doorbell heap.\n"); return r; } /* Create a boorbell page for kernel usages */ r = amdgpu_doorbell_create_kernel_doorbells(adev); if (r) { - DRM_ERROR("Failed to initialize kernel doorbells.\n"); + dev_err(adev->dev, "Failed to initialize kernel doorbells.\n"); return r; } /* Initialize preemptible memory pool */ r = amdgpu_preempt_mgr_init(adev); if (r) { - DRM_ERROR("Failed initializing PREEMPT heap.\n"); + dev_err(adev->dev, "Failed initializing PREEMPT heap.\n"); return r; } /* Initialize various on-chip memory pools */ r = amdgpu_ttm_init_on_chip(adev, AMDGPU_PL_GDS, adev->gds.gds_size); if (r) { - DRM_ERROR("Failed initializing GDS heap.\n"); + dev_err(adev->dev, "Failed initializing GDS heap.\n"); return r; } r = amdgpu_ttm_init_on_chip(adev, AMDGPU_PL_GWS, adev->gds.gws_size); if (r) { - DRM_ERROR("Failed initializing gws heap.\n"); + dev_err(adev->dev, "Failed initializing gws heap.\n"); return r; } r = amdgpu_ttm_init_on_chip(adev, AMDGPU_PL_OA, adev->gds.oa_size); if (r) { - DRM_ERROR("Failed initializing oa heap.\n"); + dev_err(adev->dev, "Failed initializing oa heap.\n"); return r; } if (amdgpu_bo_create_kernel(adev, PAGE_SIZE, PAGE_SIZE, @@ -2060,6 +2062,8 @@ void amdgpu_ttm_fini(struct amdgpu_device *adev) /* return the FW reserved memory back to VRAM */ amdgpu_bo_free_kernel(&adev->mman.fw_reserved_memory, NULL, NULL); + amdgpu_bo_free_kernel(&adev->mman.fw_reserved_memory_extend, NULL, + NULL); if (adev->mman.stolen_reserved_size) amdgpu_bo_free_kernel(&adev->mman.stolen_reserved_memory, NULL, NULL); @@ -2089,7 +2093,7 @@ void amdgpu_ttm_fini(struct amdgpu_device *adev) ttm_range_man_fini(&adev->mman.bdev, AMDGPU_PL_DOORBELL); ttm_device_fini(&adev->mman.bdev); adev->mman.initialized = false; - DRM_INFO("amdgpu: ttm finalized\n"); + dev_info(adev->dev, "amdgpu: ttm finalized\n"); } /** @@ -2121,8 +2125,9 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable) DRM_SCHED_PRIORITY_KERNEL, &sched, 1, NULL); if (r) { - DRM_ERROR("Failed setting up TTM BO move entity (%d)\n", - r); + dev_err(adev->dev, + "Failed setting up TTM BO move entity (%d)\n", + r); return; } @@ -2130,8 +2135,9 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable) DRM_SCHED_PRIORITY_NORMAL, &sched, 1, NULL); if (r) { - DRM_ERROR("Failed setting up TTM BO move entity (%d)\n", - r); + dev_err(adev->dev, + "Failed setting up TTM BO move entity (%d)\n", + r); goto error_free_entity; } } else { @@ -2202,7 +2208,8 @@ int amdgpu_copy_buffer(struct amdgpu_ring *ring, uint64_t src_offset, int r; if (!direct_submit && !ring->sched.ready) { - DRM_ERROR("Trying to move memory with ring turned off.\n"); + dev_err(adev->dev, + "Trying to move memory with ring turned off.\n"); return -EINVAL; } @@ -2237,7 +2244,7 @@ int amdgpu_copy_buffer(struct amdgpu_ring *ring, uint64_t src_offset, error_free: amdgpu_job_free(job); - DRM_ERROR("Error scheduling IBs (%d)\n", r); + dev_err(adev->dev, "Error scheduling IBs (%d)\n", r); return r; } @@ -2356,7 +2363,8 @@ int amdgpu_fill_buffer(struct amdgpu_bo *bo, int r; if (!adev->mman.buffer_funcs_enabled) { - DRM_ERROR("Trying to clear memory with ring turned off.\n"); + dev_err(adev->dev, + "Trying to clear memory with ring turned off.\n"); return -EINVAL; } @@ -2416,7 +2424,7 @@ int amdgpu_ttm_evict_resources(struct amdgpu_device *adev, int mem_type) man = ttm_manager_type(&adev->mman.bdev, mem_type); break; default: - DRM_ERROR("Trying to evict invalid memory type\n"); + dev_err(adev->dev, "Trying to evict invalid memory type\n"); return -EINVAL; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h index 208b7d1d8a277bd8463b836a7d3d0a49822325dc..215c198e4aff7b995309c37c138685b81551352a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h @@ -86,6 +86,7 @@ struct amdgpu_mman { uint32_t discovery_tmr_size; /* fw reserved memory */ struct amdgpu_bo *fw_reserved_memory; + struct amdgpu_bo *fw_reserved_memory_extend; /* firmware VRAM reservation */ u64 fw_vram_usage_start_offset; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c index 2505c46a9c3dec63f420e0c34b9d1e920627ea3a..a0b50a8ac9c4e344e6385b2b8f92e642d3a06172 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c @@ -30,6 +30,10 @@ #define AMDGPU_UCODE_NAME_MAX (128) +static const struct kicker_device kicker_device_list[] = { + {0x744B, 0x00}, +}; + static void amdgpu_ucode_print_common_hdr(const struct common_firmware_header *hdr) { DRM_DEBUG("size_bytes: %u\n", le32_to_cpu(hdr->size_bytes)); @@ -1155,6 +1159,9 @@ int amdgpu_ucode_init_bo(struct amdgpu_device *adev) adev->firmware.max_ucodes = AMDGPU_UCODE_ID_MAXIMUM; } + if (amdgpu_virt_xgmi_migrate_enabled(adev) && adev->firmware.fw_buf) + adev->firmware.fw_buf_mc = amdgpu_bo_fb_aper_addr(adev->firmware.fw_buf); + for (i = 0; i < adev->firmware.max_ucodes; i++) { ucode = &adev->firmware.ucode[i]; if (ucode->fw) { @@ -1387,6 +1394,19 @@ static const char *amdgpu_ucode_legacy_naming(struct amdgpu_device *adev, int bl return NULL; } +bool amdgpu_is_kicker_fw(struct amdgpu_device *adev) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(kicker_device_list); i++) { + if (adev->pdev->device == kicker_device_list[i].device && + adev->pdev->revision == kicker_device_list[i].revision) + return true; + } + + return false; +} + void amdgpu_ucode_ip_version_decode(struct amdgpu_device *adev, int block_type, char *ucode_prefix, int len) { int maj, min, rev; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h index 9e89c3487be5bf998581a9b9f7b370b86e991d92..6349aad6da35bca0eddeb299a6e51edc650685e1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h @@ -605,6 +605,11 @@ struct amdgpu_firmware { uint32_t pldm_version; }; +struct kicker_device{ + unsigned short device; + u8 revision; +}; + void amdgpu_ucode_print_mc_hdr(const struct common_firmware_header *hdr); void amdgpu_ucode_print_smc_hdr(const struct common_firmware_header *hdr); void amdgpu_ucode_print_imu_hdr(const struct common_firmware_header *hdr); @@ -632,5 +637,6 @@ amdgpu_ucode_get_load_type(struct amdgpu_device *adev, int load_type); const char *amdgpu_ucode_name(enum AMDGPU_UCODE_ID ucode_id); void amdgpu_ucode_ip_version_decode(struct amdgpu_device *adev, int block_type, char *ucode_prefix, int len); +bool amdgpu_is_kicker_fw(struct amdgpu_device *adev); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h index 577c6194db78c64919fa653591002940c463d95c..9320461bb4860e45ff3617aeb566f7f7934f5d64 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h @@ -301,6 +301,9 @@ struct amdgpu_virt { union amd_sriov_ras_caps ras_telemetry_en_caps; struct amdgpu_virt_ras ras; struct amd_sriov_ras_telemetry_error_count count_cache; + + /* hibernate and resume with different VF feature for xgmi enabled system */ + bool is_xgmi_node_migrate_enabled; }; struct amdgpu_video_codec_info; @@ -386,6 +389,10 @@ static inline bool is_virtual_machine(void) ((adev)->virt.gim_feature & AMDGIM_FEATURE_VCN_RB_DECOUPLE) #define amdgpu_sriov_is_mes_info_enable(adev) \ ((adev)->virt.gim_feature & AMDGIM_FEATURE_MES_INFO_ENABLE) + +#define amdgpu_virt_xgmi_migrate_enabled(adev) \ + ((adev)->virt.is_xgmi_node_migrate_enabled && (adev)->gmc.xgmi.node_segment_size != 0) + bool amdgpu_virt_mmio_blocked(struct amdgpu_device *adev); void amdgpu_virt_init_setting(struct amdgpu_device *adev); int amdgpu_virt_request_full_gpu(struct amdgpu_device *adev, bool init); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index 0ff95a56c2cee40f087777d88c6c9ff88c1f5a8a..af0f655dfd5b4a0c1ac16befda6d5bda39569722 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -1271,8 +1271,8 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va, } else { struct drm_gem_object *obj = &bo->tbo.base; - if (obj->import_attach && bo_va->is_xgmi) { - struct dma_buf *dma_buf = obj->import_attach->dmabuf; + if (drm_gem_is_imported(obj) && bo_va->is_xgmi) { + struct dma_buf *dma_buf = obj->dma_buf; struct drm_gem_object *gobj = dma_buf->priv; struct amdgpu_bo *abo = gem_to_amdgpu_bo(gobj); @@ -1631,7 +1631,7 @@ int amdgpu_vm_handle_moved(struct amdgpu_device *adev, * validation */ if (vm->is_compute_context && - bo_va->base.bo->tbo.base.import_attach && + drm_gem_is_imported(&bo_va->base.bo->tbo.base) && (!bo_va->base.bo->tbo.resource || bo_va->base.bo->tbo.resource->mem_type == TTM_PL_SYSTEM)) amdgpu_vm_bo_evicted_user(&bo_va->base); @@ -2395,10 +2395,11 @@ void amdgpu_vm_adjust_size(struct amdgpu_device *adev, uint32_t min_vm_size, else adev->vm_manager.fragment_size = amdgpu_vm_fragment_size; - DRM_INFO("vm size is %u GB, %u levels, block size is %u-bit, fragment size is %u-bit\n", - vm_size, adev->vm_manager.num_level + 1, - adev->vm_manager.block_size, - adev->vm_manager.fragment_size); + dev_info( + adev->dev, + "vm size is %u GB, %u levels, block size is %u-bit, fragment size is %u-bit\n", + vm_size, adev->vm_manager.num_level + 1, + adev->vm_manager.block_size, adev->vm_manager.fragment_size); } /** @@ -2565,8 +2566,8 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm, vm->use_cpu_for_update = !!(adev->vm_manager.vm_update_mode & AMDGPU_VM_USE_CPU_FOR_GFX); - DRM_DEBUG_DRIVER("VM update mode is %s\n", - vm->use_cpu_for_update ? "CPU" : "SDMA"); + dev_dbg(adev->dev, "VM update mode is %s\n", + vm->use_cpu_for_update ? "CPU" : "SDMA"); WARN_ONCE((vm->use_cpu_for_update && !amdgpu_gmc_vram_full_visible(&adev->gmc)), "CPU update of VM recommended only for large BAR system\n"); @@ -2608,7 +2609,7 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm, r = amdgpu_vm_create_task_info(vm); if (r) - DRM_DEBUG("Failed to create task info for VM\n"); + dev_dbg(adev->dev, "Failed to create task info for VM\n"); amdgpu_bo_unreserve(vm->root.bo); amdgpu_bo_unref(&root_bo); @@ -2659,8 +2660,8 @@ int amdgpu_vm_make_compute(struct amdgpu_device *adev, struct amdgpu_vm *vm) /* Update VM state */ vm->use_cpu_for_update = !!(adev->vm_manager.vm_update_mode & AMDGPU_VM_USE_CPU_FOR_COMPUTE); - DRM_DEBUG_DRIVER("VM update mode is %s\n", - vm->use_cpu_for_update ? "CPU" : "SDMA"); + dev_dbg(adev->dev, "VM update mode is %s\n", + vm->use_cpu_for_update ? "CPU" : "SDMA"); WARN_ONCE((vm->use_cpu_for_update && !amdgpu_gmc_vram_full_visible(&adev->gmc)), "CPU update of VM recommended only for large BAR system\n"); @@ -2983,7 +2984,7 @@ bool amdgpu_vm_handle_fault(struct amdgpu_device *adev, u32 pasid, error_unlock: amdgpu_bo_unreserve(root); if (r < 0) - DRM_ERROR("Can't handle page fault (%d)\n", r); + dev_err(adev->dev, "Can't handle page fault (%d)\n", r); error_unref: amdgpu_bo_unref(&root); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c index 322816805bfbad593ed476d9c091aea7694e6727..c8fcafeb686450fa14e74f5df70e961d74ac9b94 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c @@ -445,6 +445,222 @@ void amdgpu_xcp_release_sched(struct amdgpu_device *adev, } } +int amdgpu_xcp_select_scheds(struct amdgpu_device *adev, + u32 hw_ip, u32 hw_prio, + struct amdgpu_fpriv *fpriv, + unsigned int *num_scheds, + struct drm_gpu_scheduler ***scheds) +{ + u32 sel_xcp_id; + int i; + struct amdgpu_xcp_mgr *xcp_mgr = adev->xcp_mgr; + + if (fpriv->xcp_id == AMDGPU_XCP_NO_PARTITION) { + u32 least_ref_cnt = ~0; + + fpriv->xcp_id = 0; + for (i = 0; i < xcp_mgr->num_xcps; i++) { + u32 total_ref_cnt; + + total_ref_cnt = atomic_read(&xcp_mgr->xcp[i].ref_cnt); + if (total_ref_cnt < least_ref_cnt) { + fpriv->xcp_id = i; + least_ref_cnt = total_ref_cnt; + } + } + } + sel_xcp_id = fpriv->xcp_id; + + if (xcp_mgr->xcp[sel_xcp_id].gpu_sched[hw_ip][hw_prio].num_scheds) { + *num_scheds = + xcp_mgr->xcp[fpriv->xcp_id].gpu_sched[hw_ip][hw_prio].num_scheds; + *scheds = + xcp_mgr->xcp[fpriv->xcp_id].gpu_sched[hw_ip][hw_prio].sched; + atomic_inc(&adev->xcp_mgr->xcp[sel_xcp_id].ref_cnt); + dev_dbg(adev->dev, "Selected partition #%d", sel_xcp_id); + } else { + dev_err(adev->dev, "Failed to schedule partition #%d.", sel_xcp_id); + return -ENOENT; + } + + return 0; +} + +static void amdgpu_set_xcp_id(struct amdgpu_device *adev, + uint32_t inst_idx, + struct amdgpu_ring *ring) +{ + int xcp_id; + enum AMDGPU_XCP_IP_BLOCK ip_blk; + uint32_t inst_mask; + + ring->xcp_id = AMDGPU_XCP_NO_PARTITION; + if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE) + adev->gfx.enforce_isolation[0].xcp_id = ring->xcp_id; + if ((adev->xcp_mgr->mode == AMDGPU_XCP_MODE_NONE) || + (ring->funcs->type == AMDGPU_RING_TYPE_CPER)) + return; + + inst_mask = 1 << inst_idx; + + switch (ring->funcs->type) { + case AMDGPU_HW_IP_GFX: + case AMDGPU_RING_TYPE_COMPUTE: + case AMDGPU_RING_TYPE_KIQ: + ip_blk = AMDGPU_XCP_GFX; + break; + case AMDGPU_RING_TYPE_SDMA: + ip_blk = AMDGPU_XCP_SDMA; + break; + case AMDGPU_RING_TYPE_VCN_ENC: + case AMDGPU_RING_TYPE_VCN_JPEG: + ip_blk = AMDGPU_XCP_VCN; + break; + default: + dev_err(adev->dev, "Not support ring type %d!", ring->funcs->type); + return; + } + + for (xcp_id = 0; xcp_id < adev->xcp_mgr->num_xcps; xcp_id++) { + if (adev->xcp_mgr->xcp[xcp_id].ip[ip_blk].inst_mask & inst_mask) { + ring->xcp_id = xcp_id; + dev_dbg(adev->dev, "ring:%s xcp_id :%u", ring->name, + ring->xcp_id); + if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE) + adev->gfx.enforce_isolation[xcp_id].xcp_id = xcp_id; + break; + } + } +} + +static void amdgpu_xcp_gpu_sched_update(struct amdgpu_device *adev, + struct amdgpu_ring *ring, + unsigned int sel_xcp_id) +{ + unsigned int *num_gpu_sched; + + num_gpu_sched = &adev->xcp_mgr->xcp[sel_xcp_id] + .gpu_sched[ring->funcs->type][ring->hw_prio].num_scheds; + adev->xcp_mgr->xcp[sel_xcp_id].gpu_sched[ring->funcs->type][ring->hw_prio] + .sched[(*num_gpu_sched)++] = &ring->sched; + dev_dbg(adev->dev, "%s :[%d] gpu_sched[%d][%d] = %d", + ring->name, sel_xcp_id, ring->funcs->type, + ring->hw_prio, *num_gpu_sched); +} + +static int amdgpu_xcp_sched_list_update(struct amdgpu_device *adev) +{ + struct amdgpu_ring *ring; + int i; + + for (i = 0; i < MAX_XCP; i++) { + atomic_set(&adev->xcp_mgr->xcp[i].ref_cnt, 0); + memset(adev->xcp_mgr->xcp[i].gpu_sched, 0, sizeof(adev->xcp_mgr->xcp->gpu_sched)); + } + + if (adev->xcp_mgr->mode == AMDGPU_XCP_MODE_NONE) + return 0; + + for (i = 0; i < AMDGPU_MAX_RINGS; i++) { + ring = adev->rings[i]; + if (!ring || !ring->sched.ready || ring->no_scheduler) + continue; + + amdgpu_xcp_gpu_sched_update(adev, ring, ring->xcp_id); + + /* VCN may be shared by two partitions under CPX MODE in certain + * configs. + */ + if ((ring->funcs->type == AMDGPU_RING_TYPE_VCN_ENC || + ring->funcs->type == AMDGPU_RING_TYPE_VCN_JPEG) && + (adev->xcp_mgr->num_xcps > adev->vcn.num_vcn_inst)) + amdgpu_xcp_gpu_sched_update(adev, ring, ring->xcp_id + 1); + } + + return 0; +} + +int amdgpu_xcp_update_partition_sched_list(struct amdgpu_device *adev) +{ + int i; + + for (i = 0; i < adev->num_rings; i++) { + struct amdgpu_ring *ring = adev->rings[i]; + + if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE || + ring->funcs->type == AMDGPU_RING_TYPE_KIQ) + amdgpu_set_xcp_id(adev, ring->xcc_id, ring); + else + amdgpu_set_xcp_id(adev, ring->me, ring); + } + + return amdgpu_xcp_sched_list_update(adev); +} + +void amdgpu_xcp_update_supported_modes(struct amdgpu_xcp_mgr *xcp_mgr) +{ + struct amdgpu_device *adev = xcp_mgr->adev; + + xcp_mgr->supp_xcp_modes = 0; + + switch (NUM_XCC(adev->gfx.xcc_mask)) { + case 8: + xcp_mgr->supp_xcp_modes = BIT(AMDGPU_SPX_PARTITION_MODE) | + BIT(AMDGPU_DPX_PARTITION_MODE) | + BIT(AMDGPU_QPX_PARTITION_MODE) | + BIT(AMDGPU_CPX_PARTITION_MODE); + break; + case 6: + xcp_mgr->supp_xcp_modes = BIT(AMDGPU_SPX_PARTITION_MODE) | + BIT(AMDGPU_TPX_PARTITION_MODE) | + BIT(AMDGPU_CPX_PARTITION_MODE); + break; + case 4: + xcp_mgr->supp_xcp_modes = BIT(AMDGPU_SPX_PARTITION_MODE) | + BIT(AMDGPU_DPX_PARTITION_MODE) | + BIT(AMDGPU_CPX_PARTITION_MODE); + break; + case 2: + xcp_mgr->supp_xcp_modes = BIT(AMDGPU_SPX_PARTITION_MODE) | + BIT(AMDGPU_CPX_PARTITION_MODE); + break; + case 1: + xcp_mgr->supp_xcp_modes = BIT(AMDGPU_SPX_PARTITION_MODE) | + BIT(AMDGPU_CPX_PARTITION_MODE); + break; + + default: + break; + } +} + +int amdgpu_xcp_pre_partition_switch(struct amdgpu_xcp_mgr *xcp_mgr, u32 flags) +{ + /* TODO: + * Stop user queues and threads, and make sure GPU is empty of work. + */ + + if (flags & AMDGPU_XCP_OPS_KFD) + amdgpu_amdkfd_device_fini_sw(xcp_mgr->adev); + + return 0; +} + +int amdgpu_xcp_post_partition_switch(struct amdgpu_xcp_mgr *xcp_mgr, u32 flags) +{ + int ret = 0; + + if (flags & AMDGPU_XCP_OPS_KFD) { + amdgpu_amdkfd_device_probe(xcp_mgr->adev); + amdgpu_amdkfd_device_init(xcp_mgr->adev); + /* If KFD init failed, return failure */ + if (!xcp_mgr->adev->kfd.init_complete) + ret = -EIO; + } + + return ret; +} + /*====================== xcp sysfs - configuration ======================*/ #define XCP_CFG_SYSFS_RES_ATTR_SHOW(_name) \ static ssize_t amdgpu_xcp_res_sysfs_##_name##_show( \ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h index 454b33f889fb65d018cc4f17e7be00b91696a317..70a0f8400b5783ee551c05e8432f2440b0c43eec 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h @@ -39,6 +39,8 @@ #define AMDGPU_XCP_NO_PARTITION (~0) +#define AMDGPU_XCP_OPS_KFD (1 << 0) + struct amdgpu_fpriv; enum AMDGPU_XCP_IP_BLOCK { @@ -144,10 +146,6 @@ struct amdgpu_xcp_mgr_funcs { int (*suspend)(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id); int (*prepare_resume)(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id); int (*resume)(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id); - int (*select_scheds)(struct amdgpu_device *adev, - u32 hw_ip, u32 hw_prio, struct amdgpu_fpriv *fpriv, - unsigned int *num_scheds, struct drm_gpu_scheduler ***scheds); - int (*update_partition_sched_list)(struct amdgpu_device *adev); }; int amdgpu_xcp_prepare_suspend(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id); @@ -176,19 +174,18 @@ int amdgpu_xcp_open_device(struct amdgpu_device *adev, struct drm_file *file_priv); void amdgpu_xcp_release_sched(struct amdgpu_device *adev, struct amdgpu_ctx_entity *entity); - +int amdgpu_xcp_select_scheds(struct amdgpu_device *adev, + u32 hw_ip, u32 hw_prio, + struct amdgpu_fpriv *fpriv, + unsigned int *num_scheds, + struct drm_gpu_scheduler ***scheds); +void amdgpu_xcp_update_supported_modes(struct amdgpu_xcp_mgr *xcp_mgr); +int amdgpu_xcp_update_partition_sched_list(struct amdgpu_device *adev); +int amdgpu_xcp_pre_partition_switch(struct amdgpu_xcp_mgr *xcp_mgr, u32 flags); +int amdgpu_xcp_post_partition_switch(struct amdgpu_xcp_mgr *xcp_mgr, u32 flags); void amdgpu_xcp_sysfs_init(struct amdgpu_device *adev); void amdgpu_xcp_sysfs_fini(struct amdgpu_device *adev); -#define amdgpu_xcp_select_scheds(adev, e, c, d, x, y) \ - ((adev)->xcp_mgr && (adev)->xcp_mgr->funcs && \ - (adev)->xcp_mgr->funcs->select_scheds ? \ - (adev)->xcp_mgr->funcs->select_scheds((adev), (e), (c), (d), (x), (y)) : -ENOENT) -#define amdgpu_xcp_update_partition_sched_list(adev) \ - ((adev)->xcp_mgr && (adev)->xcp_mgr->funcs && \ - (adev)->xcp_mgr->funcs->update_partition_sched_list ? \ - (adev)->xcp_mgr->funcs->update_partition_sched_list(adev) : 0) - static inline int amdgpu_xcp_get_num_xcp(struct amdgpu_xcp_mgr *xcp_mgr) { if (!xcp_mgr) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c index d9ad37711c3eb460de921df4e04009ea459e9c8e..1ede308a7c677396888677e3a90eed372a4d1352 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c @@ -1771,16 +1771,25 @@ void amdgpu_xgmi_early_init(struct amdgpu_device *adev) case IP_VERSION(9, 4, 0): case IP_VERSION(9, 4, 1): case IP_VERSION(9, 4, 2): - adev->gmc.xgmi.max_speed = XGMI_SPEED_25GT; + /* 25 GT/s */ + adev->gmc.xgmi.max_speed = 25; adev->gmc.xgmi.max_width = 16; break; case IP_VERSION(9, 4, 3): case IP_VERSION(9, 4, 4): case IP_VERSION(9, 5, 0): - adev->gmc.xgmi.max_speed = XGMI_SPEED_32GT; + /* 32 GT/s */ + adev->gmc.xgmi.max_speed = 32; adev->gmc.xgmi.max_width = 16; break; default: break; } } + +void amgpu_xgmi_set_max_speed_width(struct amdgpu_device *adev, + uint16_t max_speed, uint8_t max_width) +{ + adev->gmc.xgmi.max_speed = max_speed; + adev->gmc.xgmi.max_width = max_width; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h index f994be985f42da8c94ff06faf0183b0d9a62a2d4..bba0b26fee8f10c9cc126d6176cb7b58d7fb14e7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h @@ -25,12 +25,6 @@ #include #include "amdgpu_ras.h" -enum amdgpu_xgmi_link_speed { - XGMI_SPEED_16GT = 16, - XGMI_SPEED_25GT = 25, - XGMI_SPEED_32GT = 32 -}; - struct amdgpu_hive_info { struct kobject kobj; uint64_t hive_id; @@ -97,7 +91,7 @@ struct amdgpu_xgmi { struct ras_common_if *ras_if; bool connected_to_cpu; struct amdgpu_xgmi_ras *ras; - enum amdgpu_xgmi_link_speed max_speed; + uint16_t max_speed; uint8_t max_width; }; @@ -130,4 +124,6 @@ int amdgpu_xgmi_get_ext_link(struct amdgpu_device *adev, int link_num); void amdgpu_xgmi_early_init(struct amdgpu_device *adev); uint32_t amdgpu_xgmi_get_max_bandwidth(struct amdgpu_device *adev); +void amgpu_xgmi_set_max_speed_width(struct amdgpu_device *adev, + uint16_t max_speed, uint8_t max_width); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c index 1c083304ae77676aaa825609317bb66cca7de7fb..914cf4bfb03333ebf16461ec8c9d3553de25953d 100644 --- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c +++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c @@ -29,12 +29,11 @@ #include "gfx_v9_4_3.h" #include "gfxhub_v1_2.h" #include "sdma_v4_4_2.h" +#include "amdgpu_ip.h" #define XCP_INST_MASK(num_inst, xcp_id) \ (num_inst ? GENMASK(num_inst - 1, 0) << (xcp_id * num_inst) : 0) -#define AMDGPU_XCP_OPS_KFD (1 << 0) - void aqua_vanjaram_doorbell_index_init(struct amdgpu_device *adev) { int i; @@ -62,234 +61,6 @@ void aqua_vanjaram_doorbell_index_init(struct amdgpu_device *adev) adev->doorbell_index.max_assignment = AMDGPU_DOORBELL_LAYOUT1_MAX_ASSIGNMENT << 1; } -static bool aqua_vanjaram_xcp_vcn_shared(struct amdgpu_device *adev) -{ - return (adev->xcp_mgr->num_xcps > adev->vcn.num_vcn_inst); -} - -static void aqua_vanjaram_set_xcp_id(struct amdgpu_device *adev, - uint32_t inst_idx, struct amdgpu_ring *ring) -{ - int xcp_id; - enum AMDGPU_XCP_IP_BLOCK ip_blk; - uint32_t inst_mask; - - ring->xcp_id = AMDGPU_XCP_NO_PARTITION; - if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE) - adev->gfx.enforce_isolation[0].xcp_id = ring->xcp_id; - if ((adev->xcp_mgr->mode == AMDGPU_XCP_MODE_NONE) || - (ring->funcs->type == AMDGPU_RING_TYPE_CPER)) - return; - - inst_mask = 1 << inst_idx; - - switch (ring->funcs->type) { - case AMDGPU_HW_IP_GFX: - case AMDGPU_RING_TYPE_COMPUTE: - case AMDGPU_RING_TYPE_KIQ: - ip_blk = AMDGPU_XCP_GFX; - break; - case AMDGPU_RING_TYPE_SDMA: - ip_blk = AMDGPU_XCP_SDMA; - break; - case AMDGPU_RING_TYPE_VCN_ENC: - case AMDGPU_RING_TYPE_VCN_JPEG: - ip_blk = AMDGPU_XCP_VCN; - break; - default: - DRM_ERROR("Not support ring type %d!", ring->funcs->type); - return; - } - - for (xcp_id = 0; xcp_id < adev->xcp_mgr->num_xcps; xcp_id++) { - if (adev->xcp_mgr->xcp[xcp_id].ip[ip_blk].inst_mask & inst_mask) { - ring->xcp_id = xcp_id; - dev_dbg(adev->dev, "ring:%s xcp_id :%u", ring->name, - ring->xcp_id); - if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE) - adev->gfx.enforce_isolation[xcp_id].xcp_id = xcp_id; - break; - } - } -} - -static void aqua_vanjaram_xcp_gpu_sched_update( - struct amdgpu_device *adev, - struct amdgpu_ring *ring, - unsigned int sel_xcp_id) -{ - unsigned int *num_gpu_sched; - - num_gpu_sched = &adev->xcp_mgr->xcp[sel_xcp_id] - .gpu_sched[ring->funcs->type][ring->hw_prio].num_scheds; - adev->xcp_mgr->xcp[sel_xcp_id].gpu_sched[ring->funcs->type][ring->hw_prio] - .sched[(*num_gpu_sched)++] = &ring->sched; - DRM_DEBUG("%s :[%d] gpu_sched[%d][%d] = %d", ring->name, - sel_xcp_id, ring->funcs->type, - ring->hw_prio, *num_gpu_sched); -} - -static int aqua_vanjaram_xcp_sched_list_update( - struct amdgpu_device *adev) -{ - struct amdgpu_ring *ring; - int i; - - for (i = 0; i < MAX_XCP; i++) { - atomic_set(&adev->xcp_mgr->xcp[i].ref_cnt, 0); - memset(adev->xcp_mgr->xcp[i].gpu_sched, 0, sizeof(adev->xcp_mgr->xcp->gpu_sched)); - } - - if (adev->xcp_mgr->mode == AMDGPU_XCP_MODE_NONE) - return 0; - - for (i = 0; i < AMDGPU_MAX_RINGS; i++) { - ring = adev->rings[i]; - if (!ring || !ring->sched.ready || ring->no_scheduler) - continue; - - aqua_vanjaram_xcp_gpu_sched_update(adev, ring, ring->xcp_id); - - /* VCN may be shared by two partitions under CPX MODE in certain - * configs. - */ - if ((ring->funcs->type == AMDGPU_RING_TYPE_VCN_ENC || - ring->funcs->type == AMDGPU_RING_TYPE_VCN_JPEG) && - aqua_vanjaram_xcp_vcn_shared(adev)) - aqua_vanjaram_xcp_gpu_sched_update(adev, ring, ring->xcp_id + 1); - } - - return 0; -} - -static int aqua_vanjaram_update_partition_sched_list(struct amdgpu_device *adev) -{ - int i; - - for (i = 0; i < adev->num_rings; i++) { - struct amdgpu_ring *ring = adev->rings[i]; - - if (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE || - ring->funcs->type == AMDGPU_RING_TYPE_KIQ) - aqua_vanjaram_set_xcp_id(adev, ring->xcc_id, ring); - else - aqua_vanjaram_set_xcp_id(adev, ring->me, ring); - } - - return aqua_vanjaram_xcp_sched_list_update(adev); -} - -static int aqua_vanjaram_select_scheds( - struct amdgpu_device *adev, - u32 hw_ip, - u32 hw_prio, - struct amdgpu_fpriv *fpriv, - unsigned int *num_scheds, - struct drm_gpu_scheduler ***scheds) -{ - u32 sel_xcp_id; - int i; - - if (fpriv->xcp_id == AMDGPU_XCP_NO_PARTITION) { - u32 least_ref_cnt = ~0; - - fpriv->xcp_id = 0; - for (i = 0; i < adev->xcp_mgr->num_xcps; i++) { - u32 total_ref_cnt; - - total_ref_cnt = atomic_read(&adev->xcp_mgr->xcp[i].ref_cnt); - if (total_ref_cnt < least_ref_cnt) { - fpriv->xcp_id = i; - least_ref_cnt = total_ref_cnt; - } - } - } - sel_xcp_id = fpriv->xcp_id; - - if (adev->xcp_mgr->xcp[sel_xcp_id].gpu_sched[hw_ip][hw_prio].num_scheds) { - *num_scheds = adev->xcp_mgr->xcp[fpriv->xcp_id].gpu_sched[hw_ip][hw_prio].num_scheds; - *scheds = adev->xcp_mgr->xcp[fpriv->xcp_id].gpu_sched[hw_ip][hw_prio].sched; - atomic_inc(&adev->xcp_mgr->xcp[sel_xcp_id].ref_cnt); - DRM_DEBUG("Selected partition #%d", sel_xcp_id); - } else { - DRM_ERROR("Failed to schedule partition #%d.", sel_xcp_id); - return -ENOENT; - } - - return 0; -} - -static int8_t aqua_vanjaram_logical_to_dev_inst(struct amdgpu_device *adev, - enum amd_hw_ip_block_type block, - int8_t inst) -{ - int8_t dev_inst; - - switch (block) { - case GC_HWIP: - case SDMA0_HWIP: - /* Both JPEG and VCN as JPEG is only alias of VCN */ - case VCN_HWIP: - dev_inst = adev->ip_map.dev_inst[block][inst]; - break; - default: - /* For rest of the IPs, no look up required. - * Assume 'logical instance == physical instance' for all configs. */ - dev_inst = inst; - break; - } - - return dev_inst; -} - -static uint32_t aqua_vanjaram_logical_to_dev_mask(struct amdgpu_device *adev, - enum amd_hw_ip_block_type block, - uint32_t mask) -{ - uint32_t dev_mask = 0; - int8_t log_inst, dev_inst; - - while (mask) { - log_inst = ffs(mask) - 1; - dev_inst = aqua_vanjaram_logical_to_dev_inst(adev, block, log_inst); - dev_mask |= (1 << dev_inst); - mask &= ~(1 << log_inst); - } - - return dev_mask; -} - -static void aqua_vanjaram_populate_ip_map(struct amdgpu_device *adev, - enum amd_hw_ip_block_type ip_block, - uint32_t inst_mask) -{ - int l = 0, i; - - while (inst_mask) { - i = ffs(inst_mask) - 1; - adev->ip_map.dev_inst[ip_block][l++] = i; - inst_mask &= ~(1 << i); - } - for (; l < HWIP_MAX_INSTANCE; l++) - adev->ip_map.dev_inst[ip_block][l] = -1; -} - -void aqua_vanjaram_ip_map_init(struct amdgpu_device *adev) -{ - u32 ip_map[][2] = { - { GC_HWIP, adev->gfx.xcc_mask }, - { SDMA0_HWIP, adev->sdma.sdma_mask }, - { VCN_HWIP, adev->vcn.inst_mask }, - }; - int i; - - for (i = 0; i < ARRAY_SIZE(ip_map); ++i) - aqua_vanjaram_populate_ip_map(adev, ip_map[i][0], ip_map[i][1]); - - adev->ip_map.logical_to_dev_inst = aqua_vanjaram_logical_to_dev_inst; - adev->ip_map.logical_to_dev_mask = aqua_vanjaram_logical_to_dev_mask; -} - /* Fixed pattern for smn addressing on different AIDs: * bit[34]: indicate cross AID access * bit[33:32]: indicate target AID id @@ -353,11 +124,14 @@ static int aqua_vanjaram_query_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr) if (adev->nbio.funcs->get_compute_partition_mode) { mode = adev->nbio.funcs->get_compute_partition_mode(adev); - if (mode != derv_mode) + if (mode != derv_mode) { dev_warn( adev->dev, "Mismatch in compute partition mode - reported : %d derived : %d", mode, derv_mode); + if (derv_mode == AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE) + amdgpu_device_bus_status_check(adev); + } } return mode; @@ -593,72 +367,6 @@ static bool __aqua_vanjaram_is_valid_mode(struct amdgpu_xcp_mgr *xcp_mgr, return false; } -static int __aqua_vanjaram_pre_partition_switch(struct amdgpu_xcp_mgr *xcp_mgr, u32 flags) -{ - /* TODO: - * Stop user queues and threads, and make sure GPU is empty of work. - */ - - if (flags & AMDGPU_XCP_OPS_KFD) - amdgpu_amdkfd_device_fini_sw(xcp_mgr->adev); - - return 0; -} - -static int __aqua_vanjaram_post_partition_switch(struct amdgpu_xcp_mgr *xcp_mgr, u32 flags) -{ - int ret = 0; - - if (flags & AMDGPU_XCP_OPS_KFD) { - amdgpu_amdkfd_device_probe(xcp_mgr->adev); - amdgpu_amdkfd_device_init(xcp_mgr->adev); - /* If KFD init failed, return failure */ - if (!xcp_mgr->adev->kfd.init_complete) - ret = -EIO; - } - - return ret; -} - -static void -__aqua_vanjaram_update_supported_modes(struct amdgpu_xcp_mgr *xcp_mgr) -{ - struct amdgpu_device *adev = xcp_mgr->adev; - - xcp_mgr->supp_xcp_modes = 0; - - switch (NUM_XCC(adev->gfx.xcc_mask)) { - case 8: - xcp_mgr->supp_xcp_modes = BIT(AMDGPU_SPX_PARTITION_MODE) | - BIT(AMDGPU_DPX_PARTITION_MODE) | - BIT(AMDGPU_QPX_PARTITION_MODE) | - BIT(AMDGPU_CPX_PARTITION_MODE); - break; - case 6: - xcp_mgr->supp_xcp_modes = BIT(AMDGPU_SPX_PARTITION_MODE) | - BIT(AMDGPU_TPX_PARTITION_MODE) | - BIT(AMDGPU_CPX_PARTITION_MODE); - break; - case 4: - xcp_mgr->supp_xcp_modes = BIT(AMDGPU_SPX_PARTITION_MODE) | - BIT(AMDGPU_DPX_PARTITION_MODE) | - BIT(AMDGPU_CPX_PARTITION_MODE); - break; - /* this seems only existing in emulation phase */ - case 2: - xcp_mgr->supp_xcp_modes = BIT(AMDGPU_SPX_PARTITION_MODE) | - BIT(AMDGPU_CPX_PARTITION_MODE); - break; - case 1: - xcp_mgr->supp_xcp_modes = BIT(AMDGPU_SPX_PARTITION_MODE) | - BIT(AMDGPU_CPX_PARTITION_MODE); - break; - - default: - break; - } -} - static void __aqua_vanjaram_update_available_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr) { int mode; @@ -705,7 +413,7 @@ static int aqua_vanjaram_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr, goto out; } - ret = __aqua_vanjaram_pre_partition_switch(xcp_mgr, flags); + ret = amdgpu_xcp_pre_partition_switch(xcp_mgr, flags); if (ret) goto unlock; @@ -718,7 +426,7 @@ static int aqua_vanjaram_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr, *num_xcps = num_xcc / num_xcc_per_xcp; amdgpu_xcp_init(xcp_mgr, *num_xcps, mode); - ret = __aqua_vanjaram_post_partition_switch(xcp_mgr, flags); + ret = amdgpu_xcp_post_partition_switch(xcp_mgr, flags); if (!ret) __aqua_vanjaram_update_available_partition_mode(xcp_mgr); unlock: @@ -801,9 +509,6 @@ struct amdgpu_xcp_mgr_funcs aqua_vanjaram_xcp_funcs = { .get_ip_details = &aqua_vanjaram_get_xcp_ip_details, .get_xcp_res_info = &aqua_vanjaram_get_xcp_res_info, .get_xcp_mem_id = &aqua_vanjaram_get_xcp_mem_id, - .select_scheds = &aqua_vanjaram_select_scheds, - .update_partition_sched_list = - &aqua_vanjaram_update_partition_sched_list }; static int aqua_vanjaram_xcp_mgr_init(struct amdgpu_device *adev) @@ -818,7 +523,7 @@ static int aqua_vanjaram_xcp_mgr_init(struct amdgpu_device *adev) if (ret) return ret; - __aqua_vanjaram_update_supported_modes(adev->xcp_mgr); + amdgpu_xcp_update_supported_modes(adev->xcp_mgr); /* TODO: Default memory node affinity init */ return ret; @@ -858,7 +563,7 @@ int aqua_vanjaram_init_soc_config(struct amdgpu_device *adev) if (ret) return ret; - aqua_vanjaram_ip_map_init(adev); + amdgpu_ip_map_init(adev); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c index 75ea071744eb5e23606fd648e60b6bb14412ece1..8c377ecbb8a75d73e0d33acce7b070593afa49b0 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c @@ -9522,7 +9522,9 @@ static void gfx_v10_ring_insert_nop(struct amdgpu_ring *ring, uint32_t num_nop) amdgpu_ring_insert_nop(ring, num_nop - 1); } -static int gfx_v10_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid) +static int gfx_v10_0_reset_kgq(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; struct amdgpu_kiq *kiq = &adev->gfx.kiq[0]; @@ -9538,6 +9540,8 @@ static int gfx_v10_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid) if (!kiq->pmf || !kiq->pmf->kiq_unmap_queues) return -EINVAL; + drm_sched_wqueue_stop(&ring->sched); + spin_lock_irqsave(&kiq->ring_lock, flags); if (amdgpu_ring_alloc(kiq_ring, 5 + 7 + 7 + kiq->pmf->map_queues_size)) { @@ -9575,11 +9579,17 @@ static int gfx_v10_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid) return r; } - return amdgpu_ring_test_ring(ring); + r = amdgpu_ring_test_ring(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static int gfx_v10_0_reset_kcq(struct amdgpu_ring *ring, - unsigned int vmid) + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; struct amdgpu_kiq *kiq = &adev->gfx.kiq[0]; @@ -9593,6 +9603,8 @@ static int gfx_v10_0_reset_kcq(struct amdgpu_ring *ring, if (!kiq->pmf || !kiq->pmf->kiq_unmap_queues) return -EINVAL; + drm_sched_wqueue_stop(&ring->sched); + spin_lock_irqsave(&kiq->ring_lock, flags); if (amdgpu_ring_alloc(kiq_ring, kiq->pmf->unmap_queues_size)) { @@ -9647,7 +9659,12 @@ static int gfx_v10_0_reset_kcq(struct amdgpu_ring *ring, if (r) return r; - return amdgpu_ring_test_ring(ring); + r = amdgpu_ring_test_ring(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static void gfx_v10_ip_print(struct amdgpu_ip_block *ip_block, struct drm_printer *p) diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c index afd6d59164bfa364e1521e5f7d5fd98328eac3ae..37dcec2d078415849eaf8c08b2edd5bc81b55cac 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c @@ -85,6 +85,7 @@ MODULE_FIRMWARE("amdgpu/gc_11_0_0_pfp.bin"); MODULE_FIRMWARE("amdgpu/gc_11_0_0_me.bin"); MODULE_FIRMWARE("amdgpu/gc_11_0_0_mec.bin"); MODULE_FIRMWARE("amdgpu/gc_11_0_0_rlc.bin"); +MODULE_FIRMWARE("amdgpu/gc_11_0_0_rlc_kicker.bin"); MODULE_FIRMWARE("amdgpu/gc_11_0_0_rlc_1.bin"); MODULE_FIRMWARE("amdgpu/gc_11_0_0_toc.bin"); MODULE_FIRMWARE("amdgpu/gc_11_0_1_pfp.bin"); @@ -759,6 +760,10 @@ static int gfx_v11_0_init_microcode(struct amdgpu_device *adev) err = amdgpu_ucode_request(adev, &adev->gfx.rlc_fw, AMDGPU_UCODE_REQUIRED, "amdgpu/gc_11_0_0_rlc_1.bin"); + else if (amdgpu_is_kicker_fw(adev)) + err = amdgpu_ucode_request(adev, &adev->gfx.rlc_fw, + AMDGPU_UCODE_REQUIRED, + "amdgpu/%s_rlc_kicker.bin", ucode_prefix); else err = amdgpu_ucode_request(adev, &adev->gfx.rlc_fw, AMDGPU_UCODE_REQUIRED, @@ -6806,7 +6811,9 @@ static int gfx_v11_reset_gfx_pipe(struct amdgpu_ring *ring) return 0; } -static int gfx_v11_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid) +static int gfx_v11_0_reset_kgq(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; int r; @@ -6814,6 +6821,8 @@ static int gfx_v11_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid) if (amdgpu_sriov_vf(adev)) return -EINVAL; + drm_sched_wqueue_stop(&ring->sched); + r = amdgpu_mes_reset_legacy_queue(ring->adev, ring, vmid, false); if (r) { @@ -6835,7 +6844,12 @@ static int gfx_v11_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid) return r; } - return amdgpu_ring_test_ring(ring); + r = amdgpu_ring_test_ring(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static int gfx_v11_0_reset_compute_pipe(struct amdgpu_ring *ring) @@ -6968,7 +6982,9 @@ static int gfx_v11_0_reset_compute_pipe(struct amdgpu_ring *ring) return 0; } -static int gfx_v11_0_reset_kcq(struct amdgpu_ring *ring, unsigned int vmid) +static int gfx_v11_0_reset_kcq(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; int r = 0; @@ -6976,6 +6992,8 @@ static int gfx_v11_0_reset_kcq(struct amdgpu_ring *ring, unsigned int vmid) if (amdgpu_sriov_vf(adev)) return -EINVAL; + drm_sched_wqueue_stop(&ring->sched); + r = amdgpu_mes_reset_legacy_queue(ring->adev, ring, vmid, true); if (r) { dev_warn(adev->dev, "fail(%d) to reset kcq and try pipe reset\n", r); @@ -6995,7 +7013,12 @@ static int gfx_v11_0_reset_kcq(struct amdgpu_ring *ring, unsigned int vmid) return r; } - return amdgpu_ring_test_ring(ring); + r = amdgpu_ring_test_ring(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static void gfx_v11_ip_print(struct amdgpu_ip_block *ip_block, struct drm_printer *p) diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c index 1234c8d64e20d9e8d929bab7440ab9861acc44a4..e4fc42470cf3ef2c08f87db28b88080751c25441 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c @@ -5307,7 +5307,9 @@ static int gfx_v12_reset_gfx_pipe(struct amdgpu_ring *ring) return 0; } -static int gfx_v12_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid) +static int gfx_v12_0_reset_kgq(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; int r; @@ -5315,6 +5317,8 @@ static int gfx_v12_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid) if (amdgpu_sriov_vf(adev)) return -EINVAL; + drm_sched_wqueue_stop(&ring->sched); + r = amdgpu_mes_reset_legacy_queue(ring->adev, ring, vmid, false); if (r) { dev_warn(adev->dev, "reset via MES failed and try pipe reset %d\n", r); @@ -5335,7 +5339,12 @@ static int gfx_v12_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid) return r; } - return amdgpu_ring_test_ring(ring); + r = amdgpu_ring_test_ring(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static int gfx_v12_0_reset_compute_pipe(struct amdgpu_ring *ring) @@ -5421,7 +5430,9 @@ static int gfx_v12_0_reset_compute_pipe(struct amdgpu_ring *ring) return 0; } -static int gfx_v12_0_reset_kcq(struct amdgpu_ring *ring, unsigned int vmid) +static int gfx_v12_0_reset_kcq(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; int r; @@ -5429,6 +5440,8 @@ static int gfx_v12_0_reset_kcq(struct amdgpu_ring *ring, unsigned int vmid) if (amdgpu_sriov_vf(adev)) return -EINVAL; + drm_sched_wqueue_stop(&ring->sched); + r = amdgpu_mes_reset_legacy_queue(ring->adev, ring, vmid, true); if (r) { dev_warn(adev->dev, "fail(%d) to reset kcq and try pipe reset\n", r); @@ -5448,7 +5461,12 @@ static int gfx_v12_0_reset_kcq(struct amdgpu_ring *ring, unsigned int vmid) return r; } - return amdgpu_ring_test_ring(ring); + r = amdgpu_ring_test_ring(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static void gfx_v12_0_ring_begin_use(struct amdgpu_ring *ring) diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c index da0534ff1271a14c51f756b68d57c0493d15596c..2aa323dab34e3323fc409f797d4b45a687657182 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c @@ -4884,76 +4884,6 @@ static void gfx_v7_0_emit_mem_sync_compute(struct amdgpu_ring *ring) amdgpu_ring_write(ring, 0x0000000A); /* poll interval */ } -static void gfx_v7_0_wait_reg_mem(struct amdgpu_ring *ring, int eng_sel, - int mem_space, int opt, uint32_t addr0, - uint32_t addr1, uint32_t ref, uint32_t mask, - uint32_t inv) -{ - amdgpu_ring_write(ring, PACKET3(PACKET3_WAIT_REG_MEM, 5)); - amdgpu_ring_write(ring, - /* memory (1) or register (0) */ - (WAIT_REG_MEM_MEM_SPACE(mem_space) | - WAIT_REG_MEM_OPERATION(opt) | /* wait */ - WAIT_REG_MEM_FUNCTION(3) | /* equal */ - WAIT_REG_MEM_ENGINE(eng_sel))); - - if (mem_space) - BUG_ON(addr0 & 0x3); /* Dword align */ - amdgpu_ring_write(ring, addr0); - amdgpu_ring_write(ring, addr1); - amdgpu_ring_write(ring, ref); - amdgpu_ring_write(ring, mask); - amdgpu_ring_write(ring, inv); /* poll interval */ -} - -static void gfx_v7_0_ring_emit_reg_wait(struct amdgpu_ring *ring, uint32_t reg, - uint32_t val, uint32_t mask) -{ - gfx_v7_0_wait_reg_mem(ring, 0, 0, 0, reg, 0, val, mask, 0x20); -} - -static int gfx_v7_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid) -{ - struct amdgpu_device *adev = ring->adev; - struct amdgpu_kiq *kiq = &adev->gfx.kiq[0]; - struct amdgpu_ring *kiq_ring = &kiq->ring; - unsigned long flags; - u32 tmp; - int r; - - if (amdgpu_sriov_vf(adev)) - return -EINVAL; - - if (!kiq->pmf || !kiq->pmf->kiq_unmap_queues) - return -EINVAL; - - spin_lock_irqsave(&kiq->ring_lock, flags); - - if (amdgpu_ring_alloc(kiq_ring, 5)) { - spin_unlock_irqrestore(&kiq->ring_lock, flags); - return -ENOMEM; - } - - tmp = REG_SET_FIELD(0, CP_VMID_RESET, RESET_REQUEST, 1 << vmid); - gfx_v7_0_ring_emit_wreg(kiq_ring, mmCP_VMID_RESET, tmp); - amdgpu_ring_commit(kiq_ring); - - spin_unlock_irqrestore(&kiq->ring_lock, flags); - - r = amdgpu_ring_test_ring(kiq_ring); - if (r) - return r; - - if (amdgpu_ring_alloc(ring, 7 + 12 + 5)) - return -ENOMEM; - gfx_v7_0_ring_emit_fence_gfx(ring, ring->fence_drv.gpu_addr, - ring->fence_drv.sync_seq, AMDGPU_FENCE_FLAG_EXEC); - gfx_v7_0_ring_emit_reg_wait(ring, mmCP_VMID_RESET, 0, 0xffff); - gfx_v7_0_ring_emit_wreg(ring, mmCP_VMID_RESET, 0); - - return amdgpu_ring_test_ring(ring); -} - static const struct amd_ip_funcs gfx_v7_0_ip_funcs = { .name = "gfx_v7_0", .early_init = gfx_v7_0_early_init, @@ -5003,7 +4933,6 @@ static const struct amdgpu_ring_funcs gfx_v7_0_ring_funcs_gfx = { .emit_wreg = gfx_v7_0_ring_emit_wreg, .soft_recovery = gfx_v7_0_ring_soft_recovery, .emit_mem_sync = gfx_v7_0_emit_mem_sync, - .reset = gfx_v7_0_reset_kgq, }; static const struct amdgpu_ring_funcs gfx_v7_0_ring_funcs_compute = { diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index 5ee2237d8ee8f11a51a9a02c97237dcf26eca2fc..68c401ecb3eca4d7d47596706bbbfddb553656f6 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -6339,34 +6339,6 @@ static void gfx_v8_0_ring_emit_wreg(struct amdgpu_ring *ring, uint32_t reg, amdgpu_ring_write(ring, val); } -static void gfx_v8_0_wait_reg_mem(struct amdgpu_ring *ring, int eng_sel, - int mem_space, int opt, uint32_t addr0, - uint32_t addr1, uint32_t ref, uint32_t mask, - uint32_t inv) -{ - amdgpu_ring_write(ring, PACKET3(PACKET3_WAIT_REG_MEM, 5)); - amdgpu_ring_write(ring, - /* memory (1) or register (0) */ - (WAIT_REG_MEM_MEM_SPACE(mem_space) | - WAIT_REG_MEM_OPERATION(opt) | /* wait */ - WAIT_REG_MEM_FUNCTION(3) | /* equal */ - WAIT_REG_MEM_ENGINE(eng_sel))); - - if (mem_space) - BUG_ON(addr0 & 0x3); /* Dword align */ - amdgpu_ring_write(ring, addr0); - amdgpu_ring_write(ring, addr1); - amdgpu_ring_write(ring, ref); - amdgpu_ring_write(ring, mask); - amdgpu_ring_write(ring, inv); /* poll interval */ -} - -static void gfx_v8_0_ring_emit_reg_wait(struct amdgpu_ring *ring, uint32_t reg, - uint32_t val, uint32_t mask) -{ - gfx_v8_0_wait_reg_mem(ring, 0, 0, 0, reg, 0, val, mask, 0x20); -} - static void gfx_v8_0_ring_soft_recovery(struct amdgpu_ring *ring, unsigned vmid) { struct amdgpu_device *adev = ring->adev; @@ -6843,48 +6815,6 @@ static void gfx_v8_0_emit_wave_limit(struct amdgpu_ring *ring, bool enable) } -static int gfx_v8_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid) -{ - struct amdgpu_device *adev = ring->adev; - struct amdgpu_kiq *kiq = &adev->gfx.kiq[0]; - struct amdgpu_ring *kiq_ring = &kiq->ring; - unsigned long flags; - u32 tmp; - int r; - - if (amdgpu_sriov_vf(adev)) - return -EINVAL; - - if (!kiq->pmf || !kiq->pmf->kiq_unmap_queues) - return -EINVAL; - - spin_lock_irqsave(&kiq->ring_lock, flags); - - if (amdgpu_ring_alloc(kiq_ring, 5)) { - spin_unlock_irqrestore(&kiq->ring_lock, flags); - return -ENOMEM; - } - - tmp = REG_SET_FIELD(0, CP_VMID_RESET, RESET_REQUEST, 1 << vmid); - gfx_v8_0_ring_emit_wreg(kiq_ring, mmCP_VMID_RESET, tmp); - amdgpu_ring_commit(kiq_ring); - - spin_unlock_irqrestore(&kiq->ring_lock, flags); - - r = amdgpu_ring_test_ring(kiq_ring); - if (r) - return r; - - if (amdgpu_ring_alloc(ring, 7 + 12 + 5)) - return -ENOMEM; - gfx_v8_0_ring_emit_fence_gfx(ring, ring->fence_drv.gpu_addr, - ring->fence_drv.sync_seq, AMDGPU_FENCE_FLAG_EXEC); - gfx_v8_0_ring_emit_reg_wait(ring, mmCP_VMID_RESET, 0, 0xffff); - gfx_v8_0_ring_emit_wreg(ring, mmCP_VMID_RESET, 0); - - return amdgpu_ring_test_ring(ring); -} - static const struct amd_ip_funcs gfx_v8_0_ip_funcs = { .name = "gfx_v8_0", .early_init = gfx_v8_0_early_init, @@ -6950,7 +6880,6 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_gfx = { .emit_wreg = gfx_v8_0_ring_emit_wreg, .soft_recovery = gfx_v8_0_ring_soft_recovery, .emit_mem_sync = gfx_v8_0_emit_mem_sync, - .reset = gfx_v8_0_reset_kgq, }; static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_compute = { diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c index d377a7c57d5e13c23edcd0aaef0959760da94339..76ba664efecb3e9f9c72511b1859ca43d9fb64a2 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c @@ -2235,6 +2235,25 @@ static int gfx_v9_0_sw_init(struct amdgpu_ip_block *ip_block) } switch (amdgpu_ip_version(adev, GC_HWIP, 0)) { + case IP_VERSION(9, 0, 1): + case IP_VERSION(9, 2, 1): + case IP_VERSION(9, 4, 0): + case IP_VERSION(9, 2, 2): + case IP_VERSION(9, 1, 0): + case IP_VERSION(9, 3, 0): + adev->gfx.cleaner_shader_ptr = gfx_9_4_2_cleaner_shader_hex; + adev->gfx.cleaner_shader_size = sizeof(gfx_9_4_2_cleaner_shader_hex); + if (adev->gfx.me_fw_version >= 167 && + adev->gfx.pfp_fw_version >= 196 && + adev->gfx.mec_fw_version >= 474) { + adev->gfx.enable_cleaner_shader = true; + r = amdgpu_gfx_cleaner_shader_sw_init(adev, adev->gfx.cleaner_shader_size); + if (r) { + adev->gfx.enable_cleaner_shader = false; + dev_err(adev->dev, "Failed to initialize cleaner shader\n"); + } + } + break; case IP_VERSION(9, 4, 2): adev->gfx.cleaner_shader_ptr = gfx_9_4_2_cleaner_shader_hex; adev->gfx.cleaner_shader_size = sizeof(gfx_9_4_2_cleaner_shader_hex); @@ -7152,53 +7171,9 @@ static void gfx_v9_ring_insert_nop(struct amdgpu_ring *ring, uint32_t num_nop) amdgpu_ring_insert_nop(ring, num_nop - 1); } -static int gfx_v9_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid) -{ - struct amdgpu_device *adev = ring->adev; - struct amdgpu_kiq *kiq = &adev->gfx.kiq[0]; - struct amdgpu_ring *kiq_ring = &kiq->ring; - unsigned long flags; - u32 tmp; - int r; - - if (amdgpu_sriov_vf(adev)) - return -EINVAL; - - if (!kiq->pmf || !kiq->pmf->kiq_unmap_queues) - return -EINVAL; - - spin_lock_irqsave(&kiq->ring_lock, flags); - - if (amdgpu_ring_alloc(kiq_ring, 5)) { - spin_unlock_irqrestore(&kiq->ring_lock, flags); - return -ENOMEM; - } - - tmp = REG_SET_FIELD(0, CP_VMID_RESET, RESET_REQUEST, 1 << vmid); - gfx_v9_0_ring_emit_wreg(kiq_ring, - SOC15_REG_OFFSET(GC, 0, mmCP_VMID_RESET), tmp); - amdgpu_ring_commit(kiq_ring); - - spin_unlock_irqrestore(&kiq->ring_lock, flags); - - r = amdgpu_ring_test_ring(kiq_ring); - if (r) - return r; - - if (amdgpu_ring_alloc(ring, 7 + 7 + 5)) - return -ENOMEM; - gfx_v9_0_ring_emit_fence(ring, ring->fence_drv.gpu_addr, - ring->fence_drv.sync_seq, AMDGPU_FENCE_FLAG_EXEC); - gfx_v9_0_ring_emit_reg_wait(ring, - SOC15_REG_OFFSET(GC, 0, mmCP_VMID_RESET), 0, 0xffff); - gfx_v9_0_ring_emit_wreg(ring, - SOC15_REG_OFFSET(GC, 0, mmCP_VMID_RESET), 0); - - return amdgpu_ring_test_ring(ring); -} - static int gfx_v9_0_reset_kcq(struct amdgpu_ring *ring, - unsigned int vmid) + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; struct amdgpu_kiq *kiq = &adev->gfx.kiq[0]; @@ -7212,6 +7187,8 @@ static int gfx_v9_0_reset_kcq(struct amdgpu_ring *ring, if (!kiq->pmf || !kiq->pmf->kiq_unmap_queues) return -EINVAL; + drm_sched_wqueue_stop(&ring->sched); + spin_lock_irqsave(&kiq->ring_lock, flags); if (amdgpu_ring_alloc(kiq_ring, kiq->pmf->unmap_queues_size)) { @@ -7267,7 +7244,13 @@ static int gfx_v9_0_reset_kcq(struct amdgpu_ring *ring, DRM_ERROR("fail to remap queue\n"); return r; } - return amdgpu_ring_test_ring(ring); + + r = amdgpu_ring_test_ring(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static void gfx_v9_ip_print(struct amdgpu_ip_block *ip_block, struct drm_printer *p) @@ -7477,7 +7460,6 @@ static const struct amdgpu_ring_funcs gfx_v9_0_ring_funcs_gfx = { .emit_reg_write_reg_wait = gfx_v9_0_ring_emit_reg_write_reg_wait, .soft_recovery = gfx_v9_0_ring_soft_recovery, .emit_mem_sync = gfx_v9_0_emit_mem_sync, - .reset = gfx_v9_0_reset_kgq, .emit_cleaner_shader = gfx_v9_0_ring_emit_cleaner_shader, .begin_use = amdgpu_gfx_enforce_isolation_ring_begin_use, .end_use = amdgpu_gfx_enforce_isolation_ring_end_use, diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c index c233edf605694c8661b9270a71c5ba437a2f0f2d..daed0f187bda5b252c0a350ba75e28ae8eb3da81 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c @@ -3552,7 +3552,8 @@ static int gfx_v9_4_3_reset_hw_pipe(struct amdgpu_ring *ring) } static int gfx_v9_4_3_reset_kcq(struct amdgpu_ring *ring, - unsigned int vmid) + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; struct amdgpu_kiq *kiq = &adev->gfx.kiq[ring->xcc_id]; @@ -3566,6 +3567,8 @@ static int gfx_v9_4_3_reset_kcq(struct amdgpu_ring *ring, if (!kiq->pmf || !kiq->pmf->kiq_unmap_queues) return -EINVAL; + drm_sched_wqueue_stop(&ring->sched); + spin_lock_irqsave(&kiq->ring_lock, flags); if (amdgpu_ring_alloc(kiq_ring, kiq->pmf->unmap_queues_size)) { @@ -3619,7 +3622,13 @@ pipe_reset: dev_err(adev->dev, "fail to remap queue\n"); return r; } - return amdgpu_ring_test_ring(ring); + + r = amdgpu_ring_test_ring(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } enum amdgpu_gfx_cp_ras_mem_id { diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c index cb25f7f0dfc1ccffc77e9b4b941476321545f703..6c03bf9f1ae85ddbd0aa96c22c07139aee4ad278 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c @@ -74,6 +74,8 @@ static void gfxhub_v1_2_setup_vm_pt_regs(struct amdgpu_device *adev, static void gfxhub_v1_2_xcc_init_gart_aperture_regs(struct amdgpu_device *adev, uint32_t xcc_mask) { + uint64_t gart_start = amdgpu_virt_xgmi_migrate_enabled(adev) ? + adev->gmc.vram_start : adev->gmc.fb_start; uint64_t pt_base; int i; @@ -91,10 +93,10 @@ static void gfxhub_v1_2_xcc_init_gart_aperture_regs(struct amdgpu_device *adev, if (adev->gmc.pdb0_bo) { WREG32_SOC15(GC, GET_INST(GC, i), regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32, - (u32)(adev->gmc.fb_start >> 12)); + (u32)(gart_start >> 12)); WREG32_SOC15(GC, GET_INST(GC, i), regVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32, - (u32)(adev->gmc.fb_start >> 44)); + (u32)(gart_start >> 44)); WREG32_SOC15(GC, GET_INST(GC, i), regVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32, @@ -180,7 +182,7 @@ gfxhub_v1_2_xcc_init_system_aperture_regs(struct amdgpu_device *adev, /* In the case squeezing vram into GART aperture, we don't use * FB aperture and AGP aperture. Disable them. */ - if (adev->gmc.pdb0_bo) { + if (adev->gmc.pdb0_bo && adev->gmc.xgmi.connected_to_cpu) { WREG32_SOC15(GC, GET_INST(GC, i), regMC_VM_FB_LOCATION_TOP, 0); WREG32_SOC15(GC, GET_INST(GC, i), regMC_VM_FB_LOCATION_BASE, 0x00FFFFFF); WREG32_SOC15(GC, GET_INST(GC, i), regMC_VM_AGP_TOP, 0); diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index 78f65aea03f8b279b516bef6680dafe7d3cb4c14..f73da518a6e86d0158330901839f7ca13a9bef85 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -78,8 +78,6 @@ #define mmHUBP0_DCSURF_PRI_VIEWPORT_DIMENSION_DCN2 0x05ea #define mmHUBP0_DCSURF_PRI_VIEWPORT_DIMENSION_DCN2_BASE_IDX 2 -#define MAX_MEM_RANGES 8 - static const char * const gfxhub_client_ids[] = { "CB", "DB", @@ -411,11 +409,6 @@ static const uint32_t ecc_umc_mcumc_ctrl_mask_addrs[] = { (0x001d43e0 + 0x00001800), }; -static inline bool gmc_v9_0_is_multi_chiplet(struct amdgpu_device *adev) -{ - return !!adev->aid_mask; -} - static int gmc_v9_0_ecc_interrupt_state(struct amdgpu_device *adev, struct amdgpu_irq_src *src, unsigned int type, @@ -649,7 +642,7 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev, addr, entry->client_id, soc15_ih_clientid_name[entry->client_id]); - if (gmc_v9_0_is_multi_chiplet(adev)) + if (amdgpu_is_multi_aid(adev)) dev_err(adev->dev, " cookie node_id %d fault from die %s%d%s\n", node_id, node_id % 4 == 3 ? "RSV" : "AID", node_id / 4, node_id % 4 == 1 ? ".XCD0" : node_id % 4 == 2 ? ".XCD1" : ""); @@ -798,7 +791,7 @@ static bool gmc_v9_0_use_invalidate_semaphore(struct amdgpu_device *adev, uint32_t vmhub) { if (amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 2) || - gmc_v9_0_is_multi_chiplet(adev)) + amdgpu_is_multi_aid(adev)) return false; return ((vmhub == AMDGPU_MMHUB0(0) || @@ -1382,46 +1375,6 @@ static unsigned int gmc_v9_0_get_vbios_fb_size(struct amdgpu_device *adev) return size; } -static enum amdgpu_memory_partition -gmc_v9_0_get_memory_partition(struct amdgpu_device *adev, u32 *supp_modes) -{ - enum amdgpu_memory_partition mode = UNKNOWN_MEMORY_PARTITION_MODE; - - if (adev->nbio.funcs->get_memory_partition_mode) - mode = adev->nbio.funcs->get_memory_partition_mode(adev, - supp_modes); - - return mode; -} - -static enum amdgpu_memory_partition -gmc_v9_0_query_vf_memory_partition(struct amdgpu_device *adev) -{ - switch (adev->gmc.num_mem_partitions) { - case 0: - return UNKNOWN_MEMORY_PARTITION_MODE; - case 1: - return AMDGPU_NPS1_PARTITION_MODE; - case 2: - return AMDGPU_NPS2_PARTITION_MODE; - case 4: - return AMDGPU_NPS4_PARTITION_MODE; - default: - return AMDGPU_NPS1_PARTITION_MODE; - } - - return AMDGPU_NPS1_PARTITION_MODE; -} - -static enum amdgpu_memory_partition -gmc_v9_0_query_memory_partition(struct amdgpu_device *adev) -{ - if (amdgpu_sriov_vf(adev)) - return gmc_v9_0_query_vf_memory_partition(adev); - - return gmc_v9_0_get_memory_partition(adev, NULL); -} - static bool gmc_v9_0_need_reset_on_init(struct amdgpu_device *adev) { if (adev->nbio.funcs && adev->nbio.funcs->is_nps_switch_requested && @@ -1443,7 +1396,7 @@ static const struct amdgpu_gmc_funcs gmc_v9_0_gmc_funcs = { .get_vm_pte = gmc_v9_0_get_vm_pte, .override_vm_pte_flags = gmc_v9_0_override_vm_pte_flags, .get_vbios_fb_size = gmc_v9_0_get_vbios_fb_size, - .query_mem_partition_mode = &gmc_v9_0_query_memory_partition, + .query_mem_partition_mode = &amdgpu_gmc_query_memory_partition, .request_mem_partition_mode = &amdgpu_gmc_request_memory_partition, .need_reset_on_init = &gmc_v9_0_need_reset_on_init, }; @@ -1550,7 +1503,7 @@ static void gmc_v9_0_set_mmhub_ras_funcs(struct amdgpu_device *adev) static void gmc_v9_0_set_gfxhub_funcs(struct amdgpu_device *adev) { - if (gmc_v9_0_is_multi_chiplet(adev)) + if (amdgpu_is_multi_aid(adev)) adev->gfxhub.funcs = &gfxhub_v1_2_funcs; else adev->gfxhub.funcs = &gfxhub_v1_0_funcs; @@ -1596,7 +1549,7 @@ static void gmc_v9_0_init_nps_details(struct amdgpu_device *adev) if (amdgpu_sriov_vf(adev) || (adev->flags & AMD_IS_APU)) return; - mode = gmc_v9_0_get_memory_partition(adev, &supp_modes); + mode = amdgpu_gmc_get_memory_partition(adev, &supp_modes); /* Mode detected by hardware and supported modes available */ if ((mode != UNKNOWN_MEMORY_PARTITION_MODE) && supp_modes) { @@ -1632,7 +1585,7 @@ static int gmc_v9_0_early_init(struct amdgpu_ip_block *ip_block) */ if (amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 0) || amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 1) || - gmc_v9_0_is_multi_chiplet(adev)) + amdgpu_is_multi_aid(adev)) adev->gmc.xgmi.supported = true; if (amdgpu_ip_version(adev, XGMI_HWIP, 0) == IP_VERSION(6, 1, 0)) { @@ -1719,7 +1672,7 @@ static void gmc_v9_0_vram_gtt_location(struct amdgpu_device *adev, /* add the xgmi offset of the physical node */ base += adev->gmc.xgmi.physical_node_id * adev->gmc.xgmi.node_segment_size; - if (adev->gmc.xgmi.connected_to_cpu) { + if (amdgpu_gmc_is_pdb0_enabled(adev)) { amdgpu_gmc_sysvm_location(adev, mc); } else { amdgpu_gmc_vram_location(adev, mc, base); @@ -1834,7 +1787,7 @@ static int gmc_v9_0_gart_init(struct amdgpu_device *adev) return 0; } - if (adev->gmc.xgmi.connected_to_cpu) { + if (amdgpu_gmc_is_pdb0_enabled(adev)) { adev->gmc.vmid0_page_table_depth = 1; adev->gmc.vmid0_page_table_block_size = 12; } else { @@ -1860,7 +1813,7 @@ static int gmc_v9_0_gart_init(struct amdgpu_device *adev) if (r) return r; - if (adev->gmc.xgmi.connected_to_cpu) + if (amdgpu_gmc_is_pdb0_enabled(adev)) r = amdgpu_gmc_pdb0_alloc(adev); } @@ -1882,188 +1835,6 @@ static void gmc_v9_0_save_registers(struct amdgpu_device *adev) adev->gmc.sdpif_register = RREG32_SOC15(DCE, 0, mmDCHUBBUB_SDPIF_MMIO_CNTRL_0); } -static bool gmc_v9_0_validate_partition_info(struct amdgpu_device *adev) -{ - enum amdgpu_memory_partition mode; - u32 supp_modes; - bool valid; - - mode = gmc_v9_0_get_memory_partition(adev, &supp_modes); - - /* Mode detected by hardware not present in supported modes */ - if ((mode != UNKNOWN_MEMORY_PARTITION_MODE) && - !(BIT(mode - 1) & supp_modes)) - return false; - - switch (mode) { - case UNKNOWN_MEMORY_PARTITION_MODE: - case AMDGPU_NPS1_PARTITION_MODE: - valid = (adev->gmc.num_mem_partitions == 1); - break; - case AMDGPU_NPS2_PARTITION_MODE: - valid = (adev->gmc.num_mem_partitions == 2); - break; - case AMDGPU_NPS4_PARTITION_MODE: - valid = (adev->gmc.num_mem_partitions == 3 || - adev->gmc.num_mem_partitions == 4); - break; - default: - valid = false; - } - - return valid; -} - -static bool gmc_v9_0_is_node_present(int *node_ids, int num_ids, int nid) -{ - int i; - - /* Check if node with id 'nid' is present in 'node_ids' array */ - for (i = 0; i < num_ids; ++i) - if (node_ids[i] == nid) - return true; - - return false; -} - -static void -gmc_v9_0_init_acpi_mem_ranges(struct amdgpu_device *adev, - struct amdgpu_mem_partition_info *mem_ranges) -{ - struct amdgpu_numa_info numa_info; - int node_ids[MAX_MEM_RANGES]; - int num_ranges = 0, ret; - int num_xcc, xcc_id; - uint32_t xcc_mask; - - num_xcc = NUM_XCC(adev->gfx.xcc_mask); - xcc_mask = (1U << num_xcc) - 1; - - for_each_inst(xcc_id, xcc_mask) { - ret = amdgpu_acpi_get_mem_info(adev, xcc_id, &numa_info); - if (ret) - continue; - - if (numa_info.nid == NUMA_NO_NODE) { - mem_ranges[0].size = numa_info.size; - mem_ranges[0].numa.node = numa_info.nid; - num_ranges = 1; - break; - } - - if (gmc_v9_0_is_node_present(node_ids, num_ranges, - numa_info.nid)) - continue; - - node_ids[num_ranges] = numa_info.nid; - mem_ranges[num_ranges].numa.node = numa_info.nid; - mem_ranges[num_ranges].size = numa_info.size; - ++num_ranges; - } - - adev->gmc.num_mem_partitions = num_ranges; -} - -static void -gmc_v9_0_init_sw_mem_ranges(struct amdgpu_device *adev, - struct amdgpu_mem_partition_info *mem_ranges) -{ - enum amdgpu_memory_partition mode; - u32 start_addr = 0, size; - int i, r, l; - - mode = gmc_v9_0_query_memory_partition(adev); - - switch (mode) { - case UNKNOWN_MEMORY_PARTITION_MODE: - adev->gmc.num_mem_partitions = 0; - break; - case AMDGPU_NPS1_PARTITION_MODE: - adev->gmc.num_mem_partitions = 1; - break; - case AMDGPU_NPS2_PARTITION_MODE: - adev->gmc.num_mem_partitions = 2; - break; - case AMDGPU_NPS4_PARTITION_MODE: - if (adev->flags & AMD_IS_APU) - adev->gmc.num_mem_partitions = 3; - else - adev->gmc.num_mem_partitions = 4; - break; - default: - adev->gmc.num_mem_partitions = 1; - break; - } - - /* Use NPS range info, if populated */ - r = amdgpu_gmc_get_nps_memranges(adev, mem_ranges, - &adev->gmc.num_mem_partitions); - if (!r) { - l = 0; - for (i = 1; i < adev->gmc.num_mem_partitions; ++i) { - if (mem_ranges[i].range.lpfn > - mem_ranges[i - 1].range.lpfn) - l = i; - } - - } else { - if (!adev->gmc.num_mem_partitions) { - dev_err(adev->dev, - "Not able to detect NPS mode, fall back to NPS1"); - adev->gmc.num_mem_partitions = 1; - } - /* Fallback to sw based calculation */ - size = (adev->gmc.real_vram_size + SZ_16M) >> AMDGPU_GPU_PAGE_SHIFT; - size /= adev->gmc.num_mem_partitions; - - for (i = 0; i < adev->gmc.num_mem_partitions; ++i) { - mem_ranges[i].range.fpfn = start_addr; - mem_ranges[i].size = - ((u64)size << AMDGPU_GPU_PAGE_SHIFT); - mem_ranges[i].range.lpfn = start_addr + size - 1; - start_addr += size; - } - - l = adev->gmc.num_mem_partitions - 1; - } - - /* Adjust the last one */ - mem_ranges[l].range.lpfn = - (adev->gmc.real_vram_size >> AMDGPU_GPU_PAGE_SHIFT) - 1; - mem_ranges[l].size = - adev->gmc.real_vram_size - - ((u64)mem_ranges[l].range.fpfn << AMDGPU_GPU_PAGE_SHIFT); -} - -static int gmc_v9_0_init_mem_ranges(struct amdgpu_device *adev) -{ - bool valid; - - adev->gmc.mem_partitions = kcalloc(MAX_MEM_RANGES, - sizeof(struct amdgpu_mem_partition_info), - GFP_KERNEL); - if (!adev->gmc.mem_partitions) - return -ENOMEM; - - /* TODO : Get the range from PSP/Discovery for dGPU */ - if (adev->gmc.is_app_apu) - gmc_v9_0_init_acpi_mem_ranges(adev, adev->gmc.mem_partitions); - else - gmc_v9_0_init_sw_mem_ranges(adev, adev->gmc.mem_partitions); - - if (amdgpu_sriov_vf(adev)) - valid = true; - else - valid = gmc_v9_0_validate_partition_info(adev); - if (!valid) { - /* TODO: handle invalid case */ - dev_WARN(adev->dev, - "Mem ranges not matching with hardware config"); - } - - return 0; -} - static void gmc_v9_4_3_init_vram_info(struct amdgpu_device *adev) { adev->gmc.vram_type = AMDGPU_VRAM_TYPE_HBM; @@ -2085,7 +1856,7 @@ static int gmc_v9_0_sw_init(struct amdgpu_ip_block *ip_block) spin_lock_init(&adev->gmc.invalidate_lock); - if (gmc_v9_0_is_multi_chiplet(adev)) { + if (amdgpu_is_multi_aid(adev)) { gmc_v9_4_3_init_vram_info(adev); } else if (!adev->bios) { if (adev->flags & AMD_IS_APU) { @@ -2235,8 +2006,8 @@ static int gmc_v9_0_sw_init(struct amdgpu_ip_block *ip_block) amdgpu_gmc_get_vbios_allocations(adev); - if (gmc_v9_0_is_multi_chiplet(adev)) { - r = gmc_v9_0_init_mem_ranges(adev); + if (amdgpu_is_multi_aid(adev)) { + r = amdgpu_gmc_init_mem_ranges(adev); if (r) return r; } @@ -2264,7 +2035,7 @@ static int gmc_v9_0_sw_init(struct amdgpu_ip_block *ip_block) adev->vm_manager.first_kfd_vmid = (amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 1) || amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 2) || - gmc_v9_0_is_multi_chiplet(adev)) ? + amdgpu_is_multi_aid(adev)) ? 3 : 8; @@ -2276,7 +2047,7 @@ static int gmc_v9_0_sw_init(struct amdgpu_ip_block *ip_block) if (r) return r; - if (gmc_v9_0_is_multi_chiplet(adev)) + if (amdgpu_is_multi_aid(adev)) amdgpu_gmc_sysfs_init(adev); return 0; @@ -2286,7 +2057,7 @@ static int gmc_v9_0_sw_fini(struct amdgpu_ip_block *ip_block) { struct amdgpu_device *adev = ip_block->adev; - if (gmc_v9_0_is_multi_chiplet(adev)) + if (amdgpu_is_multi_aid(adev)) amdgpu_gmc_sysfs_fini(adev); amdgpu_gmc_ras_fini(adev); @@ -2360,7 +2131,7 @@ static int gmc_v9_0_gart_enable(struct amdgpu_device *adev) { int r; - if (adev->gmc.xgmi.connected_to_cpu) + if (amdgpu_gmc_is_pdb0_enabled(adev)) amdgpu_gmc_init_pdb0(adev); if (adev->gart.bo == NULL) { @@ -2518,7 +2289,7 @@ static int gmc_v9_0_resume(struct amdgpu_ip_block *ip_block) * information again. */ if (adev->gmc.reset_flags & AMDGPU_GMC_INIT_RESET_NPS) { - gmc_v9_0_init_sw_mem_ranges(adev, adev->gmc.mem_partitions); + amdgpu_gmc_init_sw_mem_ranges(adev, adev->gmc.mem_partitions); adev->gmc.reset_flags &= ~AMDGPU_GMC_INIT_RESET_NPS; } diff --git a/drivers/gpu/drm/amd/amdgpu/imu_v11_0.c b/drivers/gpu/drm/amd/amdgpu/imu_v11_0.c index cfa91d709d49963dffc4684ac3ce3dcbf5e60596..cc626036ed9c3dc2e707588a31f11d39cfb1d74a 100644 --- a/drivers/gpu/drm/amd/amdgpu/imu_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/imu_v11_0.c @@ -32,6 +32,7 @@ #include "gc/gc_11_0_0_sh_mask.h" MODULE_FIRMWARE("amdgpu/gc_11_0_0_imu.bin"); +MODULE_FIRMWARE("amdgpu/gc_11_0_0_imu_kicker.bin"); MODULE_FIRMWARE("amdgpu/gc_11_0_1_imu.bin"); MODULE_FIRMWARE("amdgpu/gc_11_0_2_imu.bin"); MODULE_FIRMWARE("amdgpu/gc_11_0_3_imu.bin"); @@ -51,8 +52,12 @@ static int imu_v11_0_init_microcode(struct amdgpu_device *adev) DRM_DEBUG("\n"); amdgpu_ucode_ip_version_decode(adev, GC_HWIP, ucode_prefix, sizeof(ucode_prefix)); - err = amdgpu_ucode_request(adev, &adev->gfx.imu_fw, AMDGPU_UCODE_REQUIRED, - "amdgpu/%s_imu.bin", ucode_prefix); + if (amdgpu_is_kicker_fw(adev)) + err = amdgpu_ucode_request(adev, &adev->gfx.imu_fw, AMDGPU_UCODE_REQUIRED, + "amdgpu/%s_imu_kicker.bin", ucode_prefix); + else + err = amdgpu_ucode_request(adev, &adev->gfx.imu_fw, AMDGPU_UCODE_REQUIRED, + "amdgpu/%s_imu.bin", ucode_prefix); if (err) goto out; diff --git a/drivers/gpu/drm/amd/amdgpu/isp_v4_1_1.c b/drivers/gpu/drm/amd/amdgpu/isp_v4_1_1.c index 574880d67009953d25435acb9a4b82ebbb8ae003..f857796f0297a19e91014a0825d9162e5a163cac 100644 --- a/drivers/gpu/drm/amd/amdgpu/isp_v4_1_1.c +++ b/drivers/gpu/drm/amd/amdgpu/isp_v4_1_1.c @@ -29,6 +29,12 @@ #include "amdgpu.h" #include "isp_v4_1_1.h" +#define ISP_PERFORMANCE_STATE_LOW 0 +#define ISP_PERFORMANCE_STATE_HIGH 1 + +#define ISP_HIGH_PERFORMANC_XCLK 788 +#define ISP_HIGH_PERFORMANC_ICLK 788 + static const unsigned int isp_4_1_1_int_srcid[MAX_ISP411_INT_SRC] = { ISP_4_1__SRCID__ISP_RINGBUFFER_WPT9, ISP_4_1__SRCID__ISP_RINGBUFFER_WPT10, @@ -56,6 +62,125 @@ static struct gpiod_lookup_table isp_sensor_gpio_table = { }, }; +static int isp_poweroff(struct generic_pm_domain *genpd) +{ + struct amdgpu_isp *isp = container_of(genpd, struct amdgpu_isp, ispgpd); + struct amdgpu_device *adev = isp->adev; + + return amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_ISP, true, 0); +} + +static int isp_poweron(struct generic_pm_domain *genpd) +{ + struct amdgpu_isp *isp = container_of(genpd, struct amdgpu_isp, ispgpd); + struct amdgpu_device *adev = isp->adev; + + return amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_ISP, false, 0); +} + +static int isp_set_performance_state(struct generic_pm_domain *genpd, + unsigned int state) +{ + struct amdgpu_isp *isp = container_of(genpd, struct amdgpu_isp, ispgpd); + struct amdgpu_device *adev = isp->adev; + u32 iclk, xclk; + int ret; + + switch (state) { + case ISP_PERFORMANCE_STATE_HIGH: + xclk = ISP_HIGH_PERFORMANC_XCLK; + iclk = ISP_HIGH_PERFORMANC_ICLK; + break; + case ISP_PERFORMANCE_STATE_LOW: + /* isp runs at default lowest clock-rate on power-on, do nothing */ + return 0; + default: + return -EINVAL; + } + + ret = amdgpu_dpm_set_soft_freq_range(adev, PP_ISPXCLK, xclk, 0); + if (ret) { + drm_err(&adev->ddev, "failed to set xclk %u to %u: %d\n", + xclk, state, ret); + return ret; + } + + ret = amdgpu_dpm_set_soft_freq_range(adev, PP_ISPICLK, iclk, 0); + if (ret) { + drm_err(&adev->ddev, "failed to set iclk %u to %u: %d\n", + iclk, state, ret); + return ret; + } + + return 0; +} + +static int isp_genpd_add_device(struct device *dev, void *data) +{ + struct generic_pm_domain *gpd = data; + struct platform_device *pdev = container_of(dev, struct platform_device, dev); + struct amdgpu_isp *isp = container_of(gpd, struct amdgpu_isp, ispgpd); + struct amdgpu_device *adev = isp->adev; + int ret; + + if (!pdev) + return -EINVAL; + + if (!dev->type->name) { + drm_dbg(&adev->ddev, "Invalid device type to add\n"); + goto exit; + } + + if (strcmp(dev->type->name, "mfd_device")) { + drm_dbg(&adev->ddev, "Invalid isp mfd device %s to add\n", pdev->mfd_cell->name); + goto exit; + } + + ret = pm_genpd_add_device(gpd, dev); + if (ret) { + drm_err(&adev->ddev, "Failed to add dev %s to genpd %d\n", + pdev->mfd_cell->name, ret); + return -ENODEV; + } + +exit: + /* Continue to add */ + return 0; +} + +static int isp_genpd_remove_device(struct device *dev, void *data) +{ + struct generic_pm_domain *gpd = data; + struct platform_device *pdev = container_of(dev, struct platform_device, dev); + struct amdgpu_isp *isp = container_of(gpd, struct amdgpu_isp, ispgpd); + struct amdgpu_device *adev = isp->adev; + int ret; + + if (!pdev) + return -EINVAL; + + if (!dev->type->name) { + drm_dbg(&adev->ddev, "Invalid device type to remove\n"); + goto exit; + } + + if (strcmp(dev->type->name, "mfd_device")) { + drm_dbg(&adev->ddev, "Invalid isp mfd device %s to remove\n", + pdev->mfd_cell->name); + goto exit; + } + + ret = pm_genpd_remove_device(dev); + if (ret) { + drm_err(&adev->ddev, "Failed to remove dev from genpd %d\n", ret); + return -ENODEV; + } + +exit: + /* Continue to remove */ + return 0; +} + static int isp_v4_1_1_hw_init(struct amdgpu_isp *isp) { struct amdgpu_device *adev = isp->adev; @@ -81,11 +206,21 @@ static int isp_v4_1_1_hw_init(struct amdgpu_isp *isp) isp_base = adev->rmmio_base; + isp->ispgpd.name = "ISP_v_4_1_1"; + isp->ispgpd.power_off = isp_poweroff; + isp->ispgpd.power_on = isp_poweron; + isp->ispgpd.set_performance_state = isp_set_performance_state; + + r = pm_genpd_init(&isp->ispgpd, NULL, true); + if (r) { + drm_err(&adev->ddev, "failed to initialize genpd (%d)\n", r); + return -EINVAL; + } + isp->isp_cell = kcalloc(3, sizeof(struct mfd_cell), GFP_KERNEL); if (!isp->isp_cell) { r = -ENOMEM; - drm_err(&adev->ddev, - "%s: isp mfd cell alloc failed\n", __func__); + drm_err(&adev->ddev, "isp mfd cell alloc failed (%d)\n", r); goto failure; } @@ -95,16 +230,14 @@ static int isp_v4_1_1_hw_init(struct amdgpu_isp *isp) GFP_KERNEL); if (!isp->isp_res) { r = -ENOMEM; - drm_err(&adev->ddev, - "%s: isp mfd res alloc failed\n", __func__); + drm_err(&adev->ddev, "isp mfd resource alloc failed (%d)\n", r); goto failure; } isp->isp_pdata = kzalloc(sizeof(*isp->isp_pdata), GFP_KERNEL); if (!isp->isp_pdata) { r = -ENOMEM; - drm_err(&adev->ddev, - "%s: isp platform data alloc failed\n", __func__); + drm_err(&adev->ddev, "isp platform data alloc failed (%d)\n", r); goto failure; } @@ -142,8 +275,7 @@ static int isp_v4_1_1_hw_init(struct amdgpu_isp *isp) isp->isp_i2c_res = kcalloc(1, sizeof(struct resource), GFP_KERNEL); if (!isp->isp_i2c_res) { r = -ENOMEM; - drm_err(&adev->ddev, - "%s: isp mfd res alloc failed\n", __func__); + drm_err(&adev->ddev, "isp mfd res alloc failed (%d)\n", r); goto failure; } @@ -162,8 +294,7 @@ static int isp_v4_1_1_hw_init(struct amdgpu_isp *isp) isp->isp_gpio_res = kcalloc(1, sizeof(struct resource), GFP_KERNEL); if (!isp->isp_gpio_res) { r = -ENOMEM; - drm_err(&adev->ddev, - "%s: isp gpio res alloc failed\n", __func__); + drm_err(&adev->ddev, "isp gpio resource alloc failed (%d)\n", r); goto failure; } @@ -179,10 +310,23 @@ static int isp_v4_1_1_hw_init(struct amdgpu_isp *isp) isp->isp_cell[2].platform_data = isp->isp_pdata; isp->isp_cell[2].pdata_size = sizeof(struct isp_platform_data); - r = mfd_add_hotplug_devices(isp->parent, isp->isp_cell, 3); + /* add only amd_isp_capture and amd_isp_i2c_designware to genpd */ + r = mfd_add_hotplug_devices(isp->parent, isp->isp_cell, 2); + if (r) { + drm_err(&adev->ddev, "add mfd hotplug device failed (%d)\n", r); + goto failure; + } + + r = device_for_each_child(isp->parent, &isp->ispgpd, + isp_genpd_add_device); + if (r) { + drm_err(&adev->ddev, "failed to add devices to genpd (%d)\n", r); + goto failure; + } + + r = mfd_add_hotplug_devices(isp->parent, &isp->isp_cell[2], 1); if (r) { - drm_err(&adev->ddev, - "%s: add mfd hotplug device failed\n", __func__); + drm_err(&adev->ddev, "add pinctl hotplug device failed (%d)\n", r); goto failure; } @@ -201,6 +345,9 @@ failure: static int isp_v4_1_1_hw_fini(struct amdgpu_isp *isp) { + device_for_each_child(isp->parent, NULL, + isp_genpd_remove_device); + mfd_remove_devices(isp->parent); kfree(isp->isp_res); diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c index 4cde8a8bcc837aae0ea293b2f585f8b84529c29f..781a5a8a836140d5e836a164b951d7014becae09 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c @@ -764,11 +764,21 @@ static int jpeg_v2_0_process_interrupt(struct amdgpu_device *adev, return 0; } -static int jpeg_v2_0_ring_reset(struct amdgpu_ring *ring, unsigned int vmid) +static int jpeg_v2_0_ring_reset(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { + int r; + + drm_sched_wqueue_stop(&ring->sched); jpeg_v2_0_stop(ring->adev); jpeg_v2_0_start(ring->adev); - return amdgpu_ring_test_helper(ring); + r = amdgpu_ring_test_helper(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static const struct amd_ip_funcs jpeg_v2_0_ip_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c index 8b39e114f3be144e83d70f2fc2a5c511246c9799..5be9cdcae32c18bdc8dfc1dcc4ebfad63d4323f8 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c @@ -643,11 +643,21 @@ static int jpeg_v2_5_process_interrupt(struct amdgpu_device *adev, return 0; } -static int jpeg_v2_5_ring_reset(struct amdgpu_ring *ring, unsigned int vmid) +static int jpeg_v2_5_ring_reset(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { + int r; + + drm_sched_wqueue_stop(&ring->sched); jpeg_v2_5_stop_inst(ring->adev, ring->me); jpeg_v2_5_start_inst(ring->adev, ring->me); - return amdgpu_ring_test_helper(ring); + r = amdgpu_ring_test_helper(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static const struct amd_ip_funcs jpeg_v2_5_ip_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c index 2f8510c2986b9ac73cb294d89e379a4b2f6a350f..a24bd833d644274b1fccecc8e6daa510a703c423 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c @@ -555,11 +555,21 @@ static int jpeg_v3_0_process_interrupt(struct amdgpu_device *adev, return 0; } -static int jpeg_v3_0_ring_reset(struct amdgpu_ring *ring, unsigned int vmid) +static int jpeg_v3_0_ring_reset(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { + int r; + + drm_sched_wqueue_stop(&ring->sched); jpeg_v3_0_stop(ring->adev); jpeg_v3_0_start(ring->adev); - return amdgpu_ring_test_helper(ring); + r = amdgpu_ring_test_helper(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static const struct amd_ip_funcs jpeg_v3_0_ip_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c index f17ec5414fd69d726b489eeb5d57a6b8ff6d3589..1d4edd77837d0104a480f65dc059428b64ab27b1 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c @@ -720,14 +720,24 @@ static int jpeg_v4_0_process_interrupt(struct amdgpu_device *adev, return 0; } -static int jpeg_v4_0_ring_reset(struct amdgpu_ring *ring, unsigned int vmid) +static int jpeg_v4_0_ring_reset(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { + int r; + if (amdgpu_sriov_vf(ring->adev)) return -EINVAL; + drm_sched_wqueue_stop(&ring->sched); jpeg_v4_0_stop(ring->adev); jpeg_v4_0_start(ring->adev); - return amdgpu_ring_test_helper(ring); + r = amdgpu_ring_test_helper(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static const struct amd_ip_funcs jpeg_v4_0_ip_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c index 79e342d5ab28d87f39e43afdc9dd0c0e97bf937d..78441f8fce972c91f39c15924b6ed4deb5016043 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c @@ -1143,14 +1143,24 @@ static void jpeg_v4_0_3_core_stall_reset(struct amdgpu_ring *ring) WREG32_SOC15(JPEG, jpeg_inst, regJPEG_CORE_RST_CTRL, 0x00); } -static int jpeg_v4_0_3_ring_reset(struct amdgpu_ring *ring, unsigned int vmid) +static int jpeg_v4_0_3_ring_reset(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { + int r; + if (amdgpu_sriov_vf(ring->adev)) return -EOPNOTSUPP; + drm_sched_wqueue_stop(&ring->sched); jpeg_v4_0_3_core_stall_reset(ring); jpeg_v4_0_3_start_jrbc(ring); - return amdgpu_ring_test_helper(ring); + r = amdgpu_ring_test_helper(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static const struct amd_ip_funcs jpeg_v4_0_3_ip_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c index 3b6f65a256464a949788ebaae7b1dda727de1c97..6f8a16da9d608abeaa7466aabb8c00a9315502d3 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c @@ -834,14 +834,24 @@ static void jpeg_v5_0_1_core_stall_reset(struct amdgpu_ring *ring) WREG32_SOC15(JPEG, jpeg_inst, regJPEG_CORE_RST_CTRL, 0x00); } -static int jpeg_v5_0_1_ring_reset(struct amdgpu_ring *ring, unsigned int vmid) +static int jpeg_v5_0_1_ring_reset(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { + int r; + if (amdgpu_sriov_vf(ring->adev)) return -EOPNOTSUPP; + drm_sched_wqueue_stop(&ring->sched); jpeg_v5_0_1_core_stall_reset(ring); jpeg_v5_0_1_init_jrbc(ring); - return amdgpu_ring_test_helper(ring); + r = amdgpu_ring_test_helper(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static const struct amd_ip_funcs jpeg_v5_0_1_ip_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c index c9eba537de0940ddf56d97544dca10b1392bdfeb..28eb846280dd406cb9d4f8a4b2c9ad4d05a39c4f 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c @@ -1630,10 +1630,12 @@ static int mes_v11_0_hw_init(struct amdgpu_ip_block *ip_block) if (r) goto failure; - r = mes_v11_0_set_hw_resources_1(&adev->mes); - if (r) { - DRM_ERROR("failed mes_v11_0_set_hw_resources_1, r=%d\n", r); - goto failure; + if ((adev->mes.sched_version & AMDGPU_MES_VERSION_MASK) >= 0x50) { + r = mes_v11_0_set_hw_resources_1(&adev->mes); + if (r) { + DRM_ERROR("failed mes_v11_0_set_hw_resources_1, r=%d\n", r); + goto failure; + } } r = mes_v11_0_query_sched_status(&adev->mes); diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c index b4f17332d46643e33b61e4e9558b5253feb66e9e..6b222630f3fa1d3b047122254f46bc31c06a3f3d 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c @@ -1742,7 +1742,8 @@ static int mes_v12_0_hw_init(struct amdgpu_ip_block *ip_block) if (r) goto failure; - mes_v12_0_set_hw_resources_1(&adev->mes, AMDGPU_MES_SCHED_PIPE); + if ((adev->mes.sched_version & AMDGPU_MES_VERSION_MASK) >= 0x4b) + mes_v12_0_set_hw_resources_1(&adev->mes, AMDGPU_MES_SCHED_PIPE); mes_v12_0_init_aggregated_doorbell(&adev->mes); diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c index 76167fadb292be7344f1ed7c8853f64604e36f57..cc688ae79e84bd280fb860be20298bd6938b8c43 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c @@ -76,6 +76,8 @@ static void mmhub_v1_8_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmi static void mmhub_v1_8_init_gart_aperture_regs(struct amdgpu_device *adev) { + uint64_t gart_start = amdgpu_virt_xgmi_migrate_enabled(adev) ? + adev->gmc.vram_start : adev->gmc.fb_start; uint64_t pt_base; u32 inst_mask; int i; @@ -95,10 +97,10 @@ static void mmhub_v1_8_init_gart_aperture_regs(struct amdgpu_device *adev) if (adev->gmc.pdb0_bo) { WREG32_SOC15(MMHUB, i, regVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32, - (u32)(adev->gmc.fb_start >> 12)); + (u32)(gart_start >> 12)); WREG32_SOC15(MMHUB, i, regVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32, - (u32)(adev->gmc.fb_start >> 44)); + (u32)(gart_start >> 44)); WREG32_SOC15(MMHUB, i, regVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32, diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c index a376f072700dc7b3316b85a74d4d62ba81998509..1c22bc11c1f85f5d322e971bc4843c216ff58e74 100644 --- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c +++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c @@ -31,9 +31,6 @@ #define NPS_MODE_MASK 0x000000FFL -/* Core 0 Port 0 counter */ -#define smnPCIEP_NAK_COUNTER 0x1A340218 - static void nbio_v7_9_remap_hdp_registers(struct amdgpu_device *adev) { WREG32_SOC15(NBIO, 0, regBIF_BX0_REMAP_HDP_MEM_FLUSH_CNTL, @@ -467,22 +464,6 @@ static void nbio_v7_9_init_registers(struct amdgpu_device *adev) } } -static u64 nbio_v7_9_get_pcie_replay_count(struct amdgpu_device *adev) -{ - u32 val, nak_r, nak_g; - - if (adev->flags & AMD_IS_APU) - return 0; - - /* Get the number of NAKs received and generated */ - val = RREG32_PCIE(smnPCIEP_NAK_COUNTER); - nak_r = val & 0xFFFF; - nak_g = val >> 16; - - /* Add the total number of NAKs, i.e the number of replays */ - return (nak_r + nak_g); -} - #define MMIO_REG_HOLE_OFFSET 0x1A000 static void nbio_v7_9_set_reg_remap(struct amdgpu_device *adev) @@ -524,7 +505,6 @@ const struct amdgpu_nbio_funcs nbio_v7_9_funcs = { .get_memory_partition_mode = nbio_v7_9_get_memory_partition_mode, .is_nps_switch_requested = nbio_v7_9_is_nps_switch_requested, .init_registers = nbio_v7_9_init_registers, - .get_pcie_replay_count = nbio_v7_9_get_pcie_replay_count, .set_reg_remap = nbio_v7_9_set_reg_remap, }; diff --git a/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h b/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h index f4a91b126c73c211188f59fc75a5091699a50746..73f87131a7e9f0161f953881640bc54a41859149 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h +++ b/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h @@ -106,7 +106,9 @@ enum psp_gfx_cmd_id /*IDs of performance monitoring/profiling*/ GFX_CMD_ID_CONFIG_SQ_PERFMON = 0x00000046, /* Config CGTT_SQ_CLK_CTRL */ /* Dynamic memory partitioninig (NPS mode change)*/ - GFX_CMD_ID_FB_NPS_MODE = 0x00000048, /* Configure memory partitioning mode */ + GFX_CMD_ID_FB_NPS_MODE = 0x00000048, /* Configure memory partitioning mode */ + GFX_CMD_ID_FB_FW_RESERV_ADDR = 0x00000050, /* Query FW reservation addr */ + GFX_CMD_ID_FB_FW_RESERV_EXT_ADDR = 0x00000051, /* Query FW reservation extended addr */ }; /* PSP boot config sub-commands */ @@ -404,11 +406,19 @@ struct psp_gfx_uresp_bootcfg { uint32_t boot_cfg; /* boot config data */ }; +/* Command-specific response for fw reserve info */ +struct psp_gfx_uresp_fw_reserve_info { + uint32_t reserve_base_address_hi; + uint32_t reserve_base_address_lo; + uint32_t reserve_size; +}; + /* Union of command-specific responses for GPCOM ring. */ union psp_gfx_uresp { struct psp_gfx_uresp_reserved reserved; struct psp_gfx_uresp_bootcfg boot_cfg; struct psp_gfx_uresp_fwar_db_info fwar_db_info; + struct psp_gfx_uresp_fw_reserve_info fw_reserve_info; }; /* Structure of GFX Response buffer. diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c index 145186a1e48f6b695dad38ddf557d63e68acb901..2c4ebd98927ff384993f95511cc796994b36d200 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c @@ -94,7 +94,7 @@ static int psp_v10_0_ring_create(struct psp_context *psp, /* Wait for response flag (bit 31) in C2PMSG_64 */ ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - 0x80000000, 0x8000FFFF, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); return ret; } @@ -115,7 +115,7 @@ static int psp_v10_0_ring_stop(struct psp_context *psp, /* Wait for response flag (bit 31) in C2PMSG_64 */ ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - 0x80000000, 0x80000000, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); return ret; } diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c index 215543575f477c9a981e18f8bc0f40b2ed827572..1a4a26e6ffd24cbd407d9ef68e382d41a53f7907 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c @@ -277,11 +277,13 @@ static int psp_v11_0_ring_stop(struct psp_context *psp, /* Wait for response flag (bit 31) */ if (amdgpu_sriov_vf(adev)) - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); else - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); return ret; } @@ -317,13 +319,15 @@ static int psp_v11_0_ring_create(struct psp_context *psp, mdelay(20); /* Wait for response flag (bit 31) in C2PMSG_101 */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), - 0x80000000, 0x8000FFFF, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } else { /* Wait for sOS ready for ring creation */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), + MBOX_TOS_READY_FLAG, MBOX_TOS_READY_MASK, false); if (ret) { DRM_ERROR("Failed to wait for sOS ready for ring creation\n"); return ret; @@ -347,8 +351,9 @@ static int psp_v11_0_ring_create(struct psp_context *psp, mdelay(20); /* Wait for response flag (bit 31) in C2PMSG_64 */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - 0x80000000, 0x8000FFFF, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } return ret; @@ -381,7 +386,8 @@ static int psp_v11_0_mode1_reset(struct psp_context *psp) offset = SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64); - ret = psp_wait_for(psp, offset, 0x80000000, 0x8000FFFF, false); + ret = psp_wait_for(psp, offset, MBOX_TOS_READY_FLAG, + MBOX_TOS_READY_MASK, false); if (ret) { DRM_INFO("psp is not working correctly before mode1 reset!\n"); @@ -395,7 +401,8 @@ static int psp_v11_0_mode1_reset(struct psp_context *psp) offset = SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_33); - ret = psp_wait_for(psp, offset, 0x80000000, 0x80000000, false); + ret = psp_wait_for(psp, offset, MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, + false); if (ret) { DRM_INFO("psp mode 1 reset failed!\n"); diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v11_0_8.c b/drivers/gpu/drm/amd/amdgpu/psp_v11_0_8.c index 5697760a819bc7898438b6f2ab2062087022259c..338d015c0f2ee266c72f9b754f4551bf608376db 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v11_0_8.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v11_0_8.c @@ -41,8 +41,9 @@ static int psp_v11_0_8_ring_stop(struct psp_context *psp, /* there might be handshake issue with hardware which needs delay */ mdelay(20); /* Wait for response flag (bit 31) */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } else { /* Write the ring destroy command*/ WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_64, @@ -50,8 +51,9 @@ static int psp_v11_0_8_ring_stop(struct psp_context *psp, /* there might be handshake issue with hardware which needs delay */ mdelay(20); /* Wait for response flag (bit 31) */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } return ret; @@ -87,13 +89,15 @@ static int psp_v11_0_8_ring_create(struct psp_context *psp, mdelay(20); /* Wait for response flag (bit 31) in C2PMSG_101 */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), - 0x80000000, 0x8000FFFF, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } else { /* Wait for sOS ready for ring creation */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), + MBOX_TOS_READY_FLAG, MBOX_TOS_READY_MASK, false); if (ret) { DRM_ERROR("Failed to wait for trust OS ready for ring creation\n"); return ret; @@ -117,8 +121,9 @@ static int psp_v11_0_8_ring_create(struct psp_context *psp, mdelay(20); /* Wait for response flag (bit 31) in C2PMSG_64 */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - 0x80000000, 0x8000FFFF, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } return ret; diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c index 80153f8374704a31b4dbd7d6e3a0367fca5c27fb..d54b3e0fabaf4064d69c2f67a429a3bfbed740f1 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v12_0.c @@ -163,7 +163,7 @@ static int psp_v12_0_ring_create(struct psp_context *psp, /* Wait for response flag (bit 31) in C2PMSG_64 */ ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - 0x80000000, 0x8000FFFF, false); + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); return ret; } @@ -184,11 +184,13 @@ static int psp_v12_0_ring_stop(struct psp_context *psp, /* Wait for response flag (bit 31) */ if (amdgpu_sriov_vf(adev)) - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); else - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); return ret; } @@ -219,7 +221,8 @@ static int psp_v12_0_mode1_reset(struct psp_context *psp) offset = SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64); - ret = psp_wait_for(psp, offset, 0x80000000, 0x8000FFFF, false); + ret = psp_wait_for(psp, offset, MBOX_TOS_READY_FLAG, + MBOX_TOS_READY_MASK, false); if (ret) { DRM_INFO("psp is not working correctly before mode1 reset!\n"); @@ -233,7 +236,8 @@ static int psp_v12_0_mode1_reset(struct psp_context *psp) offset = SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_33); - ret = psp_wait_for(psp, offset, 0x80000000, 0x80000000, false); + ret = psp_wait_for(psp, offset, MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, + false); if (ret) { DRM_INFO("psp mode 1 reset failed!\n"); diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c index df612fd9cc507471166fbd0fccfac8a8cc79008b..58b6b64dcd683bd693a0d06655bbb6b001b65fe2 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c @@ -42,7 +42,9 @@ MODULE_FIRMWARE("amdgpu/psp_13_0_5_ta.bin"); MODULE_FIRMWARE("amdgpu/psp_13_0_8_toc.bin"); MODULE_FIRMWARE("amdgpu/psp_13_0_8_ta.bin"); MODULE_FIRMWARE("amdgpu/psp_13_0_0_sos.bin"); +MODULE_FIRMWARE("amdgpu/psp_13_0_0_sos_kicker.bin"); MODULE_FIRMWARE("amdgpu/psp_13_0_0_ta.bin"); +MODULE_FIRMWARE("amdgpu/psp_13_0_0_ta_kicker.bin"); MODULE_FIRMWARE("amdgpu/psp_13_0_7_sos.bin"); MODULE_FIRMWARE("amdgpu/psp_13_0_7_ta.bin"); MODULE_FIRMWARE("amdgpu/psp_13_0_10_sos.bin"); @@ -382,8 +384,9 @@ static int psp_v13_0_ring_stop(struct psp_context *psp, /* there might be handshake issue with hardware which needs delay */ mdelay(20); /* Wait for response flag (bit 31) */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_101), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_101), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } else { /* Write the ring destroy command*/ WREG32_SOC15(MP0, 0, regMP0_SMN_C2PMSG_64, @@ -391,8 +394,9 @@ static int psp_v13_0_ring_stop(struct psp_context *psp, /* there might be handshake issue with hardware which needs delay */ mdelay(20); /* Wait for response flag (bit 31) */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } return ret; @@ -428,13 +432,15 @@ static int psp_v13_0_ring_create(struct psp_context *psp, mdelay(20); /* Wait for response flag (bit 31) in C2PMSG_101 */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_101), - 0x80000000, 0x8000FFFF, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_101), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } else { /* Wait for sOS ready for ring creation */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), + MBOX_TOS_READY_FLAG, MBOX_TOS_READY_MASK, false); if (ret) { DRM_ERROR("Failed to wait for trust OS ready for ring creation\n"); return ret; @@ -458,8 +464,9 @@ static int psp_v13_0_ring_create(struct psp_context *psp, mdelay(20); /* Wait for response flag (bit 31) in C2PMSG_64 */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), - 0x80000000, 0x8000FFFF, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } return ret; diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v13_0_4.c b/drivers/gpu/drm/amd/amdgpu/psp_v13_0_4.c index eaa5512a21dacded38865a72f67b8931bbe53770..f65af52c1c193989994b85b781cf613be08092eb 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v13_0_4.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v13_0_4.c @@ -204,8 +204,9 @@ static int psp_v13_0_4_ring_stop(struct psp_context *psp, /* there might be handshake issue with hardware which needs delay */ mdelay(20); /* Wait for response flag (bit 31) */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_101), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_101), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } else { /* Write the ring destroy command*/ WREG32_SOC15(MP0, 0, regMP0_SMN_C2PMSG_64, @@ -213,8 +214,9 @@ static int psp_v13_0_4_ring_stop(struct psp_context *psp, /* there might be handshake issue with hardware which needs delay */ mdelay(20); /* Wait for response flag (bit 31) */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } return ret; @@ -250,13 +252,15 @@ static int psp_v13_0_4_ring_create(struct psp_context *psp, mdelay(20); /* Wait for response flag (bit 31) in C2PMSG_101 */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_101), - 0x80000000, 0x8000FFFF, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_101), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } else { /* Wait for sOS ready for ring creation */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), + MBOX_TOS_READY_FLAG, MBOX_TOS_READY_MASK, false); if (ret) { DRM_ERROR("Failed to wait for trust OS ready for ring creation\n"); return ret; @@ -280,8 +284,9 @@ static int psp_v13_0_4_ring_create(struct psp_context *psp, mdelay(20); /* Wait for response flag (bit 31) in C2PMSG_64 */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), - 0x80000000, 0x8000FFFF, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_64), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } return ret; diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v14_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v14_0.c index 256288c6cd78ef968130f368676adea484532776..ffa47c7d24c919c8f1f5b73cafa4d8c473c37f46 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v14_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v14_0.c @@ -248,8 +248,9 @@ static int psp_v14_0_ring_stop(struct psp_context *psp, /* there might be handshake issue with hardware which needs delay */ mdelay(20); /* Wait for response flag (bit 31) */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_101), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_101), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } else { /* Write the ring destroy command*/ WREG32_SOC15(MP0, 0, regMPASP_SMN_C2PMSG_64, @@ -257,8 +258,9 @@ static int psp_v14_0_ring_stop(struct psp_context *psp, /* there might be handshake issue with hardware which needs delay */ mdelay(20); /* Wait for response flag (bit 31) */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_64), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_64), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } return ret; @@ -294,13 +296,15 @@ static int psp_v14_0_ring_create(struct psp_context *psp, mdelay(20); /* Wait for response flag (bit 31) in C2PMSG_101 */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_101), - 0x80000000, 0x8000FFFF, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_101), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } else { /* Wait for sOS ready for ring creation */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_64), - 0x80000000, 0x80000000, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_64), + MBOX_TOS_READY_FLAG, MBOX_TOS_READY_MASK, false); if (ret) { DRM_ERROR("Failed to wait for trust OS ready for ring creation\n"); return ret; @@ -324,8 +328,9 @@ static int psp_v14_0_ring_create(struct psp_context *psp, mdelay(20); /* Wait for response flag (bit 31) in C2PMSG_64 */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_64), - 0x80000000, 0x8000FFFF, false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMPASP_SMN_C2PMSG_64), + MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK, false); } return ret; diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c index bcde34e4e0a125902daae312f04f8da69c56483d..c05f3c1f50db4894ed2fc3b45623ac1734d77355 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c @@ -45,6 +45,7 @@ #include "amdgpu_ras.h" MODULE_FIRMWARE("amdgpu/sdma_4_4_2.bin"); +MODULE_FIRMWARE("amdgpu/sdma_4_4_4.bin"); MODULE_FIRMWARE("amdgpu/sdma_4_4_5.bin"); static const struct amdgpu_hwip_reg_entry sdma_reg_list_4_4_2[] = { @@ -109,6 +110,8 @@ static void sdma_v4_4_2_set_ras_funcs(struct amdgpu_device *adev); static void sdma_v4_4_2_update_reset_mask(struct amdgpu_device *adev); static int sdma_v4_4_2_stop_queue(struct amdgpu_ring *ring); static int sdma_v4_4_2_restore_queue(struct amdgpu_ring *ring); +static int sdma_v4_4_2_soft_reset_engine(struct amdgpu_device *adev, + u32 instance_id); static u32 sdma_v4_4_2_get_reg_offset(struct amdgpu_device *adev, u32 instance, u32 offset) @@ -490,7 +493,7 @@ static void sdma_v4_4_2_inst_gfx_stop(struct amdgpu_device *adev, { struct amdgpu_ring *sdma[AMDGPU_MAX_SDMA_INSTANCES]; u32 doorbell_offset, doorbell; - u32 rb_cntl, ib_cntl; + u32 rb_cntl, ib_cntl, sdma_cntl; int i; for_each_inst(i, inst_mask) { @@ -502,6 +505,9 @@ static void sdma_v4_4_2_inst_gfx_stop(struct amdgpu_device *adev, ib_cntl = RREG32_SDMA(i, regSDMA_GFX_IB_CNTL); ib_cntl = REG_SET_FIELD(ib_cntl, SDMA_GFX_IB_CNTL, IB_ENABLE, 0); WREG32_SDMA(i, regSDMA_GFX_IB_CNTL, ib_cntl); + sdma_cntl = RREG32_SDMA(i, regSDMA_CNTL); + sdma_cntl = REG_SET_FIELD(sdma_cntl, SDMA_CNTL, UTC_L1_ENABLE, 0); + WREG32_SDMA(i, regSDMA_CNTL, sdma_cntl); if (sdma[i]->use_doorbell) { doorbell = RREG32_SDMA(i, regSDMA_GFX_DOORBELL); @@ -995,6 +1001,7 @@ static int sdma_v4_4_2_inst_start(struct amdgpu_device *adev, /* set utc l1 enable flag always to 1 */ temp = RREG32_SDMA(i, regSDMA_CNTL); temp = REG_SET_FIELD(temp, SDMA_CNTL, UTC_L1_ENABLE, 1); + WREG32_SDMA(i, regSDMA_CNTL, temp); if (amdgpu_ip_version(adev, SDMA0_HWIP, 0) < IP_VERSION(4, 4, 5)) { /* enable context empty interrupt during initialization */ @@ -1337,6 +1344,7 @@ static bool sdma_v4_4_2_fw_support_paging_queue(struct amdgpu_device *adev) static const struct amdgpu_sdma_funcs sdma_v4_4_2_sdma_funcs = { .stop_kernel_queue = &sdma_v4_4_2_stop_queue, .start_kernel_queue = &sdma_v4_4_2_restore_queue, + .soft_reset_kernel_queue = &sdma_v4_4_2_soft_reset_engine, }; static int sdma_v4_4_2_early_init(struct amdgpu_ip_block *ip_block) @@ -1648,45 +1656,27 @@ static bool sdma_v4_4_2_is_queue_selected(struct amdgpu_device *adev, uint32_t i return (context_status & SDMA_GFX_CONTEXT_STATUS__SELECTED_MASK) != 0; } -static bool sdma_v4_4_2_ring_is_guilty(struct amdgpu_ring *ring) -{ - struct amdgpu_device *adev = ring->adev; - uint32_t instance_id = ring->me; - - return sdma_v4_4_2_is_queue_selected(adev, instance_id, false); -} - -static bool sdma_v4_4_2_page_ring_is_guilty(struct amdgpu_ring *ring) -{ - struct amdgpu_device *adev = ring->adev; - uint32_t instance_id = ring->me; - - if (!adev->sdma.has_page_queue) - return false; - - return sdma_v4_4_2_is_queue_selected(adev, instance_id, true); -} - -static int sdma_v4_4_2_reset_queue(struct amdgpu_ring *ring, unsigned int vmid) +static int sdma_v4_4_2_reset_queue(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; - u32 id = GET_INST(SDMA0, ring->me); + u32 id = ring->me; int r; if (!(adev->sdma.supported_reset & AMDGPU_RESET_TYPE_PER_QUEUE)) return -EOPNOTSUPP; - amdgpu_amdkfd_suspend(adev, false); + amdgpu_amdkfd_suspend(adev, true); r = amdgpu_sdma_reset_engine(adev, id); - amdgpu_amdkfd_resume(adev, false); - + amdgpu_amdkfd_resume(adev, true); return r; } static int sdma_v4_4_2_stop_queue(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; - u32 instance_id = GET_INST(SDMA0, ring->me); + u32 instance_id = ring->me; u32 inst_mask; uint64_t rptr; @@ -1724,8 +1714,8 @@ static int sdma_v4_4_2_stop_queue(struct amdgpu_ring *ring) static int sdma_v4_4_2_restore_queue(struct amdgpu_ring *ring) { struct amdgpu_device *adev = ring->adev; - u32 inst_mask; - int i; + u32 inst_mask, tmp_mask; + int i, r; inst_mask = 1 << ring->me; udelay(50); @@ -1742,7 +1732,33 @@ static int sdma_v4_4_2_restore_queue(struct amdgpu_ring *ring) return -ETIMEDOUT; } - return sdma_v4_4_2_inst_start(adev, inst_mask, true); + r = sdma_v4_4_2_inst_start(adev, inst_mask, true); + if (r) + return r; + + tmp_mask = inst_mask; + for_each_inst(i, tmp_mask) { + ring = &adev->sdma.instance[i].ring; + + amdgpu_fence_driver_force_completion(ring); + + if (adev->sdma.has_page_queue) { + struct amdgpu_ring *page = &adev->sdma.instance[i].page; + + amdgpu_fence_driver_force_completion(page); + } + } + + return r; +} + +static int sdma_v4_4_2_soft_reset_engine(struct amdgpu_device *adev, + u32 instance_id) +{ + /* For SDMA 4.x, use the existing DPM interface for backward compatibility + * we need to convert the logical instance ID to physical instance ID before reset. + */ + return amdgpu_dpm_reset_sdma(adev, 1 << GET_INST(SDMA0, instance_id)); } static int sdma_v4_4_2_set_trap_irq_state(struct amdgpu_device *adev, @@ -2139,7 +2155,6 @@ static const struct amdgpu_ring_funcs sdma_v4_4_2_ring_funcs = { .emit_reg_wait = sdma_v4_4_2_ring_emit_reg_wait, .emit_reg_write_reg_wait = amdgpu_ring_emit_reg_write_reg_wait_helper, .reset = sdma_v4_4_2_reset_queue, - .is_guilty = sdma_v4_4_2_ring_is_guilty, }; static const struct amdgpu_ring_funcs sdma_v4_4_2_page_ring_funcs = { @@ -2172,7 +2187,6 @@ static const struct amdgpu_ring_funcs sdma_v4_4_2_page_ring_funcs = { .emit_reg_wait = sdma_v4_4_2_ring_emit_reg_wait, .emit_reg_write_reg_wait = amdgpu_ring_emit_reg_write_reg_wait_helper, .reset = sdma_v4_4_2_reset_queue, - .is_guilty = sdma_v4_4_2_page_ring_is_guilty, }; static void sdma_v4_4_2_set_ring_funcs(struct amdgpu_device *adev) diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c index 9505ae96fbecca12f89beda455fa4838aa591d0f..4d72b085b3dd77c185545ff0aea20ff1f134a8a7 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c @@ -1399,6 +1399,7 @@ static int sdma_v5_0_sw_init(struct amdgpu_ip_block *ip_block) return r; for (i = 0; i < adev->sdma.num_instances; i++) { + mutex_init(&adev->sdma.instance[i].engine_reset_mutex); adev->sdma.instance[i].funcs = &sdma_v5_0_sdma_funcs; ring = &adev->sdma.instance[i].ring; ring->ring_obj = NULL; @@ -1538,12 +1539,19 @@ static int sdma_v5_0_soft_reset(struct amdgpu_ip_block *ip_block) return 0; } -static int sdma_v5_0_reset_queue(struct amdgpu_ring *ring, unsigned int vmid) +static int sdma_v5_0_reset_queue(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; u32 inst_id = ring->me; + int r; + + amdgpu_amdkfd_suspend(adev, true); + r = amdgpu_sdma_reset_engine(adev, inst_id); + amdgpu_amdkfd_resume(adev, true); - return amdgpu_sdma_reset_engine(adev, inst_id); + return r; } static int sdma_v5_0_stop_queue(struct amdgpu_ring *ring) @@ -1610,7 +1618,10 @@ static int sdma_v5_0_restore_queue(struct amdgpu_ring *ring) r = sdma_v5_0_gfx_resume_instance(adev, inst_id, true); amdgpu_gfx_rlc_exit_safe_mode(adev, 0); - return r; + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + return 0; } static int sdma_v5_0_ring_preempt_ib(struct amdgpu_ring *ring) diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c index a6e612b4a8928d56088998f3b08360cc73b8bd87..42a25150f83ac2a7a048515a5f374a1e147f2933 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c @@ -1318,6 +1318,7 @@ static int sdma_v5_2_sw_init(struct amdgpu_ip_block *ip_block) } for (i = 0; i < adev->sdma.num_instances; i++) { + mutex_init(&adev->sdma.instance[i].engine_reset_mutex); adev->sdma.instance[i].funcs = &sdma_v5_2_sdma_funcs; ring = &adev->sdma.instance[i].ring; ring->ring_obj = NULL; @@ -1451,12 +1452,19 @@ static int sdma_v5_2_wait_for_idle(struct amdgpu_ip_block *ip_block) return -ETIMEDOUT; } -static int sdma_v5_2_reset_queue(struct amdgpu_ring *ring, unsigned int vmid) +static int sdma_v5_2_reset_queue(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; u32 inst_id = ring->me; + int r; + + amdgpu_amdkfd_suspend(adev, true); + r = amdgpu_sdma_reset_engine(adev, inst_id); + amdgpu_amdkfd_resume(adev, true); - return amdgpu_sdma_reset_engine(adev, inst_id); + return r; } static int sdma_v5_2_stop_queue(struct amdgpu_ring *ring) @@ -1526,7 +1534,10 @@ static int sdma_v5_2_restore_queue(struct amdgpu_ring *ring) r = sdma_v5_2_gfx_resume_instance(adev, inst_id, true); amdgpu_gfx_rlc_exit_safe_mode(adev, 0); - return r; + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + return 0; } static int sdma_v5_2_ring_preempt_ib(struct amdgpu_ring *ring) diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c index 5a70ae17be04e4c18112f67ba2b75e828f2a86c0..d2effa53181766631155869f4637e3ca808464bb 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c @@ -1374,9 +1374,42 @@ static int sdma_v6_0_sw_init(struct amdgpu_ip_block *ip_block) else DRM_ERROR("Failed to allocated memory for SDMA IP Dump\n"); - /* add firmware version checks here */ - if (0 && !adev->sdma.disable_uq) - adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs; + switch (amdgpu_ip_version(adev, SDMA0_HWIP, 0)) { + case IP_VERSION(6, 0, 0): + if ((adev->sdma.instance[0].fw_version >= 24) && !adev->sdma.disable_uq) + adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs; + break; + case IP_VERSION(6, 0, 1): + if ((adev->sdma.instance[0].fw_version >= 18) && !adev->sdma.disable_uq) + adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs; + break; + case IP_VERSION(6, 0, 2): + if ((adev->sdma.instance[0].fw_version >= 21) && !adev->sdma.disable_uq) + adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs; + break; + case IP_VERSION(6, 0, 3): + if ((adev->sdma.instance[0].fw_version >= 25) && !adev->sdma.disable_uq) + adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs; + break; + case IP_VERSION(6, 1, 0): + if ((adev->sdma.instance[0].fw_version >= 14) && !adev->sdma.disable_uq) + adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs; + break; + case IP_VERSION(6, 1, 1): + if ((adev->sdma.instance[0].fw_version >= 17) && !adev->sdma.disable_uq) + adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs; + break; + case IP_VERSION(6, 1, 2): + if ((adev->sdma.instance[0].fw_version >= 15) && !adev->sdma.disable_uq) + adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs; + break; + case IP_VERSION(6, 1, 3): + if ((adev->sdma.instance[0].fw_version >= 10) && !adev->sdma.disable_uq) + adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs; + break; + default: + break; + } r = amdgpu_sdma_sysfs_reset_mask_init(adev); if (r) @@ -1537,7 +1570,9 @@ static int sdma_v6_0_ring_preempt_ib(struct amdgpu_ring *ring) return r; } -static int sdma_v6_0_reset_queue(struct amdgpu_ring *ring, unsigned int vmid) +static int sdma_v6_0_reset_queue(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; int i, r; @@ -1555,11 +1590,18 @@ static int sdma_v6_0_reset_queue(struct amdgpu_ring *ring, unsigned int vmid) return -EINVAL; } + drm_sched_wqueue_stop(&ring->sched); + r = amdgpu_mes_reset_legacy_queue(adev, ring, vmid, true); if (r) return r; - return sdma_v6_0_gfx_resume_instance(adev, i, true); + r = sdma_v6_0_gfx_resume_instance(adev, i, true); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static int sdma_v6_0_set_trap_irq_state(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c index ad47d0bdf77752884319c41673f032ae2915b74b..99a080bad2a3d0e86fd06b9695d429d73b680d2c 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c @@ -802,7 +802,9 @@ static bool sdma_v7_0_check_soft_reset(struct amdgpu_ip_block *ip_block) return false; } -static int sdma_v7_0_reset_queue(struct amdgpu_ring *ring, unsigned int vmid) +static int sdma_v7_0_reset_queue(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; int i, r; @@ -820,11 +822,18 @@ static int sdma_v7_0_reset_queue(struct amdgpu_ring *ring, unsigned int vmid) return -EINVAL; } + drm_sched_wqueue_stop(&ring->sched); + r = amdgpu_mes_reset_legacy_queue(adev, ring, vmid, true); if (r) return r; - return sdma_v7_0_gfx_resume_instance(adev, i, true); + r = sdma_v7_0_gfx_resume_instance(adev, i, true); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } /** @@ -1349,9 +1358,15 @@ static int sdma_v7_0_sw_init(struct amdgpu_ip_block *ip_block) else DRM_ERROR("Failed to allocated memory for SDMA IP Dump\n"); - /* add firmware version checks here */ - if (0 && !adev->sdma.disable_uq) - adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs; + switch (amdgpu_ip_version(adev, SDMA0_HWIP, 0)) { + case IP_VERSION(7, 0, 0): + case IP_VERSION(7, 0, 1): + if ((adev->sdma.instance[0].fw_version >= 7836028) && !adev->sdma.disable_uq) + adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs; + break; + default: + break; + } return r; } diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.h b/drivers/gpu/drm/amd/amdgpu/soc15.h index ef7c603b50ae32cf4b87996c607ed9ddd36371c6..c8ac11a9cdef61967871d844a1646d427199bb87 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15.h +++ b/drivers/gpu/drm/amd/amdgpu/soc15.h @@ -118,7 +118,6 @@ int vega10_reg_base_init(struct amdgpu_device *adev); int vega20_reg_base_init(struct amdgpu_device *adev); int arct_reg_base_init(struct amdgpu_device *adev); int aldebaran_reg_base_init(struct amdgpu_device *adev); -void aqua_vanjaram_ip_map_init(struct amdgpu_device *adev); u64 aqua_vanjaram_encode_ext_smn_addressing(int ext_id); int aqua_vanjaram_init_soc_config(struct amdgpu_device *adev); ssize_t aqua_vanjaram_get_reg_state(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c index b5071f77f78d233510587f8e9013d7e194b67cd0..eec9133e1b2c4fe20ef8e6f9679002dbdc2f46a1 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c @@ -1967,18 +1967,27 @@ static int vcn_v4_0_ring_patch_cs_in_place(struct amdgpu_cs_parser *p, return 0; } -static int vcn_v4_0_ring_reset(struct amdgpu_ring *ring, unsigned int vmid) +static int vcn_v4_0_ring_reset(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; struct amdgpu_vcn_inst *vinst = &adev->vcn.inst[ring->me]; + int r; if (!(adev->vcn.supported_reset & AMDGPU_RESET_TYPE_PER_QUEUE)) return -EOPNOTSUPP; + drm_sched_wqueue_stop(&ring->sched); vcn_v4_0_stop(vinst); vcn_v4_0_start(vinst); - return amdgpu_ring_test_helper(ring); + r = amdgpu_ring_test_helper(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static struct amdgpu_ring_funcs vcn_v4_0_unified_ring_vm_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c index 5a33140f5723519ac9d2b3b6726a92aee3b57d7c..d8fd32c1e38ec27cf4249c4e59cb27c3512d1ca8 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c @@ -1594,7 +1594,9 @@ static void vcn_v4_0_3_unified_ring_set_wptr(struct amdgpu_ring *ring) } } -static int vcn_v4_0_3_ring_reset(struct amdgpu_ring *ring, unsigned int vmid) +static int vcn_v4_0_3_ring_reset(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { int r = 0; int vcn_inst; @@ -1607,6 +1609,8 @@ static int vcn_v4_0_3_ring_reset(struct amdgpu_ring *ring, unsigned int vmid) if (!(adev->vcn.supported_reset & AMDGPU_RESET_TYPE_PER_QUEUE)) return -EOPNOTSUPP; + drm_sched_wqueue_stop(&ring->sched); + vcn_inst = GET_INST(VCN, ring->me); r = amdgpu_dpm_reset_vcn(adev, 1 << vcn_inst); @@ -1621,8 +1625,11 @@ static int vcn_v4_0_3_ring_reset(struct amdgpu_ring *ring, unsigned int vmid) vcn_v4_0_3_hw_init_inst(vinst); vcn_v4_0_3_start_dpg_mode(vinst, adev->vcn.inst[ring->me].indirect_sram); r = amdgpu_ring_test_helper(ring); - - return r; + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static const struct amdgpu_ring_funcs vcn_v4_0_3_unified_ring_vm_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c index 16ade84facc789e1311c8bf08e54b0a3b046f121..7e37ddea63550d89b0e0a7933735cd26feec989c 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c @@ -1465,18 +1465,27 @@ static void vcn_v4_0_5_unified_ring_set_wptr(struct amdgpu_ring *ring) } } -static int vcn_v4_0_5_ring_reset(struct amdgpu_ring *ring, unsigned int vmid) +static int vcn_v4_0_5_ring_reset(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; struct amdgpu_vcn_inst *vinst = &adev->vcn.inst[ring->me]; + int r; if (!(adev->vcn.supported_reset & AMDGPU_RESET_TYPE_PER_QUEUE)) return -EOPNOTSUPP; + drm_sched_wqueue_stop(&ring->sched); vcn_v4_0_5_stop(vinst); vcn_v4_0_5_start(vinst); - return amdgpu_ring_test_helper(ring); + r = amdgpu_ring_test_helper(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static struct amdgpu_ring_funcs vcn_v4_0_5_unified_ring_vm_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c index f8e3f0b882da563f614c60f9df2a04e9b512791c..47c0bcc9e7d801e4ca9ff9f64cb9004ea9e901b5 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c @@ -1192,18 +1192,27 @@ static void vcn_v5_0_0_unified_ring_set_wptr(struct amdgpu_ring *ring) } } -static int vcn_v5_0_0_ring_reset(struct amdgpu_ring *ring, unsigned int vmid) +static int vcn_v5_0_0_ring_reset(struct amdgpu_ring *ring, + unsigned int vmid, + struct amdgpu_fence *timedout_fence) { struct amdgpu_device *adev = ring->adev; struct amdgpu_vcn_inst *vinst = &adev->vcn.inst[ring->me]; + int r; if (!(adev->vcn.supported_reset & AMDGPU_RESET_TYPE_PER_QUEUE)) return -EOPNOTSUPP; + drm_sched_wqueue_stop(&ring->sched); vcn_v5_0_0_stop(vinst); vcn_v5_0_0_start(vinst); - return amdgpu_ring_test_helper(ring); + r = amdgpu_ring_test_helper(ring); + if (r) + return r; + amdgpu_fence_driver_force_completion(ring); + drm_sched_wqueue_start(&ring->sched); + return 0; } static const struct amdgpu_ring_funcs vcn_v5_0_0_unified_ring_vm_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c index 338cf43c45fe7ad6e8cbf3888712932553e477fb..cdefd7fcb0da607d36e947c1eefe01c5dc71abdb 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c @@ -669,6 +669,9 @@ static int vcn_v5_0_1_start_dpg_mode(struct amdgpu_vcn_inst *vinst, if (indirect) amdgpu_vcn_psp_update_sram(adev, inst_idx, AMDGPU_UCODE_ID_VCN0_RAM); + /* resetting ring, fw should not check RB ring */ + fw_shared->sq.queue_mode |= FW_QUEUE_RING_RESET; + /* Pause dpg */ vcn_v5_0_1_pause_dpg_mode(vinst, &state); @@ -681,7 +684,7 @@ static int vcn_v5_0_1_start_dpg_mode(struct amdgpu_vcn_inst *vinst, tmp = RREG32_SOC15(VCN, vcn_inst, regVCN_RB_ENABLE); tmp &= ~(VCN_RB_ENABLE__RB1_EN_MASK); WREG32_SOC15(VCN, vcn_inst, regVCN_RB_ENABLE, tmp); - fw_shared->sq.queue_mode |= FW_QUEUE_RING_RESET; + WREG32_SOC15(VCN, vcn_inst, regUVD_RB_RPTR, 0); WREG32_SOC15(VCN, vcn_inst, regUVD_RB_WPTR, 0); @@ -692,6 +695,7 @@ static int vcn_v5_0_1_start_dpg_mode(struct amdgpu_vcn_inst *vinst, tmp = RREG32_SOC15(VCN, vcn_inst, regVCN_RB_ENABLE); tmp |= VCN_RB_ENABLE__RB1_EN_MASK; WREG32_SOC15(VCN, vcn_inst, regVCN_RB_ENABLE, tmp); + /* resetting done, fw can check RB ring */ fw_shared->sq.queue_mode &= ~(FW_QUEUE_RING_RESET | FW_QUEUE_DPG_HOLD_OFF); WREG32_SOC15(VCN, vcn_inst, regVCN_RB1_DB_CTRL, diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c index a2149afa5803e2fe4f0320315848ff1d9a7e038c..828a9ceef1e763018677ef6f01aac62fe0f945fd 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c @@ -22,7 +22,6 @@ */ #include -#include #include #include #include diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c index bf0854bd55551bd01ba23dea28fca21e319c6ac6..7e749f9b6d69daa7043e7ec081db56ba20e5eef5 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c @@ -971,7 +971,7 @@ int kgd2kfd_pre_reset(struct kfd_dev *kfd, kfd_smi_event_update_gpu_reset(node, false, reset_context); } - kgd2kfd_suspend(kfd, false); + kgd2kfd_suspend(kfd, true); for (i = 0; i < kfd->num_nodes; i++) kfd_signal_reset_event(kfd->nodes[i]); @@ -1013,13 +1013,33 @@ int kgd2kfd_post_reset(struct kfd_dev *kfd) return 0; } -bool kfd_is_locked(void) +bool kfd_is_locked(struct kfd_dev *kfd) { + uint8_t id = 0; + struct kfd_node *dev; + lockdep_assert_held(&kfd_processes_mutex); - return (kfd_locked > 0); + + /* check reset/suspend lock */ + if (kfd_locked > 0) + return true; + + if (kfd) + return kfd->kfd_dev_lock > 0; + + /* check lock on all cgroup accessible devices */ + while (kfd_topology_enum_kfd_devices(id++, &dev) == 0) { + if (!dev || kfd_devcgroup_check_permission(dev)) + continue; + + if (dev->kfd->kfd_dev_lock > 0) + return true; + } + + return false; } -void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm) +void kgd2kfd_suspend(struct kfd_dev *kfd, bool suspend_proc) { struct kfd_node *node; int i; @@ -1027,14 +1047,8 @@ void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm) if (!kfd->init_complete) return; - /* for runtime suspend, skip locking kfd */ - if (!run_pm) { - mutex_lock(&kfd_processes_mutex); - /* For first KFD device suspend all the KFD processes */ - if (++kfd_locked == 1) - kfd_suspend_all_processes(); - mutex_unlock(&kfd_processes_mutex); - } + if (suspend_proc) + kgd2kfd_suspend_process(kfd); for (i = 0; i < kfd->num_nodes; i++) { node = kfd->nodes[i]; @@ -1042,7 +1056,7 @@ void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm) } } -int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm) +int kgd2kfd_resume(struct kfd_dev *kfd, bool resume_proc) { int ret, i; @@ -1055,14 +1069,36 @@ int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm) return ret; } - /* for runtime resume, skip unlocking kfd */ - if (!run_pm) { - mutex_lock(&kfd_processes_mutex); - if (--kfd_locked == 0) - ret = kfd_resume_all_processes(); - WARN_ONCE(kfd_locked < 0, "KFD suspend / resume ref. error"); - mutex_unlock(&kfd_processes_mutex); - } + if (resume_proc) + ret = kgd2kfd_resume_process(kfd); + + return ret; +} + +void kgd2kfd_suspend_process(struct kfd_dev *kfd) +{ + if (!kfd->init_complete) + return; + + mutex_lock(&kfd_processes_mutex); + /* For first KFD device suspend all the KFD processes */ + if (++kfd_locked == 1) + kfd_suspend_all_processes(); + mutex_unlock(&kfd_processes_mutex); +} + +int kgd2kfd_resume_process(struct kfd_dev *kfd) +{ + int ret = 0; + + if (!kfd->init_complete) + return 0; + + mutex_lock(&kfd_processes_mutex); + if (--kfd_locked == 0) + ret = kfd_resume_all_processes(); + WARN_ONCE(kfd_locked < 0, "KFD suspend / resume ref. error"); + mutex_unlock(&kfd_processes_mutex); return ret; } @@ -1442,24 +1478,53 @@ unsigned int kfd_get_num_xgmi_sdma_engines(struct kfd_node *node) kfd_get_num_sdma_engines(node); } -int kgd2kfd_check_and_lock_kfd(void) +int kgd2kfd_check_and_lock_kfd(struct kfd_dev *kfd) { + struct kfd_process *p; + int r = 0, temp, idx; + mutex_lock(&kfd_processes_mutex); - if (!hash_empty(kfd_processes_table) || kfd_is_locked()) { - mutex_unlock(&kfd_processes_mutex); - return -EBUSY; + + if (hash_empty(kfd_processes_table) && !kfd_is_locked(kfd)) + goto out; + + /* fail under system reset/resume or kfd device is partition switching. */ + if (kfd_is_locked(kfd)) { + r = -EBUSY; + goto out; + } + + /* + * ensure all running processes are cgroup excluded from device before mode switch. + * i.e. no pdd was created on the process socket. + */ + idx = srcu_read_lock(&kfd_processes_srcu); + hash_for_each_rcu(kfd_processes_table, temp, p, kfd_processes) { + int i; + + for (i = 0; i < p->n_pdds; i++) { + if (p->pdds[i]->dev->kfd != kfd) + continue; + + r = -EBUSY; + goto proc_check_unlock; + } } - ++kfd_locked; +proc_check_unlock: + srcu_read_unlock(&kfd_processes_srcu, idx); +out: + if (!r) + ++kfd->kfd_dev_lock; mutex_unlock(&kfd_processes_mutex); - return 0; + return r; } -void kgd2kfd_unlock_kfd(void) +void kgd2kfd_unlock_kfd(struct kfd_dev *kfd) { mutex_lock(&kfd_processes_mutex); - --kfd_locked; + --kfd->kfd_dev_lock; mutex_unlock(&kfd_processes_mutex); } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c index dbcb60eb54b244f56f3c1482303eef8c9357e636..1d170dc50df3385550800276a64bad2097f04b72 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c @@ -23,7 +23,6 @@ */ #include -#include #include #include #include diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c index 8fa6489b6f5d9fb4311177141869fa9fc66eb136..505036968a7759c752c1d81e840b539e884708ec 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c @@ -240,7 +240,7 @@ static int pm_map_queues_v9(struct packet_manager *pm, uint32_t *buffer, packet->bitfields2.engine_sel = engine_sel__mes_map_queues__compute_vi; - packet->bitfields2.gws_control_queue = q->gws ? 1 : 0; + packet->bitfields2.gws_control_queue = q->properties.is_gws ? 1 : 0; packet->bitfields2.extended_engine_sel = extended_engine_sel__mes_map_queues__legacy_engine_sel; packet->bitfields2.queue_type = diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h index d221c58dccc3ccaa650ab535a5f97192fce5fef1..67694bcd9464653fdf4ce8d0d6f070766d25b048 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h @@ -372,6 +372,9 @@ struct kfd_dev { /* bitmap for dynamic doorbell allocation from doorbell object */ unsigned long *doorbell_bitmap; + + /* for dynamic partitioning */ + int kfd_dev_lock; }; enum kfd_mempool { @@ -1536,7 +1539,7 @@ static inline bool kfd_flush_tlb_after_unmap(struct kfd_dev *dev) int kfd_send_exception_to_runtime(struct kfd_process *p, unsigned int queue_id, uint64_t error_reason); -bool kfd_is_locked(void); +bool kfd_is_locked(struct kfd_dev *kfd); /* Compute profile */ void kfd_inc_compute_active(struct kfd_node *dev); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c index 722ac1662bdc19be73382fc822b5b9bc62e3f5f0..5be28c6c4f6aa592b4b188c6e477651fff2c4d09 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c @@ -854,7 +854,7 @@ struct kfd_process *kfd_create_process(struct task_struct *thread) */ mutex_lock(&kfd_processes_mutex); - if (kfd_is_locked()) { + if (kfd_is_locked(NULL)) { pr_debug("KFD is locked! Cannot create process"); process = ERR_PTR(-EINVAL); goto out; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c index 865dca2547de5b6557925a2d2411dff8a3d65898..a0f22ea6d15af73b9c6cacb781a7354b2791f32a 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c @@ -1171,13 +1171,12 @@ svm_range_split_head(struct svm_range *prange, uint64_t new_start, } static void -svm_range_add_child(struct svm_range *prange, struct mm_struct *mm, - struct svm_range *pchild, enum svm_work_list_ops op) +svm_range_add_child(struct svm_range *prange, struct svm_range *pchild, enum svm_work_list_ops op) { pr_debug("add child 0x%p [0x%lx 0x%lx] to prange 0x%p child list %d\n", pchild, pchild->start, pchild->last, prange, op); - pchild->work_item.mm = mm; + pchild->work_item.mm = NULL; pchild->work_item.op = op; list_add_tail(&pchild->child_list, &prange->child_list); } @@ -1278,7 +1277,7 @@ svm_range_get_pte_flags(struct kfd_node *node, mapping_flags |= ext_coherent ? AMDGPU_VM_MTYPE_UC : AMDGPU_VM_MTYPE_NC; /* system memory accessed by the dGPU */ } else { - if (gc_ip_version < IP_VERSION(9, 5, 0)) + if (gc_ip_version < IP_VERSION(9, 5, 0) || ext_coherent) mapping_flags |= AMDGPU_VM_MTYPE_UC; else mapping_flags |= AMDGPU_VM_MTYPE_NC; @@ -2394,15 +2393,17 @@ svm_range_add_list_work(struct svm_range_list *svms, struct svm_range *prange, prange->work_item.op != SVM_OP_UNMAP_RANGE) prange->work_item.op = op; } else { - prange->work_item.op = op; - - /* Pairs with mmput in deferred_list_work */ - mmget(mm); - prange->work_item.mm = mm; - list_add_tail(&prange->deferred_list, - &prange->svms->deferred_range_list); - pr_debug("add prange 0x%p [0x%lx 0x%lx] to work list op %d\n", - prange, prange->start, prange->last, op); + /* Pairs with mmput in deferred_list_work. + * If process is exiting and mm is gone, don't update mmu notifier. + */ + if (mmget_not_zero(mm)) { + prange->work_item.mm = mm; + prange->work_item.op = op; + list_add_tail(&prange->deferred_list, + &prange->svms->deferred_range_list); + pr_debug("add prange 0x%p [0x%lx 0x%lx] to work list op %d\n", + prange, prange->start, prange->last, op); + } } spin_unlock(&svms->deferred_list_lock); } @@ -2416,8 +2417,7 @@ void schedule_deferred_list_work(struct svm_range_list *svms) } static void -svm_range_unmap_split(struct mm_struct *mm, struct svm_range *parent, - struct svm_range *prange, unsigned long start, +svm_range_unmap_split(struct svm_range *parent, struct svm_range *prange, unsigned long start, unsigned long last) { struct svm_range *head; @@ -2438,12 +2438,12 @@ svm_range_unmap_split(struct mm_struct *mm, struct svm_range *parent, svm_range_split(tail, last + 1, tail->last, &head); if (head != prange && tail != prange) { - svm_range_add_child(parent, mm, head, SVM_OP_UNMAP_RANGE); - svm_range_add_child(parent, mm, tail, SVM_OP_ADD_RANGE); + svm_range_add_child(parent, head, SVM_OP_UNMAP_RANGE); + svm_range_add_child(parent, tail, SVM_OP_ADD_RANGE); } else if (tail != prange) { - svm_range_add_child(parent, mm, tail, SVM_OP_UNMAP_RANGE); + svm_range_add_child(parent, tail, SVM_OP_UNMAP_RANGE); } else if (head != prange) { - svm_range_add_child(parent, mm, head, SVM_OP_UNMAP_RANGE); + svm_range_add_child(parent, head, SVM_OP_UNMAP_RANGE); } else if (parent != prange) { prange->work_item.op = SVM_OP_UNMAP_RANGE; } @@ -2520,14 +2520,14 @@ svm_range_unmap_from_cpu(struct mm_struct *mm, struct svm_range *prange, l = min(last, pchild->last); if (l >= s) svm_range_unmap_from_gpus(pchild, s, l, trigger); - svm_range_unmap_split(mm, prange, pchild, start, last); + svm_range_unmap_split(prange, pchild, start, last); mutex_unlock(&pchild->lock); } s = max(start, prange->start); l = min(last, prange->last); if (l >= s) svm_range_unmap_from_gpus(prange, s, l, trigger); - svm_range_unmap_split(mm, prange, prange, start, last); + svm_range_unmap_split(prange, prange, start, last); if (unmap_parent) svm_range_add_list_work(svms, prange, mm, SVM_OP_UNMAP_RANGE); @@ -2570,8 +2570,6 @@ svm_range_cpu_invalidate_pagetables(struct mmu_interval_notifier *mni, if (range->event == MMU_NOTIFY_RELEASE) return true; - if (!mmget_not_zero(mni->mm)) - return true; start = mni->interval_tree.start; last = mni->interval_tree.last; @@ -2598,7 +2596,6 @@ svm_range_cpu_invalidate_pagetables(struct mmu_interval_notifier *mni, } svm_range_unlock(prange); - mmput(mni->mm); return true; } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c index baa2374acdeb53ada6e52738662df4e86ac1fc80..4ec73f33535ebfc20c9b5dc8c89c52aa4c44b126 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c @@ -510,6 +510,10 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr, dev->node_props.capability |= HSA_CAP_AQL_QUEUE_DOUBLE_MAP; + if (KFD_GC_VERSION(dev->gpu) < IP_VERSION(10, 0, 0) && + (dev->gpu->adev->sdma.supported_reset & AMDGPU_RESET_TYPE_PER_QUEUE)) + dev->node_props.capability2 |= HSA_CAP2_PER_SDMA_QUEUE_RESET_SUPPORTED; + sysfs_show_32bit_prop(buffer, offs, "max_engine_clk_fcompute", dev->node_props.max_engine_clk_fcompute); @@ -2008,8 +2012,6 @@ static void kfd_topology_set_capabilities(struct kfd_topology_device *dev) if (!amdgpu_sriov_vf(dev->gpu->adev)) dev->node_props.capability |= HSA_CAP_PER_QUEUE_RESET_SUPPORTED; - if (dev->gpu->adev->sdma.supported_reset & AMDGPU_RESET_TYPE_PER_QUEUE) - dev->node_props.capability2 |= HSA_CAP2_PER_SDMA_QUEUE_RESET_SUPPORTED; } else { dev->node_props.debug_prop |= HSA_DBG_WATCH_ADDR_MASK_LO_BIT_GFX10 | HSA_DBG_WATCH_ADDR_MASK_HI_BIT; diff --git a/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.c b/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.c index faed84172dd43091cb7ff0eb34b0175ada984999..8bc36f04b1b712cb7bfc5d0fe6315e4b915b1711 100644 --- a/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.c +++ b/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.c @@ -21,6 +21,7 @@ * */ +#include #include #include #include diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index d3100f641ac6e5d0e439bf2e01a0cde16939d674..3dd4f9e9931d27e7d056f26991d19081f4eedb7b 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -1758,10 +1758,11 @@ dm_dmub_send_vbios_gpint_command(struct amdgpu_device *adev, return DMUB_STATUS_TIMEOUT; } -static struct dml2_soc_bb *dm_dmub_get_vbios_bounding_box(struct amdgpu_device *adev) +static void *dm_dmub_get_vbios_bounding_box(struct amdgpu_device *adev) { - struct dml2_soc_bb *bb; + void *bb; long long addr; + unsigned int bb_size; int i = 0; uint16_t chunk; enum dmub_gpint_command send_addrs[] = { @@ -1774,6 +1775,7 @@ static struct dml2_soc_bb *dm_dmub_get_vbios_bounding_box(struct amdgpu_device * switch (amdgpu_ip_version(adev, DCE_HWIP, 0)) { case IP_VERSION(4, 0, 1): + bb_size = sizeof(struct dml2_soc_bb); break; default: return NULL; @@ -1781,7 +1783,7 @@ static struct dml2_soc_bb *dm_dmub_get_vbios_bounding_box(struct amdgpu_device * bb = dm_allocate_gpu_mem(adev, DC_MEM_ALLOC_TYPE_GART, - sizeof(struct dml2_soc_bb), + bb_size, &addr); if (!bb) return NULL; @@ -1847,7 +1849,7 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) mutex_init(&adev->dm.audio_lock); if (amdgpu_dm_irq_init(adev)) { - drm_err(adev_to_drm(adev), "amdgpu: failed to initialize DM IRQ support.\n"); + drm_err(adev_to_drm(adev), "failed to initialize DM IRQ support.\n"); goto error; } @@ -2037,7 +2039,7 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) adev->dm.hpd_rx_offload_wq = hpd_rx_irq_create_workqueue(adev); if (!adev->dm.hpd_rx_offload_wq) { - drm_err(adev_to_drm(adev), "amdgpu: failed to create hpd rx offload workqueue.\n"); + drm_err(adev_to_drm(adev), "failed to create hpd rx offload workqueue.\n"); goto error; } @@ -2053,7 +2055,7 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) adev->dm.freesync_module = mod_freesync_create(adev->dm.dc); if (!adev->dm.freesync_module) { drm_err(adev_to_drm(adev), - "amdgpu: failed to initialize freesync_module.\n"); + "failed to initialize freesync_module.\n"); } else drm_dbg_driver(adev_to_drm(adev), "amdgpu: freesync_module init done %p.\n", adev->dm.freesync_module); @@ -2064,7 +2066,7 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) adev->dm.vblank_control_workqueue = create_singlethread_workqueue("dm_vblank_control_workqueue"); if (!adev->dm.vblank_control_workqueue) - drm_err(adev_to_drm(adev), "amdgpu: failed to initialize vblank_workqueue.\n"); + drm_err(adev_to_drm(adev), "failed to initialize vblank_workqueue.\n"); } if (adev->dm.dc->caps.ips_support && @@ -2075,7 +2077,7 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) adev->dm.hdcp_workqueue = hdcp_create_workqueue(adev, &init_params.cp_psp, adev->dm.dc); if (!adev->dm.hdcp_workqueue) - drm_err(adev_to_drm(adev), "amdgpu: failed to initialize hdcp_workqueue.\n"); + drm_err(adev_to_drm(adev), "failed to initialize hdcp_workqueue.\n"); else drm_dbg_driver(adev_to_drm(adev), "amdgpu: hdcp_workqueue init done %p.\n", adev->dm.hdcp_workqueue); @@ -2085,20 +2087,20 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) init_completion(&adev->dm.dmub_aux_transfer_done); adev->dm.dmub_notify = kzalloc(sizeof(struct dmub_notification), GFP_KERNEL); if (!adev->dm.dmub_notify) { - drm_info(adev_to_drm(adev), "amdgpu: fail to allocate adev->dm.dmub_notify"); + drm_info(adev_to_drm(adev), "fail to allocate adev->dm.dmub_notify"); goto error; } adev->dm.delayed_hpd_wq = create_singlethread_workqueue("amdgpu_dm_hpd_wq"); if (!adev->dm.delayed_hpd_wq) { - drm_err(adev_to_drm(adev), "amdgpu: failed to create hpd offload workqueue.\n"); + drm_err(adev_to_drm(adev), "failed to create hpd offload workqueue.\n"); goto error; } amdgpu_dm_outbox_init(adev); if (!register_dmub_notify_callback(adev, DMUB_NOTIFICATION_AUX_REPLY, dmub_aux_setconfig_callback, false)) { - drm_err(adev_to_drm(adev), "amdgpu: fail to register dmub aux callback"); + drm_err(adev_to_drm(adev), "fail to register dmub aux callback"); goto error; } @@ -2107,7 +2109,7 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) if (!register_dmub_notify_callback(adev, DMUB_NOTIFICATION_FUSED_IO, dmub_aux_fused_io_callback, false)) { - drm_err(adev_to_drm(adev), "amdgpu: fail to register dmub fused io callback"); + drm_err(adev_to_drm(adev), "fail to register dmub fused io callback"); goto error; } /* Enable outbox notification only after IRQ handlers are registered and DMUB is alive. @@ -2125,7 +2127,7 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) if (amdgpu_dm_initialize_drm_device(adev)) { drm_err(adev_to_drm(adev), - "amdgpu: failed to initialize sw for display support.\n"); + "failed to initialize sw for display support.\n"); goto error; } @@ -2140,14 +2142,14 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) if (drm_vblank_init(adev_to_drm(adev), adev->dm.display_indexes_num)) { drm_err(adev_to_drm(adev), - "amdgpu: failed to initialize sw for display support.\n"); + "failed to initialize sw for display support.\n"); goto error; } #if defined(CONFIG_DRM_AMD_SECURE_DISPLAY) amdgpu_dm_crtc_secure_display_create_contexts(adev); if (!adev->dm.secure_display_ctx.crtc_ctx) - drm_err(adev_to_drm(adev), "amdgpu: failed to initialize secure display contexts.\n"); + drm_err(adev_to_drm(adev), "failed to initialize secure display contexts.\n"); if (amdgpu_ip_version(adev, DCE_HWIP, 0) >= IP_VERSION(4, 0, 1)) adev->dm.secure_display_ctx.support_mul_roi = true; @@ -2404,6 +2406,7 @@ static int dm_dmub_sw_init(struct amdgpu_device *adev) DMUB_WINDOW_MEMORY_TYPE_FB, //DMUB_WINDOW_5_TRACEBUFF DMUB_WINDOW_MEMORY_TYPE_FB, //DMUB_WINDOW_6_FW_STATE DMUB_WINDOW_MEMORY_TYPE_FB, //DMUB_WINDOW_7_SCRATCH_MEM + DMUB_WINDOW_MEMORY_TYPE_FB, //DMUB_WINDOW_IB_MEM DMUB_WINDOW_MEMORY_TYPE_FB, //DMUB_WINDOW_SHARED_STATE }; int r; @@ -2570,7 +2573,7 @@ static int dm_sw_init(struct amdgpu_ip_block *ip_block) adev->dm.cgs_device = amdgpu_cgs_create_device(adev); if (!adev->dm.cgs_device) { - drm_err(adev_to_drm(adev), "amdgpu: failed to create cgs device.\n"); + drm_err(adev_to_drm(adev), "failed to create cgs device.\n"); return -EINVAL; } @@ -3060,6 +3063,77 @@ static void hpd_rx_irq_work_suspend(struct amdgpu_display_manager *dm) } } +static int dm_cache_state(struct amdgpu_device *adev) +{ + int r; + + adev->dm.cached_state = drm_atomic_helper_suspend(adev_to_drm(adev)); + if (IS_ERR(adev->dm.cached_state)) { + r = PTR_ERR(adev->dm.cached_state); + adev->dm.cached_state = NULL; + } + + return adev->dm.cached_state ? 0 : r; +} + +static void dm_destroy_cached_state(struct amdgpu_device *adev) +{ + struct amdgpu_display_manager *dm = &adev->dm; + struct drm_device *ddev = adev_to_drm(adev); + struct dm_plane_state *dm_new_plane_state; + struct drm_plane_state *new_plane_state; + struct dm_crtc_state *dm_new_crtc_state; + struct drm_crtc_state *new_crtc_state; + struct drm_plane *plane; + struct drm_crtc *crtc; + int i; + + if (!dm->cached_state) + return; + + /* Force mode set in atomic commit */ + for_each_new_crtc_in_state(dm->cached_state, crtc, new_crtc_state, i) { + new_crtc_state->active_changed = true; + dm_new_crtc_state = to_dm_crtc_state(new_crtc_state); + reset_freesync_config_for_crtc(dm_new_crtc_state); + } + + /* + * atomic_check is expected to create the dc states. We need to release + * them here, since they were duplicated as part of the suspend + * procedure. + */ + for_each_new_crtc_in_state(dm->cached_state, crtc, new_crtc_state, i) { + dm_new_crtc_state = to_dm_crtc_state(new_crtc_state); + if (dm_new_crtc_state->stream) { + WARN_ON(kref_read(&dm_new_crtc_state->stream->refcount) > 1); + dc_stream_release(dm_new_crtc_state->stream); + dm_new_crtc_state->stream = NULL; + } + dm_new_crtc_state->base.color_mgmt_changed = true; + } + + for_each_new_plane_in_state(dm->cached_state, plane, new_plane_state, i) { + dm_new_plane_state = to_dm_plane_state(new_plane_state); + if (dm_new_plane_state->dc_state) { + WARN_ON(kref_read(&dm_new_plane_state->dc_state->refcount) > 1); + dc_plane_state_release(dm_new_plane_state->dc_state); + dm_new_plane_state->dc_state = NULL; + } + } + + drm_atomic_helper_resume(ddev, dm->cached_state); + + dm->cached_state = NULL; +} + +static void dm_complete(struct amdgpu_ip_block *ip_block) +{ + struct amdgpu_device *adev = ip_block->adev; + + dm_destroy_cached_state(adev); +} + static int dm_prepare_suspend(struct amdgpu_ip_block *ip_block) { struct amdgpu_device *adev = ip_block->adev; @@ -3068,11 +3142,8 @@ static int dm_prepare_suspend(struct amdgpu_ip_block *ip_block) return 0; WARN_ON(adev->dm.cached_state); - adev->dm.cached_state = drm_atomic_helper_suspend(adev_to_drm(adev)); - if (IS_ERR(adev->dm.cached_state)) - return PTR_ERR(adev->dm.cached_state); - return 0; + return dm_cache_state(adev); } static int dm_suspend(struct amdgpu_ip_block *ip_block) @@ -3106,9 +3177,10 @@ static int dm_suspend(struct amdgpu_ip_block *ip_block) } if (!adev->dm.cached_state) { - adev->dm.cached_state = drm_atomic_helper_suspend(adev_to_drm(adev)); - if (IS_ERR(adev->dm.cached_state)) - return PTR_ERR(adev->dm.cached_state); + int r = dm_cache_state(adev); + + if (r) + return r; } s3_handle_hdmi_cec(adev_to_drm(adev), true); @@ -3295,12 +3367,6 @@ static int dm_resume(struct amdgpu_ip_block *ip_block) struct amdgpu_dm_connector *aconnector; struct drm_connector *connector; struct drm_connector_list_iter iter; - struct drm_crtc *crtc; - struct drm_crtc_state *new_crtc_state; - struct dm_crtc_state *dm_new_crtc_state; - struct drm_plane *plane; - struct drm_plane_state *new_plane_state; - struct dm_plane_state *dm_new_plane_state; struct dm_atomic_state *dm_state = to_dm_atomic_state(dm->atomic_obj.state); enum dc_connection_type new_connection_type = dc_connection_none; struct dc_state *dc_state; @@ -3457,40 +3523,7 @@ static int dm_resume(struct amdgpu_ip_block *ip_block) } drm_connector_list_iter_end(&iter); - /* Force mode set in atomic commit */ - for_each_new_crtc_in_state(dm->cached_state, crtc, new_crtc_state, i) { - new_crtc_state->active_changed = true; - dm_new_crtc_state = to_dm_crtc_state(new_crtc_state); - reset_freesync_config_for_crtc(dm_new_crtc_state); - } - - /* - * atomic_check is expected to create the dc states. We need to release - * them here, since they were duplicated as part of the suspend - * procedure. - */ - for_each_new_crtc_in_state(dm->cached_state, crtc, new_crtc_state, i) { - dm_new_crtc_state = to_dm_crtc_state(new_crtc_state); - if (dm_new_crtc_state->stream) { - WARN_ON(kref_read(&dm_new_crtc_state->stream->refcount) > 1); - dc_stream_release(dm_new_crtc_state->stream); - dm_new_crtc_state->stream = NULL; - } - dm_new_crtc_state->base.color_mgmt_changed = true; - } - - for_each_new_plane_in_state(dm->cached_state, plane, new_plane_state, i) { - dm_new_plane_state = to_dm_plane_state(new_plane_state); - if (dm_new_plane_state->dc_state) { - WARN_ON(kref_read(&dm_new_plane_state->dc_state->refcount) > 1); - dc_plane_state_release(dm_new_plane_state->dc_state); - dm_new_plane_state->dc_state = NULL; - } - } - - drm_atomic_helper_resume(ddev, dm->cached_state); - - dm->cached_state = NULL; + dm_destroy_cached_state(adev); /* Do mst topology probing after resuming cached state*/ drm_connector_list_iter_begin(ddev, &iter); @@ -3539,6 +3572,7 @@ static const struct amd_ip_funcs amdgpu_dm_funcs = { .prepare_suspend = dm_prepare_suspend, .suspend = dm_suspend, .resume = dm_resume, + .complete = dm_complete, .is_idle = dm_is_idle, .wait_for_idle = dm_wait_for_idle, .check_soft_reset = dm_check_soft_reset, @@ -3610,13 +3644,15 @@ static void update_connector_ext_caps(struct amdgpu_dm_connector *aconnector) luminance_range = &conn_base->display_info.luminance_range; - if (luminance_range->max_luminance) { - caps->aux_min_input_signal = luminance_range->min_luminance; + if (luminance_range->max_luminance) caps->aux_max_input_signal = luminance_range->max_luminance; - } else { - caps->aux_min_input_signal = 0; + else caps->aux_max_input_signal = 512; - } + + if (luminance_range->min_luminance) + caps->aux_min_input_signal = luminance_range->min_luminance; + else + caps->aux_min_input_signal = 1; min_input_signal_override = drm_get_panel_min_brightness_quirk(aconnector->drm_edid); if (min_input_signal_override >= 0) @@ -4001,19 +4037,19 @@ static int register_hpd_handlers(struct amdgpu_device *adev) if (dc_is_dmub_outbox_supported(adev->dm.dc)) { if (!register_dmub_notify_callback(adev, DMUB_NOTIFICATION_HPD, dmub_hpd_callback, true)) { - drm_err(adev_to_drm(adev), "amdgpu: fail to register dmub hpd callback"); + drm_err(adev_to_drm(adev), "fail to register dmub hpd callback"); return -EINVAL; } if (!register_dmub_notify_callback(adev, DMUB_NOTIFICATION_HPD_IRQ, dmub_hpd_callback, true)) { - drm_err(adev_to_drm(adev), "amdgpu: fail to register dmub hpd callback"); + drm_err(adev_to_drm(adev), "fail to register dmub hpd callback"); return -EINVAL; } if (!register_dmub_notify_callback(adev, DMUB_NOTIFICATION_HPD_SENSE_NOTIFY, dmub_hpd_sense_callback, true)) { - drm_err(adev_to_drm(adev), "amdgpu: fail to register dmub hpd sense callback"); + drm_err(adev_to_drm(adev), "fail to register dmub hpd sense callback"); return -EINVAL; } } @@ -4718,9 +4754,23 @@ static int get_brightness_range(const struct amdgpu_dm_backlight_caps *caps, return 1; } +/* Rescale from [min..max] to [0..MAX_BACKLIGHT_LEVEL] */ +static inline u32 scale_input_to_fw(int min, int max, u64 input) +{ + return DIV_ROUND_CLOSEST_ULL(input * MAX_BACKLIGHT_LEVEL, max - min); +} + +/* Rescale from [0..MAX_BACKLIGHT_LEVEL] to [min..max] */ +static inline u32 scale_fw_to_input(int min, int max, u64 input) +{ + return min + DIV_ROUND_CLOSEST_ULL(input * (max - min), MAX_BACKLIGHT_LEVEL); +} + static void convert_custom_brightness(const struct amdgpu_dm_backlight_caps *caps, - uint32_t *brightness) + unsigned int min, unsigned int max, + uint32_t *user_brightness) { + u32 brightness = scale_input_to_fw(min, max, *user_brightness); u8 prev_signal = 0, prev_lum = 0; int i = 0; @@ -4731,7 +4781,7 @@ static void convert_custom_brightness(const struct amdgpu_dm_backlight_caps *cap return; /* choose start to run less interpolation steps */ - if (caps->luminance_data[caps->data_points/2].input_signal > *brightness) + if (caps->luminance_data[caps->data_points/2].input_signal > brightness) i = caps->data_points/2; do { u8 signal = caps->luminance_data[i].input_signal; @@ -4742,17 +4792,18 @@ static void convert_custom_brightness(const struct amdgpu_dm_backlight_caps *cap * brightness < signal: interpolate between previous and current luminance numerator * brightness > signal: find next data point */ - if (*brightness > signal) { + if (brightness > signal) { prev_signal = signal; prev_lum = lum; i++; continue; } - if (*brightness < signal) + if (brightness < signal) lum = prev_lum + DIV_ROUND_CLOSEST((lum - prev_lum) * - (*brightness - prev_signal), + (brightness - prev_signal), signal - prev_signal); - *brightness = DIV_ROUND_CLOSEST(lum * *brightness, 101); + *user_brightness = scale_fw_to_input(min, max, + DIV_ROUND_CLOSEST(lum * brightness, 101)); return; } while (i < caps->data_points); } @@ -4765,11 +4816,10 @@ static u32 convert_brightness_from_user(const struct amdgpu_dm_backlight_caps *c if (!get_brightness_range(caps, &min, &max)) return brightness; - convert_custom_brightness(caps, &brightness); + convert_custom_brightness(caps, min, max, &brightness); - // Rescale 0..255 to min..max - return min + DIV_ROUND_CLOSEST((max - min) * brightness, - AMDGPU_MAX_BL_LEVEL); + // Rescale 0..max to min..max + return min + DIV_ROUND_CLOSEST_ULL((u64)(max - min) * brightness, max); } static u32 convert_brightness_to_user(const struct amdgpu_dm_backlight_caps *caps, @@ -4782,8 +4832,8 @@ static u32 convert_brightness_to_user(const struct amdgpu_dm_backlight_caps *cap if (brightness < min) return 0; - // Rescale min..max to 0..255 - return DIV_ROUND_CLOSEST(AMDGPU_MAX_BL_LEVEL * (brightness - min), + // Rescale min..max to 0..max + return DIV_ROUND_CLOSEST_ULL((u64)max * (brightness - min), max - min); } @@ -4813,6 +4863,14 @@ static void amdgpu_dm_backlight_set_level(struct amdgpu_display_manager *dm, reallow_idle = true; } + if (trace_amdgpu_dm_brightness_enabled()) { + trace_amdgpu_dm_brightness(__builtin_return_address(0), + user_brightness, + brightness, + caps->aux_support, + power_supply_is_system_supplied() > 0); + } + if (caps->aux_support) { rc = dc_link_set_backlight_level_nits(link, true, brightness, AUX_BL_DEFAULT_TRANSITION_TIME_MS); @@ -4908,7 +4966,7 @@ amdgpu_dm_register_backlight_device(struct amdgpu_dm_connector *aconnector) struct drm_device *drm = aconnector->base.dev; struct amdgpu_display_manager *dm = &drm_to_adev(drm)->dm; struct backlight_properties props = { 0 }; - struct amdgpu_dm_backlight_caps caps = { 0 }; + struct amdgpu_dm_backlight_caps *caps; char bl_name[16]; int min, max; @@ -4922,22 +4980,21 @@ amdgpu_dm_register_backlight_device(struct amdgpu_dm_connector *aconnector) return; } - amdgpu_acpi_get_backlight_caps(&caps); - if (caps.caps_valid && get_brightness_range(&caps, &min, &max)) { + caps = &dm->backlight_caps[aconnector->bl_idx]; + if (get_brightness_range(caps, &min, &max)) { if (power_supply_is_system_supplied() > 0) - props.brightness = (max - min) * DIV_ROUND_CLOSEST(caps.ac_level, 100); + props.brightness = (max - min) * DIV_ROUND_CLOSEST(caps->ac_level, 100); else - props.brightness = (max - min) * DIV_ROUND_CLOSEST(caps.dc_level, 100); + props.brightness = (max - min) * DIV_ROUND_CLOSEST(caps->dc_level, 100); /* min is zero, so max needs to be adjusted */ props.max_brightness = max - min; drm_dbg(drm, "Backlight caps: min: %d, max: %d, ac %d, dc %d\n", min, max, - caps.ac_level, caps.dc_level); + caps->ac_level, caps->dc_level); } else - props.brightness = AMDGPU_MAX_BL_LEVEL; + props.brightness = props.max_brightness = MAX_BACKLIGHT_LEVEL; - if (caps.data_points && !(amdgpu_dc_debug_mask & DC_DISABLE_CUSTOM_BRIGHTNESS_CURVE)) + if (caps->data_points && !(amdgpu_dc_debug_mask & DC_DISABLE_CUSTOM_BRIGHTNESS_CURVE)) drm_info(drm, "Using custom brightness curve\n"); - props.max_brightness = AMDGPU_MAX_BL_LEVEL; props.type = BACKLIGHT_RAW; snprintf(bl_name, sizeof(bl_name), "amdgpu_bl%d", @@ -7519,7 +7576,7 @@ static enum dc_status dm_validate_stream_and_context(struct dc *dc, dc_result = DC_FAIL_ATTACH_SURFACES; if (dc_result == DC_OK) - dc_result = dc_validate_global_state(dc, dc_state, true); + dc_result = dc_validate_global_state(dc, dc_state, DC_VALIDATE_MODE_ONLY); cleanup: if (dc_state) @@ -7577,7 +7634,7 @@ create_validate_stream_for_sink(struct drm_connector *connector, dc_result = dm_validate_stream_and_context(adev->dm.dc, stream); if (dc_result != DC_OK) { - DRM_DEBUG_KMS("Mode %dx%d (clk %d) pixel_encoding:%s color_depth:%s failed validation -- %s\n", + DRM_DEBUG_KMS("Pruned mode %d x %d (clk %d) %s %s -- %s\n", drm_mode->hdisplay, drm_mode->vdisplay, drm_mode->clock, @@ -7844,6 +7901,22 @@ static int dm_encoder_helper_atomic_check(struct drm_encoder *encoder, int clock, bpp = 0; bool is_y420 = false; + if (connector->connector_type == DRM_MODE_CONNECTOR_eDP) { + struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder); + struct drm_display_mode *native_mode = &amdgpu_encoder->native_mode; + enum drm_mode_status result; + + result = drm_crtc_helper_mode_valid_fixed(encoder->crtc, adjusted_mode, native_mode); + if (result != MODE_OK && dm_new_connector_state->scaling == RMX_OFF) { + drm_dbg_driver(encoder->dev, + "mode %dx%d@%dHz is not native, enabling scaling\n", + adjusted_mode->hdisplay, adjusted_mode->vdisplay, + drm_mode_vrefresh(adjusted_mode)); + dm_new_connector_state->scaling = RMX_FULL; + } + return 0; + } + if (!aconnector->mst_output_port) return 0; @@ -8301,7 +8374,7 @@ static int amdgpu_dm_connector_get_modes(struct drm_connector *connector) drm_add_modes_noedid(connector, 1920, 1080); } else { amdgpu_dm_connector_ddc_get_modes(connector, drm_edid); - if (encoder) + if (encoder && connector->connector_type != DRM_MODE_CONNECTOR_eDP) amdgpu_dm_connector_add_common_modes(encoder, connector); amdgpu_dm_connector_add_freesync_modes(connector, drm_edid); } @@ -12141,7 +12214,7 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev, drm_dbg_atomic(dev, "MST drm_dp_mst_atomic_check() failed\n"); goto fail; } - status = dc_validate_global_state(dc, dm_state->context, true); + status = dc_validate_global_state(dc, dm_state->context, DC_VALIDATE_MODE_ONLY); if (status != DC_OK) { drm_dbg_atomic(dev, "DC global validation failure: %s (%d)", dc_status_to_str(status), status); diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h index d7d92f9911e465d46c01acd97d18982eed2471ee..b937da0a4e4a002d3b599efb9d3689e0f0335eec 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h @@ -636,8 +636,9 @@ struct amdgpu_display_manager { * @bb_from_dmub: * * Bounding box data read from dmub during early initialization for DCN4+ + * Data is stored as a byte array that should be casted to the appropriate bb struct */ - struct dml2_soc_bb *bb_from_dmub; + void *bb_from_dmub; /** * @oem_i2c: diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c index d4395b92fb85645d083dc124bd542fe9f76616ac..9e3e51a2dc4968c0dcf7b4532ff7cc00e43ad730 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c @@ -1029,6 +1029,10 @@ enum dc_edid_status dm_helpers_read_local_edid( return EDID_NO_RESPONSE; edid = drm_edid_raw(drm_edid); // FIXME: Get rid of drm_edid_raw() + if (!edid || + edid->extensions >= sizeof(sink->dc_edid.raw_edid) / EDID_LENGTH) + return EDID_BAD_INPUT; + sink->dc_edid.length = EDID_LENGTH * (edid->extensions + 1); memmove(sink->dc_edid.raw_edid, (uint8_t *)edid, sink->dc_edid.length); diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c index 25e8befbcc479a3d70cf6239838acacf3cb99134..7187d5aedf0a501256b9e2316890b6d639338efb 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c @@ -107,7 +107,7 @@ static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux, if (payload.write && result >= 0) { if (result) { /*one byte indicating partially written bytes*/ - drm_dbg_dp(adev_to_drm(adev), "amdgpu: AUX partially written\n"); + drm_dbg_dp(adev_to_drm(adev), "AUX partially written\n"); result = payload.data[0]; } else if (!payload.reply[0]) /*I2C_ACK|AUX_ACK*/ @@ -133,11 +133,11 @@ static ssize_t dm_dp_aux_transfer(struct drm_dp_aux *aux, break; } - drm_dbg_dp(adev_to_drm(adev), "amdgpu: DP AUX transfer fail:%d\n", operation_result); + drm_dbg_dp(adev_to_drm(adev), "DP AUX transfer fail:%d\n", operation_result); } if (payload.reply[0]) - drm_dbg_dp(adev_to_drm(adev), "amdgpu: AUX reply command not ACK: 0x%02x.", + drm_dbg_dp(adev_to_drm(adev), "AUX reply command not ACK: 0x%02x.", payload.reply[0]); return result; diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h index 4686d4b0cbad29892f317eacafb130e88d843704..95f890fda8aaae138591c7fb3bf48c4d180e05c2 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h @@ -726,6 +726,32 @@ TRACE_EVENT(dcn_optc_lock_unlock_state, ) ); +TRACE_EVENT(amdgpu_dm_brightness, + TP_PROTO(void *function, u32 user_brightness, u32 converted_brightness, bool aux, bool ac), + TP_ARGS(function, user_brightness, converted_brightness, aux, ac), + TP_STRUCT__entry( + __field(void *, function) + __field(u32, user_brightness) + __field(u32, converted_brightness) + __field(bool, aux) + __field(bool, ac) + ), + TP_fast_assign( + __entry->function = function; + __entry->user_brightness = user_brightness; + __entry->converted_brightness = converted_brightness; + __entry->aux = aux; + __entry->ac = ac; + ), + TP_printk("%ps: brightness requested=%u converted=%u aux=%s power=%s", + (void *)__entry->function, + (u32)__entry->user_brightness, + (u32)__entry->converted_brightness, + (__entry->aux) ? "true" : "false", + (__entry->ac) ? "AC" : "DC" + ) +); + #endif /* _AMDGPU_DM_TRACE_H_ */ #undef TRACE_INCLUDE_PATH diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table2.c b/drivers/gpu/drm/amd/display/dc/bios/command_table2.c index 2c645dffec185afef5884551f5fa434d26207f92..f2b1720a6a66006e62c82891fa127ff9fea376b2 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/command_table2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/command_table2.c @@ -396,6 +396,7 @@ static enum bp_result transmitter_control_v1_7( process_phy_transition_init_params.display_port_link_rate = link->cur_link_settings.link_rate; process_phy_transition_init_params.transition_bitmask = link->phy_transition_bitmask; } + dig_v1_7.skip_phy_ssc_reduction = link->wa_flags.skip_phy_ssc_reduction; } // Handle PRE_OFF_TO_ON: Process ACPI PHY Transition Interlock diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.c index a3b8e3d4a429e337e485ba85fa9421d805247873..514a5efda1021607ec3f3a1ccf8026424be70549 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.c @@ -22,8 +22,6 @@ #include "dcn/dcn_4_1_0_offset.h" #include "dcn/dcn_4_1_0_sh_mask.h" -#include "dml/dcn401/dcn401_fpu.h" - #define DCN_BASE__INST0_SEG1 0x000000C0 #define mmCLK01_CLK0_CLK_PLL_REQ 0x16E37 @@ -183,43 +181,36 @@ static void dcn401_init_single_clock(struct clk_mgr_internal *clk_mgr, PPCLK_e c static void dcn401_build_wm_range_table(struct clk_mgr *clk_mgr) { - /* legacy */ - DC_FP_START(); - dcn401_build_wm_range_table_fpu(clk_mgr); - DC_FP_END(); - - if (clk_mgr->ctx->dc->debug.using_dml21) { - /* For min clocks use as reported by PM FW and report those as min */ - uint16_t min_uclk_mhz = clk_mgr->bw_params->clk_table.entries[0].memclk_mhz; - uint16_t min_dcfclk_mhz = clk_mgr->bw_params->clk_table.entries[0].dcfclk_mhz; - - /* Set A - Normal - default values */ - clk_mgr->bw_params->wm_table.nv_entries[WM_A].valid = true; - clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.wm_type = WATERMARKS_CLOCK_RANGE; - clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.min_dcfclk = min_dcfclk_mhz; - clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.max_dcfclk = 0xFFFF; - clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.min_uclk = min_uclk_mhz; - clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.max_uclk = 0xFFFF; - - /* Set B - Unused on dcn4 */ - clk_mgr->bw_params->wm_table.nv_entries[WM_B].valid = false; - - /* Set 1A - Dummy P-State - P-State latency set to "dummy p-state" value */ - /* 'DalDummyClockChangeLatencyNs' registry key option set to 0x7FFFFFFF can be used to disable Set C for dummy p-state */ - if (clk_mgr->ctx->dc->bb_overrides.dummy_clock_change_latency_ns != 0x7FFFFFFF) { - clk_mgr->bw_params->wm_table.nv_entries[WM_1A].valid = true; - clk_mgr->bw_params->wm_table.nv_entries[WM_1A].pmfw_breakdown.wm_type = WATERMARKS_DUMMY_PSTATE; - clk_mgr->bw_params->wm_table.nv_entries[WM_1A].pmfw_breakdown.min_dcfclk = min_dcfclk_mhz; - clk_mgr->bw_params->wm_table.nv_entries[WM_1A].pmfw_breakdown.max_dcfclk = 0xFFFF; - clk_mgr->bw_params->wm_table.nv_entries[WM_1A].pmfw_breakdown.min_uclk = min_uclk_mhz; - clk_mgr->bw_params->wm_table.nv_entries[WM_1A].pmfw_breakdown.max_uclk = 0xFFFF; - } else { - clk_mgr->bw_params->wm_table.nv_entries[WM_1A].valid = false; - } - - /* Set 1B - Unused on dcn4 */ - clk_mgr->bw_params->wm_table.nv_entries[WM_1B].valid = false; + /* For min clocks use as reported by PM FW and report those as min */ + uint16_t min_uclk_mhz = clk_mgr->bw_params->clk_table.entries[0].memclk_mhz; + uint16_t min_dcfclk_mhz = clk_mgr->bw_params->clk_table.entries[0].dcfclk_mhz; + + /* Set A - Normal - default values */ + clk_mgr->bw_params->wm_table.nv_entries[WM_A].valid = true; + clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.wm_type = WATERMARKS_CLOCK_RANGE; + clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.min_dcfclk = min_dcfclk_mhz; + clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.max_dcfclk = 0xFFFF; + clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.min_uclk = min_uclk_mhz; + clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.max_uclk = 0xFFFF; + + /* Set B - Unused on dcn4 */ + clk_mgr->bw_params->wm_table.nv_entries[WM_B].valid = false; + + /* Set 1A - Dummy P-State - P-State latency set to "dummy p-state" value */ + /* 'DalDummyClockChangeLatencyNs' registry key option set to 0x7FFFFFFF can be used to disable Set C for dummy p-state */ + if (clk_mgr->ctx->dc->bb_overrides.dummy_clock_change_latency_ns != 0x7FFFFFFF) { + clk_mgr->bw_params->wm_table.nv_entries[WM_1A].valid = true; + clk_mgr->bw_params->wm_table.nv_entries[WM_1A].pmfw_breakdown.wm_type = WATERMARKS_DUMMY_PSTATE; + clk_mgr->bw_params->wm_table.nv_entries[WM_1A].pmfw_breakdown.min_dcfclk = min_dcfclk_mhz; + clk_mgr->bw_params->wm_table.nv_entries[WM_1A].pmfw_breakdown.max_dcfclk = 0xFFFF; + clk_mgr->bw_params->wm_table.nv_entries[WM_1A].pmfw_breakdown.min_uclk = min_uclk_mhz; + clk_mgr->bw_params->wm_table.nv_entries[WM_1A].pmfw_breakdown.max_uclk = 0xFFFF; + } else { + clk_mgr->bw_params->wm_table.nv_entries[WM_1A].valid = false; } + + /* Set 1B - Unused on dcn4 */ + clk_mgr->bw_params->wm_table.nv_entries[WM_1B].valid = false; } void dcn401_init_clocks(struct clk_mgr *clk_mgr_base) @@ -320,6 +311,25 @@ void dcn401_init_clocks(struct clk_mgr *clk_mgr_base) dcn401_build_wm_range_table(clk_mgr_base); } +bool dcn401_is_dc_mode_present(struct clk_mgr *clk_mgr_base) +{ + struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base); + + return clk_mgr->smu_present && clk_mgr->dpm_present && + ((clk_mgr_base->bw_params->clk_table.num_entries_per_clk.num_dcfclk_levels && + clk_mgr_base->bw_params->dc_mode_limit.dcfclk_mhz) || + (clk_mgr_base->bw_params->clk_table.num_entries_per_clk.num_dispclk_levels && + clk_mgr_base->bw_params->dc_mode_limit.dispclk_mhz) || + (clk_mgr_base->bw_params->clk_table.num_entries_per_clk.num_dtbclk_levels && + clk_mgr_base->bw_params->dc_mode_limit.dtbclk_mhz) || + (clk_mgr_base->bw_params->clk_table.num_entries_per_clk.num_fclk_levels && + clk_mgr_base->bw_params->dc_mode_limit.fclk_mhz) || + (clk_mgr_base->bw_params->clk_table.num_entries_per_clk.num_memclk_levels && + clk_mgr_base->bw_params->dc_mode_limit.memclk_mhz) || + (clk_mgr_base->bw_params->clk_table.num_entries_per_clk.num_socclk_levels && + clk_mgr_base->bw_params->dc_mode_limit.socclk_mhz)); +} + static void dcn401_dump_clk_registers(struct clk_state_registers_and_bypass *regs_and_bypass, struct clk_mgr *clk_mgr_base, struct clk_log_info *log_info) { @@ -1505,6 +1515,7 @@ static struct clk_mgr_funcs dcn401_funcs = { .get_dispclk_from_dentist = dcn401_get_dispclk_from_dentist, .get_hard_min_memclk = dcn401_get_hard_min_memclk, .get_hard_min_fclk = dcn401_get_hard_min_fclk, + .is_dc_mode_present = dcn401_is_dc_mode_present, }; struct clk_mgr_internal *dcn401_clk_mgr_construct( diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.h b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.h index 6c9ae5ca2c7e96975e384d3b76799ffd65c839d4..616e964df96df07b4e19fdf3af29d26fa08954fd 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.h +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.h @@ -105,6 +105,7 @@ struct dcn401_clk_mgr { }; void dcn401_init_clocks(struct clk_mgr *clk_mgr_base); +bool dcn401_is_dc_mode_present(struct clk_mgr *clk_mgr_base); struct clk_mgr_internal *dcn401_clk_mgr_construct(struct dc_context *ctx, struct dccg *dccg); diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index 56d011a1323cfde0827016aa33846839afb6c762..c744aa9d830f6b3c30273d89a82a11fdfcb889f9 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -241,6 +241,7 @@ static bool create_links( DC_LOG_DC("BIOS object table - end"); /* Create a link for each usb4 dpia port */ + dc->lowest_dpia_link_index = MAX_LINKS; for (i = 0; i < dc->res_pool->usb4_dpia_count; i++) { struct link_init_data link_init_params = {0}; struct dc_link *link; @@ -253,6 +254,9 @@ static bool create_links( link = dc->link_srv->create_link(&link_init_params); if (link) { + if (dc->lowest_dpia_link_index > dc->link_count) + dc->lowest_dpia_link_index = dc->link_count; + dc->links[dc->link_count] = link; link->dc = dc; ++dc->link_count; @@ -2377,7 +2381,7 @@ enum dc_status dc_commit_streams(struct dc *dc, struct dc_commit_streams_params context->power_source = params->power_source; - res = dc_validate_with_context(dc, set, params->stream_count, context, false); + res = dc_validate_with_context(dc, set, params->stream_count, context, DC_VALIDATE_MODE_AND_PROGRAMMING); /* * Only update link encoder to stream assignment after bandwidth validation passed. @@ -3300,7 +3304,8 @@ static void copy_stream_update_to_stream(struct dc *dc, if (dsc_validate_context) { stream->timing.dsc_cfg = *update->dsc_config; stream->timing.flags.DSC = enable_dsc; - if (dc->res_pool->funcs->validate_bandwidth(dc, dsc_validate_context, true) != DC_OK) { + if (dc->res_pool->funcs->validate_bandwidth(dc, dsc_validate_context, + DC_VALIDATE_MODE_ONLY) != DC_OK) { stream->timing.dsc_cfg = old_dsc_cfg; stream->timing.flags.DSC = old_dsc_enabled; update->dsc_config = NULL; @@ -3522,7 +3527,7 @@ static bool update_planes_and_stream_state(struct dc *dc, } if (update_type == UPDATE_TYPE_FULL) { - if (dc->res_pool->funcs->validate_bandwidth(dc, context, false) != DC_OK) { + if (dc->res_pool->funcs->validate_bandwidth(dc, context, DC_VALIDATE_MODE_AND_PROGRAMMING) != DC_OK) { BREAK_TO_DEBUGGER(); goto fail; } @@ -4628,7 +4633,8 @@ static struct dc_state *create_minimal_transition_state(struct dc *dc, backup_and_set_minimal_pipe_split_policy(dc, base_context, policy); /* commit minimal state */ - if (dc->res_pool->funcs->validate_bandwidth(dc, minimal_transition_context, false) == DC_OK) { + if (dc->res_pool->funcs->validate_bandwidth(dc, minimal_transition_context, + DC_VALIDATE_MODE_AND_PROGRAMMING) == DC_OK) { /* prevent underflow and corruption when reconfiguring pipes */ force_vsync_flip_in_minimal_transition_context(minimal_transition_context); } else { @@ -5151,7 +5157,7 @@ static bool update_planes_and_stream_v1(struct dc *dc, copy_stream_update_to_stream(dc, context, stream, stream_update); if (update_type >= UPDATE_TYPE_FULL) { - if (dc->res_pool->funcs->validate_bandwidth(dc, context, false) != DC_OK) { + if (dc->res_pool->funcs->validate_bandwidth(dc, context, DC_VALIDATE_MODE_AND_PROGRAMMING) != DC_OK) { DC_ERROR("Mode validation failed for stream update!\n"); dc_state_release(context); return false; @@ -6337,13 +6343,14 @@ void dc_set_edp_power(const struct dc *dc, struct dc_link *edp_link, edp_link->dc->link_srv->edp_set_panel_power(edp_link, powerOn); } -/* - ***************************************************************************** +/** * dc_get_power_profile_for_dc_state() - extracts power profile from dc state * * Called when DM wants to make power policy decisions based on dc_state * - ***************************************************************************** + * @context: Pointer to the dc_state from which the power profile is extracted. + * + * Return: The power profile structure containing the power level information. */ struct dc_power_profile dc_get_power_profile_for_dc_state(const struct dc_state *context) { @@ -6359,13 +6366,14 @@ struct dc_power_profile dc_get_power_profile_for_dc_state(const struct dc_state return profile; } -/* - ********************************************************************************** +/** * dc_get_det_buffer_size_from_state() - extracts detile buffer size from dc state * - * Called when DM wants to log detile buffer size from dc_state + * This function is called to log the detile buffer size from the dc_state. + * + * @context: a pointer to the dc_state from which the detile buffer size is extracted. * - ********************************************************************************** + * Return: the size of the detile buffer, or 0 if not available. */ unsigned int dc_get_det_buffer_size_from_state(const struct dc_state *context) { @@ -6377,6 +6385,36 @@ unsigned int dc_get_det_buffer_size_from_state(const struct dc_state *context) return 0; } +/** + * dc_get_host_router_index: Get index of host router from a dpia link + * + * This function return a host router index of the target link. If the target link is dpia link. + * + * @link: Pointer to the target link (input) + * @host_router_index: Pointer to store the host router index of the target link (output). + * + * Return: true if the host router index is found and valid. + * + */ +bool dc_get_host_router_index(const struct dc_link *link, unsigned int *host_router_index) +{ + struct dc *dc; + + if (!link || !host_router_index || link->ep_type != DISPLAY_ENDPOINT_USB4_DPIA) + return false; + + dc = link->ctx->dc; + + if (link->link_index < dc->lowest_dpia_link_index) + return false; + + *host_router_index = (link->link_index - dc->lowest_dpia_link_index) / dc->caps.num_of_dpias_per_host_router; + if (*host_router_index < dc->caps.num_of_host_routers) + return true; + else + return false; +} + bool dc_is_cursor_limit_pending(struct dc *dc) { uint32_t i; diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_debug.c b/drivers/gpu/drm/amd/display/dc/core/dc_debug.c index 7551d0a3fe8223b2396357da0741b7c45e57d552..bbce751b485fdcfd2df0aeb03165ce9b7e818af1 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_debug.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_debug.c @@ -268,6 +268,8 @@ char *dc_status_to_str(enum dc_status status) return "Insufficient DP link bandwidth"; case DC_FAIL_HW_CURSOR_SUPPORT: return "HW Cursor not supported"; + case DC_FAIL_DP_TUNNEL_BW_VALIDATE: + return "Fail DP Tunnel BW validation"; case DC_ERROR_UNEXPECTED: return "Unexpected error"; } diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c index 71e15da4bb69af110d9f90c70457cb875f0ced7a..130455f2802a71575b94d0fb790443adb7f2c548 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c @@ -515,7 +515,8 @@ void dc_link_enable_hpd_filter(struct dc_link *link, bool enable) link->dc->link_srv->enable_hpd_filter(link, enable); } -bool dc_link_dp_dpia_validate(struct dc *dc, const struct dc_stream_state *streams, const unsigned int count) +enum dc_status dc_link_validate_dp_tunneling_bandwidth(const struct dc *dc, const struct dc_state *new_ctx) { - return dc->link_srv->validate_dpia_bandwidth(streams, count); + return dc->link_srv->validate_dp_tunnel_bandwidth(dc, new_ctx); } + diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c index 3da25bd8b5788acf4ad20d56d20cfc2c9bf3405c..854fc51f159c1a3e58c73d1cb08e48986cdaaa4f 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c @@ -4053,7 +4053,7 @@ static bool add_all_planes_for_stream( * @set: An array of dc_validation_set with all the current streams reference * @set_count: Total of streams * @context: New context - * @fast_validate: Enable or disable fast validation + * @validate_mode: identify the validation mode * * This function updates the potential new stream in the context object. It * creates multiple lists for the add, remove, and unchanged streams. In @@ -4068,7 +4068,7 @@ enum dc_status dc_validate_with_context(struct dc *dc, const struct dc_validation_set set[], int set_count, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { struct dc_stream_state *unchanged_streams[MAX_PIPES] = { 0 }; struct dc_stream_state *del_streams[MAX_PIPES] = { 0 }; @@ -4242,7 +4242,7 @@ enum dc_status dc_validate_with_context(struct dc *dc, dc_state_set_stream_subvp_cursor_limit(context->streams[i], context, false); } - res = dc_validate_global_state(dc, context, fast_validate); + res = dc_validate_global_state(dc, context, validate_mode); /* calculate pixel rate divider after deciding pxiel clock & odm combine */ if ((dc->hwss.calculate_pix_rate_divider) && (res == DC_OK)) { @@ -4299,7 +4299,7 @@ static void decide_hblank_borrow(struct pipe_ctx *pipe_ctx) * * @dc: dc struct for this driver * @new_ctx: state to be validated - * @fast_validate: set to true if only yes/no to support matters + * @validate_mode: identify the validation mode * * Checks hardware resource availability and bandwidth requirement. * @@ -4309,7 +4309,7 @@ static void decide_hblank_borrow(struct pipe_ctx *pipe_ctx) enum dc_status dc_validate_global_state( struct dc *dc, struct dc_state *new_ctx, - bool fast_validate) + enum dc_validate_mode validate_mode) { enum dc_status result = DC_ERROR_UNEXPECTED; int i, j; @@ -4368,7 +4368,7 @@ enum dc_status dc_validate_global_state( result = resource_build_scaling_params_for_context(dc, new_ctx); if (result == DC_OK) - result = dc->res_pool->funcs->validate_bandwidth(dc, new_ctx, fast_validate); + result = dc->res_pool->funcs->validate_bandwidth(dc, new_ctx, validate_mode); return result; } diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_state.c b/drivers/gpu/drm/amd/display/dc/core/dc_state.c index 4db7383720fd0853e340e06f52994f843e326b73..47712a4aec55ba89e325378af19bd875f284e001 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_state.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_state.c @@ -194,11 +194,6 @@ static void init_state(struct dc *dc, struct dc_state *state) struct dc_state *dc_state_create(struct dc *dc, struct dc_state_create_params *params) { struct dc_state *state; -#ifdef CONFIG_DRM_AMD_DC_FP - struct dml2_configuration_options *dml2_opt = &dc->dml2_tmp; - - memcpy(dml2_opt, &dc->dml2_options, sizeof(dc->dml2_options)); -#endif state = kvzalloc(sizeof(struct dc_state), GFP_KERNEL); @@ -211,14 +206,12 @@ struct dc_state *dc_state_create(struct dc *dc, struct dc_state_create_params *p #ifdef CONFIG_DRM_AMD_DC_FP if (dc->debug.using_dml2) { - dml2_opt->use_clock_dc_limits = false; - if (!dml2_create(dc, dml2_opt, &state->bw_ctx.dml2)) { + if (!dml2_create(dc, &dc->dml2_options, &state->bw_ctx.dml2)) { dc_state_release(state); return NULL; } - dml2_opt->use_clock_dc_limits = true; - if (!dml2_create(dc, dml2_opt, &state->bw_ctx.dml2_dc_power_source)) { + if (!dml2_create(dc, &dc->dml2_dc_power_options, &state->bw_ctx.dml2_dc_power_source)) { dc_state_release(state); return NULL; } diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h index 1d917be36fc4376fd55ab856d9b2fb8126c8bbcd..ce1957c7862f9f2a7284cc50a8800c536afb0ce2 100644 --- a/drivers/gpu/drm/amd/display/dc/dc.h +++ b/drivers/gpu/drm/amd/display/dc/dc.h @@ -46,6 +46,8 @@ #include "dmub/inc/dmub_cmd.h" +#include "sspl/dc_spl_types.h" + struct abm_save_restore; /* forward declaration */ @@ -53,7 +55,7 @@ struct aux_payload; struct set_config_cmd_payload; struct dmub_notification; -#define DC_VER "3.2.334" +#define DC_VER "3.2.340" /** * MAX_SURFACES - representative of the upper bound of surfaces that can be piped to a single CRTC @@ -66,7 +68,11 @@ struct dmub_notification; #define MAX_STREAMS 6 #define MIN_VIEWPORT_SIZE 12 #define MAX_NUM_EDP 2 -#define MAX_HOST_ROUTERS_NUM 2 +#define MAX_SUPPORTED_FORMATS 7 + +#define MAX_HOST_ROUTERS_NUM 3 +#define MAX_DPIA_PER_HOST_ROUTER 3 +#define MAX_DPIA_NUM (MAX_HOST_ROUTERS_NUM * MAX_DPIA_PER_HOST_ROUTER) /* Display Core Interfaces */ struct dc_versions { @@ -192,6 +198,34 @@ struct dpp_color_caps { struct rom_curve_caps ogam_rom_caps; }; +/* Below structure is to describe the HW support for mem layout, extend support + range to match what OS could handle in the roadmap */ +struct lut3d_caps { + uint32_t dma_3d_lut : 1; /*< DMA mode support for 3D LUT */ + struct { + uint32_t swizzle_3d_rgb : 1; + uint32_t swizzle_3d_bgr : 1; + uint32_t linear_1d : 1; + } mem_layout_support; + struct { + uint32_t unorm_12msb : 1; + uint32_t unorm_12lsb : 1; + uint32_t float_fp1_5_10 : 1; + } mem_format_support; + struct { + uint32_t order_rgba : 1; + uint32_t order_bgra : 1; + } mem_pixel_order_support; + /*< size options are 9, 17, 33, 45, 65 */ + struct { + uint32_t dim_9 : 1; /* 3D LUT support for 9x9x9 */ + uint32_t dim_17 : 1; /* 3D LUT support for 17x17x17 */ + uint32_t dim_33 : 1; /* 3D LUT support for 33x33x33 */ + uint32_t dim_45 : 1; /* 3D LUT support for 45x45x45 */ + uint32_t dim_65 : 1; /* 3D LUT support for 65x65x65 */ + } lut_dim_caps; +}; + /** * struct mpc_color_caps - color pipeline capabilities for multiple pipe and * plane combined blocks @@ -203,6 +237,9 @@ struct dpp_color_caps { * @shared_3d_lut: shared 3D LUT flag. Can be either DPP or MPC, but single * instance * @ogam_rom_caps: pre-definied curve caps for regamma 1D LUT + * @mcm_3d_lut_caps: HW support cap for MCM LUT memory + * @rmcm_3d_lut_caps: HW support cap for RMCM LUT memory + * @preblend: whether color manager supports preblend with MPC */ struct mpc_color_caps { uint16_t gamut_remap : 1; @@ -211,6 +248,9 @@ struct mpc_color_caps { uint16_t num_3dluts : 3; uint16_t shared_3d_lut:1; struct rom_curve_caps ogam_rom_caps; + struct lut3d_caps mcm_3d_lut_caps; + struct lut3d_caps rmcm_3d_lut_caps; + bool preblend; }; /** @@ -305,6 +345,8 @@ struct dc_caps { /* Conservative limit for DCC cases which require ODM4:1 to support*/ uint32_t dcc_plane_width_limit; struct dc_scl_caps scl_caps; + uint8_t num_of_host_routers; + uint8_t num_of_dpias_per_host_router; }; struct dc_bug_wa { @@ -481,6 +523,8 @@ struct dc_config { bool set_pipe_unlock_order; bool enable_dpia_pre_training; bool unify_link_enc_assignment; + struct spl_sharpness_range dcn_sharpness_range; + struct spl_sharpness_range dcn_override_sharpness_range; }; enum visual_confirm { @@ -786,10 +830,8 @@ union dpia_debug_options { uint32_t disable_mst_dsc_work_around:1; /* bit 3 */ uint32_t enable_force_tbt3_work_around:1; /* bit 4 */ uint32_t disable_usb4_pm_support:1; /* bit 5 */ - uint32_t enable_consolidated_dpia_dp_lt:1; /* bit 6 */ - uint32_t enable_dpia_pre_training:1; /* bit 7 */ - uint32_t unify_link_enc_assignment:1; /* bit 8 */ - uint32_t reserved:24; + uint32_t enable_usb4_bw_zero_alloc_patch:1; /* bit 6 */ + uint32_t reserved:25; } bits; uint32_t raw; }; @@ -1151,7 +1193,7 @@ struct dc_init_data { uint32_t *dcn_reg_offsets; uint32_t *nbio_reg_offsets; uint32_t *clk_reg_offsets; - struct dml2_soc_bb *bb_from_dmub; + void *bb_from_dmub; }; struct dc_callback_init { @@ -1389,6 +1431,8 @@ struct dc_plane_state { int sharpness_level; enum linear_light_scaling linear_light_scaling; unsigned int sdr_white_level_nits; + struct spl_sharpness_range sharpness_range; + enum sharpness_range_source sharpness_source; }; struct dc_plane_info { @@ -1570,6 +1614,7 @@ struct dc_scratch_space { bool blank_stream_on_ocs_change; bool read_dpcd204h_on_irq_hpd; bool force_dp_ffe_preset; + bool skip_phy_ssc_reduction; } wa_flags; union dc_dp_ffe_preset forced_dp_ffe_preset; struct link_mst_stream_allocation_table mst_stream_alloc_table; @@ -1603,6 +1648,7 @@ struct dc { uint8_t link_count; struct dc_link *links[MAX_LINKS]; + uint8_t lowest_dpia_link_index; struct link_service *link_srv; struct dc_state *current_state; @@ -1662,7 +1708,7 @@ struct dc { } scratch; struct dml2_configuration_options dml2_options; - struct dml2_configuration_options dml2_tmp; + struct dml2_configuration_options dml2_dc_power_options; enum dc_acpi_cm_power_state power_state; }; @@ -1767,19 +1813,15 @@ enum dc_status dc_validate_with_context(struct dc *dc, const struct dc_validation_set set[], int set_count, struct dc_state *context, - bool fast_validate); + enum dc_validate_mode validate_mode); bool dc_set_generic_gpio_for_stereo(bool enable, struct gpio_service *gpio_service); -/* - * fast_validate: we return after determining if we can support the new state, - * but before we populate the programming info - */ enum dc_status dc_validate_global_state( struct dc *dc, struct dc_state *new_ctx, - bool fast_validate); + enum dc_validate_mode validate_mode); bool dc_acquire_release_mpc_3dlut( struct dc *dc, bool acquire, @@ -2375,17 +2417,12 @@ void dc_link_dp_dpia_handle_usb4_bandwidth_allocation_for_link( struct dc_link *link, int peak_bw); /* - * Validate the BW of all the valid DPIA links to make sure it doesn't exceed - * available BW for each host router - * - * @dc: pointer to dc struct - * @stream: pointer to all possible streams - * @count: number of valid DPIA streams + * Calculates the DP tunneling bandwidth required for the stream timing + * and aggregates the stream bandwidth for the respective DP tunneling link * - * return: TRUE if bw used by DPIAs doesn't exceed available BW else return FALSE + * return: dc_status */ -bool dc_link_dp_dpia_validate(struct dc *dc, const struct dc_stream_state *streams, - const unsigned int count); +enum dc_status dc_link_validate_dp_tunneling_bandwidth(const struct dc *dc, const struct dc_state *new_ctx); /* Sink Interfaces - A sink corresponds to a display output device */ @@ -2595,6 +2632,8 @@ struct dc_power_profile dc_get_power_profile_for_dc_state(const struct dc_state unsigned int dc_get_det_buffer_size_from_state(const struct dc_state *context); +bool dc_get_host_router_index(const struct dc_link *link, unsigned int *host_router_index); + /* DSC Interfaces */ #include "dc_dsc.h" diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c index afbcf866520e0232f8ddd4ed7a71c05cc0ab1f51..0a47d1a3515b634790d541be4e41e7867ce419ba 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c +++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c @@ -1656,7 +1656,7 @@ bool dc_wake_and_execute_gpint(const struct dc_context *ctx, enum dmub_gpint_com return result; } -void dc_dmub_srv_fams2_update_config(struct dc *dc, +static void dc_dmub_srv_rb_based_fams2_update_config(struct dc *dc, struct dc_state *context, bool enable) { @@ -1722,6 +1722,63 @@ void dc_dmub_srv_fams2_update_config(struct dc *dc, dm_execute_dmub_cmd_list(dc->ctx, num_cmds, cmd, DM_DMUB_WAIT_TYPE_WAIT); } +static void dc_dmub_srv_ib_based_fams2_update_config(struct dc *dc, + struct dc_state *context, + bool enable) +{ + struct dmub_fams2_config_v2 *config = (struct dmub_fams2_config_v2 *)dc->ctx->dmub_srv->dmub->ib_mem_gart.cpu_addr; + union dmub_rb_cmd cmd; + uint32_t i; + + memset(config, 0, sizeof(*config)); + memset(&cmd, 0, sizeof(cmd)); + + cmd.ib_fams2_config.header.type = DMUB_CMD__FW_ASSISTED_MCLK_SWITCH; + cmd.ib_fams2_config.header.sub_type = DMUB_CMD__FAMS2_IB_CONFIG; + + cmd.ib_fams2_config.ib_data.src.quad_part = dc->ctx->dmub_srv->dmub->ib_mem_gart.gpu_addr; + cmd.ib_fams2_config.ib_data.size = sizeof(*config); + + if (enable && context->bw_ctx.bw.dcn.fams2_global_config.features.bits.enable) { + /* copy static feature configuration overrides */ + config->global.features.bits.enable_stall_recovery = dc->debug.fams2_config.bits.enable_stall_recovery; + config->global.features.bits.enable_offload_flip = dc->debug.fams2_config.bits.enable_offload_flip; + config->global.features.bits.enable_debug = dc->debug.fams2_config.bits.enable_debug; + + /* send global configuration parameters */ + memcpy(&config->global, &context->bw_ctx.bw.dcn.fams2_global_config, + sizeof(struct dmub_cmd_fams2_global_config)); + + /* construct per-stream configs */ + for (i = 0; i < context->bw_ctx.bw.dcn.fams2_global_config.num_streams; i++) { + /* copy stream static base state */ + memcpy(&config->stream_v1[i].base, + &context->bw_ctx.bw.dcn.fams2_stream_base_params[i], + sizeof(config->stream_v1[i].base)); + + /* copy stream static sub-state */ + memcpy(&config->stream_v1[i].sub_state, + &context->bw_ctx.bw.dcn.fams2_stream_sub_params_v2[i], + sizeof(config->stream_v1[i].sub_state)); + } + } + + config->global.features.bits.enable_visual_confirm = dc->debug.visual_confirm == VISUAL_CONFIRM_FAMS2; + config->global.features.bits.enable = enable; + + dm_execute_dmub_cmd_list(dc->ctx, 1, &cmd, DM_DMUB_WAIT_TYPE_WAIT); +} + +void dc_dmub_srv_fams2_update_config(struct dc *dc, + struct dc_state *context, + bool enable) +{ + if (dc->debug.fams_version.major == 2) + dc_dmub_srv_rb_based_fams2_update_config(dc, context, enable); + if (dc->debug.fams_version.major == 3) + dc_dmub_srv_ib_based_fams2_update_config(dc, context, enable); +} + void dc_dmub_srv_fams2_drr_update(struct dc *dc, uint32_t tg_inst, uint32_t vtotal_min, @@ -1847,83 +1904,250 @@ void dc_dmub_srv_fams2_passthrough_flip( } } -bool dc_dmub_srv_ips_residency_cntl(struct dc_dmub_srv *dc_dmub_srv, bool start_measurement) + +bool dc_dmub_srv_ips_residency_cntl(const struct dc_context *ctx, uint8_t panel_inst, bool start_measurement) { - bool result; + union dmub_rb_cmd cmd; - if (!dc_dmub_srv || !dc_dmub_srv->dmub) + memset(&cmd, 0, sizeof(cmd)); + + cmd.ips_residency_cntl.header.type = DMUB_CMD__IPS; + cmd.ips_residency_cntl.header.sub_type = DMUB_CMD__IPS_RESIDENCY_CNTL; + cmd.ips_residency_cntl.header.payload_bytes = sizeof(struct dmub_cmd_ips_residency_cntl_data); + + // only panel_inst=0 is supported at the moment + cmd.ips_residency_cntl.cntl_data.panel_inst = panel_inst; + cmd.ips_residency_cntl.cntl_data.start_measurement = start_measurement; + + if (!dc_wake_and_execute_dmub_cmd(ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) return false; - result = dc_wake_and_execute_gpint(dc_dmub_srv->ctx, DMUB_GPINT__IPS_RESIDENCY, - start_measurement, NULL, DM_DMUB_WAIT_TYPE_WAIT); + return true; +} + +bool dc_dmub_srv_ips_query_residency_info(const struct dc_context *ctx, uint8_t panel_inst, struct dmub_ips_residency_info *driver_info, + enum ips_residency_mode ips_mode) +{ + union dmub_rb_cmd cmd; + uint32_t bytes = sizeof(struct dmub_ips_residency_info); + + dmub_flush_buffer_mem(&ctx->dmub_srv->dmub->scratch_mem_fb); + memset(&cmd, 0, sizeof(cmd)); + + cmd.ips_query_residency_info.header.type = DMUB_CMD__IPS; + cmd.ips_query_residency_info.header.sub_type = DMUB_CMD__IPS_QUERY_RESIDENCY_INFO; + cmd.ips_query_residency_info.header.payload_bytes = sizeof(struct dmub_cmd_ips_query_residency_info_data); + + cmd.ips_query_residency_info.info_data.dest.quad_part = ctx->dmub_srv->dmub->scratch_mem_fb.gpu_addr; + cmd.ips_query_residency_info.info_data.size = bytes; + cmd.ips_query_residency_info.info_data.panel_inst = panel_inst; + cmd.ips_query_residency_info.info_data.ips_mode = (uint32_t)ips_mode; + + if (!dc_wake_and_execute_dmub_cmd(ctx, &cmd, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY) || + cmd.ips_query_residency_info.header.ret_status == 0) + return false; + + // copy the result to the output since ret_status != 0 means the command returned data + memcpy(driver_info, ctx->dmub_srv->dmub->scratch_mem_fb.cpu_addr, bytes); + + return true; +} + +bool dmub_lsdma_init(struct dc_dmub_srv *dc_dmub_srv) +{ + struct dc_context *dc_ctx = dc_dmub_srv->ctx; + union dmub_rb_cmd cmd; + enum dm_dmub_wait_type wait_type; + struct dmub_cmd_lsdma_data *lsdma_data = &cmd.lsdma.lsdma_data; + bool result; + + memset(&cmd, 0, sizeof(cmd)); + + cmd.cmd_common.header.type = DMUB_CMD__LSDMA; + cmd.cmd_common.header.sub_type = DMUB_CMD__LSDMA_INIT_CONFIG; + wait_type = DM_DMUB_WAIT_TYPE_NO_WAIT; + + lsdma_data->u.init_data.gpu_addr_base.quad_part = dc_ctx->dmub_srv->dmub->lsdma_rb_fb.gpu_addr; + lsdma_data->u.init_data.ring_size = dc_ctx->dmub_srv->dmub->lsdma_rb_fb.size; + + result = dc_wake_and_execute_dmub_cmd(dc_ctx, &cmd, wait_type); + + if (!result) + DC_ERROR("LSDMA Init failed in DMUB"); return result; } -void dc_dmub_srv_ips_query_residency_info(struct dc_dmub_srv *dc_dmub_srv, struct ips_residency_info *output) +bool dmub_lsdma_send_linear_copy_packet( + struct dc_dmub_srv *dc_dmub_srv, + uint64_t src_addr, + uint64_t dst_addr, + uint32_t count) { - uint32_t i; - enum dmub_gpint_command command_code; + struct dc_context *dc_ctx = dc_dmub_srv->ctx; + union dmub_rb_cmd cmd; + enum dm_dmub_wait_type wait_type; + struct dmub_cmd_lsdma_data *lsdma_data = &cmd.lsdma.lsdma_data; + bool result; - if (!dc_dmub_srv || !dc_dmub_srv->dmub) - return; + memset(&cmd, 0, sizeof(cmd)); - switch (output->ips_mode) { - case DMUB_IPS_MODE_IPS1_MAX: - command_code = DMUB_GPINT__GET_IPS1_HISTOGRAM_COUNTER; - break; - case DMUB_IPS_MODE_IPS2: - command_code = DMUB_GPINT__GET_IPS2_HISTOGRAM_COUNTER; - break; - case DMUB_IPS_MODE_IPS1_RCG: - command_code = DMUB_GPINT__GET_IPS1_RCG_HISTOGRAM_COUNTER; - break; - case DMUB_IPS_MODE_IPS1_ONO2_ON: - command_code = DMUB_GPINT__GET_IPS1_ONO2_ON_HISTOGRAM_COUNTER; - break; - default: - command_code = DMUB_GPINT__INVALID_COMMAND; - break; - } + cmd.cmd_common.header.type = DMUB_CMD__LSDMA; + cmd.cmd_common.header.sub_type = DMUB_CMD__LSDMA_LINEAR_COPY; + wait_type = DM_DMUB_WAIT_TYPE_NO_WAIT; - if (command_code == DMUB_GPINT__INVALID_COMMAND) - return; + lsdma_data->u.linear_copy_data.count = count - 1; // LSDMA controller expects bytes to copy -1 + lsdma_data->u.linear_copy_data.src_lo = src_addr & 0xFFFFFFFF; + lsdma_data->u.linear_copy_data.src_hi = (src_addr >> 32) & 0xFFFFFFFF; + lsdma_data->u.linear_copy_data.dst_lo = dst_addr & 0xFFFFFFFF; + lsdma_data->u.linear_copy_data.dst_hi = (dst_addr >> 32) & 0xFFFFFFFF; - for (i = 0; i < GPINT_RETRY_NUM; i++) { - // false could mean GPINT timeout, in which case we should retry - if (dc_wake_and_execute_gpint(dc_dmub_srv->ctx, DMUB_GPINT__GET_IPS_RESIDENCY_PERCENT, - (uint16_t)(output->ips_mode), &output->residency_percent, - DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) - break; - udelay(100); - } + result = dc_wake_and_execute_dmub_cmd(dc_ctx, &cmd, wait_type); + + if (!result) + DC_ERROR("LSDMA Linear Copy failed in DMUB"); + + return result; +} + +bool dmub_lsdma_send_tiled_to_tiled_copy_command( + struct dc_dmub_srv *dc_dmub_srv, + struct lsdma_send_tiled_to_tiled_copy_command_params params) +{ + struct dc_context *dc_ctx = dc_dmub_srv->ctx; + union dmub_rb_cmd cmd; + enum dm_dmub_wait_type wait_type; + struct dmub_cmd_lsdma_data *lsdma_data = &cmd.lsdma.lsdma_data; + bool result; + + memset(&cmd, 0, sizeof(cmd)); + + cmd.cmd_common.header.type = DMUB_CMD__LSDMA; + cmd.cmd_common.header.sub_type = DMUB_CMD__LSDMA_TILED_TO_TILED_COPY; + wait_type = DM_DMUB_WAIT_TYPE_NO_WAIT; + + lsdma_data->u.tiled_copy_data.src_addr_lo = params.src_addr & 0xFFFFFFFF; + lsdma_data->u.tiled_copy_data.src_addr_hi = (params.src_addr >> 32) & 0xFFFFFFFF; + lsdma_data->u.tiled_copy_data.dst_addr_lo = params.dst_addr & 0xFFFFFFFF; + lsdma_data->u.tiled_copy_data.dst_addr_hi = (params.dst_addr >> 32) & 0xFFFFFFFF; + lsdma_data->u.tiled_copy_data.src_x = params.src_x; + lsdma_data->u.tiled_copy_data.src_y = params.src_y; + lsdma_data->u.tiled_copy_data.dst_x = params.dst_x; + lsdma_data->u.tiled_copy_data.dst_y = params.dst_y; + lsdma_data->u.tiled_copy_data.src_width = params.src_width - 1; // LSDMA controller expects width -1 + lsdma_data->u.tiled_copy_data.dst_width = params.dst_width - 1; // LSDMA controller expects width -1 + lsdma_data->u.tiled_copy_data.src_swizzle_mode = params.swizzle_mode; + lsdma_data->u.tiled_copy_data.dst_swizzle_mode = params.swizzle_mode; + lsdma_data->u.tiled_copy_data.src_element_size = params.element_size; + lsdma_data->u.tiled_copy_data.dst_element_size = params.element_size; + lsdma_data->u.tiled_copy_data.rect_x = params.rect_x; + lsdma_data->u.tiled_copy_data.rect_y = params.rect_y; + lsdma_data->u.tiled_copy_data.dcc = params.dcc; + lsdma_data->u.tiled_copy_data.tmz = params.tmz; + lsdma_data->u.tiled_copy_data.read_compress = params.read_compress; + lsdma_data->u.tiled_copy_data.write_compress = params.write_compress; + lsdma_data->u.tiled_copy_data.src_height = params.src_height - 1; // LSDMA controller expects height -1 + lsdma_data->u.tiled_copy_data.dst_height = params.dst_height - 1; // LSDMA controller expects height -1 + lsdma_data->u.tiled_copy_data.data_format = params.data_format; + lsdma_data->u.tiled_copy_data.max_com = params.max_com; + lsdma_data->u.tiled_copy_data.max_uncom = params.max_uncom; + + result = dc_wake_and_execute_dmub_cmd(dc_ctx, &cmd, wait_type); + + if (!result) + DC_ERROR("LSDMA Tiled to Tiled Copy failed in DMUB"); + + return result; +} + +bool dmub_lsdma_send_pio_copy_command( + struct dc_dmub_srv *dc_dmub_srv, + uint64_t src_addr, + uint64_t dst_addr, + uint32_t byte_count, + uint32_t overlap_disable) +{ + struct dc_context *dc_ctx = dc_dmub_srv->ctx; + union dmub_rb_cmd cmd; + enum dm_dmub_wait_type wait_type; + struct dmub_cmd_lsdma_data *lsdma_data = &cmd.lsdma.lsdma_data; + bool result; + + memset(&cmd, 0, sizeof(cmd)); + + cmd.cmd_common.header.type = DMUB_CMD__LSDMA; + cmd.cmd_common.header.sub_type = DMUB_CMD__LSDMA_PIO_COPY; + wait_type = DM_DMUB_WAIT_TYPE_NO_WAIT; - if (!dc_wake_and_execute_gpint(dc_dmub_srv->ctx, DMUB_GPINT__GET_IPS_RESIDENCY_ENTRY_COUNTER, - (uint16_t)(output->ips_mode), - &output->entry_counter, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) - output->entry_counter = 0; - - if (!dc_wake_and_execute_gpint(dc_dmub_srv->ctx, DMUB_GPINT__GET_IPS_RESIDENCY_DURATION_US_LO, - (uint16_t)(output->ips_mode), - &output->total_active_time_us[0], DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) - output->total_active_time_us[0] = 0; - if (!dc_wake_and_execute_gpint(dc_dmub_srv->ctx, DMUB_GPINT__GET_IPS_RESIDENCY_DURATION_US_HI, - (uint16_t)(output->ips_mode), - &output->total_active_time_us[1], DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) - output->total_active_time_us[1] = 0; - - if (!dc_wake_and_execute_gpint(dc_dmub_srv->ctx, DMUB_GPINT__GET_IPS_INACTIVE_RESIDENCY_DURATION_US_LO, - (uint16_t)(output->ips_mode), - &output->total_inactive_time_us[0], DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) - output->total_inactive_time_us[0] = 0; - if (!dc_wake_and_execute_gpint(dc_dmub_srv->ctx, DMUB_GPINT__GET_IPS_INACTIVE_RESIDENCY_DURATION_US_HI, - (uint16_t)(output->ips_mode), - &output->total_inactive_time_us[1], DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) - output->total_inactive_time_us[1] = 0; - - // NUM_IPS_HISTOGRAM_BUCKETS = 16 - for (i = 0; i < 16; i++) - if (!dc_wake_and_execute_gpint(dc_dmub_srv->ctx, command_code, i, &output->histogram[i], - DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) - output->histogram[i] = 0; + lsdma_data->u.pio_copy_data.packet.fields.byte_count = byte_count; + lsdma_data->u.pio_copy_data.packet.fields.overlap_disable = overlap_disable; + lsdma_data->u.pio_copy_data.src_lo = src_addr & 0xFFFFFFFF; + lsdma_data->u.pio_copy_data.src_hi = (src_addr >> 32) & 0xFFFFFFFF; + lsdma_data->u.pio_copy_data.dst_lo = dst_addr & 0xFFFFFFFF; + lsdma_data->u.pio_copy_data.dst_hi = (dst_addr >> 32) & 0xFFFFFFFF; + + result = dc_wake_and_execute_dmub_cmd(dc_ctx, &cmd, wait_type); + + if (!result) + DC_ERROR("LSDMA PIO Copy failed in DMUB"); + + return result; } + +bool dmub_lsdma_send_pio_constfill_command( + struct dc_dmub_srv *dc_dmub_srv, + uint64_t dst_addr, + uint32_t byte_count, + uint32_t data) +{ + struct dc_context *dc_ctx = dc_dmub_srv->ctx; + union dmub_rb_cmd cmd; + enum dm_dmub_wait_type wait_type; + struct dmub_cmd_lsdma_data *lsdma_data = &cmd.lsdma.lsdma_data; + bool result; + + memset(&cmd, 0, sizeof(cmd)); + + cmd.cmd_common.header.type = DMUB_CMD__LSDMA; + cmd.cmd_common.header.sub_type = DMUB_CMD__LSDMA_PIO_CONSTFILL; + wait_type = DM_DMUB_WAIT_TYPE_NO_WAIT; + + lsdma_data->u.pio_constfill_data.packet.fields.constant_fill = 1; + lsdma_data->u.pio_constfill_data.packet.fields.byte_count = byte_count; + lsdma_data->u.pio_constfill_data.dst_lo = dst_addr & 0xFFFFFFFF; + lsdma_data->u.pio_constfill_data.dst_hi = (dst_addr >> 32) & 0xFFFFFFFF; + lsdma_data->u.pio_constfill_data.data = data; + + result = dc_wake_and_execute_dmub_cmd(dc_ctx, &cmd, wait_type); + + if (!result) + DC_ERROR("LSDMA PIO Constfill failed in DMUB"); + + return result; +} + +bool dmub_lsdma_send_poll_reg_write_command(struct dc_dmub_srv *dc_dmub_srv, uint32_t reg_addr, uint32_t reg_data) +{ + struct dc_context *dc_ctx = dc_dmub_srv->ctx; + union dmub_rb_cmd cmd; + enum dm_dmub_wait_type wait_type; + struct dmub_cmd_lsdma_data *lsdma_data = &cmd.lsdma.lsdma_data; + bool result; + + memset(&cmd, 0, sizeof(cmd)); + + cmd.cmd_common.header.type = DMUB_CMD__LSDMA; + cmd.cmd_common.header.sub_type = DMUB_CMD__LSDMA_POLL_REG_WRITE; + wait_type = DM_DMUB_WAIT_TYPE_NO_WAIT; + + lsdma_data->u.reg_write_data.reg_addr = reg_addr; + lsdma_data->u.reg_write_data.reg_data = reg_data; + + result = dc_wake_and_execute_dmub_cmd(dc_ctx, &cmd, wait_type); + + if (!result) + DC_ERROR("LSDMA Poll Reg failed in DMUB"); + + return result; +} + diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h index ada5c2fb2db300b13de4c2f4e3de61ceac490deb..1f1c155be30e938af72b90e28d3f88598982d3e7 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h +++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h @@ -210,6 +210,60 @@ void dc_dmub_srv_fams2_passthrough_flip( struct dc_surface_update *srf_updates, int surface_count); +bool dmub_lsdma_init(struct dc_dmub_srv *dc_dmub_srv); +bool dmub_lsdma_send_linear_copy_packet( + struct dc_dmub_srv *dc_dmub_srv, + uint64_t src_addr, + uint64_t dst_addr, + uint32_t count); +bool dmub_lsdma_send_pio_copy_command( + struct dc_dmub_srv *dc_dmub_srv, + uint64_t src_addr, + uint64_t dst_addr, + uint32_t byte_count, + uint32_t overlap_disable); +bool dmub_lsdma_send_pio_constfill_command( + struct dc_dmub_srv *dc_dmub_srv, + uint64_t dst_addr, + uint32_t byte_count, + uint32_t data); + +struct lsdma_send_tiled_to_tiled_copy_command_params { + uint64_t src_addr; + uint64_t dst_addr; + + uint32_t src_x : 16; + uint32_t src_y : 16; + + uint32_t dst_x : 16; + uint32_t dst_y : 16; + + uint32_t src_width : 16; + uint32_t dst_width : 16; + + uint32_t rect_x : 16; + uint32_t rect_y : 16; + + uint32_t src_height : 16; + uint32_t dst_height : 16; + + uint32_t data_format : 6; + uint32_t swizzle_mode : 5; + uint32_t element_size : 3; + uint32_t dcc : 1; + uint32_t tmz : 1; + uint32_t read_compress : 2; + uint32_t write_compress : 2; + uint32_t max_com : 2; + uint32_t max_uncom : 1; + uint32_t padding : 9; +}; + +bool dmub_lsdma_send_tiled_to_tiled_copy_command( + struct dc_dmub_srv *dc_dmub_srv, + struct lsdma_send_tiled_to_tiled_copy_command_params params); +bool dmub_lsdma_send_poll_reg_write_command(struct dc_dmub_srv *dc_dmub_srv, uint32_t reg_addr, uint32_t reg_data); + /** * struct ips_residency_info - struct containing info from dmub_ips_residency_stats * @@ -223,7 +277,7 @@ void dc_dmub_srv_fams2_passthrough_flip( * @histogram: Histogram of given IPS state durations - bucket definitions in dmub_ips.c */ struct ips_residency_info { - enum dmub_ips_mode ips_mode; + enum ips_residency_mode ips_mode; unsigned int residency_percent; unsigned int entry_counter; unsigned int total_active_time_us[2]; @@ -231,21 +285,10 @@ struct ips_residency_info { unsigned int histogram[16]; }; -/** - * bool dc_dmub_srv_ips_residency_cntl() - Controls IPS residency measurement status - * - * @dc_dmub_srv: The DC DMUB service pointer - * @start_measurement: Describes whether to start or stop measurement - * - * Return: true if GPINT was sent successfully, false otherwise - */ -bool dc_dmub_srv_ips_residency_cntl(struct dc_dmub_srv *dc_dmub_srv, bool start_measurement); +bool dc_dmub_srv_ips_residency_cntl(const struct dc_context *ctx, uint8_t panel_inst, bool start_measurement); + +bool dc_dmub_srv_ips_query_residency_info(const struct dc_context *ctx, uint8_t panel_inst, + struct dmub_ips_residency_info *driver_info, + enum ips_residency_mode ips_mode); -/** - * bool dc_dmub_srv_ips_query_residency_info() - Queries DMCUB for residency info - * - * @dc_dmub_srv: The DC DMUB service pointer - * @output: Output struct to copy the the residency info to - */ -void dc_dmub_srv_ips_query_residency_info(struct dc_dmub_srv *dc_dmub_srv, struct ips_residency_info *output); #endif /* _DMUB_DC_SRV_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h index 0bad8304ccf60e091ae637c6d9e9f1b6f7673a5e..5ce1be362534dc97329409c6b08adeb64e80e84f 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h @@ -162,6 +162,11 @@ struct dc_link_settings { struct dc_tunnel_settings { bool should_enable_dp_tunneling; bool should_use_dp_bw_allocation; + uint8_t cm_id; + uint8_t group_id; + uint32_t bw_granularity; + uint32_t estimated_bw; + uint32_t allocated_bw; }; union dc_dp_ffe_preset { @@ -957,11 +962,21 @@ union usb4_driver_bw_cap { uint8_t raw; }; +/* DPCD[0xE0021] DP_IN_ADAPTER_TUNNEL_INFORMATION register. */ +union dpia_tunnel_info { + struct { + uint8_t group_id :3; + uint8_t rsvd :5; + } bits; + uint8_t raw; +}; + /* DP Tunneling over USB4 */ struct dpcd_usb4_dp_tunneling_info { union dp_tun_cap_support dp_tun_cap; union dpia_info dpia_info; union usb4_driver_bw_cap driver_bw_cap; + union dpia_tunnel_info dpia_tunnel_info; uint8_t usb4_driver_id; uint8_t usb4_topology_id[DPCD_USB4_TOPOLOGY_ID_LEN]; }; @@ -1172,8 +1187,8 @@ struct dc_lttpr_caps { union dp_128b_132b_supported_lttpr_link_rates supported_128b_132b_rates; union dp_alpm_lttpr_cap alpm; uint8_t aux_rd_interval[MAX_REPEATER_CNT - 1]; - uint8_t lttpr_ieee_oui[3]; - uint8_t lttpr_device_id[6]; + uint8_t lttpr_ieee_oui[3]; // Always read from closest LTTPR to host + uint8_t lttpr_device_id[6]; // Always read from closest LTTPR to host }; struct dc_dongle_dfp_cap_ext { diff --git a/drivers/gpu/drm/amd/display/dc/dc_hw_types.h b/drivers/gpu/drm/amd/display/dc/dc_hw_types.h index d562ddeca51260db98fc92ec3a4823d8164a9e53..6678525172461f80d6bd57b31cee36ad44fa0886 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_hw_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_hw_types.h @@ -68,7 +68,7 @@ enum dc_plane_addr_type { struct dc_plane_address { enum dc_plane_addr_type type; - bool tmz_surface; + uint8_t tmz_surface; union { struct{ PHYSICAL_ADDRESS_LOC addr; @@ -974,6 +974,7 @@ struct dc_crtc_timing { uint32_t pix_clk_100hz; uint32_t min_refresh_in_uhz; + uint32_t max_refresh_in_uhz; uint32_t vic; uint32_t hdmi_vic; @@ -1103,7 +1104,8 @@ enum mpcc_gamut_remap_mode_select { enum mpcc_gamut_remap_id { MPCC_OGAM_GAMUT_REMAP, MPCC_MCM_FIRST_GAMUT_REMAP, - MPCC_MCM_SECOND_GAMUT_REMAP + MPCC_MCM_SECOND_GAMUT_REMAP, + MPCC_RMCM_GAMUT_REMAP, }; enum cursor_matrix_mode { diff --git a/drivers/gpu/drm/amd/display/dc/dc_spl_translate.c b/drivers/gpu/drm/amd/display/dc/dc_spl_translate.c index e3a8283b4098c7cbe3c506286f23b5b6b2658255..7f57661433eb57ff0d05a3ee020e89b1fca8693f 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_spl_translate.c +++ b/drivers/gpu/drm/amd/display/dc/dc_spl_translate.c @@ -156,15 +156,16 @@ void translate_SPL_in_params_from_pipe_ctx(struct pipe_ctx *pipe_ctx, struct spl spl_in->adaptive_sharpness.enable = true; spl_in->adaptive_sharpness.sharpness_level = 0; } else if (sharpness_setting == SHARPNESS_CUSTOM) { - spl_in->adaptive_sharpness.sharpness_range.sdr_rgb_min = 0; - spl_in->adaptive_sharpness.sharpness_range.sdr_rgb_max = 1750; - spl_in->adaptive_sharpness.sharpness_range.sdr_rgb_mid = 750; - spl_in->adaptive_sharpness.sharpness_range.sdr_yuv_min = 0; - spl_in->adaptive_sharpness.sharpness_range.sdr_yuv_max = 3500; - spl_in->adaptive_sharpness.sharpness_range.sdr_yuv_mid = 1500; - spl_in->adaptive_sharpness.sharpness_range.hdr_rgb_min = 0; - spl_in->adaptive_sharpness.sharpness_range.hdr_rgb_max = 2750; - spl_in->adaptive_sharpness.sharpness_range.hdr_rgb_mid = 1500; + /* SAT: read harpness_range from dc_plane_state */ + spl_in->adaptive_sharpness.sharpness_range.sdr_rgb_min = plane_state->sharpness_range.sdr_rgb_min; + spl_in->adaptive_sharpness.sharpness_range.sdr_rgb_max = plane_state->sharpness_range.sdr_rgb_max; + spl_in->adaptive_sharpness.sharpness_range.sdr_rgb_mid = plane_state->sharpness_range.sdr_rgb_mid; + spl_in->adaptive_sharpness.sharpness_range.sdr_yuv_min = plane_state->sharpness_range.sdr_yuv_min; + spl_in->adaptive_sharpness.sharpness_range.sdr_yuv_max = plane_state->sharpness_range.sdr_yuv_max; + spl_in->adaptive_sharpness.sharpness_range.sdr_yuv_mid = plane_state->sharpness_range.sdr_yuv_mid; + spl_in->adaptive_sharpness.sharpness_range.hdr_rgb_min = plane_state->sharpness_range.hdr_rgb_min; + spl_in->adaptive_sharpness.sharpness_range.hdr_rgb_max = plane_state->sharpness_range.hdr_rgb_max; + spl_in->adaptive_sharpness.sharpness_range.hdr_rgb_mid = plane_state->sharpness_range.hdr_rgb_mid; if (force_sharpness_level > 0) { if (force_sharpness_level > 10) diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h index a4cd0eb39a3af328fa5048a4984ed14ecec73640..b203ed020cd5746b35098182f822835d9f69c898 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_types.h @@ -1255,7 +1255,6 @@ enum dc_cm2_gpu_mem_layout { enum dc_cm2_gpu_mem_pixel_component_order { DC_CM2_GPU_MEM_PIXEL_COMPONENT_ORDER_RGBA, - DC_CM2_GPU_MEM_PIXEL_COMPONENT_ORDER_BGRA }; enum dc_cm2_gpu_mem_format { @@ -1277,7 +1276,6 @@ struct dc_cm2_gpu_mem_format_parameters { enum dc_cm2_gpu_mem_size { DC_CM2_GPU_MEM_SIZE_171717, - DC_CM2_GPU_MEM_SIZE_333333, DC_CM2_GPU_MEM_SIZE_TRANSFORMED, }; @@ -1372,4 +1370,19 @@ struct set_backlight_level_params { uint8_t aux_inst; }; +enum dc_validate_mode { + /* validate the mode and program HW */ + DC_VALIDATE_MODE_AND_PROGRAMMING = 0, + /* only validate the mode */ + DC_VALIDATE_MODE_ONLY = 1, + /* validate the mode and get the max state (voltage level) */ + DC_VALIDATE_MODE_AND_STATE_INDEX = 2, +}; + +struct dc_validation_dpia_set { + const struct dc_link *link; + const struct dc_tunnel_settings *tunnel_settings; + uint32_t required_bw; +}; + #endif /* DC_TYPES_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.c b/drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.c index ffd172231fdfe6304a224530539ed7997b16fb10..668ee2d405fdf520c81f3db1c4849e0abe280b84 100644 --- a/drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.c +++ b/drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.c @@ -727,7 +727,7 @@ void dccg401_init(struct dccg *dccg) } } -void dccg401_set_dto_dscclk(struct dccg *dccg, uint32_t inst) +void dccg401_set_dto_dscclk(struct dccg *dccg, uint32_t inst, uint32_t num_slices_h) { struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg); diff --git a/drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.h b/drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.h index 55e8718aad22b966ca8fa75ae90b46b002987aa6..5947a35363aace644094b6ba7261b8445ca1558b 100644 --- a/drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.h +++ b/drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.h @@ -209,7 +209,7 @@ void dccg401_disable_symclk32_le( struct dccg *dccg, int hpo_le_inst); void dccg401_disable_dpstreamclk(struct dccg *dccg, int dp_hpo_inst); -void dccg401_set_dto_dscclk(struct dccg *dccg, uint32_t inst); +void dccg401_set_dto_dscclk(struct dccg *dccg, uint32_t inst, uint32_t num_slices_h); void dccg401_set_ref_dscclk(struct dccg *dccg, uint32_t dsc_inst); void dccg401_set_src_sel( @@ -230,7 +230,6 @@ void dccg401_set_dp_dto( const struct dp_dto_params *params); void dccg401_enable_symclk_se(struct dccg *dccg, uint32_t stream_enc_inst, uint32_t link_enc_inst); void dccg401_disable_symclk_se(struct dccg *dccg, uint32_t stream_enc_inst, uint32_t link_enc_inst); -void dccg401_set_dto_dscclk(struct dccg *dccg, uint32_t inst); void dccg401_set_dtbclk_p_src( struct dccg *dccg, enum streamclk_source src, diff --git a/drivers/gpu/drm/amd/display/dc/dml/Makefile b/drivers/gpu/drm/amd/display/dc/dml/Makefile index e1d500633dfad75e4f2265552be42a3e19dee6bf..b357683b4255af219817f994fe470a759fb89fa6 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dml/Makefile @@ -114,9 +114,6 @@ CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/calcs/dcn_calcs.o := $(dml_rcflags) CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/calcs/dcn_calc_auto.o := $(dml_rcflags) CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/calcs/dcn_calc_math.o := $(dml_rcflags) -CFLAGS_$(AMDDALPATH)/dc/dml/dcn401/dcn401_fpu.o := $(dml_ccflags) -CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml/dcn401/dcn401_fpu.o := $(dml_rcflags) - ifdef CONFIG_DRM_AMD_DC_FP DML += display_mode_lib.o display_rq_dlg_helpers.o dml1_display_rq_dlg_calc.o DML += dcn10/dcn10_fpu.o @@ -137,7 +134,6 @@ DML += dcn303/dcn303_fpu.o DML += dcn314/dcn314_fpu.o DML += dcn35/dcn35_fpu.o DML += dcn351/dcn351_fpu.o -DML += dcn401/dcn401_fpu.o DML += dsc/rc_calc_fpu.o DML += calcs/dcn_calcs.o calcs/dcn_calc_math.o calcs/dcn_calc_auto.o endif diff --git a/drivers/gpu/drm/amd/display/dc/dml/calcs/dcn_calcs.c b/drivers/gpu/drm/amd/display/dc/dml/calcs/dcn_calcs.c index f1235bf9a5965f8b96875ea9856a30bc59d0ffe2..74962791302f653ae83d0b632c8d00b85eaa4c78 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/calcs/dcn_calcs.c +++ b/drivers/gpu/drm/amd/display/dc/dml/calcs/dcn_calcs.c @@ -748,7 +748,7 @@ static unsigned int get_highest_allowed_voltage_level(bool is_vmin_only_asic) bool dcn_validate_bandwidth( struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { /* * we want a breakdown of the various stages of validation, which the @@ -1119,7 +1119,7 @@ bool dcn_validate_bandwidth( BW_VAL_TRACE_END_VOLTAGE_LEVEL(); - if (v->voltage_level != number_of_states_plus_one && !fast_validate) { + if (v->voltage_level != number_of_states_plus_one && validate_mode == DC_VALIDATE_MODE_AND_PROGRAMMING) { float bw_consumed = v->total_bandwidth_consumed_gbyte_per_second; if (bw_consumed < v->fabric_and_dram_bandwidth_vmin0p65) @@ -1286,7 +1286,7 @@ bool dcn_validate_bandwidth( } } else if (v->voltage_level == number_of_states_plus_one) { BW_VAL_TRACE_SKIP(fail); - } else if (fast_validate) { + } else if (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING) { BW_VAL_TRACE_SKIP(fast); } diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c index e9fea9c2162e800247206d5448ce58549b8a8246..2a2eaf6adf26cbc74fbe0e3218c334dda8c4aee6 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c @@ -1315,7 +1315,7 @@ static void swizzle_to_dml_params( int dcn20_populate_dml_pipes_from_context(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate) + enum dc_validate_mode validate_mode) { int pipe_cnt, i; bool synchronized_vblank = true; @@ -1733,7 +1733,7 @@ void dcn20_calculate_wm(struct dc *dc, struct dc_state *context, int *out_pipe_cnt, int *pipe_split_from, int vlevel, - bool fast_validate) + enum dc_validate_mode validate_mode) { int pipe_cnt, i, pipe_idx; @@ -1780,10 +1780,10 @@ void dcn20_calculate_wm(struct dc *dc, struct dc_state *context, if (pipe_cnt != pipe_idx) { if (dc->res_pool->funcs->populate_dml_pipes) pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, - context, pipes, fast_validate); + context, pipes, validate_mode); else pipe_cnt = dcn20_populate_dml_pipes_from_context(dc, - context, pipes, fast_validate); + context, pipes, validate_mode); } *out_pipe_cnt = pipe_cnt; @@ -2027,7 +2027,7 @@ void dcn20_patch_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st } static bool dcn20_validate_bandwidth_internal(struct dc *dc, struct dc_state *context, - bool fast_validate, display_e2e_pipe_params_st *pipes) + enum dc_validate_mode validate_mode, display_e2e_pipe_params_st *pipes) { bool out = false; @@ -2040,7 +2040,7 @@ static bool dcn20_validate_bandwidth_internal(struct dc *dc, struct dc_state *co BW_VAL_TRACE_COUNT(); - out = dcn20_fast_validate_bw(dc, context, pipes, &pipe_cnt, pipe_split_from, &vlevel, fast_validate); + out = dcn20_fast_validate_bw(dc, context, pipes, &pipe_cnt, pipe_split_from, &vlevel, validate_mode); if (pipe_cnt == 0) goto validate_out; @@ -2050,12 +2050,12 @@ static bool dcn20_validate_bandwidth_internal(struct dc *dc, struct dc_state *co BW_VAL_TRACE_END_VOLTAGE_LEVEL(); - if (fast_validate) { + if (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING) { BW_VAL_TRACE_SKIP(fast); goto validate_out; } - dcn20_calculate_wm(dc, context, pipes, &pipe_cnt, pipe_split_from, vlevel, fast_validate); + dcn20_calculate_wm(dc, context, pipes, &pipe_cnt, pipe_split_from, vlevel, validate_mode); dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel); BW_VAL_TRACE_END_WATERMARKS(); @@ -2077,7 +2077,7 @@ validate_out: } bool dcn20_validate_bandwidth_fp(struct dc *dc, struct dc_state *context, - bool fast_validate, display_e2e_pipe_params_st *pipes) + enum dc_validate_mode validate_mode, display_e2e_pipe_params_st *pipes) { bool voltage_supported = false; bool full_pstate_supported = false; @@ -2095,12 +2095,11 @@ bool dcn20_validate_bandwidth_fp(struct dc *dc, struct dc_state *context, /*Unsafe due to current pipe merge and split logic*/ ASSERT(context != dc->current_state); - if (fast_validate) { - return dcn20_validate_bandwidth_internal(dc, context, true, pipes); - } + if (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING) + return dcn20_validate_bandwidth_internal(dc, context, validate_mode, pipes); // Best case, we support full UCLK switch latency - voltage_supported = dcn20_validate_bandwidth_internal(dc, context, false, pipes); + voltage_supported = dcn20_validate_bandwidth_internal(dc, context, DC_VALIDATE_MODE_AND_PROGRAMMING, pipes); full_pstate_supported = context->bw_ctx.bw.dcn.clk.p_state_change_support; if (context->bw_ctx.dml.soc.dummy_pstate_latency_us == 0 || @@ -2113,7 +2112,7 @@ bool dcn20_validate_bandwidth_fp(struct dc *dc, struct dc_state *context, context->bw_ctx.dml.soc.dram_clock_change_latency_us = context->bw_ctx.dml.soc.dummy_pstate_latency_us; memset(pipes, 0, dc->res_pool->pipe_count * sizeof(display_e2e_pipe_params_st)); - voltage_supported = dcn20_validate_bandwidth_internal(dc, context, false, pipes); + voltage_supported = dcn20_validate_bandwidth_internal(dc, context, DC_VALIDATE_MODE_AND_PROGRAMMING, pipes); dummy_pstate_supported = context->bw_ctx.bw.dcn.clk.p_state_change_support; if (voltage_supported && (dummy_pstate_supported || !(context->stream_count))) { @@ -2156,14 +2155,14 @@ void dcn20_fpu_adjust_dppclk(struct vba_vars_st *v, int dcn21_populate_dml_pipes_from_context(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate) + enum dc_validate_mode validate_mode) { uint32_t pipe_cnt; int i; dc_assert_fp_enabled(); - pipe_cnt = dcn20_populate_dml_pipes_from_context(dc, context, pipes, fast_validate); + pipe_cnt = dcn20_populate_dml_pipes_from_context(dc, context, pipes, validate_mode); for (i = 0; i < pipe_cnt; i++) { @@ -2239,7 +2238,7 @@ static void dcn21_calculate_wm(struct dc *dc, struct dc_state *context, int *out_pipe_cnt, int *pipe_split_from, int vlevel_req, - bool fast_validate) + enum dc_validate_mode validate_mode) { int pipe_cnt, i, pipe_idx; int vlevel, vlevel_max; @@ -2281,10 +2280,10 @@ static void dcn21_calculate_wm(struct dc *dc, struct dc_state *context, if (pipe_cnt != pipe_idx) { if (dc->res_pool->funcs->populate_dml_pipes) pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, - context, pipes, fast_validate); + context, pipes, validate_mode); else pipe_cnt = dcn21_populate_dml_pipes_from_context(dc, - context, pipes, fast_validate); + context, pipes, validate_mode); } *out_pipe_cnt = pipe_cnt; @@ -2319,7 +2318,7 @@ static void dcn21_calculate_wm(struct dc *dc, struct dc_state *context, } bool dcn21_validate_bandwidth_fp(struct dc *dc, struct dc_state *context, - bool fast_validate, display_e2e_pipe_params_st *pipes) + enum dc_validate_mode validate_mode, display_e2e_pipe_params_st *pipes) { bool out = false; @@ -2337,7 +2336,7 @@ bool dcn21_validate_bandwidth_fp(struct dc *dc, struct dc_state *context, /*Unsafe due to current pipe merge and split logic*/ ASSERT(context != dc->current_state); - out = dcn21_fast_validate_bw(dc, context, pipes, &pipe_cnt, pipe_split_from, &vlevel, fast_validate); + out = dcn21_fast_validate_bw(dc, context, pipes, &pipe_cnt, pipe_split_from, &vlevel, validate_mode); if (pipe_cnt == 0) goto validate_out; @@ -2347,12 +2346,12 @@ bool dcn21_validate_bandwidth_fp(struct dc *dc, struct dc_state *context, BW_VAL_TRACE_END_VOLTAGE_LEVEL(); - if (fast_validate) { + if (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING) { BW_VAL_TRACE_SKIP(fast); goto validate_out; } - dcn21_calculate_wm(dc, context, pipes, &pipe_cnt, pipe_split_from, vlevel, fast_validate); + dcn21_calculate_wm(dc, context, pipes, &pipe_cnt, pipe_split_from, vlevel, validate_mode); dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel); BW_VAL_TRACE_END_WATERMARKS(); diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.h index b6c34198ddc8615cd88ef31c0158dc9c78ac6571..aed00039ca62510cb7ee6674332bdb3ee0509a14 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.h @@ -44,14 +44,14 @@ void dcn20_calculate_dlg_params(struct dc *dc, int dcn20_populate_dml_pipes_from_context(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate); + enum dc_validate_mode validate_mode); void dcn20_calculate_wm(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, int *out_pipe_cnt, int *pipe_split_from, int vlevel, - bool fast_validate); + enum dc_validate_mode validate_mode); void dcn20_cap_soc_clocks(struct _vcs_dpi_soc_bounding_box_st *bb, struct pp_smu_nv_clock_table max_clocks); void dcn20_update_bounding_box(struct dc *dc, @@ -62,7 +62,7 @@ void dcn20_update_bounding_box(struct dc *dc, void dcn20_patch_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_st *bb); bool dcn20_validate_bandwidth_fp(struct dc *dc, struct dc_state *context, - bool fast_validate, display_e2e_pipe_params_st *pipes); + enum dc_validate_mode validate_mode, display_e2e_pipe_params_st *pipes); void dcn20_fpu_set_wm_ranges(int i, struct pp_smu_wm_range_sets *ranges, struct _vcs_dpi_soc_bounding_box_st *loaded_bb); @@ -75,9 +75,9 @@ void dcn20_fpu_adjust_dppclk(struct vba_vars_st *v, int dcn21_populate_dml_pipes_from_context(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate); -bool dcn21_validate_bandwidth_fp(struct dc *dc, struct dc_state *context, bool - fast_validate, display_e2e_pipe_params_st *pipes); + enum dc_validate_mode validate_mode); +bool dcn21_validate_bandwidth_fp(struct dc *dc, struct dc_state *context, enum + dc_validate_mode, display_e2e_pipe_params_st *pipes); void dcn21_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params); void dcn21_clk_mgr_set_bw_params_wm_table(struct clk_bw_params *bw_params); diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c index 88789987bdbcb99497f5db4dde0b141e9f01e8f9..e5f5c0663750c0aeee2f702cd1322eb01e2b1007 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn30/dcn30_fpu.c @@ -339,7 +339,8 @@ void dcn30_fpu_calculate_wm_and_dlg( * newly found dummy_latency_index */ context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.pstate_latency_us; - dcn30_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, false, true); + dcn30_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, + DC_VALIDATE_MODE_AND_PROGRAMMING, true); maxMpcComb = context->bw_ctx.dml.vba.maxMpcComb; dcfclk = context->bw_ctx.dml.vba.DCFCLKState[vlevel][context->bw_ctx.dml.vba.maxMpcComb]; pstate_en = context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][maxMpcComb] != dm_dram_clock_change_unsupported; @@ -630,7 +631,8 @@ int dcn30_find_dummy_latency_index_for_fw_based_mclk_switch(struct dc *dc, while (dummy_latency_index < max_latency_table_entries) { context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->dummy_pstate_table[dummy_latency_index].dummy_pstate_latency_us; - dcn30_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, false, true); + dcn30_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, + DC_VALIDATE_MODE_AND_PROGRAMMING, true); if (context->bw_ctx.dml.soc.allow_dram_self_refresh_or_dram_clock_change_in_vblank == dm_allow_self_refresh_and_mclk_switch) diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.h index d2ae43a82ba568fe8d7913d88cf568c7a97f8ecc..dfcc5d50071e57cce21ebb19c9c13b153bff5fc1 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/dcn31_fpu.h @@ -55,5 +55,5 @@ int dcn_get_approx_det_segs_required_for_pstate( int dcn31x_populate_dml_pipes_from_context(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate); + enum dc_validate_mode validate_mode); #endif /* __DCN31_FPU_H__*/ diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c index 5ed117e11aa2ad2c2bc28c6955f2b06839e7c063..df9d50b9b57cd39170f059d69e152c75cf6d5819 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c @@ -306,7 +306,7 @@ static unsigned int get_vertical_back_porch(struct dc_crtc_timing *timing) int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate) + enum dc_validate_mode validate_mode) { int i, pipe_cnt; struct resource_context *res_ctx = &context->res_ctx; @@ -316,7 +316,7 @@ int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *c dc_assert_fp_enabled(); - dcn31x_populate_dml_pipes_from_context(dc, context, pipes, fast_validate); + dcn31x_populate_dml_pipes_from_context(dc, context, pipes, validate_mode); for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { struct dc_crtc_timing *timing; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.h index d32c5bb99f4c98907a945900b77d98c48c78a076..362ac79184ea2a728757b69fa9a83d14860cd62c 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.h @@ -35,6 +35,6 @@ void dcn314_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_params); int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate); + enum dc_validate_mode validate_mode); #endif diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c index b0fc1fd202084df430eefd2e7b33b2528d033ad4..6160952245b4332c3e015a9e08f0a93ad22856a8 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c @@ -290,7 +290,7 @@ int dcn32_find_dummy_latency_index_for_fw_based_mclk_switch(struct dc *dc, vba->DRAMClockChangeSupport[vlevel][context->bw_ctx.dml.vba.maxMpcComb] = temp_clock_change_support; context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->dummy_pstate_table[dummy_latency_index].dummy_pstate_latency_us; - dcn32_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, false); + dcn32_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, DC_VALIDATE_MODE_AND_PROGRAMMING); /* for subvp + DRR case, if subvp pipes are still present we support pstate */ if (vba->DRAMClockChangeSupport[vlevel][vba->maxMpcComb] == dm_dram_clock_change_unsupported && @@ -1479,7 +1479,7 @@ static bool dcn32_full_validate_bw_helper(struct dc *dc, /* Conditions for setting up phantom pipes for SubVP: * 1. Not force disable SubVP - * 2. Full update (i.e. !fast_validate) + * 2. Full update (i.e. DC_VALIDATE_MODE_AND_PROGRAMMING) * 3. Enough pipes are available to support SubVP (TODO: Which pipes will use VACTIVE / VBLANK / SUBVP?) * 4. Display configuration passes validation * 5. (Config doesn't support MCLK in VACTIVE/VBLANK || dc->debug.force_subvp_mclk_switch) @@ -1517,7 +1517,8 @@ static bool dcn32_full_validate_bw_helper(struct dc *dc, dc->res_pool->funcs->add_phantom_pipes(dc, context, pipes, *pipe_cnt, dc_pipe_idx); - *pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, false); + *pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, + DC_VALIDATE_MODE_AND_PROGRAMMING); // Populate dppclk to trigger a recalculate in dml_get_voltage_level // so the phantom pipe DLG params can be assigned correctly. pipes[0].clks_cfg.dppclk_mhz = get_dppclk_calculated(&context->bw_ctx.dml, pipes, *pipe_cnt, 0); @@ -1560,7 +1561,8 @@ static bool dcn32_full_validate_bw_helper(struct dc *dc, dc_state_remove_phantom_streams_and_planes(dc, context); dc_state_release_phantom_streams_and_planes(dc, context); vba->DRAMClockChangeSupport[*vlevel][vba->maxMpcComb] = dm_dram_clock_change_unsupported; - *pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, false); + *pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, + DC_VALIDATE_MODE_AND_PROGRAMMING); *vlevel = dml_get_voltage_level(&context->bw_ctx.dml, pipes, *pipe_cnt); /* This may adjust vlevel and maxMpcComb */ @@ -2138,7 +2140,7 @@ bool dcn32_internal_validate_bw(struct dc *dc, display_e2e_pipe_params_st *pipes, int *pipe_cnt_out, int *vlevel_out, - bool fast_validate) + enum dc_validate_mode validate_mode) { bool out = false; bool repopulate_pipes = false; @@ -2162,7 +2164,7 @@ bool dcn32_internal_validate_bw(struct dc *dc, for (i = 0; i < context->stream_count; i++) resource_update_pipes_for_stream_with_slice_count(context, dc->current_state, dc->res_pool, context->streams[i], 1); - pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, fast_validate); + pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, validate_mode); if (!pipe_cnt) { out = true; @@ -2172,13 +2174,13 @@ bool dcn32_internal_validate_bw(struct dc *dc, dml_log_pipe_params(&context->bw_ctx.dml, pipes, pipe_cnt); context->bw_ctx.dml.soc.max_vratio_pre = dcn32_determine_max_vratio_prefetch(dc, context); - if (!fast_validate) { + if (validate_mode == DC_VALIDATE_MODE_AND_PROGRAMMING) { if (!dcn32_full_validate_bw_helper(dc, context, pipes, &vlevel, split, merge, &pipe_cnt, &repopulate_pipes)) goto validate_fail; } - if (fast_validate || + if (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING || (dc->debug.dml_disallow_alternate_prefetch_modes && (vlevel == context->bw_ctx.dml.soc.num_states || vba->DRAMClockChangeSupport[vlevel][vba->maxMpcComb] == dm_dram_clock_change_unsupported))) { @@ -2195,7 +2197,7 @@ bool dcn32_internal_validate_bw(struct dc *dc, context->bw_ctx.dml.soc.allow_for_pstate_or_stutter_in_vblank_final = dm_prefetch_support_none; - context->bw_ctx.dml.validate_max_state = fast_validate; + context->bw_ctx.dml.validate_max_state = (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING); vlevel = dml_get_voltage_level(&context->bw_ctx.dml, pipes, pipe_cnt); context->bw_ctx.dml.validate_max_state = false; @@ -2247,7 +2249,7 @@ bool dcn32_internal_validate_bw(struct dc *dc, int flag_vlevel = vlevel; int i; - pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, fast_validate); + pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, validate_mode); if (!dc->config.enable_windowed_mpo_odm) dcn32_update_dml_pipes_odm_policy_based_on_context(dc, context, pipes); @@ -2343,7 +2345,7 @@ void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context, } context->bw_ctx.dml.soc.dram_clock_change_latency_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.pstate_latency_us; - dcn32_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, false); + dcn32_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, DC_VALIDATE_MODE_AND_PROGRAMMING); maxMpcComb = context->bw_ctx.dml.vba.maxMpcComb; if (is_subvp_p_drr) { context->bw_ctx.dml.vba.DRAMClockChangeSupport[vlevel][maxMpcComb] = dm_dram_clock_change_vblank_w_mall_sub_vp; @@ -2389,7 +2391,8 @@ void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context, context->bw_ctx.dml.soc.fclk_change_latency_us = dc->clk_mgr->bw_params->dummy_pstate_table[dummy_latency_index].dummy_pstate_latency_us; } - dcn32_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel_temp, false); + dcn32_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel_temp, + DC_VALIDATE_MODE_AND_PROGRAMMING); if (vlevel_temp < vlevel) { vlevel = vlevel_temp; maxMpcComb = context->bw_ctx.dml.vba.maxMpcComb; @@ -2410,7 +2413,8 @@ void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context, stream_status->fpo_in_use = false; } context->bw_ctx.dml.soc.fclk_change_latency_us = dc->clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.fclk_change_latency_us; - dcn32_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, false); + dcn32_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, + DC_VALIDATE_MODE_AND_PROGRAMMING); } } } diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h index 276e90e4e0cea12ca3334a46268e7db9aeac9e95..273d2bd79d85b904f8009676be708018e79eae99 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h @@ -49,7 +49,7 @@ bool dcn32_internal_validate_bw(struct dc *dc, display_e2e_pipe_params_st *pipes, int *pipe_cnt_out, int *vlevel_out, - bool fast_validate); + enum dc_validate_mode validate_mode); void dcn32_calculate_wm_and_dlg_fpu(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c index 92f0a099d089ac1c0b82ee8427cebb03871f1182..5d73efa2f0c9094c4a1bd4fe456115fbd3154aa5 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c @@ -437,7 +437,7 @@ static unsigned int get_vertical_back_porch(struct dc_crtc_timing *timing) int dcn35_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate) + enum dc_validate_mode validate_mode) { int i, pipe_cnt; struct resource_context *res_ctx = &context->res_ctx; @@ -446,7 +446,7 @@ int dcn35_populate_dml_pipes_from_context_fpu(struct dc *dc, const unsigned int max_allowed_vblank_nom = 1023; dcn31_populate_dml_pipes_from_context(dc, context, pipes, - fast_validate); + validate_mode); for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { struct dc_crtc_timing *timing; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.h index 067480fc3691322154f419e57479333be51b6b66..d121c5afce7186cee5f738d0ab1e14cae0019065 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.h @@ -37,7 +37,7 @@ void dcn35_update_bw_bounding_box_fpu(struct dc *dc, int dcn35_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate); + enum dc_validate_mode validate_mode); void dcn35_decide_zstate_support(struct dc *dc, struct dc_state *context); diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.c index 17d0b4923b0cc42835475b3889a4fa9d43c93b42..6f516af82956449183686e220e124f463d9de3c7 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.c @@ -470,7 +470,7 @@ static unsigned int get_vertical_back_porch(struct dc_crtc_timing *timing) int dcn351_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate) + enum dc_validate_mode validate_mode) { int i, pipe_cnt; struct resource_context *res_ctx = &context->res_ctx; @@ -479,7 +479,7 @@ int dcn351_populate_dml_pipes_from_context_fpu(struct dc *dc, const unsigned int max_allowed_vblank_nom = 1023; dcn31_populate_dml_pipes_from_context(dc, context, pipes, - fast_validate); + validate_mode); for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { struct dc_crtc_timing *timing; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.h index f93efab9a668439380b7c4149a251996c4a9cef8..f71d9d8d0759622d8706bd76dcafdf27aac11eab 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.h +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.h @@ -12,7 +12,7 @@ void dcn351_update_bw_bounding_box_fpu(struct dc *dc, int dcn351_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate); + enum dc_validate_mode validate_mode); void dcn351_decide_zstate_support(struct dc *dc, struct dc_state *context); diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn401/dcn401_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn401/dcn401_fpu.c deleted file mode 100644 index 4fbecb5ff3497d4e4ee665857b6d78ac663117bc..0000000000000000000000000000000000000000 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn401/dcn401_fpu.c +++ /dev/null @@ -1,239 +0,0 @@ -// SPDX-License-Identifier: MIT -// -// Copyright 2024 Advanced Micro Devices, Inc. - -#include "dcn401_fpu.h" -#include "dcn401/dcn401_resource.h" -// We need this includes for WATERMARKS_* defines -#include "clk_mgr/dcn401/dcn401_smu14_driver_if.h" -#include "link.h" - -#define DC_LOGGER_INIT(logger) - -void dcn401_build_wm_range_table_fpu(struct clk_mgr *clk_mgr) -{ - /* defaults */ - double pstate_latency_us = clk_mgr->ctx->dc->dml.soc.dram_clock_change_latency_us; - double fclk_change_latency_us = clk_mgr->ctx->dc->dml.soc.fclk_change_latency_us; - double sr_exit_time_us = clk_mgr->ctx->dc->dml.soc.sr_exit_time_us; - double sr_enter_plus_exit_time_us = clk_mgr->ctx->dc->dml.soc.sr_enter_plus_exit_time_us; - /* For min clocks use as reported by PM FW and report those as min */ - uint16_t min_uclk_mhz = clk_mgr->bw_params->clk_table.entries[0].memclk_mhz; - uint16_t min_dcfclk_mhz = clk_mgr->bw_params->clk_table.entries[0].dcfclk_mhz; - uint16_t setb_min_uclk_mhz = min_uclk_mhz; - uint16_t dcfclk_mhz_for_the_second_state = clk_mgr->ctx->dc->dml.soc.clock_limits[2].dcfclk_mhz; - - dc_assert_fp_enabled(); - - /* For Set B ranges use min clocks state 2 when available, and report those to PM FW */ - if (dcfclk_mhz_for_the_second_state) - clk_mgr->bw_params->wm_table.nv_entries[WM_B].pmfw_breakdown.min_dcfclk = dcfclk_mhz_for_the_second_state; - else - clk_mgr->bw_params->wm_table.nv_entries[WM_B].pmfw_breakdown.min_dcfclk = clk_mgr->bw_params->clk_table.entries[0].dcfclk_mhz; - - if (clk_mgr->bw_params->clk_table.entries[2].memclk_mhz) - setb_min_uclk_mhz = clk_mgr->bw_params->clk_table.entries[2].memclk_mhz; - - /* Set A - Normal - default values */ - clk_mgr->bw_params->wm_table.nv_entries[WM_A].valid = true; - clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.pstate_latency_us = pstate_latency_us; - clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.fclk_change_latency_us = fclk_change_latency_us; - clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.sr_exit_time_us = sr_exit_time_us; - clk_mgr->bw_params->wm_table.nv_entries[WM_A].dml_input.sr_enter_plus_exit_time_us = sr_enter_plus_exit_time_us; - clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.wm_type = WATERMARKS_CLOCK_RANGE; - clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.min_dcfclk = min_dcfclk_mhz; - clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.max_dcfclk = 0xFFFF; - clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.min_uclk = min_uclk_mhz; - clk_mgr->bw_params->wm_table.nv_entries[WM_A].pmfw_breakdown.max_uclk = 0xFFFF; - - /* Set B - Performance - higher clocks, using DPM[2] DCFCLK and UCLK */ - clk_mgr->bw_params->wm_table.nv_entries[WM_B].valid = true; - clk_mgr->bw_params->wm_table.nv_entries[WM_B].dml_input.pstate_latency_us = pstate_latency_us; - clk_mgr->bw_params->wm_table.nv_entries[WM_B].dml_input.fclk_change_latency_us = fclk_change_latency_us; - clk_mgr->bw_params->wm_table.nv_entries[WM_B].dml_input.sr_exit_time_us = sr_exit_time_us; - clk_mgr->bw_params->wm_table.nv_entries[WM_B].dml_input.sr_enter_plus_exit_time_us = sr_enter_plus_exit_time_us; - clk_mgr->bw_params->wm_table.nv_entries[WM_B].pmfw_breakdown.wm_type = WATERMARKS_CLOCK_RANGE; - clk_mgr->bw_params->wm_table.nv_entries[WM_B].pmfw_breakdown.max_dcfclk = 0xFFFF; - clk_mgr->bw_params->wm_table.nv_entries[WM_B].pmfw_breakdown.min_uclk = setb_min_uclk_mhz; - clk_mgr->bw_params->wm_table.nv_entries[WM_B].pmfw_breakdown.max_uclk = 0xFFFF; - - /* Set C - Dummy P-State - P-State latency set to "dummy p-state" value */ - /* 'DalDummyClockChangeLatencyNs' registry key option set to 0x7FFFFFFF can be used to disable Set C for dummy p-state */ - if (clk_mgr->ctx->dc->bb_overrides.dummy_clock_change_latency_ns != 0x7FFFFFFF) { - clk_mgr->bw_params->wm_table.nv_entries[WM_C].valid = true; - clk_mgr->bw_params->wm_table.nv_entries[WM_C].dml_input.pstate_latency_us = 50; - clk_mgr->bw_params->wm_table.nv_entries[WM_C].dml_input.fclk_change_latency_us = fclk_change_latency_us; - clk_mgr->bw_params->wm_table.nv_entries[WM_C].dml_input.sr_exit_time_us = sr_exit_time_us; - clk_mgr->bw_params->wm_table.nv_entries[WM_C].dml_input.sr_enter_plus_exit_time_us = sr_enter_plus_exit_time_us; - clk_mgr->bw_params->wm_table.nv_entries[WM_C].pmfw_breakdown.wm_type = WATERMARKS_DUMMY_PSTATE; - clk_mgr->bw_params->wm_table.nv_entries[WM_C].pmfw_breakdown.min_dcfclk = min_dcfclk_mhz; - clk_mgr->bw_params->wm_table.nv_entries[WM_C].pmfw_breakdown.max_dcfclk = 0xFFFF; - clk_mgr->bw_params->wm_table.nv_entries[WM_C].pmfw_breakdown.min_uclk = min_uclk_mhz; - clk_mgr->bw_params->wm_table.nv_entries[WM_C].pmfw_breakdown.max_uclk = 0xFFFF; - clk_mgr->bw_params->dummy_pstate_table[0].dram_speed_mts = clk_mgr->bw_params->clk_table.entries[0].memclk_mhz * 16; - clk_mgr->bw_params->dummy_pstate_table[0].dummy_pstate_latency_us = 50; - clk_mgr->bw_params->dummy_pstate_table[1].dram_speed_mts = clk_mgr->bw_params->clk_table.entries[1].memclk_mhz * 16; - clk_mgr->bw_params->dummy_pstate_table[1].dummy_pstate_latency_us = 9; - clk_mgr->bw_params->dummy_pstate_table[2].dram_speed_mts = clk_mgr->bw_params->clk_table.entries[2].memclk_mhz * 16; - clk_mgr->bw_params->dummy_pstate_table[2].dummy_pstate_latency_us = 8; - clk_mgr->bw_params->dummy_pstate_table[3].dram_speed_mts = clk_mgr->bw_params->clk_table.entries[3].memclk_mhz * 16; - clk_mgr->bw_params->dummy_pstate_table[3].dummy_pstate_latency_us = 5; - } - /* Set D - MALL - SR enter and exit time specific to MALL, TBD after bringup or later phase for now use DRAM values / 2 */ - /* For MALL DRAM clock change latency is N/A, for watermak calculations use lowest value dummy P state latency */ - clk_mgr->bw_params->wm_table.nv_entries[WM_D].valid = true; - clk_mgr->bw_params->wm_table.nv_entries[WM_D].dml_input.pstate_latency_us = clk_mgr->bw_params->dummy_pstate_table[3].dummy_pstate_latency_us; - clk_mgr->bw_params->wm_table.nv_entries[WM_D].dml_input.fclk_change_latency_us = fclk_change_latency_us; - clk_mgr->bw_params->wm_table.nv_entries[WM_D].dml_input.sr_exit_time_us = sr_exit_time_us / 2; // TBD - clk_mgr->bw_params->wm_table.nv_entries[WM_D].dml_input.sr_enter_plus_exit_time_us = sr_enter_plus_exit_time_us / 2; // TBD - clk_mgr->bw_params->wm_table.nv_entries[WM_D].pmfw_breakdown.wm_type = WATERMARKS_MALL; - clk_mgr->bw_params->wm_table.nv_entries[WM_D].pmfw_breakdown.min_dcfclk = min_dcfclk_mhz; - clk_mgr->bw_params->wm_table.nv_entries[WM_D].pmfw_breakdown.max_dcfclk = 0xFFFF; - clk_mgr->bw_params->wm_table.nv_entries[WM_D].pmfw_breakdown.min_uclk = min_uclk_mhz; - clk_mgr->bw_params->wm_table.nv_entries[WM_D].pmfw_breakdown.max_uclk = 0xFFFF; -} - -/* - * dcn401_update_bw_bounding_box - * - * This would override some dcn4_01 ip_or_soc initial parameters hardcoded from - * spreadsheet with actual values as per dGPU SKU: - * - with passed few options from dc->config - * - with dentist_vco_frequency from Clk Mgr (currently hardcoded, but might - * need to get it from PM FW) - * - with passed latency values (passed in ns units) in dc-> bb override for - * debugging purposes - * - with passed latencies from VBIOS (in 100_ns units) if available for - * certain dGPU SKU - * - with number of DRAM channels from VBIOS (which differ for certain dGPU SKU - * of the same ASIC) - * - clocks levels with passed clk_table entries from Clk Mgr as reported by PM - * FW for different clocks (which might differ for certain dGPU SKU of the - * same ASIC) - */ -void dcn401_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_params) -{ - dc_assert_fp_enabled(); - - /* Override from passed dc->bb_overrides if available*/ - if (dc->bb_overrides.sr_exit_time_ns) - dc->dml2_options.bbox_overrides.sr_exit_latency_us = - dc->bb_overrides.sr_exit_time_ns / 1000.0; - - if (dc->bb_overrides.sr_enter_plus_exit_time_ns) - dc->dml2_options.bbox_overrides.sr_enter_plus_exit_latency_us = - dc->bb_overrides.sr_enter_plus_exit_time_ns / 1000.0; - - if (dc->bb_overrides.urgent_latency_ns) - dc->dml2_options.bbox_overrides.urgent_latency_us = - dc->bb_overrides.urgent_latency_ns / 1000.0; - - if (dc->bb_overrides.dram_clock_change_latency_ns) - dc->dml2_options.bbox_overrides.dram_clock_change_latency_us = - dc->bb_overrides.dram_clock_change_latency_ns / 1000.0; - - if (dc->bb_overrides.fclk_clock_change_latency_ns) - dc->dml2_options.bbox_overrides.fclk_change_latency_us = - dc->bb_overrides.fclk_clock_change_latency_ns / 1000; - - /* Override from VBIOS if VBIOS bb_info available */ - if (dc->ctx->dc_bios->funcs->get_soc_bb_info) { - struct bp_soc_bb_info bb_info = {0}; - if (dc->ctx->dc_bios->funcs->get_soc_bb_info(dc->ctx->dc_bios, &bb_info) == BP_RESULT_OK) { - if (bb_info.dram_clock_change_latency_100ns > 0) - dc->dml2_options.bbox_overrides.dram_clock_change_latency_us = - bb_info.dram_clock_change_latency_100ns * 10; - - if (bb_info.dram_sr_enter_exit_latency_100ns > 0) - dc->dml2_options.bbox_overrides.sr_enter_plus_exit_latency_us = - bb_info.dram_sr_enter_exit_latency_100ns * 10; - - if (bb_info.dram_sr_exit_latency_100ns > 0) - dc->dml2_options.bbox_overrides.sr_exit_latency_us = - bb_info.dram_sr_exit_latency_100ns * 10; - } - } - - /* Override from VBIOS for num_chan */ - if (dc->ctx->dc_bios->vram_info.num_chans) { - dc->dml2_options.bbox_overrides.dram_num_chan = - dc->ctx->dc_bios->vram_info.num_chans; - - } - - if (dc->ctx->dc_bios->vram_info.dram_channel_width_bytes) - dc->dml2_options.bbox_overrides.dram_chanel_width_bytes = - dc->ctx->dc_bios->vram_info.dram_channel_width_bytes; - - dc->dml2_options.bbox_overrides.disp_pll_vco_speed_mhz = dc->clk_mgr->dentist_vco_freq_khz / 1000.0; - dc->dml2_options.bbox_overrides.xtalclk_mhz = dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency / 1000.0; - dc->dml2_options.bbox_overrides.dchub_refclk_mhz = dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000.0; - dc->dml2_options.bbox_overrides.dprefclk_mhz = dc->clk_mgr->dprefclk_khz / 1000.0; - - if (dc->clk_mgr->bw_params->clk_table.num_entries > 1) { - unsigned int i = 0; - - dc->dml2_options.bbox_overrides.clks_table.num_states = dc->clk_mgr->bw_params->clk_table.num_entries; - - dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_dcfclk_levels = - dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_dcfclk_levels; - - dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_fclk_levels = - dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_fclk_levels; - - dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_memclk_levels = - dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_memclk_levels; - - dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_socclk_levels = - dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_socclk_levels; - - dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_dtbclk_levels = - dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_dtbclk_levels; - - dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_dispclk_levels = - dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_dispclk_levels; - - dc->dml2_options.bbox_overrides.clks_table.num_entries_per_clk.num_dppclk_levels = - dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_dppclk_levels; - - for (i = 0; i < dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_dcfclk_levels; i++) { - if (dc->clk_mgr->bw_params->clk_table.entries[i].dcfclk_mhz) - dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].dcfclk_mhz = - dc->clk_mgr->bw_params->clk_table.entries[i].dcfclk_mhz; - } - - for (i = 0; i < dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_fclk_levels; i++) { - if (dc->clk_mgr->bw_params->clk_table.entries[i].fclk_mhz) - dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].fclk_mhz = - dc->clk_mgr->bw_params->clk_table.entries[i].fclk_mhz; - } - - for (i = 0; i < dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_memclk_levels; i++) { - if (dc->clk_mgr->bw_params->clk_table.entries[i].memclk_mhz) - dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].memclk_mhz = - dc->clk_mgr->bw_params->clk_table.entries[i].memclk_mhz; - } - - for (i = 0; i < dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_socclk_levels; i++) { - if (dc->clk_mgr->bw_params->clk_table.entries[i].socclk_mhz) - dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].socclk_mhz = - dc->clk_mgr->bw_params->clk_table.entries[i].socclk_mhz; - } - - for (i = 0; i < dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_dtbclk_levels; i++) { - if (dc->clk_mgr->bw_params->clk_table.entries[i].dtbclk_mhz) - dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].dtbclk_mhz = - dc->clk_mgr->bw_params->clk_table.entries[i].dtbclk_mhz; - } - - for (i = 0; i < dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_dispclk_levels; i++) { - if (dc->clk_mgr->bw_params->clk_table.entries[i].dispclk_mhz) { - dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].dispclk_mhz = - dc->clk_mgr->bw_params->clk_table.entries[i].dispclk_mhz; - dc->dml2_options.bbox_overrides.clks_table.clk_entries[i].dppclk_mhz = - dc->clk_mgr->bw_params->clk_table.entries[i].dispclk_mhz; - } - } - } -} - diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn401/dcn401_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn401/dcn401_fpu.h deleted file mode 100644 index 329f1788843c97555e2cf45a0063d474c94c1b5d..0000000000000000000000000000000000000000 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn401/dcn401_fpu.h +++ /dev/null @@ -1,14 +0,0 @@ -// SPDX-License-Identifier: MIT -// -// Copyright 2024 Advanced Micro Devices, Inc. - -#ifndef __DCN401_FPU_H__ -#define __DCN401_FPU_H__ - -#include "clk_mgr.h" - -void dcn401_build_wm_range_table_fpu(struct clk_mgr *clk_mgr); - -void dcn401_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_params); - -#endif diff --git a/drivers/gpu/drm/amd/display/dc/dml2/Makefile b/drivers/gpu/drm/amd/display/dc/dml2/Makefile index 157ecf008d6cdc7f9dd51278cca0bd2772fba725..4c21ce42054c5a5141eb4dfe7517796111527fbf 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dml2/Makefile @@ -81,10 +81,11 @@ AMD_DAL_DML2 = $(addprefix $(AMDDALPATH)/dc/dml2/,$(DML2)) AMD_DISPLAY_FILES += $(AMD_DAL_DML2) -CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_top/dml2_top_interfaces.o := $(dml2_ccflags) -CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_top/dml2_top_soc15.o := $(dml2_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4.o := $(dml2_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.o := $(dml2_ccflags) $(frame_warn_flag) +CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_core/dml2_core_utils.o := $(dml2_ccflags) $(frame_warn_flag) +CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_top/dml2_top_interfaces.o := $(dml2_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_top/dml2_top_soc15.o := $(dml2_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_core/dml2_core_factory.o := $(dml2_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_dcn4.o := $(dml2_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_factory.o := $(dml2_ccflags) @@ -94,17 +95,16 @@ CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_dcn3.o := $(dml2_ccflag CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_dcn4_fams2.o := $(dml2_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_factory.o := $(dml2_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_standalone_libraries/lib_float_math.o := $(dml2_ccflags) - - CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/src/dml21_wrapper.o := $(dml2_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/dml21_translation_helper.o := $(dml2_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml2/dml21/dml21_utils.o := $(dml2_ccflags) -CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_top/dml2_top_interfaces.o := $(dml2_rcflags) -CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_top/dml2_top_soc15.o := $(dml2_rcflags) CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4.o := $(dml2_rcflags) CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.o := $(dml2_rcflags) CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_core/dml2_core_factory.o := $(dml2_rcflags) +CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_core/dml2_core_utils.o := $(dml2_rcflags) +CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_top/dml2_top_interfaces.o := $(dml2_rcflags) +CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_top/dml2_top_soc15.o := $(dml2_rcflags) CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_dcn4.o := $(dml2_rcflags) CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_factory.o := $(dml2_rcflags) CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/src/dml2_mcg/dml2_mcg_dcn4.o := $(dml2_rcflags) @@ -120,6 +120,7 @@ CFLAGS_REMOVE_$(AMDDALPATH)/dc/dml2/dml21/dml21_utils.o := $(dml2_rcflags) DML21 := src/dml2_top/dml2_top_interfaces.o DML21 += src/dml2_top/dml2_top_soc15.o DML21 += src/dml2_core/dml2_core_dcn4.o +DML21 += src/dml2_core/dml2_core_utils.o DML21 += src/dml2_core/dml2_core_factory.o DML21 += src/dml2_core/dml2_core_dcn4_calcs.o DML21 += src/dml2_dpmm/dml2_dpmm_dcn4.o diff --git a/drivers/gpu/drm/amd/display/dc/dml2/display_mode_core.c b/drivers/gpu/drm/amd/display/dc/dml2/display_mode_core.c index 7ae9c0ba0c9e6aa22848591f209a39c147257ee3..715f9019a33e27faede7b7bd4ce6d55087c88606 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/display_mode_core.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/display_mode_core.c @@ -10189,7 +10189,7 @@ dml_uint_t dml_mode_support_ex(struct dml_mode_support_ex_params_st *in_out_para result = mode_support_pwr_states(&in_out_params->out_lowest_state_idx, in_out_params->mode_lib, in_out_params->in_display_cfg, - 0, + in_out_params->in_start_state_idx, in_out_params->mode_lib->states.num_states - 1); if (result) diff --git a/drivers/gpu/drm/amd/display/dc/dml2/display_mode_core_structs.h b/drivers/gpu/drm/amd/display/dc/dml2/display_mode_core_structs.h index 0670e4dc4fd91029aff4806563ac1d167e11d478..dbeb084660922fc448afbf537e52f23619103fad 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/display_mode_core_structs.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/display_mode_core_structs.h @@ -1917,6 +1917,7 @@ struct display_mode_lib_st { struct dml_mode_support_ex_params_st { struct display_mode_lib_st *mode_lib; const struct dml_display_cfg_st *in_display_cfg; + dml_uint_t in_start_state_idx; dml_uint_t out_lowest_state_idx; struct dml_mode_support_info_st *out_evaluation_info; }; diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c index d47cacfdb6955e8df3a4f6352fdc190c2ac0233b..a06217a9eef60a2ea14151ca4a68c13e073bfb9b 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c @@ -2,7 +2,6 @@ // // Copyright 2024 Advanced Micro Devices, Inc. - #include "dml21_wrapper.h" #include "dml2_core_dcn4_calcs.h" #include "dml2_internal_shared_types.h" @@ -11,277 +10,263 @@ #include "dml21_translation_helper.h" #include "bounding_boxes/dcn4_soc_bb.h" -static void dml21_init_socbb_params(struct dml2_initialize_instance_in_out *dml_init, - const struct dml2_configuration_options *config, - const struct dc *in_dc) -{ - const struct dml2_soc_bb *soc_bb; - const struct dml2_soc_qos_parameters *qos_params; - - switch (in_dc->ctx->dce_version) { - case DCN_VERSION_4_01: - default: - if (config->bb_from_dmub) - soc_bb = config->bb_from_dmub; - else - soc_bb = &dml2_socbb_dcn401; - - qos_params = &dml_dcn4_variant_a_soc_qos_params; - } - - /* patch soc bb */ - memcpy(&dml_init->soc_bb, soc_bb, sizeof(struct dml2_soc_bb)); - - /* patch qos params */ - memcpy(&dml_init->soc_bb.qos_parameters, qos_params, sizeof(struct dml2_soc_qos_parameters)); -} - -static void dml21_external_socbb_params(struct dml2_initialize_instance_in_out *dml_init, - const struct dml2_configuration_options *config) -{ - memcpy(&dml_init->soc_bb, &config->external_socbb_ip_params->soc_bb, sizeof(struct dml2_soc_bb)); -} - -static void dml21_external_ip_params(struct dml2_initialize_instance_in_out *dml_init, +static void dml21_populate_pmo_options(struct dml2_pmo_options *pmo_options, + const struct dc *in_dc, const struct dml2_configuration_options *config) { - memcpy(&dml_init->ip_caps, &config->external_socbb_ip_params->ip_params, sizeof(struct dml2_ip_capabilities)); + bool disable_fams2 = !in_dc->debug.fams2_config.bits.enable; + + /* ODM options */ + pmo_options->disable_dyn_odm = !config->minimize_dispclk_using_odm; + pmo_options->disable_dyn_odm_for_multi_stream = true; + pmo_options->disable_dyn_odm_for_stream_with_svp = true; + + pmo_options->disable_vblank = ((in_dc->debug.dml21_disable_pstate_method_mask >> 1) & 1); + + /* NOTE: DRR and SubVP Require FAMS2 */ + pmo_options->disable_svp = ((in_dc->debug.dml21_disable_pstate_method_mask >> 2) & 1) || + in_dc->debug.force_disable_subvp || + disable_fams2; + pmo_options->disable_drr_clamped = ((in_dc->debug.dml21_disable_pstate_method_mask >> 3) & 1) || + disable_fams2; + pmo_options->disable_drr_var = ((in_dc->debug.dml21_disable_pstate_method_mask >> 4) & 1) || + disable_fams2; + pmo_options->disable_fams2 = disable_fams2; + + pmo_options->disable_drr_var_when_var_active = in_dc->debug.disable_fams_gaming == INGAME_FAMS_DISABLE || + in_dc->debug.disable_fams_gaming == INGAME_FAMS_MULTI_DISP_CLAMPED_ONLY; + pmo_options->disable_drr_clamped_when_var_active = in_dc->debug.disable_fams_gaming == INGAME_FAMS_DISABLE; } -static void dml21_init_ip_params(struct dml2_initialize_instance_in_out *dml_init, +/* + * Populate dml_init based on default static values in soc bb. The default + * values are for reference and support at least minimal operation of current + * SoC and DCN hardware. The values could be modifed by subsequent override + * functions to reflect our true hardware capability. + */ +static void populate_default_dml_init_params(struct dml2_initialize_instance_in_out *dml_init, const struct dml2_configuration_options *config, const struct dc *in_dc) { - const struct dml2_ip_capabilities *ip_caps; - switch (in_dc->ctx->dce_version) { case DCN_VERSION_4_01: + dml_init->options.project_id = dml2_project_dcn4x_stage2_auto_drr_svp; + dml21_populate_pmo_options(&dml_init->options.pmo_options, in_dc, config); + dml_init->soc_bb = dml2_socbb_dcn401; + dml_init->soc_bb.qos_parameters = dml_dcn4_variant_a_soc_qos_params; + dml_init->ip_caps = dml2_dcn401_max_ip_caps; + break; default: - ip_caps = &dml2_dcn401_max_ip_caps; + memset(dml_init, 0, sizeof(*dml_init)); + DC_ERR("unsupported dcn version for DML21!"); + return; } - - memcpy(&dml_init->ip_caps, ip_caps, sizeof(struct dml2_ip_capabilities)); } -void dml21_initialize_soc_bb_params(struct dml2_initialize_instance_in_out *dml_init, +static void override_dml_init_with_values_from_hardware_default(struct dml2_initialize_instance_in_out *dml_init, const struct dml2_configuration_options *config, const struct dc *in_dc) { - if (config->use_native_soc_bb_construction) - dml21_init_socbb_params(dml_init, config, in_dc); - else - dml21_external_socbb_params(dml_init, config); + dml_init->soc_bb.dchub_refclk_mhz = in_dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000; + dml_init->soc_bb.dprefclk_mhz = in_dc->clk_mgr->dprefclk_khz / 1000; + dml_init->soc_bb.dispclk_dppclk_vco_speed_mhz = in_dc->clk_mgr->dentist_vco_freq_khz / 1000.0; } -void dml21_initialize_ip_params(struct dml2_initialize_instance_in_out *dml_init, +/* + * SMU stands for System Management Unit. It is a power management processor. + * It owns the initialization of dc's clock table and programming of clock values + * based on dc's requests. + * Our clock values in base soc bb is a dummy placeholder. The real clock values + * are retrieved from SMU firmware to dc clock table at runtime. + * This function overrides our dummy placeholder values with real values in dc + * clock table. + */ +static void override_dml_init_with_values_from_smu( + struct dml2_initialize_instance_in_out *dml_init, const struct dml2_configuration_options *config, const struct dc *in_dc) -{ - if (config->use_native_soc_bb_construction) - dml21_init_ip_params(dml_init, config, in_dc); - else - dml21_external_ip_params(dml_init, config); -} - -void dml21_apply_soc_bb_overrides(struct dml2_initialize_instance_in_out *dml_init, - const struct dml2_configuration_options *config, const struct dc *in_dc) { int i; - const struct clk_bw_params *dc_bw_params = in_dc->clk_mgr->bw_params; const struct clk_limit_table *dc_clk_table = &dc_bw_params->clk_table; - struct dml2_soc_bb *dml_soc_bb = &dml_init->soc_bb; - struct dml2_soc_state_table *dml_clk_table = &dml_soc_bb->clk_table; - - /* override clocks if smu is present */ - if (in_dc->clk_mgr->funcs->is_smu_present && in_dc->clk_mgr->funcs->is_smu_present(in_dc->clk_mgr)) { - /* dcfclk */ - if (dc_clk_table->num_entries_per_clk.num_dcfclk_levels) { - dml_clk_table->dcfclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dcfclk_levels; - for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { - if (i < dml_clk_table->dcfclk.num_clk_values) { - if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.dcfclk_mhz && - dc_clk_table->entries[i].dcfclk_mhz > dc_bw_params->dc_mode_limit.dcfclk_mhz) { - if (i == 0 || dc_clk_table->entries[i-1].dcfclk_mhz < dc_bw_params->dc_mode_limit.dcfclk_mhz) { - dml_clk_table->dcfclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dcfclk_mhz * 1000; - dml_clk_table->dcfclk.num_clk_values = i + 1; - } else { - dml_clk_table->dcfclk.clk_values_khz[i] = 0; - dml_clk_table->dcfclk.num_clk_values = i; - } + struct dml2_soc_state_table *dml_clk_table = &dml_init->soc_bb.clk_table; + + if (!in_dc->clk_mgr->funcs->is_smu_present || + !in_dc->clk_mgr->funcs->is_smu_present(in_dc->clk_mgr)) + /* skip if smu is not present */ + return; + + /* dcfclk */ + if (dc_clk_table->num_entries_per_clk.num_dcfclk_levels) { + dml_clk_table->dcfclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dcfclk_levels; + for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { + if (i < dml_clk_table->dcfclk.num_clk_values) { + if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.dcfclk_mhz && + dc_clk_table->entries[i].dcfclk_mhz > dc_bw_params->dc_mode_limit.dcfclk_mhz) { + if (i == 0 || dc_clk_table->entries[i-1].dcfclk_mhz < dc_bw_params->dc_mode_limit.dcfclk_mhz) { + dml_clk_table->dcfclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dcfclk_mhz * 1000; + dml_clk_table->dcfclk.num_clk_values = i + 1; } else { - dml_clk_table->dcfclk.clk_values_khz[i] = dc_clk_table->entries[i].dcfclk_mhz * 1000; + dml_clk_table->dcfclk.clk_values_khz[i] = 0; + dml_clk_table->dcfclk.num_clk_values = i; } } else { - dml_clk_table->dcfclk.clk_values_khz[i] = 0; + dml_clk_table->dcfclk.clk_values_khz[i] = dc_clk_table->entries[i].dcfclk_mhz * 1000; } + } else { + dml_clk_table->dcfclk.clk_values_khz[i] = 0; } } + } - /* fclk */ - if (dc_clk_table->num_entries_per_clk.num_fclk_levels) { - dml_clk_table->fclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_fclk_levels; - for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { - if (i < dml_clk_table->fclk.num_clk_values) { - if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.fclk_mhz && - dc_clk_table->entries[i].fclk_mhz > dc_bw_params->dc_mode_limit.fclk_mhz) { - if (i == 0 || dc_clk_table->entries[i-1].fclk_mhz < dc_bw_params->dc_mode_limit.fclk_mhz) { - dml_clk_table->fclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.fclk_mhz * 1000; - dml_clk_table->fclk.num_clk_values = i + 1; - } else { - dml_clk_table->fclk.clk_values_khz[i] = 0; - dml_clk_table->fclk.num_clk_values = i; - } + /* fclk */ + if (dc_clk_table->num_entries_per_clk.num_fclk_levels) { + dml_clk_table->fclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_fclk_levels; + for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { + if (i < dml_clk_table->fclk.num_clk_values) { + if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.fclk_mhz && + dc_clk_table->entries[i].fclk_mhz > dc_bw_params->dc_mode_limit.fclk_mhz) { + if (i == 0 || dc_clk_table->entries[i-1].fclk_mhz < dc_bw_params->dc_mode_limit.fclk_mhz) { + dml_clk_table->fclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.fclk_mhz * 1000; + dml_clk_table->fclk.num_clk_values = i + 1; } else { - dml_clk_table->fclk.clk_values_khz[i] = dc_clk_table->entries[i].fclk_mhz * 1000; + dml_clk_table->fclk.clk_values_khz[i] = 0; + dml_clk_table->fclk.num_clk_values = i; } } else { - dml_clk_table->fclk.clk_values_khz[i] = 0; + dml_clk_table->fclk.clk_values_khz[i] = dc_clk_table->entries[i].fclk_mhz * 1000; } + } else { + dml_clk_table->fclk.clk_values_khz[i] = 0; } } + } - /* uclk */ - if (dc_clk_table->num_entries_per_clk.num_memclk_levels) { - dml_clk_table->uclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_memclk_levels; - for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { - if (i < dml_clk_table->uclk.num_clk_values) { - if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.memclk_mhz && - dc_clk_table->entries[i].memclk_mhz > dc_bw_params->dc_mode_limit.memclk_mhz) { - if (i == 0 || dc_clk_table->entries[i-1].memclk_mhz < dc_bw_params->dc_mode_limit.memclk_mhz) { - dml_clk_table->uclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.memclk_mhz * 1000; - dml_clk_table->uclk.num_clk_values = i + 1; - } else { - dml_clk_table->uclk.clk_values_khz[i] = 0; - dml_clk_table->uclk.num_clk_values = i; - } + /* uclk */ + if (dc_clk_table->num_entries_per_clk.num_memclk_levels) { + dml_clk_table->uclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_memclk_levels; + for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { + if (i < dml_clk_table->uclk.num_clk_values) { + if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.memclk_mhz && + dc_clk_table->entries[i].memclk_mhz > dc_bw_params->dc_mode_limit.memclk_mhz) { + if (i == 0 || dc_clk_table->entries[i-1].memclk_mhz < dc_bw_params->dc_mode_limit.memclk_mhz) { + dml_clk_table->uclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.memclk_mhz * 1000; + dml_clk_table->uclk.num_clk_values = i + 1; } else { - dml_clk_table->uclk.clk_values_khz[i] = dc_clk_table->entries[i].memclk_mhz * 1000; + dml_clk_table->uclk.clk_values_khz[i] = 0; + dml_clk_table->uclk.num_clk_values = i; } } else { - dml_clk_table->uclk.clk_values_khz[i] = 0; + dml_clk_table->uclk.clk_values_khz[i] = dc_clk_table->entries[i].memclk_mhz * 1000; } + } else { + dml_clk_table->uclk.clk_values_khz[i] = 0; } } + } - /* dispclk */ - if (dc_clk_table->num_entries_per_clk.num_dispclk_levels) { - dml_clk_table->dispclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dispclk_levels; - for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { - if (i < dml_clk_table->dispclk.num_clk_values) { - if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.dispclk_mhz && - dc_clk_table->entries[i].dispclk_mhz > dc_bw_params->dc_mode_limit.dispclk_mhz) { - if (i == 0 || dc_clk_table->entries[i-1].dispclk_mhz < dc_bw_params->dc_mode_limit.dispclk_mhz) { - dml_clk_table->dispclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dispclk_mhz * 1000; - dml_clk_table->dispclk.num_clk_values = i + 1; - } else { - dml_clk_table->dispclk.clk_values_khz[i] = 0; - dml_clk_table->dispclk.num_clk_values = i; - } + /* dispclk */ + if (dc_clk_table->num_entries_per_clk.num_dispclk_levels) { + dml_clk_table->dispclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dispclk_levels; + for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { + if (i < dml_clk_table->dispclk.num_clk_values) { + if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.dispclk_mhz && + dc_clk_table->entries[i].dispclk_mhz > dc_bw_params->dc_mode_limit.dispclk_mhz) { + if (i == 0 || dc_clk_table->entries[i-1].dispclk_mhz < dc_bw_params->dc_mode_limit.dispclk_mhz) { + dml_clk_table->dispclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dispclk_mhz * 1000; + dml_clk_table->dispclk.num_clk_values = i + 1; } else { - dml_clk_table->dispclk.clk_values_khz[i] = dc_clk_table->entries[i].dispclk_mhz * 1000; + dml_clk_table->dispclk.clk_values_khz[i] = 0; + dml_clk_table->dispclk.num_clk_values = i; } } else { - dml_clk_table->dispclk.clk_values_khz[i] = 0; + dml_clk_table->dispclk.clk_values_khz[i] = dc_clk_table->entries[i].dispclk_mhz * 1000; } + } else { + dml_clk_table->dispclk.clk_values_khz[i] = 0; } } + } - /* dppclk */ - if (dc_clk_table->num_entries_per_clk.num_dppclk_levels) { - dml_clk_table->dppclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dppclk_levels; - for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { - if (i < dml_clk_table->dppclk.num_clk_values) { - if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.dppclk_mhz && - dc_clk_table->entries[i].dppclk_mhz > dc_bw_params->dc_mode_limit.dppclk_mhz) { - if (i == 0 || dc_clk_table->entries[i-1].dppclk_mhz < dc_bw_params->dc_mode_limit.dppclk_mhz) { - dml_clk_table->dppclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dppclk_mhz * 1000; - dml_clk_table->dppclk.num_clk_values = i + 1; - } else { - dml_clk_table->dppclk.clk_values_khz[i] = 0; - dml_clk_table->dppclk.num_clk_values = i; - } + /* dppclk */ + if (dc_clk_table->num_entries_per_clk.num_dppclk_levels) { + dml_clk_table->dppclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dppclk_levels; + for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { + if (i < dml_clk_table->dppclk.num_clk_values) { + if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.dppclk_mhz && + dc_clk_table->entries[i].dppclk_mhz > dc_bw_params->dc_mode_limit.dppclk_mhz) { + if (i == 0 || dc_clk_table->entries[i-1].dppclk_mhz < dc_bw_params->dc_mode_limit.dppclk_mhz) { + dml_clk_table->dppclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dppclk_mhz * 1000; + dml_clk_table->dppclk.num_clk_values = i + 1; } else { - dml_clk_table->dppclk.clk_values_khz[i] = dc_clk_table->entries[i].dppclk_mhz * 1000; + dml_clk_table->dppclk.clk_values_khz[i] = 0; + dml_clk_table->dppclk.num_clk_values = i; } } else { - dml_clk_table->dppclk.clk_values_khz[i] = 0; + dml_clk_table->dppclk.clk_values_khz[i] = dc_clk_table->entries[i].dppclk_mhz * 1000; } + } else { + dml_clk_table->dppclk.clk_values_khz[i] = 0; } } + } - /* dtbclk */ - if (dc_clk_table->num_entries_per_clk.num_dtbclk_levels) { - dml_clk_table->dtbclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dtbclk_levels; - for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { - if (i < dml_clk_table->dtbclk.num_clk_values) { - if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.dtbclk_mhz && - dc_clk_table->entries[i].dtbclk_mhz > dc_bw_params->dc_mode_limit.dtbclk_mhz) { - if (i == 0 || dc_clk_table->entries[i-1].dtbclk_mhz < dc_bw_params->dc_mode_limit.dtbclk_mhz) { - dml_clk_table->dtbclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dtbclk_mhz * 1000; - dml_clk_table->dtbclk.num_clk_values = i + 1; - } else { - dml_clk_table->dtbclk.clk_values_khz[i] = 0; - dml_clk_table->dtbclk.num_clk_values = i; - } + /* dtbclk */ + if (dc_clk_table->num_entries_per_clk.num_dtbclk_levels) { + dml_clk_table->dtbclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dtbclk_levels; + for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { + if (i < dml_clk_table->dtbclk.num_clk_values) { + if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.dtbclk_mhz && + dc_clk_table->entries[i].dtbclk_mhz > dc_bw_params->dc_mode_limit.dtbclk_mhz) { + if (i == 0 || dc_clk_table->entries[i-1].dtbclk_mhz < dc_bw_params->dc_mode_limit.dtbclk_mhz) { + dml_clk_table->dtbclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dtbclk_mhz * 1000; + dml_clk_table->dtbclk.num_clk_values = i + 1; } else { - dml_clk_table->dtbclk.clk_values_khz[i] = dc_clk_table->entries[i].dtbclk_mhz * 1000; + dml_clk_table->dtbclk.clk_values_khz[i] = 0; + dml_clk_table->dtbclk.num_clk_values = i; } } else { - dml_clk_table->dtbclk.clk_values_khz[i] = 0; + dml_clk_table->dtbclk.clk_values_khz[i] = dc_clk_table->entries[i].dtbclk_mhz * 1000; } + } else { + dml_clk_table->dtbclk.clk_values_khz[i] = 0; } } + } - /* socclk */ - if (dc_clk_table->num_entries_per_clk.num_socclk_levels) { - dml_clk_table->socclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_socclk_levels; - for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { - if (i < dml_clk_table->socclk.num_clk_values) { - if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.socclk_mhz && - dc_clk_table->entries[i].socclk_mhz > dc_bw_params->dc_mode_limit.socclk_mhz) { - if (i == 0 || dc_clk_table->entries[i-1].socclk_mhz < dc_bw_params->dc_mode_limit.socclk_mhz) { - dml_clk_table->socclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.socclk_mhz * 1000; - dml_clk_table->socclk.num_clk_values = i + 1; - } else { - dml_clk_table->socclk.clk_values_khz[i] = 0; - dml_clk_table->socclk.num_clk_values = i; - } + /* socclk */ + if (dc_clk_table->num_entries_per_clk.num_socclk_levels) { + dml_clk_table->socclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_socclk_levels; + for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) { + if (i < dml_clk_table->socclk.num_clk_values) { + if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.socclk_mhz && + dc_clk_table->entries[i].socclk_mhz > dc_bw_params->dc_mode_limit.socclk_mhz) { + if (i == 0 || dc_clk_table->entries[i-1].socclk_mhz < dc_bw_params->dc_mode_limit.socclk_mhz) { + dml_clk_table->socclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.socclk_mhz * 1000; + dml_clk_table->socclk.num_clk_values = i + 1; } else { - dml_clk_table->socclk.clk_values_khz[i] = dc_clk_table->entries[i].socclk_mhz * 1000; + dml_clk_table->socclk.clk_values_khz[i] = 0; + dml_clk_table->socclk.num_clk_values = i; } } else { - dml_clk_table->socclk.clk_values_khz[i] = 0; + dml_clk_table->socclk.clk_values_khz[i] = dc_clk_table->entries[i].socclk_mhz * 1000; } + } else { + dml_clk_table->socclk.clk_values_khz[i] = 0; } } - - /* do not override phyclks for now */ - /* phyclk */ - // dml_clk_table->phyclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_phyclk_levels; - // for (i = 0; i < DML_MAX_CLK_TABLE_SIZE; i++) { - // dml_clk_table->phyclk.clk_values_khz[i] = dc_clk_table->entries[i].phyclk_mhz * 1000; - // } - - /* phyclk_d18 */ - // dml_clk_table->phyclk_d18.num_clk_values = dc_clk_table->num_entries_per_clk.num_phyclk_d18_levels; - // for (i = 0; i < DML_MAX_CLK_TABLE_SIZE; i++) { - // dml_clk_table->phyclk_d18.clk_values_khz[i] = dc_clk_table->entries[i].phyclk_d18_mhz * 1000; - // } - - /* phyclk_d32 */ - // dml_clk_table->phyclk_d32.num_clk_values = dc_clk_table->num_entries_per_clk.num_phyclk_d32_levels; - // for (i = 0; i < DML_MAX_CLK_TABLE_SIZE; i++) { - // dml_clk_table->phyclk_d32.clk_values_khz[i] = dc_clk_table->entries[i].phyclk_d32_mhz * 1000; - // } } +} - dml_soc_bb->dchub_refclk_mhz = in_dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000; - dml_soc_bb->dprefclk_mhz = in_dc->clk_mgr->dprefclk_khz / 1000; - dml_soc_bb->xtalclk_mhz = in_dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency / 1000; - dml_soc_bb->dispclk_dppclk_vco_speed_mhz = in_dc->clk_mgr->dentist_vco_freq_khz / 1000.0; +static void override_dml_init_with_values_from_vbios( + struct dml2_initialize_instance_in_out *dml_init, + const struct dml2_configuration_options *config, + const struct dc *in_dc) +{ + const struct clk_bw_params *dc_bw_params = in_dc->clk_mgr->bw_params; + struct dml2_soc_bb *dml_soc_bb = &dml_init->soc_bb; + struct dml2_soc_state_table *dml_clk_table = &dml_init->soc_bb.clk_table; - /* override bounding box paramters from VBIOS */ if (in_dc->ctx->dc_bios->bb_info.dram_clock_change_latency_100ns > 0) dml_soc_bb->power_management_parameters.dram_clk_change_blackout_us = (in_dc->ctx->dc_bios->bb_info.dram_clock_change_latency_100ns + 9) / 10; @@ -308,32 +293,120 @@ void dml21_apply_soc_bb_overrides(struct dml2_initialize_instance_in_out *dml_in dml_clk_table->dram_config.channel_width_bytes = in_dc->ctx->dc_bios->vram_info.dram_channel_width_bytes; } - /* override bounding box paramters from DC config */ - if (in_dc->bb_overrides.sr_exit_time_ns) { - dml_soc_bb->power_management_parameters.stutter_exit_latency_us = - in_dc->bb_overrides.sr_exit_time_ns / 1000.0; + dml_init->soc_bb.xtalclk_mhz = in_dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency / 1000; +} + + +static void override_dml_init_with_values_from_dmub(struct dml2_initialize_instance_in_out *dml_init, + const struct dml2_configuration_options *config, + const struct dc *in_dc) +{ + /* + * TODO - There seems to be overlaps between the values overriden from + * dmub and vbios. Investigate and identify the values that DMUB needs + * to own. + */ +// const struct dmub_soc_bb_params *dmub_bb_params = +// (const struct dmub_soc_bb_params *)config->bb_from_dmub; + +// if (dmub_bb_params == NULL) +// return; + +// if (dmub_bb_params->dram_clk_change_blackout_ns > 0) +// dml_init->soc_bb.power_management_parameters.dram_clk_change_blackout_us = +// (double) dmub_bb_params->dram_clk_change_blackout_ns / 1000.0; +// if (dmub_bb_params->dram_clk_change_read_only_ns > 0) +// dml_init->soc_bb.power_management_parameters.dram_clk_change_read_only_us = +// (double) dmub_bb_params->dram_clk_change_read_only_ns / 1000.0; +// if (dmub_bb_params->dram_clk_change_write_only_ns > 0) +// dml_init->soc_bb.power_management_parameters.dram_clk_change_write_only_us = +// (double) dmub_bb_params->dram_clk_change_write_only_ns / 1000.0; +// if (dmub_bb_params->fclk_change_blackout_ns > 0) +// dml_init->soc_bb.power_management_parameters.fclk_change_blackout_us = +// (double) dmub_bb_params->fclk_change_blackout_ns / 1000.0; +// if (dmub_bb_params->g7_ppt_blackout_ns > 0) +// dml_init->soc_bb.power_management_parameters.g7_ppt_blackout_us = +// (double) dmub_bb_params->g7_ppt_blackout_ns / 1000.0; +// if (dmub_bb_params->stutter_enter_plus_exit_latency_ns > 0) +// dml_init->soc_bb.power_management_parameters.stutter_enter_plus_exit_latency_us = +// (double) dmub_bb_params->stutter_enter_plus_exit_latency_ns / 1000.0; +// if (dmub_bb_params->stutter_exit_latency_ns > 0) +// dml_init->soc_bb.power_management_parameters.stutter_exit_latency_us = +// (double) dmub_bb_params->stutter_exit_latency_ns / 1000.0; +// if (dmub_bb_params->z8_stutter_enter_plus_exit_latency_ns > 0) +// dml_init->soc_bb.power_management_parameters.z8_stutter_enter_plus_exit_latency_us = +// (double) dmub_bb_params->z8_stutter_enter_plus_exit_latency_ns / 1000.0; +// if (dmub_bb_params->z8_stutter_exit_latency_ns > 0) +// dml_init->soc_bb.power_management_parameters.z8_stutter_exit_latency_us = +// (double) dmub_bb_params->z8_stutter_exit_latency_ns / 1000.0; +// if (dmub_bb_params->z8_min_idle_time_ns > 0) +// dml_init->soc_bb.power_management_parameters.z8_min_idle_time = +// (double) dmub_bb_params->z8_min_idle_time_ns / 1000.0; +// #ifndef TRIM_DML2_DCN6B_IP_SENSITIVE +// if (dmub_bb_params->type_b_dram_clk_change_blackout_ns > 0) +// dml_init->soc_bb.power_management_parameters.lpddr5_dram_clk_change_blackout_us = +// (double) dmub_bb_params->type_b_dram_clk_change_blackout_ns / 1000.0; +// if (dmub_bb_params->type_b_ppt_blackout_ns > 0) +// dml_init->soc_bb.power_management_parameters.lpddr5_ppt_blackout_us = +// (double) dmub_bb_params->type_b_ppt_blackout_ns / 1000.0; +// #else +// if (dmub_bb_params->type_b_dram_clk_change_blackout_ns > 0) +// dml_init->soc_bb.power_management_parameters.type_b_dram_clk_change_blackout_us = +// (double) dmub_bb_params->type_b_dram_clk_change_blackout_ns / 1000.0; +// if (dmub_bb_params->type_b_ppt_blackout_ns > 0) +// dml_init->soc_bb.power_management_parameters.type_b_ppt_blackout_us = +// (double) dmub_bb_params->type_b_ppt_blackout_ns / 1000.0; +// #endif +// if (dmub_bb_params->vmin_limit_dispclk_khz > 0) +// dml_init->soc_bb.vmin_limit.dispclk_khz = dmub_bb_params->vmin_limit_dispclk_khz; +// if (dmub_bb_params->vmin_limit_dcfclk_khz > 0) +// dml_init->soc_bb.vmin_limit.dcfclk_khz = dmub_bb_params->vmin_limit_dcfclk_khz; +// if (dmub_bb_params->g7_temperature_read_blackout_ns > 0) +// dml_init->soc_bb.power_management_parameters.g7_temperature_read_blackout_us = +// (double) dmub_bb_params->g7_temperature_read_blackout_ns / 1000.0; +} + +static void override_dml_init_with_values_from_software_policy(struct dml2_initialize_instance_in_out *dml_init, + const struct dml2_configuration_options *config, + const struct dc *in_dc) +{ + if (!config->use_native_soc_bb_construction) { + dml_init->soc_bb = config->external_socbb_ip_params->soc_bb; + dml_init->ip_caps = config->external_socbb_ip_params->ip_params; } - if (in_dc->bb_overrides.sr_enter_plus_exit_time_ns) { - dml_soc_bb->power_management_parameters.stutter_enter_plus_exit_latency_us = + if (in_dc->bb_overrides.sr_exit_time_ns) + dml_init->soc_bb.power_management_parameters.stutter_exit_latency_us = + in_dc->bb_overrides.sr_exit_time_ns / 1000.0; + + if (in_dc->bb_overrides.sr_enter_plus_exit_time_ns) + dml_init->soc_bb.power_management_parameters.stutter_enter_plus_exit_latency_us = in_dc->bb_overrides.sr_enter_plus_exit_time_ns / 1000.0; - } - if (in_dc->bb_overrides.dram_clock_change_latency_ns) { - dml_soc_bb->power_management_parameters.dram_clk_change_blackout_us = + if (in_dc->bb_overrides.dram_clock_change_latency_ns) + dml_init->soc_bb.power_management_parameters.dram_clk_change_blackout_us = in_dc->bb_overrides.dram_clock_change_latency_ns / 1000.0; - } - if (in_dc->bb_overrides.fclk_clock_change_latency_ns) { - dml_soc_bb->power_management_parameters.fclk_change_blackout_us = + if (in_dc->bb_overrides.fclk_clock_change_latency_ns) + dml_init->soc_bb.power_management_parameters.fclk_change_blackout_us = in_dc->bb_overrides.fclk_clock_change_latency_ns / 1000.0; - } +} - //TODO - // if (in_dc->bb_overrides.dummy_clock_change_latency_ns) { - // dml_soc_bb->power_management_parameters.dram_clk_change_blackout_us = - // in_dc->bb_overrides.dram_clock_change_latency_ns / 1000.0; - // } +void dml21_populate_dml_init_params(struct dml2_initialize_instance_in_out *dml_init, + const struct dml2_configuration_options *config, + const struct dc *in_dc) +{ + populate_default_dml_init_params(dml_init, config, in_dc); + + override_dml_init_with_values_from_hardware_default(dml_init, config, in_dc); + + override_dml_init_with_values_from_smu(dml_init, config, in_dc); + + override_dml_init_with_values_from_vbios(dml_init, config, in_dc); + + override_dml_init_with_values_from_dmub(dml_init, config, in_dc); + + override_dml_init_with_values_from_software_policy(dml_init, config, in_dc); } static unsigned int calc_max_hardware_v_total(const struct dc_stream_state *stream) @@ -726,7 +799,6 @@ static void populate_dml21_surface_config_from_plane_state( switch (plane_state->tiling_info.gfxversion) { case DcGfxVersion7: case DcGfxVersion8: - // Placeholder for programming the array_mode break; case DcGfxVersion9: case DcGfxVersion10: @@ -788,6 +860,7 @@ static void populate_dml21_plane_config_from_plane_state(struct dml2_context *dm plane->pixel_format = dml2_420_10; break; case SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616: + case SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616: case SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616F: case SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616F: plane->pixel_format = dml2_444_64; @@ -888,10 +961,8 @@ static void populate_dml21_plane_config_from_plane_state(struct dml2_context *dm case DC_CM2_GPU_MEM_SIZE_171717: plane->tdlut.tdlut_width_mode = dml2_tdlut_width_17_cube; break; - case DC_CM2_GPU_MEM_SIZE_333333: - plane->tdlut.tdlut_width_mode = dml2_tdlut_width_33_cube; - break; case DC_CM2_GPU_MEM_SIZE_TRANSFORMED: + default: //plane->tdlut.tdlut_width_mode = dml2_tdlut_width_flatten; // dml2_tdlut_width_flatten undefined break; } diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.h index 73a013be1e48c52578ed8fabca4d76a467ce9217..9880d3e0398e2f8603c947b9819d18bb2d86ad46 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.h @@ -17,9 +17,7 @@ struct dml2_context; struct dml2_configuration_options; struct dml2_initialize_instance_in_out; -void dml21_apply_soc_bb_overrides(struct dml2_initialize_instance_in_out *dml_init, const struct dml2_configuration_options *config, const struct dc *in_dc); -void dml21_initialize_soc_bb_params(struct dml2_initialize_instance_in_out *dml_init, const struct dml2_configuration_options *config, const struct dc *in_dc); -void dml21_initialize_ip_params(struct dml2_initialize_instance_in_out *dml_init, const struct dml2_configuration_options *config, const struct dc *in_dc); +void dml21_populate_dml_init_params(struct dml2_initialize_instance_in_out *dml_init, const struct dml2_configuration_options *config, const struct dc *in_dc); bool dml21_map_dc_state_into_dml_display_cfg(const struct dc *in_dc, struct dc_state *context, struct dml2_context *dml_ctx); void dml21_copy_clocks_to_dc_state(struct dml2_context *in_ctx, struct dc_state *context); void dml21_extract_watermark_sets(const struct dc *in_dc, union dcn_watermark_set *watermarks, struct dml2_context *in_ctx); diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_utils.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_utils.c index 930e86cdb88a2fc81ca873cee3a8043a0fe25e98..ee721606b883962dc1775c92b06f1ecafb264250 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_utils.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_utils.c @@ -384,6 +384,7 @@ void dml21_build_fams2_programming(const struct dc *dc, /* reset fams2 data */ memset(&context->bw_ctx.bw.dcn.fams2_stream_base_params, 0, sizeof(union dmub_cmd_fams2_config) * DML2_MAX_PLANES); memset(&context->bw_ctx.bw.dcn.fams2_stream_sub_params, 0, sizeof(union dmub_cmd_fams2_config) * DML2_MAX_PLANES); + memset(&context->bw_ctx.bw.dcn.fams2_stream_sub_params_v2, 0, sizeof(union dmub_fams2_stream_static_sub_state_v2) * DML2_MAX_PLANES); memset(&context->bw_ctx.bw.dcn.fams2_global_config, 0, sizeof(struct dmub_cmd_fams2_global_config)); if (dml_ctx->v21.mode_programming.programming->fams2_required) { @@ -414,9 +415,16 @@ void dml21_build_fams2_programming(const struct dc *dc, memcpy(static_base_state, &dml_ctx->v21.mode_programming.programming->stream_programming[dml_stream_idx].fams2_base_params, sizeof(union dmub_cmd_fams2_config)); - memcpy(static_sub_state, - &dml_ctx->v21.mode_programming.programming->stream_programming[dml_stream_idx].fams2_sub_params, - sizeof(union dmub_cmd_fams2_config)); + + if (dc->debug.fams_version.major == 3) { + memcpy(&context->bw_ctx.bw.dcn.fams2_stream_sub_params_v2[num_fams2_streams], + &dml_ctx->v21.mode_programming.programming->stream_programming[dml_stream_idx].fams2_sub_params_v2, + sizeof(union dmub_fams2_stream_static_sub_state_v2)); + } else { + memcpy(static_sub_state, + &dml_ctx->v21.mode_programming.programming->stream_programming[dml_stream_idx].fams2_sub_params, + sizeof(union dmub_cmd_fams2_config)); + } switch (dc->debug.fams_version.minor) { case 1: diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c index 208d3651b6bafbbf9460ed0a30cc6cf19237d2c9..03de3cf06ae59ae79cce3e8a143894ebb9c522b5 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c @@ -2,8 +2,6 @@ // // Copyright 2024 Advanced Micro Devices, Inc. -#include - #include "dml2_internal_types.h" #include "dml_top.h" #include "dml2_core_dcn4_calcs.h" @@ -37,15 +35,11 @@ static bool dml21_allocate_memory(struct dml2_context **dml_ctx) return true; } -static void dml21_apply_debug_options(const struct dc *in_dc, struct dml2_context *dml_ctx, const struct dml2_configuration_options *config) +static void dml21_populate_configuration_options(const struct dc *in_dc, + struct dml2_context *dml_ctx, + const struct dml2_configuration_options *config) { - bool disable_fams2; - struct dml2_pmo_options *pmo_options = &dml_ctx->v21.dml_init.options.pmo_options; - - /* ODM options */ - pmo_options->disable_dyn_odm = !config->minimize_dispclk_using_odm; - pmo_options->disable_dyn_odm_for_multi_stream = true; - pmo_options->disable_dyn_odm_for_stream_with_svp = true; + dml_ctx->config = *config; /* UCLK P-State options */ if (in_dc->debug.dml21_force_pstate_method) { @@ -55,52 +49,20 @@ static void dml21_apply_debug_options(const struct dc *in_dc, struct dml2_contex } else { dml_ctx->config.pmo.force_pstate_method_enable = false; } - - pmo_options->disable_vblank = ((in_dc->debug.dml21_disable_pstate_method_mask >> 1) & 1); - - /* NOTE: DRR and SubVP Require FAMS2 */ - disable_fams2 = !in_dc->debug.fams2_config.bits.enable; - pmo_options->disable_svp = ((in_dc->debug.dml21_disable_pstate_method_mask >> 2) & 1) || - in_dc->debug.force_disable_subvp || - disable_fams2; - pmo_options->disable_drr_clamped = ((in_dc->debug.dml21_disable_pstate_method_mask >> 3) & 1) || - disable_fams2; - pmo_options->disable_drr_var = ((in_dc->debug.dml21_disable_pstate_method_mask >> 4) & 1) || - disable_fams2; - pmo_options->disable_fams2 = disable_fams2; - - pmo_options->disable_drr_var_when_var_active = in_dc->debug.disable_fams_gaming == INGAME_FAMS_DISABLE || - in_dc->debug.disable_fams_gaming == INGAME_FAMS_MULTI_DISP_CLAMPED_ONLY; - pmo_options->disable_drr_clamped_when_var_active = in_dc->debug.disable_fams_gaming == INGAME_FAMS_DISABLE; } -static void dml21_init(const struct dc *in_dc, struct dml2_context **dml_ctx, const struct dml2_configuration_options *config) +static void dml21_init(const struct dc *in_dc, struct dml2_context *dml_ctx, const struct dml2_configuration_options *config) { - switch (in_dc->ctx->dce_version) { - case DCN_VERSION_4_01: - (*dml_ctx)->v21.dml_init.options.project_id = dml2_project_dcn4x_stage2_auto_drr_svp; - break; - default: - (*dml_ctx)->v21.dml_init.options.project_id = dml2_project_invalid; - } - (*dml_ctx)->architecture = dml2_architecture_21; + dml_ctx->architecture = dml2_architecture_21; - /* Store configuration options */ - (*dml_ctx)->config = *config; + dml21_populate_configuration_options(in_dc, dml_ctx, config); DC_FP_START(); - /*Initialize SOCBB and DCNIP params */ - dml21_initialize_soc_bb_params(&(*dml_ctx)->v21.dml_init, config, in_dc); - dml21_initialize_ip_params(&(*dml_ctx)->v21.dml_init, config, in_dc); - dml21_apply_soc_bb_overrides(&(*dml_ctx)->v21.dml_init, config, in_dc); - - /* apply debug overrides */ - dml21_apply_debug_options(in_dc, *dml_ctx, config); + dml21_populate_dml_init_params(&dml_ctx->v21.dml_init, config, in_dc); - /*Initialize DML21 instance */ - dml2_initialize_instance(&(*dml_ctx)->v21.dml_init); + dml2_initialize_instance(&dml_ctx->v21.dml_init); DC_FP_END(); } @@ -111,7 +73,7 @@ bool dml21_create(const struct dc *in_dc, struct dml2_context **dml_ctx, const s if (!dml21_allocate_memory(dml_ctx)) return false; - dml21_init(in_dc, dml_ctx, config); + dml21_init(in_dc, *dml_ctx, config); return true; } @@ -328,12 +290,13 @@ static bool dml21_check_mode_support(const struct dc *in_dc, struct dc_state *co return true; } -bool dml21_validate(const struct dc *in_dc, struct dc_state *context, struct dml2_context *dml_ctx, bool fast_validate) +bool dml21_validate(const struct dc *in_dc, struct dc_state *context, struct dml2_context *dml_ctx, + enum dc_validate_mode validate_mode) { bool out = false; - /* Use dml_validate_only for fast_validate path */ - if (fast_validate) + /* Use dml21_check_mode_support for DC_VALIDATE_MODE_ONLY and DC_VALIDATE_MODE_AND_STATE_INDEX path */ + if (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING) out = dml21_check_mode_support(in_dc, context, dml_ctx); else out = dml21_mode_check_and_programming(in_dc, context, dml_ctx); @@ -496,7 +459,7 @@ bool dml21_create_copy(struct dml2_context **dst_dml_ctx, return true; } -void dml21_reinit(const struct dc *in_dc, struct dml2_context **dml_ctx, const struct dml2_configuration_options *config) +void dml21_reinit(const struct dc *in_dc, struct dml2_context *dml_ctx, const struct dml2_configuration_options *config) { dml21_init(in_dc, dml_ctx, config); } diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.h index 42e715024bc91842f29d53b92ec042ed163ee22b..15f92029d2e59dd1c789d2fe1dee7e1715b29f23 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.h @@ -14,6 +14,7 @@ struct dc; struct dc_state; struct dml2_configuration_options; struct dml2_context; +enum dc_validate_mode; /** * dml2_create - Creates dml21_context. @@ -33,22 +34,23 @@ void dml21_copy(struct dml2_context *dst_dml_ctx, struct dml2_context *src_dml_ctx); bool dml21_create_copy(struct dml2_context **dst_dml_ctx, struct dml2_context *src_dml_ctx); -void dml21_reinit(const struct dc *in_dc, struct dml2_context **dml_ctx, const struct dml2_configuration_options *config); +void dml21_reinit(const struct dc *in_dc, struct dml2_context *dml_ctx, const struct dml2_configuration_options *config); /** * dml21_validate - Determines if a display configuration is supported or not. * @in_dc: dc. * @context: dc_state to be validated. - * @fast_validate: Fast validate will not populate context.res_ctx. + * @validate_mode: DC_VALIDATE_MODE_ONLY and DC_VALIDATE_MODE_AND_STATE_INDEX + * will not populate context.res_ctx. * * Based on fast_validate option internally would call: * - * -dml21_mode_check_and_programming - for non fast_validate option + * -dml21_mode_check_and_programming - for DC_VALIDATE_MODE_AND_PROGRAMMING option * Calculates if dc_state can be supported on the input display * configuration. If supported, generates the necessary HW * programming for the new dc_state. * - * -dml21_check_mode_support - for fast_validate option + * -dml21_check_mode_support - for DC_VALIDATE_MODE_ONLY and DC_VALIDATE_MODE_AND_STATE_INDEX option * Calculates if dc_state can be supported for the input display * config. @@ -56,7 +58,8 @@ void dml21_reinit(const struct dc *in_dc, struct dml2_context **dml_ctx, const s * separate dc_states for validation. * Return: True if mode is supported, false otherwise. */ -bool dml21_validate(const struct dc *in_dc, struct dc_state *context, struct dml2_context *dml_ctx, bool fast_validate); +bool dml21_validate(const struct dc *in_dc, struct dc_state *context, struct dml2_context *dml_ctx, + enum dc_validate_mode validate_mode); /* Prepare hubp mcache_regs for hubp mcache ID and split coordinate programming */ void dml21_prepare_mcache_programming(struct dc *in_dc, struct dc_state *context, struct dml2_context *dml_ctx); diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top.h index c047d56527c47a19d5038ca8fd5320b1f365563e..a64ec4dcf11abe141be34b3da1f9e528d0b836ea 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top.h @@ -43,5 +43,4 @@ bool dml2_build_mode_programming(struct dml2_build_mode_programming_in_out *in_o */ bool dml2_build_mcache_programming(struct dml2_build_mcache_programming_in_out *in_out); - #endif diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_dchub_registers.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_dchub_registers.h index 84c90050668c1921991ce5a4b797469d40ccc228..b05030926ce854575f902c2442b3a49087de716c 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_dchub_registers.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_dchub_registers.h @@ -46,6 +46,7 @@ struct dml2_display_dlg_regs { uint32_t dst_y_delta_drq_limit; uint32_t refcyc_per_vm_dmdata; uint32_t dmdata_dl_delta; + uint32_t dst_y_svp_drq_limit; // MRQ uint32_t refcyc_per_meta_chunk_vblank_l; diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_display_cfg_types.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_display_cfg_types.h index 255f05de362cf959bdc8f13b6fce109a750f7e5f..e8dc6471c0becbb10c9b5b7563cda25cde24273c 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_display_cfg_types.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_display_cfg_types.h @@ -222,6 +222,7 @@ struct dml2_composition_cfg { struct { bool enabled; + bool upsp_enabled; struct { double h_ratio; double v_ratio; @@ -426,6 +427,7 @@ struct dml2_stream_parameters { struct dml2_display_cfg { bool gpuvm_enable; + bool ffbm_enable; bool hostvm_enable; // Allocate DET proportionally between streams based on pixel rate diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_soc_parameter_types.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_soc_parameter_types.h index 5f0bc42d1d2f7757d235ae7b504d34f21e7746f8..8c9f414aa6bf9c3e456ae6708ca93f55d2345ad5 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_soc_parameter_types.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_soc_parameter_types.h @@ -93,12 +93,15 @@ struct dml2_soc_power_management_parameters { double dram_clk_change_write_only_us; double fclk_change_blackout_us; double g7_ppt_blackout_us; + double g7_temperature_read_blackout_us; double stutter_enter_plus_exit_latency_us; double stutter_exit_latency_us; double z8_stutter_enter_plus_exit_latency_us; double z8_stutter_exit_latency_us; double z8_min_idle_time; double g6_temp_read_blackout_us[DML_MAX_CLK_TABLE_SIZE]; + double type_b_dram_clk_change_blackout_us; + double type_b_ppt_blackout_us; }; struct dml2_clk_table { @@ -130,6 +133,7 @@ struct dml2_soc_state_table { struct dml2_soc_vmin_clock_limits { unsigned long dispclk_khz; + unsigned long dcfclk_khz; }; struct dml2_soc_bb { @@ -138,6 +142,7 @@ struct dml2_soc_bb { struct dml2_soc_power_management_parameters power_management_parameters; struct dml2_soc_vmin_clock_limits vmin_limit; + double lower_bound_bandwidth_dchub; unsigned int dprefclk_mhz; unsigned int xtalclk_mhz; unsigned int pcie_refclk_mhz; diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_types.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_types.h index 0dbf886d8926100a7f65e4efbb75915548848f81..98c0234e2f474e004d3eeaa739f21613ee0d98d8 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_types.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_types.h @@ -53,7 +53,9 @@ enum dml2_output_type_and_rate__rate { dml2_output_rate_hdmi_rate_6x4 = 9, dml2_output_rate_hdmi_rate_8x4 = 10, dml2_output_rate_hdmi_rate_10x4 = 11, - dml2_output_rate_hdmi_rate_12x4 = 12 + dml2_output_rate_hdmi_rate_12x4 = 12, + dml2_output_rate_hdmi_rate_16x4 = 13, + dml2_output_rate_hdmi_rate_20x4 = 14 }; struct dml2_pmo_options { @@ -279,7 +281,10 @@ struct dml2_per_stream_programming { } phantom_stream; union dmub_cmd_fams2_config fams2_base_params; - union dmub_cmd_fams2_config fams2_sub_params; + union { + union dmub_cmd_fams2_config fams2_sub_params; + union dmub_fams2_stream_static_sub_state_v2 fams2_sub_params_v2; + }; }; //----------------- @@ -674,9 +679,14 @@ struct dml2_display_cfg_programming { // unlimited # of mcache struct dml2_mcache_surface_allocation non_optimized_mcache_allocation[DML2_MAX_PLANES]; + bool failed_prefetch; + bool failed_uclk_pstate; bool failed_mcache_validation; bool failed_dpmm; bool failed_mode_programming; + bool failed_mode_programming_dcfclk; + bool failed_mode_programming_prefetch; + bool failed_mode_programming_flip; bool failed_map_watermarks; } informative; }; diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.c index c4dad7164d31432a72ba03c93d04937e6075424a..b9cff2198511091c084efeeb6bcfba1ac88b0e93 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.c @@ -4685,7 +4685,10 @@ static void calculate_tdlut_setting( //the tdlut is fetched during the 2 row times of prefetch. if (p->setup_for_tdlut) { *p->tdlut_groups_per_2row_ub = (unsigned int)math_ceil2((double) *p->tdlut_bytes_per_frame / *p->tdlut_bytes_per_group, 1); - *p->tdlut_opt_time = (*p->tdlut_bytes_per_frame - p->cursor_buffer_size * 1024) / tdlut_drain_rate; + if (*p->tdlut_bytes_per_frame > p->cursor_buffer_size * 1024) + *p->tdlut_opt_time = (*p->tdlut_bytes_per_frame - p->cursor_buffer_size * 1024) / tdlut_drain_rate; + else + *p->tdlut_opt_time = 0; *p->tdlut_drain_time = p->cursor_buffer_size * 1024 / tdlut_drain_rate; *p->tdlut_bytes_to_deliver = (unsigned int) (p->cursor_buffer_size * 1024.0); } @@ -4858,7 +4861,7 @@ static double get_urgent_bandwidth_required( double ReadBandwidthChroma[], double PrefetchBandwidthLuma[], double PrefetchBandwidthChroma[], - double PrefetchBandwidthOto[], + double PrefetchBandwidthMax[], double excess_vactive_fill_bw_l[], double excess_vactive_fill_bw_c[], double cursor_bw[], @@ -4922,9 +4925,9 @@ static double get_urgent_bandwidth_required( l->vm_row_bw = NumberOfDPP[k] * prefetch_vmrow_bw[k]; l->flip_and_active_bw = l->per_plane_flip_bw[k] + ReadBandwidthLuma[k] * l->adj_factor_p0 + ReadBandwidthChroma[k] * l->adj_factor_p1 + cursor_bw[k] * l->adj_factor_cur; l->flip_and_prefetch_bw = l->per_plane_flip_bw[k] + NumberOfDPP[k] * (PrefetchBandwidthLuma[k] * l->adj_factor_p0_pre + PrefetchBandwidthChroma[k] * l->adj_factor_p1_pre) + prefetch_cursor_bw[k] * l->adj_factor_cur_pre; - l->flip_and_prefetch_bw_oto = l->per_plane_flip_bw[k] + NumberOfDPP[k] * (PrefetchBandwidthOto[k] * l->adj_factor_p0_pre + PrefetchBandwidthChroma[k] * l->adj_factor_p1_pre) + prefetch_cursor_bw[k] * l->adj_factor_cur_pre; + l->flip_and_prefetch_bw_max = l->per_plane_flip_bw[k] + NumberOfDPP[k] * (PrefetchBandwidthMax[k] * l->adj_factor_p0_pre + PrefetchBandwidthChroma[k] * l->adj_factor_p1_pre) + prefetch_cursor_bw[k] * l->adj_factor_cur_pre; l->active_and_excess_bw = (ReadBandwidthLuma[k] + excess_vactive_fill_bw_l[k]) * l->tmp_nom_adj_factor_p0 + (ReadBandwidthChroma[k] + excess_vactive_fill_bw_c[k]) * l->tmp_nom_adj_factor_p1 + dpte_row_bw[k] + meta_row_bw[k]; - surface_required_bw[k] = math_max5(l->vm_row_bw, l->flip_and_active_bw, l->flip_and_prefetch_bw, l->active_and_excess_bw, l->flip_and_prefetch_bw_oto); + surface_required_bw[k] = math_max5(l->vm_row_bw, l->flip_and_active_bw, l->flip_and_prefetch_bw, l->active_and_excess_bw, l->flip_and_prefetch_bw_max); /* export peak required bandwidth for the surface */ surface_peak_required_bw[k] = math_max2(surface_required_bw[k], surface_peak_required_bw[k]); @@ -5122,7 +5125,7 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch s->Tsw_est3 = 0.0; s->cursor_prefetch_bytes = 0; *p->prefetch_cursor_bw = 0; - *p->RequiredPrefetchBWOTO = 0.0; + *p->RequiredPrefetchBWMax = 0.0; dcc_mrq_enable = (p->dcc_enable && p->mrq_present); @@ -5353,7 +5356,7 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch * mp will fail if ms decides to use equ schedule and mp decides to use oto schedule * and the required bandwidth increases when going from ms to mp */ - *p->RequiredPrefetchBWOTO = s->prefetch_bw_oto; + *p->RequiredPrefetchBWMax = s->prefetch_bw_oto; #ifdef __DML_VBA_DEBUG__ DML_LOG_VERBOSE("DML::%s: vactive_sw_bw_l = %f\n", __func__, p->vactive_sw_bw_l); @@ -5715,8 +5718,14 @@ static bool CalculatePrefetchSchedule(struct dml2_core_internal_scratch *scratch s->TimeForFetchingVM = s->Tvm_equ; s->TimeForFetchingRowInVBlank = s->Tr0_equ; - *p->dst_y_per_vm_vblank = math_ceil2(4.0 * s->TimeForFetchingVM / s->LineTime, 1.0) / 4.0; - *p->dst_y_per_row_vblank = math_ceil2(4.0 * s->TimeForFetchingRowInVBlank / s->LineTime, 1.0) / 4.0; + *p->dst_y_per_vm_vblank = math_ceil2(4.0 * s->TimeForFetchingVM / s->LineTime, 1.0) / 4.0; + *p->dst_y_per_row_vblank = math_ceil2(4.0 * s->TimeForFetchingRowInVBlank / s->LineTime, 1.0) / 4.0; + + /* equ bw should be propagated so a ceiling of the equ bw is accounted for prior to mode programming. + * Overall bandwidth may be lower when going from mode support to mode programming but final pixel data + * bandwidth may end up higher than what was calculated in mode support. + */ + *p->RequiredPrefetchBWMax = math_max2(s->prefetch_bw_equ, *p->RequiredPrefetchBWMax); #ifdef __DML_VBA_DEBUG__ DML_LOG_VERBOSE("DML::%s: Using equ bw scheduling for prefetch\n", __func__); @@ -6112,7 +6121,7 @@ static void calculate_peak_bandwidth_required( p->surface_read_bandwidth_c, l->zero_array, //PrefetchBandwidthLuma, l->zero_array, //PrefetchBandwidthChroma, - l->zero_array, //PrefetchBWOTO + l->zero_array, //PrefetchBWMax l->zero_array, l->zero_array, l->zero_array, @@ -6149,7 +6158,7 @@ static void calculate_peak_bandwidth_required( p->surface_read_bandwidth_c, l->zero_array, //PrefetchBandwidthLuma, l->zero_array, //PrefetchBandwidthChroma, - l->zero_array, //PrefetchBWOTO + l->zero_array, //PrefetchBWMax p->excess_vactive_fill_bw_l, p->excess_vactive_fill_bw_c, p->cursor_bw, @@ -6186,7 +6195,7 @@ static void calculate_peak_bandwidth_required( p->surface_read_bandwidth_c, p->prefetch_bandwidth_l, p->prefetch_bandwidth_c, - p->prefetch_bandwidth_oto, // to prevent ms/mp mismatch when oto bw > total vactive bw + p->prefetch_bandwidth_max, // to prevent ms/mp mismatches where mp prefetch bw > ms prefetch bw p->excess_vactive_fill_bw_l, p->excess_vactive_fill_bw_c, p->cursor_bw, @@ -6223,7 +6232,7 @@ static void calculate_peak_bandwidth_required( p->surface_read_bandwidth_c, p->prefetch_bandwidth_l, p->prefetch_bandwidth_c, - p->prefetch_bandwidth_oto, // to prevent ms/mp mismatch when oto bw > total vactive bw + p->prefetch_bandwidth_max, // to prevent ms/mp mismatch where mp prefetch bw > ms prefetch bw p->excess_vactive_fill_bw_l, p->excess_vactive_fill_bw_c, p->cursor_bw, @@ -6260,7 +6269,7 @@ static void calculate_peak_bandwidth_required( p->surface_read_bandwidth_c, p->prefetch_bandwidth_l, p->prefetch_bandwidth_c, - p->prefetch_bandwidth_oto, // to prevent ms/mp mismatch when oto bw > total vactive bw + p->prefetch_bandwidth_max, // to prevent ms/mp mismatches where mp prefetch bw > ms prefetch bw p->excess_vactive_fill_bw_l, p->excess_vactive_fill_bw_c, p->cursor_bw, @@ -7487,7 +7496,7 @@ static noinline_for_stack void dml_core_ms_prefetch_check(struct dml2_core_inter CalculatePrefetchSchedule_params->VRatioPrefetchC = &mode_lib->ms.VRatioPreC[k]; CalculatePrefetchSchedule_params->RequiredPrefetchPixelDataBWLuma = &mode_lib->ms.RequiredPrefetchPixelDataBWLuma[k]; // prefetch_sw_bw_l CalculatePrefetchSchedule_params->RequiredPrefetchPixelDataBWChroma = &mode_lib->ms.RequiredPrefetchPixelDataBWChroma[k]; // prefetch_sw_bw_c - CalculatePrefetchSchedule_params->RequiredPrefetchBWOTO = &mode_lib->ms.RequiredPrefetchBWOTO[k]; + CalculatePrefetchSchedule_params->RequiredPrefetchBWMax = &mode_lib->ms.RequiredPrefetchBWMax[k]; CalculatePrefetchSchedule_params->NotEnoughTimeForDynamicMetadata = &mode_lib->ms.NoTimeForDynamicMetadata[k]; CalculatePrefetchSchedule_params->Tno_bw = &mode_lib->ms.Tno_bw[k]; CalculatePrefetchSchedule_params->Tno_bw_flip = &mode_lib->ms.Tno_bw_flip[k]; @@ -7632,7 +7641,7 @@ static noinline_for_stack void dml_core_ms_prefetch_check(struct dml2_core_inter calculate_peak_bandwidth_params->surface_read_bandwidth_c = mode_lib->ms.vactive_sw_bw_c; calculate_peak_bandwidth_params->prefetch_bandwidth_l = mode_lib->ms.RequiredPrefetchPixelDataBWLuma; calculate_peak_bandwidth_params->prefetch_bandwidth_c = mode_lib->ms.RequiredPrefetchPixelDataBWChroma; - calculate_peak_bandwidth_params->prefetch_bandwidth_oto = mode_lib->ms.RequiredPrefetchBWOTO; + calculate_peak_bandwidth_params->prefetch_bandwidth_max = mode_lib->ms.RequiredPrefetchBWMax; calculate_peak_bandwidth_params->excess_vactive_fill_bw_l = mode_lib->ms.excess_vactive_fill_bw_l; calculate_peak_bandwidth_params->excess_vactive_fill_bw_c = mode_lib->ms.excess_vactive_fill_bw_c; calculate_peak_bandwidth_params->cursor_bw = mode_lib->ms.cursor_bw; @@ -7799,7 +7808,7 @@ static noinline_for_stack void dml_core_ms_prefetch_check(struct dml2_core_inter calculate_peak_bandwidth_params->surface_read_bandwidth_c = mode_lib->ms.vactive_sw_bw_c; calculate_peak_bandwidth_params->prefetch_bandwidth_l = mode_lib->ms.RequiredPrefetchPixelDataBWLuma; calculate_peak_bandwidth_params->prefetch_bandwidth_c = mode_lib->ms.RequiredPrefetchPixelDataBWChroma; - calculate_peak_bandwidth_params->prefetch_bandwidth_oto = mode_lib->ms.RequiredPrefetchBWOTO; + calculate_peak_bandwidth_params->prefetch_bandwidth_max = mode_lib->ms.RequiredPrefetchBWMax; calculate_peak_bandwidth_params->excess_vactive_fill_bw_l = mode_lib->ms.excess_vactive_fill_bw_l; calculate_peak_bandwidth_params->excess_vactive_fill_bw_c = mode_lib->ms.excess_vactive_fill_bw_c; calculate_peak_bandwidth_params->cursor_bw = mode_lib->ms.cursor_bw; @@ -7905,6 +7914,7 @@ static noinline_for_stack void dml_core_ms_prefetch_check(struct dml2_core_inter } + static bool dml_core_mode_support(struct dml2_core_calcs_mode_support_ex *in_out_params) { struct dml2_core_internal_display_mode_lib *mode_lib = in_out_params->mode_lib; @@ -11253,7 +11263,7 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex CalculatePrefetchSchedule_params->VRatioPrefetchC = &mode_lib->mp.VRatioPrefetchC[k]; CalculatePrefetchSchedule_params->RequiredPrefetchPixelDataBWLuma = &mode_lib->mp.RequiredPrefetchPixelDataBWLuma[k]; CalculatePrefetchSchedule_params->RequiredPrefetchPixelDataBWChroma = &mode_lib->mp.RequiredPrefetchPixelDataBWChroma[k]; - CalculatePrefetchSchedule_params->RequiredPrefetchBWOTO = &s->dummy_single_array[0][k]; + CalculatePrefetchSchedule_params->RequiredPrefetchBWMax = &s->dummy_single_array[0][k]; CalculatePrefetchSchedule_params->NotEnoughTimeForDynamicMetadata = &mode_lib->mp.NotEnoughTimeForDynamicMetadata[k]; CalculatePrefetchSchedule_params->Tno_bw = &mode_lib->mp.Tno_bw[k]; CalculatePrefetchSchedule_params->Tno_bw_flip = &mode_lib->mp.Tno_bw_flip[k]; @@ -11396,7 +11406,7 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex calculate_peak_bandwidth_params->surface_read_bandwidth_c = mode_lib->mp.vactive_sw_bw_c; calculate_peak_bandwidth_params->prefetch_bandwidth_l = mode_lib->mp.RequiredPrefetchPixelDataBWLuma; calculate_peak_bandwidth_params->prefetch_bandwidth_c = mode_lib->mp.RequiredPrefetchPixelDataBWChroma; - calculate_peak_bandwidth_params->prefetch_bandwidth_oto = s->dummy_single_array[0]; + calculate_peak_bandwidth_params->prefetch_bandwidth_max = s->dummy_single_array[0]; calculate_peak_bandwidth_params->excess_vactive_fill_bw_l = mode_lib->mp.excess_vactive_fill_bw_l; calculate_peak_bandwidth_params->excess_vactive_fill_bw_c = mode_lib->mp.excess_vactive_fill_bw_c; calculate_peak_bandwidth_params->cursor_bw = mode_lib->mp.cursor_bw; @@ -11536,7 +11546,7 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex calculate_peak_bandwidth_params->meta_row_bw = mode_lib->mp.meta_row_bw; calculate_peak_bandwidth_params->prefetch_cursor_bw = mode_lib->mp.prefetch_cursor_bw; calculate_peak_bandwidth_params->prefetch_vmrow_bw = mode_lib->mp.prefetch_vmrow_bw; - calculate_peak_bandwidth_params->prefetch_bandwidth_oto = s->dummy_single_array[0]; + calculate_peak_bandwidth_params->prefetch_bandwidth_max = s->dummy_single_array[0]; calculate_peak_bandwidth_params->flip_bw = mode_lib->mp.final_flip_bw; calculate_peak_bandwidth_params->urgent_burst_factor_l = mode_lib->mp.UrgentBurstFactorLuma; calculate_peak_bandwidth_params->urgent_burst_factor_c = mode_lib->mp.UrgentBurstFactorChroma; @@ -11880,7 +11890,7 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex } //Maximum Bandwidth Used - s->TotalWRBandwidth = 0; + mode_lib->mp.TotalWRBandwidth = 0; for (k = 0; k < display_cfg->num_streams; ++k) { s->WRBandwidth = 0; if (display_cfg->stream_descriptors[k].writeback.active_writebacks_per_stream > 0) { @@ -11889,7 +11899,7 @@ static bool dml_core_mode_programming(struct dml2_core_calcs_mode_programming_ex (display_cfg->stream_descriptors[k].timing.h_total * display_cfg->stream_descriptors[k].writeback.writeback_stream[0].input_height / ((double)display_cfg->stream_descriptors[k].timing.pixel_clock_khz / 1000)) * (display_cfg->stream_descriptors[k].writeback.writeback_stream[0].pixel_format == dml2_444_32 ? 4.0 : 8.0); - s->TotalWRBandwidth = s->TotalWRBandwidth + s->WRBandwidth; + mode_lib->mp.TotalWRBandwidth = mode_lib->mp.TotalWRBandwidth + s->WRBandwidth; } } @@ -13059,6 +13069,10 @@ void dml2_core_calcs_get_informative(const struct dml2_core_internal_display_mod out->informative.mode_support_info.OutputRate[k] = dml2_output_rate_hdmi_rate_10x4; else if (mode_lib->ms.support.OutputRate[k] == dml2_core_internal_output_rate_hdmi_rate_12x4) out->informative.mode_support_info.OutputRate[k] = dml2_output_rate_hdmi_rate_12x4; + else if (mode_lib->ms.support.OutputRate[k] == dml2_core_internal_output_rate_hdmi_rate_16x4) + out->informative.mode_support_info.OutputRate[k] = dml2_output_rate_hdmi_rate_16x4; + else if (mode_lib->ms.support.OutputRate[k] == dml2_core_internal_output_rate_hdmi_rate_20x4) + out->informative.mode_support_info.OutputRate[k] = dml2_output_rate_hdmi_rate_20x4; out->informative.mode_support_info.AlignedYPitch[k] = mode_lib->ms.support.AlignedYPitch[k]; out->informative.mode_support_info.AlignedCPitch[k] = mode_lib->ms.support.AlignedCPitch[k]; @@ -13243,7 +13257,7 @@ void dml2_core_calcs_get_informative(const struct dml2_core_internal_display_mod out->informative.misc.DisplayPipeLineDeliveryTimeLumaPrefetch[k] = mode_lib->mp.DisplayPipeLineDeliveryTimeLumaPrefetch[k]; out->informative.misc.DisplayPipeLineDeliveryTimeChromaPrefetch[k] = mode_lib->mp.DisplayPipeLineDeliveryTimeChromaPrefetch[k]; - out->informative.misc.WritebackRequiredBandwidth = mode_lib->scratch.dml_core_mode_programming_locals.TotalWRBandwidth / 1000.0; + out->informative.misc.WritebackRequiredBandwidth = mode_lib->mp.TotalWRBandwidth / 1000.0; out->informative.misc.WritebackAllowDRAMClockChangeEndPosition[k] = mode_lib->mp.WritebackAllowDRAMClockChangeEndPosition[k]; out->informative.misc.WritebackAllowFCLKChangeEndPosition[k] = mode_lib->mp.WritebackAllowFCLKChangeEndPosition[k]; out->informative.misc.DSCCLK_calculated[k] = mode_lib->mp.DSCCLK[k]; diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_shared_types.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_shared_types.h index bdee6ad7bc59c0b9d38c63b287d01f16cad9641b..28687565ac222ff0c9d670ac806afcfeb0c396d9 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_shared_types.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_shared_types.h @@ -102,6 +102,7 @@ struct dml2_core_internal_DmlPipe { double DCFClkDeepSleep; unsigned int DPPPerSurface; bool ScalerEnabled; + bool UPSPEnabled; enum dml2_rotation_angle RotationAngle; bool mirrored; unsigned int ViewportHeight; @@ -186,7 +187,9 @@ enum dml2_core_internal_output_type_rate { dml2_core_internal_output_rate_hdmi_rate_6x4 = 9, dml2_core_internal_output_rate_hdmi_rate_8x4 = 10, dml2_core_internal_output_rate_hdmi_rate_10x4 = 11, - dml2_core_internal_output_rate_hdmi_rate_12x4 = 12 + dml2_core_internal_output_rate_hdmi_rate_12x4 = 12, + dml2_core_internal_output_rate_hdmi_rate_16x4 = 13, + dml2_core_internal_output_rate_hdmi_rate_20x4 = 14 }; struct dml2_core_internal_watermarks { @@ -260,12 +263,14 @@ struct dml2_core_internal_mode_support_info { bool AvgBandwidthSupport; bool UrgVactiveBandwidthSupport; bool EnoughUrgentLatencyHidingSupport; + bool PrefetchScheduleSupported; bool PrefetchSupported; bool PrefetchBandwidthSupported; bool DynamicMetadataSupported; bool VRatioInPrefetchSupported; bool DISPCLK_DPPCLK_Support; bool TotalAvailablePipesSupport; + bool ODMSupport; bool ModeSupport; bool ViewportSizeSupport; @@ -314,9 +319,7 @@ struct dml2_core_internal_mode_support_info { double non_urg_bandwidth_required[dml2_core_internal_soc_state_max][dml2_core_internal_bw_max]; // same as urg_bandwidth, except not scaled by urg burst factor double non_urg_bandwidth_required_flip[dml2_core_internal_soc_state_max][dml2_core_internal_bw_max]; - bool avg_bandwidth_support_ok[dml2_core_internal_soc_state_max][dml2_core_internal_bw_max]; - double max_urgent_latency_us; double max_non_urgent_latency_us; double avg_non_urgent_latency_us; @@ -329,6 +332,8 @@ struct dml2_core_internal_mode_support_info { bool temp_read_or_ppt_support; struct dml2_core_internal_watermarks watermarks; + bool dcfclk_support; + bool qos_bandwidth_support; }; struct dml2_core_internal_mode_support { @@ -350,9 +355,11 @@ struct dml2_core_internal_mode_support { double SOCCLK; /// min_clocks.dcn4x.dispclk_khz, &state_table->dispclk); - if (result) - result = round_up_to_next_dpm(&display_cfg->min_clocks.dcn4x.deepsleep_dcfclk_khz, &state_table->dcfclk); - for (i = 0; i < DML2_MAX_DCN_PIPES; i++) { if (result) result = round_up_to_next_dpm(&display_cfg->plane_programming[i].min_clocks.dcn4x.dppclk_khz, &state_table->dppclk); diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/inc/dml2_debug.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/inc/dml2_debug.h index b226225103c34aea3b0a5c2cbf87dea72c6184b4..611c80f4f1bf141d1cbab33ebc551705d59af1e9 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/inc/dml2_debug.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/inc/dml2_debug.h @@ -10,15 +10,74 @@ #define DML_LOG_LEVEL_DEFAULT DML_LOG_LEVEL_WARN #define DML_LOG_INTERNAL(fmt, ...) dm_output_to_console(fmt, ## __VA_ARGS__) -/* ASSERT with message output */ -#define DML_ASSERT_MSG(condition, fmt, ...) \ - do { \ - if (!(condition)) { \ - DML_LOG_ERROR("DML ASSERT hit in %s line %d\n", __func__, __LINE__); \ - DML_LOG_ERROR(fmt, ## __VA_ARGS__); \ - DML_ASSERT(condition); \ - } \ - } while (0) +/* private helper macros */ +#define _BOOL_FORMAT(field) "%s", field ? "true" : "false" +#define _UINT_FORMAT(field) "%u", field +#define _INT_FORMAT(field) "%d", field +#define _DOUBLE_FORMAT(field) "%lf", field +#define _ELEMENT_FUNC "function" +#define _ELEMENT_COMP_IF "component_interface" +#define _ELEMENT_TOP_IF "top_interface" +#define _LOG_ENTRY(element) do { \ + DML_LOG_INTERNAL("<"element" name=\""); \ + DML_LOG_INTERNAL(__func__); \ + DML_LOG_INTERNAL("\">\n"); \ +} while (0) +#define _LOG_EXIT(element) DML_LOG_INTERNAL("\n") +#define _LOG_SCALAR(field, format) do { \ + DML_LOG_INTERNAL(#field" = "format(field)); \ + DML_LOG_INTERNAL("\n"); \ +} while (0) +#define _LOG_ARRAY(field, size, format) do { \ + DML_LOG_INTERNAL(#field " = ["); \ + for (int _i = 0; _i < (int) size; _i++) { \ + DML_LOG_INTERNAL(format(field[_i])); \ + if (_i + 1 == (int) size) \ + DML_LOG_INTERNAL("]\n"); \ + else \ + DML_LOG_INTERNAL(", "); \ +}} while (0) +#define _LOG_2D_ARRAY(field, size0, size1, format) do { \ + DML_LOG_INTERNAL(#field" = ["); \ + for (int _i = 0; _i < (int) size0; _i++) { \ + DML_LOG_INTERNAL("\n\t["); \ + for (int _j = 0; _j < (int) size1; _j++) { \ + DML_LOG_INTERNAL(format(field[_i][_j])); \ + if (_j + 1 == (int) size1) \ + DML_LOG_INTERNAL("]"); \ + else \ + DML_LOG_INTERNAL(", "); \ + } \ + if (_i + 1 == (int) size0) \ + DML_LOG_INTERNAL("]\n"); \ + else \ + DML_LOG_INTERNAL(", "); \ + } \ +} while (0) +#define _LOG_3D_ARRAY(field, size0, size1, size2, format) do { \ + DML_LOG_INTERNAL(#field" = ["); \ + for (int _i = 0; _i < (int) size0; _i++) { \ + DML_LOG_INTERNAL("\n\t["); \ + for (int _j = 0; _j < (int) size1; _j++) { \ + DML_LOG_INTERNAL("["); \ + for (int _k = 0; _k < (int) size2; _k++) { \ + DML_LOG_INTERNAL(format(field[_i][_j][_k])); \ + if (_k + 1 == (int) size2) \ + DML_LOG_INTERNAL("]"); \ + else \ + DML_LOG_INTERNAL(", "); \ + } \ + if (_j + 1 == (int) size1) \ + DML_LOG_INTERNAL("]"); \ + else \ + DML_LOG_INTERNAL(", "); \ + } \ + if (_i + 1 == (int) size0) \ + DML_LOG_INTERNAL("]\n"); \ + else \ + DML_LOG_INTERNAL(", "); \ + } \ +} while (0) /* fatal errors for unrecoverable DML states until a full reset */ #define DML_LOG_LEVEL_FATAL 0 @@ -28,7 +87,7 @@ #define DML_LOG_LEVEL_WARN 2 /* high level tracing of DML interfaces */ #define DML_LOG_LEVEL_INFO 3 -/* detailed tracing of DML internal components */ +/* tracing of DML internal executions */ #define DML_LOG_LEVEL_DEBUG 4 /* detailed tracing of DML calculation procedure */ #define DML_LOG_LEVEL_VERBOSE 5 @@ -37,30 +96,94 @@ #define DML_LOG_LEVEL DML_LOG_LEVEL_DEFAULT #endif /* #ifndef DML_LOG_LEVEL */ +/* public macros for DML_LOG_LEVEL_FATAL and up */ #define DML_LOG_FATAL(fmt, ...) DML_LOG_INTERNAL("[DML FATAL] " fmt, ## __VA_ARGS__) + +/* public macros for DML_LOG_LEVEL_ERROR and up */ #if DML_LOG_LEVEL >= DML_LOG_LEVEL_ERROR #define DML_LOG_ERROR(fmt, ...) DML_LOG_INTERNAL("[DML ERROR] "fmt, ## __VA_ARGS__) +#define DML_ASSERT_MSG(condition, fmt, ...) \ + do { \ + if (!(condition)) { \ + DML_LOG_ERROR("ASSERT hit in %s line %d\n", __func__, __LINE__); \ + DML_LOG_ERROR(fmt, ## __VA_ARGS__); \ + DML_ASSERT(condition); \ + } \ + } while (0) #else #define DML_LOG_ERROR(fmt, ...) ((void)0) +#define DML_ASSERT_MSG(condition, fmt, ...) ((void)0) #endif + +/* public macros for DML_LOG_LEVEL_WARN and up */ #if DML_LOG_LEVEL >= DML_LOG_LEVEL_WARN #define DML_LOG_WARN(fmt, ...) DML_LOG_INTERNAL("[DML WARN] "fmt, ## __VA_ARGS__) #else #define DML_LOG_WARN(fmt, ...) ((void)0) #endif + +/* public macros for DML_LOG_LEVEL_INFO and up */ #if DML_LOG_LEVEL >= DML_LOG_LEVEL_INFO #define DML_LOG_INFO(fmt, ...) DML_LOG_INTERNAL("[DML INFO] "fmt, ## __VA_ARGS__) +#define DML_LOG_TOP_IF_ENTER() _LOG_ENTRY(_ELEMENT_TOP_IF) +#define DML_LOG_TOP_IF_EXIT() _LOG_EXIT(_ELEMENT_TOP_IF) #else #define DML_LOG_INFO(fmt, ...) ((void)0) +#define DML_LOG_TOP_IF_ENTER() ((void)0) +#define DML_LOG_TOP_IF_EXIT() ((void)0) #endif + +/* public macros for DML_LOG_LEVEL_DEBUG and up */ #if DML_LOG_LEVEL >= DML_LOG_LEVEL_DEBUG -#define DML_LOG_DEBUG(fmt, ...) DML_LOG_INTERNAL("[DML DEBUG] "fmt, ## __VA_ARGS__) +#define DML_LOG_DEBUG(fmt, ...) DML_LOG_INTERNAL(fmt, ## __VA_ARGS__) +#define DML_LOG_COMP_IF_ENTER() _LOG_ENTRY(_ELEMENT_COMP_IF) +#define DML_LOG_COMP_IF_EXIT() _LOG_EXIT(_ELEMENT_COMP_IF) +#define DML_LOG_FUNC_ENTER() _LOG_ENTRY(_ELEMENT_FUNC) +#define DML_LOG_FUNC_EXIT() _LOG_EXIT(_ELEMENT_FUNC) +#define DML_LOG_DEBUG_BOOL(field) _LOG_SCALAR(field, _BOOL_FORMAT) +#define DML_LOG_DEBUG_UINT(field) _LOG_SCALAR(field, _UINT_FORMAT) +#define DML_LOG_DEBUG_INT(field) _LOG_SCALAR(field, _INT_FORMAT) +#define DML_LOG_DEBUG_DOUBLE(field) _LOG_SCALAR(field, _DOUBLE_FORMAT) +#define DML_LOG_DEBUG_ARRAY_BOOL(field, size) _LOG_ARRAY(field, size, _BOOL_FORMAT) +#define DML_LOG_DEBUG_ARRAY_UINT(field, size) _LOG_ARRAY(field, size, _UINT_FORMAT) +#define DML_LOG_DEBUG_ARRAY_INT(field, size) _LOG_ARRAY(field, size, _INT_FORMAT) +#define DML_LOG_DEBUG_ARRAY_DOUBLE(field, size) _LOG_ARRAY(field, size, _DOUBLE_FORMAT) +#define DML_LOG_DEBUG_2D_ARRAY_BOOL(field, size0, size1) _LOG_2D_ARRAY(field, size0, size1, _BOOL_FORMAT) +#define DML_LOG_DEBUG_2D_ARRAY_UINT(field, size0, size1) _LOG_2D_ARRAY(field, size0, size1, _UINT_FORMAT) +#define DML_LOG_DEBUG_2D_ARRAY_INT(field, size0, size1) _LOG_2D_ARRAY(field, size0, size1, _INT_FORMAT) +#define DML_LOG_DEBUG_2D_ARRAY_DOUBLE(field, size0, size1) _LOG_2D_ARRAY(field, size0, size1, _DOUBLE_FORMAT) +#define DML_LOG_DEBUG_3D_ARRAY_BOOL(field, size0, size1, size2) _LOG_3D_ARRAY(field, size0, size1, size2, _BOOL_FORMAT) +#define DML_LOG_DEBUG_3D_ARRAY_UINT(field, size0, size1, size2) _LOG_3D_ARRAY(field, size0, size1, size2, _UINT_FORMAT) +#define DML_LOG_DEBUG_3D_ARRAY_INT(field, size0, size1, size2) _LOG_3D_ARRAY(field, size0, size1, size2, _INT_FORMAT) +#define DML_LOG_DEBUG_3D_ARRAY_DOUBLE(field, size0, size1, size2) _LOG_3D_ARRAY(field, size0, size1, size2, _DOUBLE_FORMAT) #else #define DML_LOG_DEBUG(fmt, ...) ((void)0) +#define DML_LOG_COMP_IF_ENTER() ((void)0) +#define DML_LOG_COMP_IF_EXIT() ((void)0) +#define DML_LOG_FUNC_ENTER() ((void)0) +#define DML_LOG_FUNC_EXIT() ((void)0) +#define DML_LOG_DEBUG_BOOL(field) ((void)0) +#define DML_LOG_DEBUG_UINT(field) ((void)0) +#define DML_LOG_DEBUG_INT(field) ((void)0) +#define DML_LOG_DEBUG_DOUBLE(field) ((void)0) +#define DML_LOG_DEBUG_ARRAY_BOOL(field, size) ((void)0) +#define DML_LOG_DEBUG_ARRAY_UINT(field, size) ((void)0) +#define DML_LOG_DEBUG_ARRAY_INT(field, size) ((void)0) +#define DML_LOG_DEBUG_ARRAY_DOUBLE(field, size) ((void)0) +#define DML_LOG_DEBUG_2D_ARRAY_BOOL(field, size0, size1) ((void)0) +#define DML_LOG_DEBUG_2D_ARRAY_UINT(field, size0, size1) ((void)0) +#define DML_LOG_DEBUG_2D_ARRAY_INT(field, size0, size1) ((void)0) +#define DML_LOG_DEBUG_2D_ARRAY_DOUBLE(field, size0, size1) ((void)0) +#define DML_LOG_DEBUG_3D_ARRAY_BOOL(field, size0, size1, size2) ((void)0) +#define DML_LOG_DEBUG_3D_ARRAY_UINT(field, size0, size1, size2) ((void)0) +#define DML_LOG_DEBUG_3D_ARRAY_INT(field, size0, size1, size2) ((void)0) +#define DML_LOG_DEBUG_3D_ARRAY_DOUBLE(field, size0, size1, size2) ((void)0) #endif + +/* public macros for DML_LOG_LEVEL_VERBOSE */ #if DML_LOG_LEVEL >= DML_LOG_LEVEL_VERBOSE -#define DML_LOG_VERBOSE(fmt, ...) DML_LOG_INTERNAL("[DML VERBOSE] "fmt, ## __VA_ARGS__) +#define DML_LOG_VERBOSE(fmt, ...) DML_LOG_INTERNAL(fmt, ## __VA_ARGS__) #else #define DML_LOG_VERBOSE(fmt, ...) ((void)0) -#endif +#endif /* #if DML_LOG_LEVEL >= DML_LOG_LEVEL_VERBOSE */ #endif /* __DML2_DEBUG_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/inc/dml2_internal_shared_types.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/inc/dml2_internal_shared_types.h index 00688b9f1df4c6b766d0faebfe76834db47d71a4..d52aa82283b3696ab4591b68d557ef743591eb62 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/inc/dml2_internal_shared_types.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/inc/dml2_internal_shared_types.h @@ -202,6 +202,8 @@ struct dml2_core_mode_support_result { } active; unsigned int dispclk_khz; + unsigned int dpprefclk_khz; + unsigned int dtbrefclk_khz; unsigned int dcfclk_deepsleep_khz; unsigned int socclk_khz; @@ -446,13 +448,17 @@ struct dml2_core_internal_state_intermediates { }; struct dml2_core_mode_support_locals { - struct dml2_core_calcs_mode_support_ex mode_support_ex_params; + union { + struct dml2_core_calcs_mode_support_ex mode_support_ex_params; + }; struct dml2_display_cfg svp_expanded_display_cfg; struct dml2_calculate_mcache_allocation_in_out calc_mcache_allocation_params; }; struct dml2_core_mode_programming_locals { - struct dml2_core_calcs_mode_programming_ex mode_programming_ex_params; + union { + struct dml2_core_calcs_mode_programming_ex mode_programming_ex_params; + }; struct dml2_display_cfg svp_expanded_display_cfg; }; diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c index 6b3b8803e0aee2094e6486554e283aac4149d52e..a56e75cdf712378054d653b759c815f7d6cf7b6f 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c @@ -868,7 +868,7 @@ bool dml2_svp_remove_all_phantom_pipes(struct dml2_context *ctx, struct dc_state /* Conditions for setting up phantom pipes for SubVP: * 1. Not force disable SubVP - * 2. Full update (i.e. !fast_validate) + * 2. Full update (i.e. DC_VALIDATE_MODE_AND_PROGRAMMING) * 3. Enough pipes are available to support SubVP (TODO: Which pipes will use VACTIVE / VBLANK / SUBVP?) * 4. Display configuration passes validation * 5. (Config doesn't support MCLK in VACTIVE/VBLANK || dc->debug.force_subvp_mclk_switch) diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c index 5de775fd8fceef1a47b66de5c5beb24f8628fc96..3b866e876bf4df40bee841f5485dbf1a8f47804d 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_translation_helper.c @@ -953,6 +953,7 @@ static void populate_dml_surface_cfg_from_plane_state(enum dml_project_id dml2_p out->SourcePixelFormat[location] = dml_420_10; break; case SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616: + case SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616: case SURFACE_PIXEL_FORMAT_GRPH_ARGB16161616F: case SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616F: out->SourcePixelFormat[location] = dml_444_64; @@ -1188,22 +1189,6 @@ static unsigned int map_plane_to_dml_display_cfg(const struct dml2_context *dml2 return location; } -static void apply_legacy_svp_drr_settings(struct dml2_context *dml2, const struct dc_state *state, struct dml_display_cfg_st *dml_dispcfg) -{ - int i; - - if (state->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) { - ASSERT(state->stream_count == 1); - dml_dispcfg->timing.DRRDisplay[0] = true; - } else if (state->bw_ctx.bw.dcn.legacy_svp_drr_stream_index_valid) { - - for (i = 0; i < dml_dispcfg->num_timings; i++) { - if (dml2->v20.scratch.dml_to_dc_pipe_mapping.disp_cfg_to_stream_id[i] == state->streams[state->bw_ctx.bw.dcn.legacy_svp_drr_stream_index]->stream_id) - dml_dispcfg->timing.DRRDisplay[i] = true; - } - } -} - static void dml2_populate_pipe_to_plane_index_mapping(struct dml2_context *dml2, struct dc_state *state) { unsigned int i; @@ -1436,9 +1421,6 @@ void map_dc_state_into_dml_display_cfg(struct dml2_context *dml2, struct dc_stat } } } - - if (!dml2->config.use_native_pstate_optimization) - apply_legacy_svp_drr_settings(dml2, context, dml_dispcfg); } void dml2_update_pipe_ctx_dchub_regs(struct _vcs_dpi_dml_display_rq_regs_st *rq_regs, diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c index 525b7d04bf84cd10e60fc5e39f43b9d644930c45..0318260370edfec86e67104d3958adef529394fa 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c @@ -24,8 +24,6 @@ * */ -#include - #include "display_mode_core.h" #include "dml2_internal_types.h" #include "dml2_utils.h" @@ -95,12 +93,17 @@ static void map_hw_resources(struct dml2_context *dml2, static unsigned int pack_and_call_dml_mode_support_ex(struct dml2_context *dml2, const struct dml_display_cfg_st *display_cfg, - struct dml_mode_support_info_st *evaluation_info) + struct dml_mode_support_info_st *evaluation_info, + enum dc_validate_mode validate_mode) { struct dml2_wrapper_scratch *s = &dml2->v20.scratch; s->mode_support_params.mode_lib = &dml2->v20.dml_core_ctx; s->mode_support_params.in_display_cfg = display_cfg; + if (validate_mode == DC_VALIDATE_MODE_ONLY) + s->mode_support_params.in_start_state_idx = dml2->v20.dml_core_ctx.states.num_states - 1; + else + s->mode_support_params.in_start_state_idx = 0; s->mode_support_params.out_evaluation_info = evaluation_info; memset(evaluation_info, 0, sizeof(struct dml_mode_support_info_st)); @@ -112,10 +115,8 @@ static unsigned int pack_and_call_dml_mode_support_ex(struct dml2_context *dml2, static bool optimize_configuration(struct dml2_context *dml2, struct dml2_wrapper_optimize_configuration_params *p) { int unused_dpps = p->ip_params->max_num_dpp; - int i, j; - int odms_needed, refresh_rate_hz, dpps_needed, subvp_height, pstate_width_fw_delay_lines, surface_count; - int subvp_timing_to_add, new_timing_index, subvp_surface_to_add, new_surface_index; - float frame_time_sec, max_frame_time_sec; + int i; + int odms_needed; int largest_blend_and_timing = 0; bool optimization_done = false; @@ -130,79 +131,6 @@ static bool optimize_configuration(struct dml2_context *dml2, struct dml2_wrappe if (p->new_display_config != p->cur_display_config) *p->new_display_config = *p->cur_display_config; - // Optimize P-State Support - if (dml2->config.use_native_pstate_optimization) { - if (p->cur_mode_support_info->DRAMClockChangeSupport[0] == dml_dram_clock_change_unsupported) { - // Find a display with < 120Hz refresh rate with maximal refresh rate that's not already subvp - subvp_timing_to_add = -1; - subvp_surface_to_add = -1; - max_frame_time_sec = 0; - surface_count = 0; - for (i = 0; i < (int) p->cur_display_config->num_timings; i++) { - refresh_rate_hz = (int)div_u64((unsigned long long) p->cur_display_config->timing.PixelClock[i] * 1000 * 1000, - (p->cur_display_config->timing.HTotal[i] * p->cur_display_config->timing.VTotal[i])); - if (refresh_rate_hz < 120) { - // Check its upstream surfaces to see if this one could be converted to subvp. - dpps_needed = 0; - for (j = 0; j < (int) p->cur_display_config->num_surfaces; j++) { - if (p->cur_display_config->plane.BlendingAndTiming[j] == i && - p->cur_display_config->plane.UseMALLForPStateChange[j] == dml_use_mall_pstate_change_disable) { - dpps_needed += p->cur_mode_support_info->DPPPerSurface[j]; - subvp_surface_to_add = j; - surface_count++; - } - } - - if (surface_count == 1 && dpps_needed > 0 && dpps_needed <= unused_dpps) { - frame_time_sec = (float)1 / refresh_rate_hz; - if (frame_time_sec > max_frame_time_sec) { - max_frame_time_sec = frame_time_sec; - subvp_timing_to_add = i; - } - } - } - } - if (subvp_timing_to_add >= 0) { - new_timing_index = p->new_display_config->num_timings++; - new_surface_index = p->new_display_config->num_surfaces++; - // Add a phantom pipe reflecting the main pipe's timing - dml2_util_copy_dml_timing(&p->new_display_config->timing, new_timing_index, subvp_timing_to_add); - - pstate_width_fw_delay_lines = (int)(((double)(p->config->svp_pstate.subvp_fw_processing_delay_us + - p->config->svp_pstate.subvp_pstate_allow_width_us) / 1000000) * - (p->new_display_config->timing.PixelClock[subvp_timing_to_add] * 1000 * 1000) / - (double)p->new_display_config->timing.HTotal[subvp_timing_to_add]); - - subvp_height = p->cur_mode_support_info->SubViewportLinesNeededInMALL[subvp_timing_to_add] + pstate_width_fw_delay_lines; - - p->new_display_config->timing.VActive[new_timing_index] = subvp_height; - p->new_display_config->timing.VTotal[new_timing_index] = subvp_height + - p->new_display_config->timing.VTotal[subvp_timing_to_add] - p->new_display_config->timing.VActive[subvp_timing_to_add]; - - p->new_display_config->output.OutputDisabled[new_timing_index] = true; - - p->new_display_config->plane.UseMALLForPStateChange[subvp_surface_to_add] = dml_use_mall_pstate_change_sub_viewport; - - dml2_util_copy_dml_plane(&p->new_display_config->plane, new_surface_index, subvp_surface_to_add); - dml2_util_copy_dml_surface(&p->new_display_config->surface, new_surface_index, subvp_surface_to_add); - - p->new_display_config->plane.ViewportHeight[new_surface_index] = subvp_height; - p->new_display_config->plane.ViewportHeightChroma[new_surface_index] = subvp_height; - p->new_display_config->plane.ViewportStationary[new_surface_index] = false; - - p->new_display_config->plane.UseMALLForStaticScreen[new_surface_index] = dml_use_mall_static_screen_disable; - p->new_display_config->plane.UseMALLForPStateChange[new_surface_index] = dml_use_mall_pstate_change_phantom_pipe; - - p->new_display_config->plane.NumberOfCursors[new_surface_index] = 0; - - p->new_policy->ImmediateFlipRequirement[new_surface_index] = dml_immediate_flip_not_required; - - p->new_display_config->plane.BlendingAndTiming[new_surface_index] = new_timing_index; - - optimization_done = true; - } - } - } // Optimize Clocks if (!optimization_done) { @@ -226,7 +154,8 @@ static bool optimize_configuration(struct dml2_context *dml2, struct dml2_wrappe return optimization_done; } -static int calculate_lowest_supported_state_for_temp_read(struct dml2_context *dml2, struct dc_state *display_state) +static int calculate_lowest_supported_state_for_temp_read(struct dml2_context *dml2, struct dc_state *display_state, + enum dc_validate_mode validate_mode) { struct dml2_calculate_lowest_supported_state_for_temp_read_scratch *s = &dml2->v20.scratch.dml2_calculate_lowest_supported_state_for_temp_read_scratch; struct dml2_wrapper_scratch *s_global = &dml2->v20.scratch; @@ -268,7 +197,8 @@ static int calculate_lowest_supported_state_for_temp_read(struct dml2_context *d dml2->v20.dml_core_ctx.states.state_array[j].dram_clock_change_latency_us = s_global->dummy_pstate_table[i].dummy_pstate_latency_us; } - dml_result = pack_and_call_dml_mode_support_ex(dml2, &s->cur_display_config, &s->evaluation_info); + dml_result = pack_and_call_dml_mode_support_ex(dml2, &s->cur_display_config, &s->evaluation_info, + validate_mode); if (dml_result && s->evaluation_info.DRAMClockChangeSupport[0] == dml_dram_clock_change_vactive) { map_hw_resources(dml2, &s->cur_display_config, &s->evaluation_info); @@ -333,7 +263,8 @@ static bool does_configuration_meet_sw_policies(struct dml2_context *ctx, const } static bool dml_mode_support_wrapper(struct dml2_context *dml2, - struct dc_state *display_state) + struct dc_state *display_state, + enum dc_validate_mode validate_mode) { struct dml2_wrapper_scratch *s = &dml2->v20.scratch; unsigned int result = 0, i; @@ -369,7 +300,8 @@ static bool dml_mode_support_wrapper(struct dml2_context *dml2, result = pack_and_call_dml_mode_support_ex(dml2, &s->cur_display_config, - &s->mode_support_info); + &s->mode_support_info, + validate_mode); if (result) result = does_configuration_meet_sw_policies(dml2, &s->cur_display_config, &s->mode_support_info); @@ -390,7 +322,8 @@ static bool dml_mode_support_wrapper(struct dml2_context *dml2, dml2->v20.dml_core_ctx.policy = s->new_policy; optimized_result = pack_and_call_dml_mode_support_ex(dml2, &s->new_display_config, - &s->mode_support_info); + &s->mode_support_info, + validate_mode); if (optimized_result) optimized_result = does_configuration_meet_sw_policies(dml2, &s->new_display_config, &s->mode_support_info); @@ -409,7 +342,8 @@ static bool dml_mode_support_wrapper(struct dml2_context *dml2, if (!optimized_result) { result = pack_and_call_dml_mode_support_ex(dml2, &s->cur_display_config, - &s->mode_support_info); + &s->mode_support_info, + validate_mode); } } @@ -419,118 +353,7 @@ static bool dml_mode_support_wrapper(struct dml2_context *dml2, return result; } -static int find_drr_eligible_stream(struct dc_state *display_state) -{ - int i; - - for (i = 0; i < display_state->stream_count; i++) { - if (dc_state_get_stream_subvp_type(display_state, display_state->streams[i]) == SUBVP_NONE - && display_state->streams[i]->ignore_msa_timing_param) { - // Use ignore_msa_timing_param flag to identify as DRR - return i; - } - } - - return -1; -} - -static bool optimize_pstate_with_svp_and_drr(struct dml2_context *dml2, struct dc_state *display_state) -{ - struct dml2_wrapper_scratch *s = &dml2->v20.scratch; - bool pstate_optimization_done = false; - bool pstate_optimization_success = false; - bool result = false; - int drr_display_index = 0, non_svp_streams = 0; - bool force_svp = dml2->config.svp_pstate.force_enable_subvp; - - display_state->bw_ctx.bw.dcn.clk.fw_based_mclk_switching = false; - display_state->bw_ctx.bw.dcn.legacy_svp_drr_stream_index_valid = false; - - result = dml_mode_support_wrapper(dml2, display_state); - - if (!result) { - pstate_optimization_done = true; - } else if (s->mode_support_info.DRAMClockChangeSupport[0] != dml_dram_clock_change_unsupported && !force_svp) { - pstate_optimization_success = true; - pstate_optimization_done = true; - } - - if (display_state->stream_count == 1 && dml2->config.callbacks.can_support_mclk_switch_using_fw_based_vblank_stretch(dml2->config.callbacks.dc, display_state)) { - display_state->bw_ctx.bw.dcn.clk.fw_based_mclk_switching = true; - - result = dml_mode_support_wrapper(dml2, display_state); - } else { - non_svp_streams = display_state->stream_count; - - while (!pstate_optimization_done) { - result = dml_mode_programming(&dml2->v20.dml_core_ctx, s->mode_support_params.out_lowest_state_idx, &s->cur_display_config, true); - - // Always try adding SVP first - if (result) - result = dml2_svp_add_phantom_pipe_to_dc_state(dml2, display_state, &s->mode_support_info); - else - pstate_optimization_done = true; - - - if (result) { - result = dml_mode_support_wrapper(dml2, display_state); - } else { - pstate_optimization_done = true; - } - - if (result) { - non_svp_streams--; - - if (s->mode_support_info.DRAMClockChangeSupport[0] != dml_dram_clock_change_unsupported) { - if (dml2_svp_validate_static_schedulability(dml2, display_state, s->mode_support_info.DRAMClockChangeSupport[0])) { - pstate_optimization_success = true; - pstate_optimization_done = true; - } else { - pstate_optimization_success = false; - pstate_optimization_done = false; - } - } else { - drr_display_index = find_drr_eligible_stream(display_state); - - // If there is only 1 remaining non SubVP pipe that is DRR, check static - // schedulability for SubVP + DRR. - if (non_svp_streams == 1 && drr_display_index >= 0) { - if (dml2_svp_drr_schedulable(dml2, display_state, &display_state->streams[drr_display_index]->timing)) { - display_state->bw_ctx.bw.dcn.legacy_svp_drr_stream_index_valid = true; - display_state->bw_ctx.bw.dcn.legacy_svp_drr_stream_index = drr_display_index; - result = dml_mode_support_wrapper(dml2, display_state); - } - - if (result && s->mode_support_info.DRAMClockChangeSupport[0] != dml_dram_clock_change_unsupported) { - pstate_optimization_success = true; - pstate_optimization_done = true; - } else { - pstate_optimization_success = false; - pstate_optimization_done = false; - } - } - - if (pstate_optimization_success) { - pstate_optimization_done = true; - } else { - pstate_optimization_done = false; - } - } - } - } - } - - if (!pstate_optimization_success) { - dml2_svp_remove_all_phantom_pipes(dml2, display_state); - display_state->bw_ctx.bw.dcn.clk.fw_based_mclk_switching = false; - display_state->bw_ctx.bw.dcn.legacy_svp_drr_stream_index_valid = false; - result = dml_mode_support_wrapper(dml2, display_state); - } - - return result; -} - -static bool call_dml_mode_support_and_programming(struct dc_state *context) +static bool call_dml_mode_support_and_programming(struct dc_state *context, enum dc_validate_mode validate_mode) { unsigned int result = 0; unsigned int min_state = 0; @@ -544,16 +367,13 @@ static bool call_dml_mode_support_and_programming(struct dc_state *context) struct dml2_wrapper_scratch *s = &dml2->v20.scratch; if (!context->streams[0]->sink->link->dc->caps.is_apu) { - min_state_for_g6_temp_read = calculate_lowest_supported_state_for_temp_read(dml2, context); + min_state_for_g6_temp_read = calculate_lowest_supported_state_for_temp_read(dml2, context, + validate_mode); ASSERT(min_state_for_g6_temp_read >= 0); } - if (!dml2->config.use_native_pstate_optimization) { - result = optimize_pstate_with_svp_and_drr(dml2, context); - } else { - result = dml_mode_support_wrapper(dml2, context); - } + result = dml_mode_support_wrapper(dml2, context, validate_mode); /* Upon trying to sett certain frequencies in FRL, min_state_for_g6_temp_read is reported as -1. This leads to an invalid value of min_state causing crashes later on. * Use the default logic for min_state only when min_state_for_g6_temp_read is a valid value. In other cases, use the value calculated by the DML directly. @@ -575,7 +395,8 @@ static bool call_dml_mode_support_and_programming(struct dc_state *context) return result; } -static bool dml2_validate_and_build_resource(const struct dc *in_dc, struct dc_state *context) +static bool dml2_validate_and_build_resource(const struct dc *in_dc, struct dc_state *context, + enum dc_validate_mode validate_mode) { struct dml2_context *dml2 = context->bw_ctx.dml2; struct dml2_wrapper_scratch *s = &dml2->v20.scratch; @@ -611,7 +432,7 @@ static bool dml2_validate_and_build_resource(const struct dc *in_dc, struct dc_s copy_dummy_pstate_table(s->dummy_pstate_table, in_dc->clk_mgr->bw_params->dummy_pstate_table, 4); - result = call_dml_mode_support_and_programming(context); + result = call_dml_mode_support_and_programming(context, validate_mode); /* Call map dc pipes to map the pipes based on the DML output. For correctly determining if recalculation * is required or not, the resource context needs to correctly reflect the number of active pipes. We would * only know the correct number if active pipes after dml2_map_dc_pipes is called. @@ -628,7 +449,7 @@ static bool dml2_validate_and_build_resource(const struct dc *in_dc, struct dc_s need_recalculation = dml2_verify_det_buffer_configuration(dml2, context, &dml2->det_helper_scratch); if (need_recalculation) { /* Engage the DML again if recalculation is required. */ - call_dml_mode_support_and_programming(context); + call_dml_mode_support_and_programming(context, validate_mode); if (!dml2->config.skip_hw_state_mapping) { dml2_map_dc_pipes(dml2, context, &s->cur_display_config, &s->dml_to_dc_pipe_mapping, in_dc->current_state); } @@ -684,7 +505,7 @@ static bool dml2_validate_and_build_resource(const struct dc *in_dc, struct dc_s return result; } -static bool dml2_validate_only(struct dc_state *context) +static bool dml2_validate_only(struct dc_state *context, enum dc_validate_mode validate_mode) { struct dml2_context *dml2; unsigned int result = 0; @@ -708,7 +529,8 @@ static bool dml2_validate_only(struct dc_state *context) result = pack_and_call_dml_mode_support_ex(dml2, &dml2->v20.scratch.cur_display_config, - &dml2->v20.scratch.mode_support_info); + &dml2->v20.scratch.mode_support_info, + validate_mode); if (result) result = does_configuration_meet_sw_policies(dml2, &dml2->v20.scratch.cur_display_config, &dml2->v20.scratch.mode_support_info); @@ -723,7 +545,8 @@ static void dml2_apply_debug_options(const struct dc *dc, struct dml2_context *d } } -bool dml2_validate(const struct dc *in_dc, struct dc_state *context, struct dml2_context *dml2, bool fast_validate) +bool dml2_validate(const struct dc *in_dc, struct dc_state *context, struct dml2_context *dml2, + enum dc_validate_mode validate_mode) { bool out = false; @@ -733,17 +556,17 @@ bool dml2_validate(const struct dc *in_dc, struct dc_state *context, struct dml2 /* DML2.1 validation path */ if (dml2->architecture == dml2_architecture_21) { - out = dml21_validate(in_dc, context, dml2, fast_validate); + out = dml21_validate(in_dc, context, dml2, validate_mode); return out; } DC_FP_START(); - /* Use dml_validate_only for fast_validate path */ - if (fast_validate) - out = dml2_validate_only(context); + /* Use dml_validate_only for DC_VALIDATE_MODE_ONLY and DC_VALIDATE_MODE_AND_STATE_INDEX path */ + if (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING) + out = dml2_validate_only(context, validate_mode); else - out = dml2_validate_and_build_resource(in_dc, context); + out = dml2_validate_and_build_resource(in_dc, context, validate_mode); DC_FP_END(); @@ -757,8 +580,8 @@ static inline struct dml2_context *dml2_allocate_memory(void) static void dml2_init(const struct dc *in_dc, const struct dml2_configuration_options *config, struct dml2_context **dml2) { - if ((in_dc->debug.using_dml21) && (in_dc->ctx->dce_version == DCN_VERSION_4_01)) { - dml21_reinit(in_dc, dml2, config); + if ((in_dc->debug.using_dml21) && (in_dc->ctx->dce_version >= DCN_VERSION_4_01)) { + dml21_reinit(in_dc, *dml2, config); return; } @@ -803,9 +626,7 @@ static void dml2_init(const struct dc *in_dc, const struct dml2_configuration_op bool dml2_create(const struct dc *in_dc, const struct dml2_configuration_options *config, struct dml2_context **dml2) { // TODO : Temporarily add DCN_VERSION_3_2 for N-1 validation. Remove DCN_VERSION_3_2 after N-1 validation phase is complete. - if ((in_dc->debug.using_dml21) - && (in_dc->ctx->dce_version == DCN_VERSION_4_01 - )) + if ((in_dc->debug.using_dml21) && (in_dc->ctx->dce_version >= DCN_VERSION_4_01)) return dml21_create(in_dc, dml2, config); // Allocate Mode Lib Ctx @@ -874,8 +695,8 @@ void dml2_reinit(const struct dc *in_dc, const struct dml2_configuration_options *config, struct dml2_context **dml2) { - if ((in_dc->debug.using_dml21) && (in_dc->ctx->dce_version == DCN_VERSION_4_01)) { - dml21_reinit(in_dc, dml2, config); + if ((in_dc->debug.using_dml21) && (in_dc->ctx->dce_version >= DCN_VERSION_4_01)) { + dml21_reinit(in_dc, *dml2, config); return; } diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.h b/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.h index 5100f269368e794b7c5b2f6c79e87eb0ea10f6a6..c384e141cebc8f750ca7451ca5b5016780a1598f 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.h +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.h @@ -240,7 +240,7 @@ struct dml2_configuration_options { bool use_clock_dc_limits; bool gpuvm_enable; bool force_tdlut_enable; - struct dml2_soc_bb *bb_from_dmub; + void *bb_from_dmub; }; /* @@ -272,7 +272,7 @@ void dml2_reinit(const struct dc *in_dc, * dml2_validate - Determines if a display configuration is supported or not. * @in_dc: dc. * @context: dc_state to be validated. - * @fast_validate: Fast validate will not populate context.res_ctx. + * @validate_mode: DC_VALIDATE_MODE_ONLY and DC_VALIDATE_MODE_AND_STATE_INDEX will not populate context.res_ctx. * * DML1.0 compatible interface for validation. * @@ -295,7 +295,7 @@ void dml2_reinit(const struct dc *in_dc, bool dml2_validate(const struct dc *in_dc, struct dc_state *context, struct dml2_context *dml2, - bool fast_validate); + enum dc_validate_mode validate_mode); /* * dml2_extract_dram_and_fclk_change_support - Extracts the FCLK and UCLK change support info. diff --git a/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp.c b/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp.c index 97bf26fa3573898027ada5f6cac701b9b103724c..36187f890d5d0bc26949b215cb7f6617157ce2fe 100644 --- a/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp.c +++ b/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp.c @@ -231,7 +231,7 @@ static struct dpp_funcs dcn401_dpp_funcs = { .dpp_program_regamma_pwl = NULL, .dpp_set_pre_degam = dpp3_set_pre_degam, .dpp_program_input_lut = NULL, - .dpp_full_bypass = dpp401_full_bypass, + .dpp_full_bypass = NULL, .dpp_setup = dpp401_dpp_setup, .dpp_program_degamma_pwl = NULL, .dpp_program_cm_dealpha = dpp3_program_cm_dealpha, diff --git a/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp.h b/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp.h index ecaa976e1f5251b7139d51a75cfa0a41e2832d87..5a6a861402b3c4eaa740f39084831d1c7928b8a5 100644 --- a/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp.h +++ b/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp.h @@ -641,6 +641,7 @@ uint32_t ISHARP_DELTA_DATA; \ uint32_t ISHARP_DELTA_INDEX; \ uint32_t ISHARP_NLDELTA_SOFT_CLIP + struct dcn401_dpp_registers { DPP_REG_VARIABLE_LIST_DCN401; }; @@ -683,8 +684,6 @@ void dpp401_dscl_set_scaler_manual_scale( struct dpp *dpp_base, const struct scaler_data *scl_data); -void dpp401_full_bypass(struct dpp *dpp_base); - void dpp401_dpp_setup( struct dpp *dpp_base, enum surface_pixel_format format, diff --git a/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_cm.c b/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_cm.c index 712aff7e17f7a0f727f7751676216c9168d875a2..7aab77b588694de74542c24f87dc2b1ff8e159bf 100644 --- a/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_cm.c +++ b/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_cm.c @@ -88,30 +88,6 @@ enum dscl_mode_sel { DSCL_MODE_DSCL_BYPASS = 6 }; -void dpp401_full_bypass(struct dpp *dpp_base) -{ - struct dcn401_dpp *dpp = TO_DCN401_DPP(dpp_base); - - /* Input pixel format: ARGB8888 */ - REG_SET(CNVC_SURFACE_PIXEL_FORMAT, 0, - CNVC_SURFACE_PIXEL_FORMAT, 0x8); - - /* Zero expansion */ - REG_SET_3(FORMAT_CONTROL, 0, - CNVC_BYPASS, 0, - FORMAT_CONTROL__ALPHA_EN, 0, - FORMAT_EXPANSION_MODE, 0); - - /* COLOR_KEYER_CONTROL.COLOR_KEYER_EN = 0 this should be default */ - if (dpp->tf_mask->CM_BYPASS_EN) - REG_SET(CM_CONTROL, 0, CM_BYPASS_EN, 1); - else - REG_SET(CM_CONTROL, 0, CM_BYPASS, 1); - - /* Setting degamma bypass for now */ - REG_SET(CM_DGAM_CONTROL, 0, CM_DGAM_LUT_MODE, 0); -} - void dpp401_set_cursor_attributes( struct dpp *dpp_base, struct dc_cursor_attributes *cursor_attributes) diff --git a/drivers/gpu/drm/amd/display/dc/dsc/dcn401/dcn401_dsc.c b/drivers/gpu/drm/amd/display/dc/dsc/dcn401/dcn401_dsc.c index 4222679fd4c9ad00dc5f8a742347b3749514f5ef..af5de564faec570dcfec3415597115c2ced32e11 100644 --- a/drivers/gpu/drm/amd/display/dc/dsc/dcn401/dcn401_dsc.c +++ b/drivers/gpu/drm/amd/display/dc/dsc/dcn401/dcn401_dsc.c @@ -17,8 +17,6 @@ static void dsc_write_to_registers(struct display_stream_compressor *dsc, const /* Object I/F functions */ //static void dsc401_get_enc_caps(struct dsc_enc_caps *dsc_enc_caps, int pixel_clock_100Hz); //static bool dsc401_get_packed_pps(struct display_stream_compressor *dsc, const struct dsc_config *dsc_cfg, uint8_t *dsc_packed_pps); -static void dsc401_wait_disconnect_pending_clear(struct display_stream_compressor *dsc); -static void dsc401_get_enc_caps(struct dsc_enc_caps *dsc_enc_caps, int pixel_clock_100Hz); static const struct dsc_funcs dcn401_dsc_funcs = { .dsc_get_enc_caps = dsc401_get_enc_caps, @@ -66,7 +64,7 @@ void dsc401_construct(struct dcn401_dsc *dsc, dsc->max_image_width = 5184; } -static void dsc401_get_enc_caps(struct dsc_enc_caps *dsc_enc_caps, int pixel_clock_100Hz) +void dsc401_get_enc_caps(struct dsc_enc_caps *dsc_enc_caps, int pixel_clock_100Hz) { int min_dsc_unit_required = (pixel_clock_100Hz + MAX_THROUGHPUT_PER_DSC_100HZ - 1) / MAX_THROUGHPUT_PER_DSC_100HZ; @@ -191,7 +189,7 @@ void dsc401_disable(struct display_stream_compressor *dsc) DSC_CLOCK_EN, 0); } -static void dsc401_wait_disconnect_pending_clear(struct display_stream_compressor *dsc) +void dsc401_wait_disconnect_pending_clear(struct display_stream_compressor *dsc) { struct dcn401_dsc *dsc401 = TO_DCN401_DSC(dsc); diff --git a/drivers/gpu/drm/amd/display/dc/dsc/dcn401/dcn401_dsc.h b/drivers/gpu/drm/amd/display/dc/dsc/dcn401/dcn401_dsc.h index e3ca70058e643176503b353b0aea0a373b3ed292..1d927d8e83bf3ea90ee92fc6a91643a1e86d7995 100644 --- a/drivers/gpu/drm/amd/display/dc/dsc/dcn401/dcn401_dsc.h +++ b/drivers/gpu/drm/amd/display/dc/dsc/dcn401/dcn401_dsc.h @@ -341,5 +341,7 @@ void dsc401_set_config(struct display_stream_compressor *dsc, const struct dsc_c void dsc401_enable(struct display_stream_compressor *dsc, int opp_pipe); void dsc401_disable(struct display_stream_compressor *dsc); void dsc401_disconnect(struct display_stream_compressor *dsc); +void dsc401_get_enc_caps(struct dsc_enc_caps *dsc_enc_caps, int pixel_clock_100Hz); +void dsc401_wait_disconnect_pending_clear(struct display_stream_compressor *dsc); #endif diff --git a/drivers/gpu/drm/amd/display/dc/hubp/dcn10/dcn10_hubp.h b/drivers/gpu/drm/amd/display/dc/hubp/dcn10/dcn10_hubp.h index c7765e6f09e6dab7afad9fba34e7e02c40228dfb..f8f991785d4f8b7369365fe6e86f913707b84304 100644 --- a/drivers/gpu/drm/amd/display/dc/hubp/dcn10/dcn10_hubp.h +++ b/drivers/gpu/drm/amd/display/dc/hubp/dcn10/dcn10_hubp.h @@ -666,10 +666,29 @@ struct dcn_mi_mask { DCN_HUBP_REG_FIELD_LIST(uint32_t); }; +struct dcn_fl_regs_st { + uint32_t lut_enable; + uint32_t lut_done; + uint32_t lut_addr_mode; + uint32_t lut_width; + uint32_t lut_tmz; + uint32_t lut_crossbar_sel_r; + uint32_t lut_crossbar_sel_g; + uint32_t lut_crossbar_sel_b; + uint32_t lut_addr_hi; + uint32_t lut_addr_lo; + uint32_t refcyc_3dlut_group; + uint32_t lut_fl_bias; + uint32_t lut_fl_scale; + uint32_t lut_fl_mode; + uint32_t lut_fl_format; +}; + struct dcn_hubp_state { struct _vcs_dpi_display_dlg_regs_st dlg_attr; struct _vcs_dpi_display_ttu_regs_st ttu_attr; struct _vcs_dpi_display_rq_regs_st rq_regs; + struct dcn_fl_regs_st fl_regs; uint32_t pixel_format; uint32_t inuse_addr_hi; uint32_t inuse_addr_lo; diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c index e8730cc40edbe865e60cbb59d65e2942d66c124a..252e862449a2c53dcd55073f2d3b142de16591d3 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c @@ -1186,8 +1186,10 @@ void dce110_disable_stream(struct pipe_ctx *pipe_ctx) if (dccg) { dccg->funcs->disable_symclk32_se(dccg, dp_hpo_inst); dccg->funcs->set_dpstreamclk(dccg, REFCLK, tg->inst, dp_hpo_inst); - if (dccg && dccg->funcs->set_dtbclk_dto) - dccg->funcs->set_dtbclk_dto(dccg, &dto_params); + if (!(dc->ctx->dce_version >= DCN_VERSION_3_5)) { + if (dccg && dccg->funcs->set_dtbclk_dto) + dccg->funcs->set_dtbclk_dto(dccg, &dto_params); + } } } else if (dccg && dccg->funcs->disable_symclk_se) { dccg->funcs->disable_symclk_se(dccg, stream_enc->stream_enc_inst, @@ -1225,7 +1227,7 @@ void dce110_blank_stream(struct pipe_ctx *pipe_ctx) return; if (link->local_sink && link->local_sink->sink_signal == SIGNAL_TYPE_EDP) { - if (!link->skip_implict_edp_power_control) + if (!link->skip_implict_edp_power_control && hws) hws->funcs.edp_backlight_control(link, false); link->dc->hwss.set_abm_immediate_disable(pipe_ctx); } @@ -1379,7 +1381,7 @@ static void populate_audio_dp_link_info( } } -static void build_audio_output( +void build_audio_output( struct dc_state *state, const struct pipe_ctx *pipe_ctx, struct audio_output *audio_output) diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h index 06789ac3a2245da193b502e898e02b1c6882fa25..7cd8c157698816f3fa089809fdf13599da7da753 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h @@ -110,5 +110,9 @@ void dce110_enable_dp_link_output( enum signal_type signal, enum clock_source_id clock_source, const struct dc_link_settings *link_settings); +void build_audio_output( + struct dc_state *state, + const struct pipe_ctx *pipe_ctx, + struct audio_output *audio_output); #endif /* __DC_HWSS_DCE110_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c index f9ee55998b6b2442e67d88a002efdf2764ab87bf..9ad29be925e2fe8725e115d2a8f0663602ee12bf 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c @@ -327,6 +327,35 @@ static void dcn10_log_hubp_states(struct dc *dc, void *log_ctx) } } + DTN_INFO("\n=======HUBP FL======\n"); + DTN_INFO( + "HUBP FL: Enabled Done adr_mode width tmz xbar_sel_R xbar_sel_G xbar_sel_B adr_hi adr_low REFCYC Bias Scale Mode Format\n"); + for (i = 0; i < pool->pipe_count; i++) { + struct dcn_hubp_state *s = &(TO_DCN10_HUBP(pool->hubps[i])->state); + struct dcn_fl_regs_st *fl_regs = &s->fl_regs; + + if (!s->blank_en) { + DTN_INFO("[%2d]: %5xh %6xh %5d %6d %2xh %2xh %6xh %6d %8d %8d %7d %8xh %5x %5x %5x", + pool->hubps[i]->inst, + fl_regs->lut_enable, + fl_regs->lut_done, + fl_regs->lut_addr_mode, + fl_regs->lut_width, + fl_regs->lut_tmz, + fl_regs->lut_crossbar_sel_r, + fl_regs->lut_crossbar_sel_g, + fl_regs->lut_crossbar_sel_b, + fl_regs->lut_addr_hi, + fl_regs->lut_addr_lo, + fl_regs->refcyc_3dlut_group, + fl_regs->lut_fl_bias, + fl_regs->lut_fl_scale, + fl_regs->lut_fl_mode, + fl_regs->lut_fl_format); + DTN_INFO("\n"); + } + } + DTN_INFO("\n=========RQ========\n"); DTN_INFO("HUBP: drq_exp_m prq_exp_m mrq_exp_m crq_exp_m plane1_ba L:chunk_s min_chu_s meta_ch_s" " min_m_c_s dpte_gr_s mpte_gr_s swath_hei pte_row_h C:chunk_s min_chu_s meta_ch_s" @@ -511,6 +540,36 @@ static void dcn10_log_color_state(struct dc *dc, dc->caps.color.mpc.num_3dluts, dc->caps.color.mpc.ogam_ram, dc->caps.color.mpc.ocsc); + DTN_INFO("===== MPC RMCM 3DLUT =====\n"); + DTN_INFO("MPCC: SIZE MODE MODE_CUR RD_SEL 30BIT_EN WR_EN_MASK RAM_SEL OUT_NORM_FACTOR FL_SEL OUT_OFFSET OUT_SCALE FL_DONE SOFT_UNDERFLOW HARD_UNDERFLOW MEM_PWR_ST FORCE DIS MODE\n"); + for (i = 0; i < pool->mpcc_count; i++) { + struct mpcc_state s = {0}; + + pool->mpc->funcs->read_mpcc_state(pool->mpc, i, &s); + if (s.opp_id != 0xf) + DTN_INFO("[%2d]: %4xh %4xh %6xh %4x %4x %4x %4x %4x %4xh %4xh %6xh %4x %4x %4x %4x %4x %4x %4x\n", + i, s.rmcm_regs.rmcm_3dlut_size, s.rmcm_regs.rmcm_3dlut_mode, s.rmcm_regs.rmcm_3dlut_mode_cur, + s.rmcm_regs.rmcm_3dlut_read_sel, s.rmcm_regs.rmcm_3dlut_30bit_en, s.rmcm_regs.rmcm_3dlut_wr_en_mask, + s.rmcm_regs.rmcm_3dlut_ram_sel, s.rmcm_regs.rmcm_3dlut_out_norm_factor, s.rmcm_regs.rmcm_3dlut_fl_sel, + s.rmcm_regs.rmcm_3dlut_out_offset_r, s.rmcm_regs.rmcm_3dlut_out_scale_r, s.rmcm_regs.rmcm_3dlut_fl_done, + s.rmcm_regs.rmcm_3dlut_fl_soft_underflow, s.rmcm_regs.rmcm_3dlut_fl_hard_underflow, s.rmcm_regs.rmcm_3dlut_mem_pwr_state, + s.rmcm_regs.rmcm_3dlut_mem_pwr_force, s.rmcm_regs.rmcm_3dlut_mem_pwr_dis, s.rmcm_regs.rmcm_3dlut_mem_pwr_mode); + } + DTN_INFO("\n"); + DTN_INFO("===== MPC RMCM Shaper =====\n"); + DTN_INFO("MPCC: CNTL LUT_MODE MODE_CUR WR_EN_MASK WR_SEL OFFSET SCALE START_B START_SEG_B END_B END_BASE_B MEM_PWR_ST FORCE DIS MODE\n"); + for (i = 0; i < pool->mpcc_count; i++) { + struct mpcc_state s = {0}; + + pool->mpc->funcs->read_mpcc_state(pool->mpc, i, &s); + if (s.opp_id != 0xf) + DTN_INFO("[%2d]: %4xh %4xh %6xh %4x %4x %4x %4x %4x %4xh %4xh %6xh %4x %4x %4x %4x\n", + i, s.rmcm_regs.rmcm_cntl, s.rmcm_regs.rmcm_shaper_lut_mode, s.rmcm_regs.rmcm_shaper_mode_cur, + s.rmcm_regs.rmcm_shaper_lut_write_en_mask, s.rmcm_regs.rmcm_shaper_lut_write_sel, s.rmcm_regs.rmcm_shaper_offset_b, + s.rmcm_regs.rmcm_shaper_scale_b, s.rmcm_regs.rmcm_shaper_rama_exp_region_start_b, s.rmcm_regs.rmcm_shaper_rama_exp_region_start_seg_b, + s.rmcm_regs.rmcm_shaper_rama_exp_region_end_b, s.rmcm_regs.rmcm_shaper_rama_exp_region_end_base_b, s.rmcm_regs.rmcm_shaper_mem_pwr_state, + s.rmcm_regs.rmcm_shaper_mem_pwr_force, s.rmcm_regs.rmcm_shaper_mem_pwr_dis, s.rmcm_regs.rmcm_shaper_mem_pwr_mode); + } } void dcn10_log_hw_state(struct dc *dc, diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c index c277df12c8172e4db9832d9fe14b71712a62f0d8..3207addbd4ebb3ce2640ebbec02b27a74c7f21da 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c @@ -283,14 +283,13 @@ void dcn20_setup_gsl_group_as_lock( } /* at this point we want to program whether it's to enable or disable */ - if (pipe_ctx->stream_res.tg->funcs->set_gsl != NULL && - pipe_ctx->stream_res.tg->funcs->set_gsl_source_select != NULL) { + if (pipe_ctx->stream_res.tg->funcs->set_gsl != NULL) { pipe_ctx->stream_res.tg->funcs->set_gsl( pipe_ctx->stream_res.tg, &gsl); - - pipe_ctx->stream_res.tg->funcs->set_gsl_source_select( - pipe_ctx->stream_res.tg, group_idx, enable ? 4 : 0); + if (pipe_ctx->stream_res.tg->funcs->set_gsl_source_select != NULL) + pipe_ctx->stream_res.tg->funcs->set_gsl_source_select( + pipe_ctx->stream_res.tg, group_idx, enable ? 4 : 0); } else BREAK_TO_DEBUGGER(); } @@ -956,7 +955,7 @@ enum dc_status dcn20_enable_stream_timing( return DC_ERROR_UNEXPECTED; } - hws->funcs.wait_for_blank_complete(pipe_ctx->stream_res.opp); + fsleep(stream->timing.v_total * (stream->timing.h_total * 10000u / stream->timing.pix_clk_100hz)); params.vertical_total_min = stream->adjust.v_total_min; params.vertical_total_max = stream->adjust.v_total_max; @@ -1971,14 +1970,6 @@ static void dcn20_program_pipe( pipe_ctx->plane_state->update_flags.bits.hdr_mult)) hws->funcs.set_hdr_multiplier(pipe_ctx); - if (hws->funcs.populate_mcm_luts) { - if (pipe_ctx->plane_state) { - hws->funcs.populate_mcm_luts(dc, pipe_ctx, pipe_ctx->plane_state->mcm_luts, - pipe_ctx->plane_state->lut_bank_a); - pipe_ctx->plane_state->lut_bank_a = !pipe_ctx->plane_state->lut_bank_a; - } - } - if (pipe_ctx->plane_state && (pipe_ctx->plane_state->update_flags.bits.in_transfer_func_change || pipe_ctx->plane_state->update_flags.bits.gamma_change || @@ -2492,7 +2483,7 @@ bool dcn20_update_bandwidth( struct dce_hwseq *hws = dc->hwseq; /* recalculate DML parameters */ - if (dc->res_pool->funcs->validate_bandwidth(dc, context, false) != DC_OK) + if (dc->res_pool->funcs->validate_bandwidth(dc, context, DC_VALIDATE_MODE_AND_PROGRAMMING) != DC_OK) return false; /* apply updated bandwidth parameters */ @@ -2816,6 +2807,8 @@ void dcn20_reset_back_end_for_pipe( { struct dc_link *link = pipe_ctx->stream->link; const struct link_hwss *link_hwss = get_link_hwss(link, &pipe_ctx->link_res); + struct dccg *dccg = dc->res_pool->dccg; + struct dtbclk_dto_params dto_params = {0}; DC_LOGGER_INIT(dc->ctx->logger); if (pipe_ctx->stream_res.stream_enc == NULL) { @@ -2876,6 +2869,13 @@ void dcn20_reset_back_end_for_pipe( &pipe_ctx->link_res, pipe_ctx->stream->signal); link->phy_state.symclk_state = SYMCLK_OFF_TX_OFF; } + if (dc->link_srv->dp_is_128b_132b_signal(pipe_ctx) && dccg + && dc->ctx->dce_version >= DCN_VERSION_3_5) { + dto_params.otg_inst = pipe_ctx->stream_res.tg->inst; + dto_params.timing = &pipe_ctx->stream->timing; + if (dccg && dccg->funcs->set_dtbclk_dto) + dccg->funcs->set_dtbclk_dto(dccg, &dto_params); + } } /* diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c index e68f21fd5f0fb498699d8ecd688fb552e1577a86..a40e119d8582e6f932e6c86794d793e7c517efcf 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c @@ -55,15 +55,15 @@ #include "dcn20/dcn20_optc.h" #include "dcn30/dcn30_cm_common.h" -#define DC_LOGGER_INIT(logger) +#define DC_LOGGER_INIT(logger) \ + struct dal_logger *dc_logger = logger #define CTX \ hws->ctx #define REG(reg)\ hws->regs->reg #define DC_LOGGER \ - stream->ctx->logger - + dc_logger #undef FN #define FN(reg_name, field_name) \ @@ -76,6 +76,8 @@ static void update_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable) struct pipe_ctx *odm_pipe; int opp_cnt = 1; + DC_LOGGER_INIT(stream->ctx->logger); + ASSERT(dsc); for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) opp_cnt++; @@ -528,3 +530,32 @@ void dcn314_disable_link_output(struct dc_link *link, apply_symclk_on_tx_off_wa(link); } + + +void dcn314_plane_atomic_power_down(struct dc *dc, + struct dpp *dpp, + struct hubp *hubp) +{ + struct dce_hwseq *hws = dc->hwseq; + DC_LOGGER_INIT(dc->ctx->logger); + + if (REG(DC_IP_REQUEST_CNTL)) { + REG_SET(DC_IP_REQUEST_CNTL, 0, IP_REQUEST_EN, 1); + + if (hws->funcs.dpp_pg_control) { + hws->funcs.dpp_pg_control(hws, dpp->inst, false); + dpp->funcs->dpp_reset(dpp); + } + + if (hws->funcs.hubp_pg_control) { + hws->funcs.hubp_pg_control(hws, hubp->inst, false); + hubp->funcs->hubp_reset(hubp); + } + + REG_SET(DC_IP_REQUEST_CNTL, 0, IP_REQUEST_EN, 0); + DC_LOG_DEBUG("Power gated front end %d\n", hubp->inst); + } + + if (hws->funcs.dpp_root_clock_control) + hws->funcs.dpp_root_clock_control(hws, dpp->inst, false); +} diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.h index 2305ad282f218b3d304ca92b0e5e7e2d61695d6a..12a57b79edfbb27a35f2418831f87b2078db5e80 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.h @@ -47,4 +47,6 @@ void dcn314_dpp_root_clock_control(struct dce_hwseq *hws, unsigned int dpp_inst, void dcn314_disable_link_output(struct dc_link *link, const struct link_resource *link_res, enum signal_type signal); +void dcn314_plane_atomic_power_down(struct dc *dc, struct dpp *dpp, struct hubp *hubp); + #endif /* __DC_HWSS_DCN314_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_init.c index f5112742edf9b4e7722bbbbb041e81551644e121..6963d25608ac0d0f2172fd2cf7643a53100b5774 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_init.c @@ -137,7 +137,7 @@ static const struct hwseq_private_funcs dcn314_private_funcs = { .disable_vga = dcn20_disable_vga, .bios_golden_init = dcn10_bios_golden_init, .plane_atomic_disable = dcn20_plane_atomic_disable, - .plane_atomic_power_down = dcn10_plane_atomic_power_down, + .plane_atomic_power_down = dcn314_plane_atomic_power_down, .enable_power_gating_plane = dcn314_enable_power_gating_plane, .dpp_root_clock_control = dcn314_dpp_root_clock_control, .hubp_pg_control = dcn31_hubp_pg_control, diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c index a0b05b9ef660898a20073352ba8d0c03d83889e9..416b1dca3dac919f868074427c449d5c3ef1d5a2 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c @@ -1063,15 +1063,17 @@ void dcn32_update_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable) dsc_cfg.dc_dsc_cfg.num_slices_h /= opp_cnt; if (should_use_dto_dscclk) - dccg->funcs->set_dto_dscclk(dccg, dsc->inst); + dccg->funcs->set_dto_dscclk(dccg, dsc->inst, dsc_cfg.dc_dsc_cfg.num_slices_h); dsc->funcs->dsc_set_config(dsc, &dsc_cfg, &dsc_optc_cfg); dsc->funcs->dsc_enable(dsc, pipe_ctx->stream_res.opp->inst); for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { struct display_stream_compressor *odm_dsc = odm_pipe->stream_res.dsc; ASSERT(odm_dsc); + if (!odm_dsc) + continue; if (should_use_dto_dscclk) - dccg->funcs->set_dto_dscclk(dccg, odm_dsc->inst); + dccg->funcs->set_dto_dscclk(dccg, odm_dsc->inst, dsc_cfg.dc_dsc_cfg.num_slices_h); odm_dsc->funcs->dsc_set_config(odm_dsc, &dsc_cfg, &dsc_optc_cfg); odm_dsc->funcs->dsc_enable(odm_dsc, odm_pipe->stream_res.opp->inst); } diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c index c814d957305adae7e20db81d37b2b8df068f0f98..a267f574b61937cbc8ce67169c76ea199c86137e 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c @@ -1047,6 +1047,15 @@ void dcn35_calc_blocks_to_gate(struct dc *dc, struct dc_state *context, if (dc->caps.sequential_ono) { update_state->pg_pipe_res_update[PG_HUBP][pipe_ctx->stream_res.dsc->inst] = false; update_state->pg_pipe_res_update[PG_DPP][pipe_ctx->stream_res.dsc->inst] = false; + + /* All HUBP/DPP instances must be powered if the DSC inst != HUBP inst */ + if (!pipe_ctx->top_pipe && pipe_ctx->plane_res.hubp && + pipe_ctx->plane_res.hubp->inst != pipe_ctx->stream_res.dsc->inst) { + for (j = 0; j < dc->res_pool->pipe_count; ++j) { + update_state->pg_pipe_res_update[PG_HUBP][j] = false; + update_state->pg_pipe_res_update[PG_DPP][j] = false; + } + } } } @@ -1193,6 +1202,25 @@ void dcn35_calc_blocks_to_ungate(struct dc *dc, struct dc_state *context, update_state->pg_pipe_res_update[PG_HDMISTREAM][0] = true; if (dc->caps.sequential_ono) { + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *new_pipe = &context->res_ctx.pipe_ctx[i]; + + if (new_pipe->stream_res.dsc && !new_pipe->top_pipe && + update_state->pg_pipe_res_update[PG_DSC][new_pipe->stream_res.dsc->inst]) { + update_state->pg_pipe_res_update[PG_HUBP][new_pipe->stream_res.dsc->inst] = true; + update_state->pg_pipe_res_update[PG_DPP][new_pipe->stream_res.dsc->inst] = true; + + /* All HUBP/DPP instances must be powered if the DSC inst != HUBP inst */ + if (new_pipe->plane_res.hubp && + new_pipe->plane_res.hubp->inst != new_pipe->stream_res.dsc->inst) { + for (j = 0; j < dc->res_pool->pipe_count; ++j) { + update_state->pg_pipe_res_update[PG_HUBP][j] = true; + update_state->pg_pipe_res_update[PG_DPP][j] = true; + } + } + } + } + for (i = dc->res_pool->pipe_count - 1; i >= 0; i--) { if (update_state->pg_pipe_res_update[PG_HUBP][i] && update_state->pg_pipe_res_update[PG_DPP][i]) { diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c index c4177a9a662facb404c3f252d3bb99244595b3e6..a0d61df07f22a6c6f22ab6b057bc82aa70a98465 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c @@ -2,6 +2,8 @@ // // Copyright 2024 Advanced Micro Devices, Inc. + +#include "os_types.h" #include "dm_services.h" #include "basics/dc_common.h" #include "dm_helpers.h" @@ -143,13 +145,8 @@ void dcn401_init_hw(struct dc *dc) dc->clk_mgr->funcs->init_clocks(dc->clk_mgr); // mark dcmode limits present if any clock has distinct AC and DC values from SMU - dc->caps.dcmode_power_limits_present = - (dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_dcfclk_levels && dc->clk_mgr->bw_params->dc_mode_limit.dcfclk_mhz) || - (dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_dispclk_levels && dc->clk_mgr->bw_params->dc_mode_limit.dispclk_mhz) || - (dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_dtbclk_levels && dc->clk_mgr->bw_params->dc_mode_limit.dtbclk_mhz) || - (dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_fclk_levels && dc->clk_mgr->bw_params->dc_mode_limit.fclk_mhz) || - (dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_memclk_levels && dc->clk_mgr->bw_params->dc_mode_limit.memclk_mhz) || - (dc->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_socclk_levels && dc->clk_mgr->bw_params->dc_mode_limit.socclk_mhz); + dc->caps.dcmode_power_limits_present = dc->clk_mgr->funcs->is_dc_mode_present && + dc->clk_mgr->funcs->is_dc_mode_present(dc->clk_mgr); } // Initialize the dccg @@ -396,249 +393,6 @@ static void dcn401_get_mcm_lut_xable_from_pipe_ctx(struct dc *dc, struct pipe_ct } } -static void dcn401_set_mcm_location_post_blend(struct dc *dc, struct pipe_ctx *pipe_ctx, bool bPostBlend) -{ - struct mpc *mpc = dc->res_pool->mpc; - int mpcc_id = pipe_ctx->plane_res.hubp->inst; - - if (!pipe_ctx->plane_state) - return; - - mpc->funcs->set_movable_cm_location(mpc, MPCC_MOVABLE_CM_LOCATION_BEFORE, mpcc_id); - pipe_ctx->plane_state->mcm_location = (bPostBlend) ? - MPCC_MOVABLE_CM_LOCATION_AFTER : - MPCC_MOVABLE_CM_LOCATION_BEFORE; -} - -static void dc_get_lut_mode( - enum dc_cm2_gpu_mem_layout layout, - enum hubp_3dlut_fl_mode *mode, - enum hubp_3dlut_fl_addressing_mode *addr_mode) -{ - switch (layout) { - case DC_CM2_GPU_MEM_LAYOUT_3D_SWIZZLE_LINEAR_RGB: - *mode = hubp_3dlut_fl_mode_native_1; - *addr_mode = hubp_3dlut_fl_addressing_mode_sw_linear; - break; - case DC_CM2_GPU_MEM_LAYOUT_3D_SWIZZLE_LINEAR_BGR: - *mode = hubp_3dlut_fl_mode_native_2; - *addr_mode = hubp_3dlut_fl_addressing_mode_sw_linear; - break; - case DC_CM2_GPU_MEM_LAYOUT_1D_PACKED_LINEAR: - *mode = hubp_3dlut_fl_mode_transform; - *addr_mode = hubp_3dlut_fl_addressing_mode_simple_linear; - break; - default: - *mode = hubp_3dlut_fl_mode_disable; - *addr_mode = hubp_3dlut_fl_addressing_mode_sw_linear; - break; - } -} - -static void dc_get_lut_format( - enum dc_cm2_gpu_mem_format dc_format, - enum hubp_3dlut_fl_format *format) -{ - switch (dc_format) { - case DC_CM2_GPU_MEM_FORMAT_16161616_UNORM_12MSB: - *format = hubp_3dlut_fl_format_unorm_12msb_bitslice; - break; - case DC_CM2_GPU_MEM_FORMAT_16161616_UNORM_12LSB: - *format = hubp_3dlut_fl_format_unorm_12lsb_bitslice; - break; - case DC_CM2_GPU_MEM_FORMAT_16161616_FLOAT_FP1_5_10: - *format = hubp_3dlut_fl_format_float_fp1_5_10; - break; - } -} - -static void dc_get_lut_xbar( - enum dc_cm2_gpu_mem_pixel_component_order order, - enum hubp_3dlut_fl_crossbar_bit_slice *cr_r, - enum hubp_3dlut_fl_crossbar_bit_slice *y_g, - enum hubp_3dlut_fl_crossbar_bit_slice *cb_b) -{ - switch (order) { - case DC_CM2_GPU_MEM_PIXEL_COMPONENT_ORDER_RGBA: - *cr_r = hubp_3dlut_fl_crossbar_bit_slice_32_47; - *y_g = hubp_3dlut_fl_crossbar_bit_slice_16_31; - *cb_b = hubp_3dlut_fl_crossbar_bit_slice_0_15; - break; - case DC_CM2_GPU_MEM_PIXEL_COMPONENT_ORDER_BGRA: - *cr_r = hubp_3dlut_fl_crossbar_bit_slice_0_15; - *y_g = hubp_3dlut_fl_crossbar_bit_slice_16_31; - *cb_b = hubp_3dlut_fl_crossbar_bit_slice_32_47; - break; - } -} - -static void dc_get_lut_width( - enum dc_cm2_gpu_mem_size size, - enum hubp_3dlut_fl_width *width) -{ - switch (size) { - case DC_CM2_GPU_MEM_SIZE_333333: - *width = hubp_3dlut_fl_width_33; - break; - case DC_CM2_GPU_MEM_SIZE_171717: - *width = hubp_3dlut_fl_width_17; - break; - case DC_CM2_GPU_MEM_SIZE_TRANSFORMED: - *width = hubp_3dlut_fl_width_transformed; - break; - } -} -static bool dc_is_rmcm_3dlut_supported(struct hubp *hubp, struct mpc *mpc) -{ - if (mpc->funcs->rmcm.update_3dlut_fast_load_select && - mpc->funcs->rmcm.program_lut_read_write_control && - hubp->funcs->hubp_program_3dlut_fl_addr && - mpc->funcs->rmcm.program_bit_depth && - hubp->funcs->hubp_program_3dlut_fl_mode && - hubp->funcs->hubp_program_3dlut_fl_addressing_mode && - hubp->funcs->hubp_program_3dlut_fl_format && - hubp->funcs->hubp_update_3dlut_fl_bias_scale && - mpc->funcs->rmcm.program_bias_scale && - hubp->funcs->hubp_program_3dlut_fl_crossbar && - hubp->funcs->hubp_program_3dlut_fl_width && - mpc->funcs->rmcm.update_3dlut_fast_load_select && - mpc->funcs->rmcm.populate_lut && - mpc->funcs->rmcm.program_lut_mode && - hubp->funcs->hubp_enable_3dlut_fl && - mpc->funcs->rmcm.enable_3dlut_fl) - return true; - - return false; -} - -bool dcn401_program_rmcm_luts( - struct hubp *hubp, - struct pipe_ctx *pipe_ctx, - enum dc_cm2_transfer_func_source lut3d_src, - struct dc_cm2_func_luts *mcm_luts, - struct mpc *mpc, - bool lut_bank_a, - int mpcc_id) -{ - struct dpp *dpp_base = pipe_ctx->plane_res.dpp; - union mcm_lut_params m_lut_params; - enum MCM_LUT_XABLE shaper_xable, lut3d_xable = MCM_LUT_DISABLE, lut1d_xable; - enum hubp_3dlut_fl_mode mode; - enum hubp_3dlut_fl_addressing_mode addr_mode; - enum hubp_3dlut_fl_format format = 0; - enum hubp_3dlut_fl_crossbar_bit_slice crossbar_bit_slice_y_g = 0; - enum hubp_3dlut_fl_crossbar_bit_slice crossbar_bit_slice_cb_b = 0; - enum hubp_3dlut_fl_crossbar_bit_slice crossbar_bit_slice_cr_r = 0; - enum hubp_3dlut_fl_width width = 0; - struct dc *dc = hubp->ctx->dc; - - bool bypass_rmcm_3dlut = false; - bool bypass_rmcm_shaper = false; - - dcn401_get_mcm_lut_xable_from_pipe_ctx(dc, pipe_ctx, &shaper_xable, &lut3d_xable, &lut1d_xable); - - /* 3DLUT */ - switch (lut3d_src) { - case DC_CM2_TRANSFER_FUNC_SOURCE_SYSMEM: - memset(&m_lut_params, 0, sizeof(m_lut_params)); - // Don't know what to do in this case. - //case DC_CM2_TRANSFER_FUNC_SOURCE_SYSMEM: - break; - case DC_CM2_TRANSFER_FUNC_SOURCE_VIDMEM: - dc_get_lut_width(mcm_luts->lut3d_data.gpu_mem_params.size, &width); - if (!dc_is_rmcm_3dlut_supported(hubp, mpc) || - !mpc->funcs->rmcm.is_config_supported(width)) - return false; - - //0. disable fl on mpc - mpc->funcs->update_3dlut_fast_load_select(mpc, mpcc_id, 0xF); - - //1. power down the block - mpc->funcs->rmcm.power_on_shaper_3dlut(mpc, mpcc_id, false); - - //2. program RMCM - //2a. 3dlut reg programming - mpc->funcs->rmcm.program_lut_read_write_control(mpc, MCM_LUT_3DLUT, lut_bank_a, - (!bypass_rmcm_3dlut) && lut3d_xable != MCM_LUT_DISABLE, mpcc_id); - - hubp->funcs->hubp_program_3dlut_fl_addr(hubp, - mcm_luts->lut3d_data.gpu_mem_params.addr); - - mpc->funcs->rmcm.program_bit_depth(mpc, - mcm_luts->lut3d_data.gpu_mem_params.bit_depth, mpcc_id); - - // setting native or transformed mode, - dc_get_lut_mode(mcm_luts->lut3d_data.gpu_mem_params.layout, &mode, &addr_mode); - - //these program the mcm 3dlut - hubp->funcs->hubp_program_3dlut_fl_mode(hubp, mode); - - hubp->funcs->hubp_program_3dlut_fl_addressing_mode(hubp, addr_mode); - - //seems to be only for the MCM - dc_get_lut_format(mcm_luts->lut3d_data.gpu_mem_params.format_params.format, &format); - hubp->funcs->hubp_program_3dlut_fl_format(hubp, format); - - mpc->funcs->rmcm.program_bias_scale(mpc, - mcm_luts->lut3d_data.gpu_mem_params.format_params.float_params.bias, - mcm_luts->lut3d_data.gpu_mem_params.format_params.float_params.scale, - mpcc_id); - hubp->funcs->hubp_update_3dlut_fl_bias_scale(hubp, - mcm_luts->lut3d_data.gpu_mem_params.format_params.float_params.bias, - mcm_luts->lut3d_data.gpu_mem_params.format_params.float_params.scale); - - dc_get_lut_xbar( - mcm_luts->lut3d_data.gpu_mem_params.component_order, - &crossbar_bit_slice_cr_r, - &crossbar_bit_slice_y_g, - &crossbar_bit_slice_cb_b); - - hubp->funcs->hubp_program_3dlut_fl_crossbar(hubp, - crossbar_bit_slice_cr_r, - crossbar_bit_slice_y_g, - crossbar_bit_slice_cb_b); - - mpc->funcs->rmcm.program_3dlut_size(mpc, width, mpcc_id); - - mpc->funcs->update_3dlut_fast_load_select(mpc, mpcc_id, hubp->inst); - - //2b. shaper reg programming - memset(&m_lut_params, 0, sizeof(m_lut_params)); - - if (mcm_luts->shaper->type == TF_TYPE_HWPWL) { - m_lut_params.pwl = &mcm_luts->shaper->pwl; - } else if (mcm_luts->shaper->type == TF_TYPE_DISTRIBUTED_POINTS) { - ASSERT(false); - cm_helper_translate_curve_to_hw_format( - dc->ctx, - mcm_luts->shaper, - &dpp_base->regamma_params, true); - m_lut_params.pwl = &dpp_base->regamma_params; - } - if (m_lut_params.pwl) { - mpc->funcs->rmcm.populate_lut(mpc, m_lut_params, lut_bank_a, mpcc_id); - mpc->funcs->rmcm.program_lut_mode(mpc, !bypass_rmcm_shaper, lut_bank_a, mpcc_id); - } else { - //RMCM 3dlut won't work without its shaper - return false; - } - - //3. Select the hubp connected to this RMCM - hubp->funcs->hubp_enable_3dlut_fl(hubp, true); - mpc->funcs->rmcm.enable_3dlut_fl(mpc, true, mpcc_id); - - //4. power on the block - if (m_lut_params.pwl) - mpc->funcs->rmcm.power_on_shaper_3dlut(mpc, mpcc_id, true); - - break; - default: - return false; - } - - return true; -} - void dcn401_populate_mcm_luts(struct dc *dc, struct pipe_ctx *pipe_ctx, struct dc_cm2_func_luts mcm_luts, @@ -664,25 +418,6 @@ void dcn401_populate_mcm_luts(struct dc *dc, dcn401_get_mcm_lut_xable_from_pipe_ctx(dc, pipe_ctx, &shaper_xable, &lut3d_xable, &lut1d_xable); - //MCM - setting its location (Before/After) blender - //set to post blend (true) - dcn401_set_mcm_location_post_blend( - dc, - pipe_ctx, - mcm_luts.lut3d_data.mpc_mcm_post_blend); - - //RMCM - 3dLUT+Shaper - if (mcm_luts.lut3d_data.rmcm_3dlut_enable) { - dcn401_program_rmcm_luts( - hubp, - pipe_ctx, - lut3d_src, - &mcm_luts, - mpc, - lut_bank_a, - mpcc_id); - } - /* 1D LUT */ if (mcm_luts.lut1d_func) { memset(&m_lut_params, 0, sizeof(m_lut_params)); @@ -740,15 +475,15 @@ void dcn401_populate_mcm_luts(struct dc *dc, break; case DC_CM2_TRANSFER_FUNC_SOURCE_VIDMEM: switch (mcm_luts.lut3d_data.gpu_mem_params.size) { - case DC_CM2_GPU_MEM_SIZE_333333: - width = hubp_3dlut_fl_width_33; - break; case DC_CM2_GPU_MEM_SIZE_171717: width = hubp_3dlut_fl_width_17; break; case DC_CM2_GPU_MEM_SIZE_TRANSFORMED: width = hubp_3dlut_fl_width_transformed; break; + default: + //TODO: handle default case + break; } //check for support @@ -817,11 +552,14 @@ void dcn401_populate_mcm_luts(struct dc *dc, //navi 4x has a bug and r and blue are swapped and need to be worked around here in //TODO: need to make a method for get_xbar per asic OR do the workaround in program_crossbar for 4x - dc_get_lut_xbar( - mcm_luts.lut3d_data.gpu_mem_params.component_order, - &crossbar_bit_slice_cr_r, - &crossbar_bit_slice_y_g, - &crossbar_bit_slice_cb_b); + switch (mcm_luts.lut3d_data.gpu_mem_params.component_order) { + case DC_CM2_GPU_MEM_PIXEL_COMPONENT_ORDER_RGBA: + default: + crossbar_bit_slice_cr_r = hubp_3dlut_fl_crossbar_bit_slice_0_15; + crossbar_bit_slice_y_g = hubp_3dlut_fl_crossbar_bit_slice_16_31; + crossbar_bit_slice_cb_b = hubp_3dlut_fl_crossbar_bit_slice_32_47; + break; + } if (hubp->funcs->hubp_program_3dlut_fl_crossbar) hubp->funcs->hubp_program_3dlut_fl_crossbar(hubp, @@ -2269,14 +2007,6 @@ void dcn401_program_pipe( pipe_ctx->plane_state->update_flags.bits.hdr_mult)) hws->funcs.set_hdr_multiplier(pipe_ctx); - if (hws->funcs.populate_mcm_luts) { - if (pipe_ctx->plane_state) { - hws->funcs.populate_mcm_luts(dc, pipe_ctx, pipe_ctx->plane_state->mcm_luts, - pipe_ctx->plane_state->lut_bank_a); - pipe_ctx->plane_state->lut_bank_a = !pipe_ctx->plane_state->lut_bank_a; - } - } - if (pipe_ctx->plane_state && (pipe_ctx->plane_state->update_flags.bits.in_transfer_func_change || pipe_ctx->plane_state->update_flags.bits.gamma_change || @@ -2651,7 +2381,7 @@ bool dcn401_update_bandwidth( struct dce_hwseq *hws = dc->hwseq; /* recalculate DML parameters */ - if (dc->res_pool->funcs->validate_bandwidth(dc, context, false) != DC_OK) + if (dc->res_pool->funcs->validate_bandwidth(dc, context, DC_VALIDATE_MODE_AND_PROGRAMMING) != DC_OK) return false; /* apply updated bandwidth parameters */ diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.h index ce65b4f6c67273d7d218a61488630e4d29db1390..781cf0efccc6cdf395c971fade567691242623ac 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.h @@ -109,12 +109,4 @@ void dcn401_detect_pipe_changes( void dcn401_plane_atomic_power_down(struct dc *dc, struct dpp *dpp, struct hubp *hubp); -bool dcn401_program_rmcm_luts( - struct hubp *hubp, - struct pipe_ctx *pipe_ctx, - enum dc_cm2_transfer_func_source lut3d_src, - struct dc_cm2_func_luts *mcm_luts, - struct mpc *mpc, - bool lut_bank_a, - int mpcc_id); #endif /* __DC_HWSS_DCN401_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_status.h b/drivers/gpu/drm/amd/display/dc/inc/core_status.h index f3696143590c446d31cfd69f4cc19f655e450e9f..82085d9c3f407801f3b7921e6e8afd3d92889bc0 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/core_status.h +++ b/drivers/gpu/drm/amd/display/dc/inc/core_status.h @@ -59,6 +59,7 @@ enum dc_status { DC_FAIL_DP_PAYLOAD_ALLOCATION = 27, DC_FAIL_DP_LINK_BANDWIDTH = 28, DC_FAIL_HW_CURSOR_SUPPORT = 29, + DC_FAIL_DP_TUNNEL_BW_VALIDATE = 30, DC_ERROR_UNEXPECTED = -1 }; diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h index 0cf349cafb3e56b70315dc048daf5902cf585d60..c9454fe1cd058e45fa711e61f453f93e1a5d854c 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h +++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h @@ -82,7 +82,7 @@ struct resource_funcs { enum dc_status (*validate_bandwidth)( struct dc *dc, struct dc_state *context, - bool fast_validate); + enum dc_validate_mode validate_mode); void (*calculate_wm_and_dlg)( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, @@ -107,7 +107,7 @@ struct resource_funcs { struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate); + enum dc_validate_mode validate_mode); /* * Algorithm for assigning available link encoders to links. @@ -223,6 +223,11 @@ struct resource_funcs { const struct dc_stream_state *stream); bool (*program_mcache_pipe_config)(struct dc_state *context, const struct dc_mcache_params *mcache_params); + enum dc_status (*update_dc_state_for_encoder_switch)(struct dc_link *link, + struct dc_link_settings *link_setting, + uint8_t pipe_count, + struct pipe_ctx *pipes, + struct audio_output *audio_output); }; struct audio_support{ @@ -556,7 +561,10 @@ struct dcn_bw_output { struct dml2_mcache_surface_allocation mcache_allocations[DML2_MAX_PLANES]; struct dmub_cmd_fams2_global_config fams2_global_config; union dmub_cmd_fams2_config fams2_stream_base_params[DML2_MAX_PLANES]; - union dmub_cmd_fams2_config fams2_stream_sub_params[DML2_MAX_PLANES]; + union { + union dmub_cmd_fams2_config fams2_stream_sub_params[DML2_MAX_PLANES]; + union dmub_fams2_stream_static_sub_state_v2 fams2_stream_sub_params_v2[DML2_MAX_PLANES]; + }; struct dml2_display_arb_regs arb_regs; }; diff --git a/drivers/gpu/drm/amd/display/dc/inc/dcn_calcs.h b/drivers/gpu/drm/amd/display/dc/inc/dcn_calcs.h index d19a595c2be408347719b7f712f5175487a253b1..134091d5842d0ddc393cd4b48145b72f9e9db279 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/dcn_calcs.h +++ b/drivers/gpu/drm/amd/display/dc/inc/dcn_calcs.h @@ -622,7 +622,7 @@ extern const struct dcn_ip_params dcn10_ip_defaults; bool dcn_validate_bandwidth( struct dc *dc, struct dc_state *context, - bool fast_validate); + enum dc_validate_mode validate_mode); void dcn_get_soc_clks( struct dc *dc, diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h index c14d64687a3d421a84060a19fc74db50ffe3c20a..3b736f4687a683ddf3bcddb75bb173590f350271 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h @@ -324,6 +324,8 @@ struct clk_mgr_funcs { int (*get_dispclk_from_dentist)(struct clk_mgr *clk_mgr_base); + bool (*is_dc_mode_present)(struct clk_mgr *clk_mgr); + }; struct clk_mgr { diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h index e94e9ba60f55a6906acf83aea6896f8c6c289f97..61c4d2a7db1cebd8ec49ad7f94c419bd89f37a06 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h @@ -211,7 +211,7 @@ struct dccg_funcs { struct dccg *dccg, enum streamclk_source src, uint32_t otg_inst); - void (*set_dto_dscclk)(struct dccg *dccg, uint32_t dsc_inst); + void (*set_dto_dscclk)(struct dccg *dccg, uint32_t dsc_inst, uint32_t num_slices_h); void (*set_ref_dscclk)(struct dccg *dccg, uint32_t dsc_inst); void (*dccg_root_gate_disable_control)(struct dccg *dccg, uint32_t pipe_idx, uint32_t disable_clock_gating); }; diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h b/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h index 6e303b81bfb0ff16629d0ac67c57847ae9db2a34..7641439f6ca0622d48cc74c48a55a07f1859238c 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h @@ -190,6 +190,42 @@ struct mpc_grph_gamut_adjustment { enum mpcc_gamut_remap_id mpcc_gamut_remap_block_id; }; +struct mpc_rmcm_regs { + uint32_t rmcm_3dlut_mem_pwr_state; + uint32_t rmcm_3dlut_mem_pwr_force; + uint32_t rmcm_3dlut_mem_pwr_dis; + uint32_t rmcm_3dlut_mem_pwr_mode; + uint32_t rmcm_3dlut_size; + uint32_t rmcm_3dlut_mode; + uint32_t rmcm_3dlut_mode_cur; + uint32_t rmcm_3dlut_read_sel; + uint32_t rmcm_3dlut_30bit_en; + uint32_t rmcm_3dlut_wr_en_mask; + uint32_t rmcm_3dlut_ram_sel; + uint32_t rmcm_3dlut_out_norm_factor; + uint32_t rmcm_3dlut_fl_sel; + uint32_t rmcm_3dlut_out_offset_r; + uint32_t rmcm_3dlut_out_scale_r; + uint32_t rmcm_3dlut_fl_done; + uint32_t rmcm_3dlut_fl_soft_underflow; + uint32_t rmcm_3dlut_fl_hard_underflow; + uint32_t rmcm_cntl; + uint32_t rmcm_shaper_mem_pwr_state; + uint32_t rmcm_shaper_mem_pwr_force; + uint32_t rmcm_shaper_mem_pwr_dis; + uint32_t rmcm_shaper_mem_pwr_mode; + uint32_t rmcm_shaper_lut_mode; + uint32_t rmcm_shaper_mode_cur; + uint32_t rmcm_shaper_lut_write_en_mask; + uint32_t rmcm_shaper_lut_write_sel; + uint32_t rmcm_shaper_offset_b; + uint32_t rmcm_shaper_scale_b; + uint32_t rmcm_shaper_rama_exp_region_start_b; + uint32_t rmcm_shaper_rama_exp_region_start_seg_b; + uint32_t rmcm_shaper_rama_exp_region_end_b; + uint32_t rmcm_shaper_rama_exp_region_end_base_b; +}; + struct mpcc_sm_cfg { bool enable; /* 0-single plane,2-row subsampling,4-column subsampling,6-checkboard subsampling */ @@ -301,6 +337,7 @@ struct mpcc_state { uint32_t rgam_mode; uint32_t rgam_lut; struct mpc_grph_gamut_adjustment gamut_remap; + struct mpc_rmcm_regs rmcm_regs; }; /** @@ -1038,6 +1075,11 @@ struct mpc_funcs { */ void (*program_3dlut_size)(struct mpc *mpc, bool is_17x17x17, int mpcc_id); + /** + * @mcm: + * + * MPC MCM new HW sequential programming functions + */ struct { void (*program_3dlut_size)(struct mpc *mpc, uint32_t width, int mpcc_id); void (*program_bias_scale)(struct mpc *mpc, uint16_t bias, uint16_t scale, int mpcc_id); @@ -1050,6 +1092,11 @@ struct mpc_funcs { bool lut_bank_a, int mpcc_id); } mcm; + /** + * @rmcm: + * + * MPC RMCM new HW sequential programming functions + */ struct { void (*enable_3dlut_fl)(struct mpc *mpc, bool enable, int mpcc_id); void (*update_3dlut_fast_load_select)(struct mpc *mpc, int mpcc_id, int hubp_idx); diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h b/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h index fe7f3137f2285e77072acf166b315c12e42cfa63..27f950ae45ee51bc3687635b02ca25a3591bfd97 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h @@ -117,6 +117,7 @@ struct stream_encoder { uint32_t stream_enc_inst; struct vpg *vpg; struct afmt *afmt; + struct apg *apg; }; struct enc_state { diff --git a/drivers/gpu/drm/amd/display/dc/inc/link.h b/drivers/gpu/drm/amd/display/dc/inc/link.h index 7d16351bba9902cc1ede31094d5884b578cb4b08..f2503402c10e74537878eaa5e5f490474a8ee13b 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/link.h +++ b/drivers/gpu/drm/amd/display/dc/inc/link.h @@ -144,9 +144,9 @@ struct link_service { uint32_t (*dp_link_bandwidth_kbps)( const struct dc_link *link, const struct dc_link_settings *link_settings); - bool (*validate_dpia_bandwidth)( - const struct dc_stream_state *stream, - const unsigned int num_streams); + enum dc_status (*validate_dp_tunnel_bandwidth)( + const struct dc *dc, + const struct dc_state *new_ctx); uint32_t (*dp_required_hblank_size_bytes)( const struct dc_link *link, diff --git a/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c b/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c index 96febabf464af0114bc5f733d6de2b16c653e464..2956c2b3ad1aad79f5142617a1529268bf6ab06a 100644 --- a/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c +++ b/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c @@ -34,6 +34,7 @@ #include "dm_helpers.h" #include "dc_dmub_srv.h" #include "dce/dmub_hw_lock_mgr.h" +#include "clk_mgr.h" #define DC_LOGGER \ link->ctx->logger @@ -67,10 +68,17 @@ static void dp_retrain_link_dp_test(struct dc_link *link, { struct pipe_ctx *pipes[MAX_PIPES]; struct dc_state *state = link->dc->current_state; + struct dc_stream_update stream_update = { 0 }; + bool dpms_off = false; + bool needs_divider_update = false; bool was_hpo_acquired = resource_is_hpo_acquired(link->dc->current_state); bool is_hpo_acquired; uint8_t count; int i; + struct audio_output audio_output[MAX_PIPES]; + + needs_divider_update = (link->dc->link_srv->dp_get_encoding_format(link_setting) != + link->dc->link_srv->dp_get_encoding_format((const struct dc_link_settings *) &link->cur_link_settings)); udelay(100); @@ -83,16 +91,59 @@ static void dp_retrain_link_dp_test(struct dc_link *link, link->dc, state, pipes[i]); + + // Disable OTG and re-enable after updating clocks + pipes[i]->stream_res.tg->funcs->disable_crtc(pipes[i]->stream_res.tg); } - if (link->dc->hwss.setup_hpo_hw_control) { - is_hpo_acquired = resource_is_hpo_acquired(state); - if (was_hpo_acquired != is_hpo_acquired) - link->dc->hwss.setup_hpo_hw_control(link->dc->hwseq, is_hpo_acquired); + if (needs_divider_update && link->dc->res_pool->funcs->update_dc_state_for_encoder_switch) { + link->dc->res_pool->funcs->update_dc_state_for_encoder_switch(link, + link_setting, count, + *pipes, &audio_output[0]); + for (i = 0; i < count; i++) { + pipes[i]->clock_source->funcs->program_pix_clk( + pipes[i]->clock_source, + &pipes[i]->stream_res.pix_clk_params, + link->dc->link_srv->dp_get_encoding_format(&pipes[i]->link_config.dp_link_settings), + &pipes[i]->pll_settings); + + if (pipes[i]->stream_res.audio != NULL) { + const struct link_hwss *link_hwss = get_link_hwss( + link, &pipes[i]->link_res); + + link_hwss->setup_audio_output(pipes[i], &audio_output[i], + pipes[i]->stream_res.audio->inst); + + pipes[i]->stream_res.audio->funcs->az_configure( + pipes[i]->stream_res.audio, + pipes[i]->stream->signal, + &audio_output[i].crtc_info, + &pipes[i]->stream->audio_info, + &audio_output[i].dp_link_info); + + if (link->dc->config.disable_hbr_audio_dp2 && + pipes[i]->stream_res.audio->funcs->az_disable_hbr_audio && + link->dc->link_srv->dp_is_128b_132b_signal(pipes[i])) + pipes[i]->stream_res.audio->funcs->az_disable_hbr_audio(pipes[i]->stream_res.audio); + } + } } - for (i = count-1; i >= 0; i--) - link_set_dpms_on(state, pipes[i]); + // Toggle on HPO I/O if necessary + is_hpo_acquired = resource_is_hpo_acquired(state); + if (was_hpo_acquired != is_hpo_acquired && link->dc->hwss.setup_hpo_hw_control) + link->dc->hwss.setup_hpo_hw_control(link->dc->hwseq, is_hpo_acquired); + + for (i = 0; i < count; i++) + pipes[i]->stream_res.tg->funcs->enable_crtc(pipes[i]->stream_res.tg); + + // Set DPMS on with stream update + for (i = 0; i < state->stream_count; i++) + if (state->streams[i] && state->streams[i]->link && state->streams[i]->link == link) { + stream_update.stream = state->streams[i]; + stream_update.dpms_off = &dpms_off; + dc_update_planes_and_stream(state->clk_mgr->ctx->dc, NULL, 0, state->streams[i], &stream_update); + } } static void dp_test_send_link_training(struct dc_link *link) diff --git a/drivers/gpu/drm/amd/display/dc/link/link_detection.c b/drivers/gpu/drm/amd/display/dc/link/link_detection.c index 9655e6fa53a4bf69d9a179410241c31b83326d67..827b630daf49a4002dcd6a5748315c0271df9d07 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_detection.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_detection.c @@ -593,8 +593,9 @@ static bool detect_dp(struct dc_link *link, if (sink_caps->transaction_type == DDC_TRANSACTION_TYPE_I2C_OVER_AUX) { sink_caps->signal = SIGNAL_TYPE_DISPLAY_PORT; - if (!detect_dp_sink_caps(link)) + if (!detect_dp_sink_caps(link)) { return false; + } if (is_dp_branch_device(link)) /* DP SST branch */ diff --git a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c index 273a3be6d593af910e4b89a259c049d960540f8e..bd51b279ad1484e911a2ab1fabe39445b1ce0fbb 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c @@ -140,7 +140,8 @@ void link_blank_dp_stream(struct dc_link *link, bool hw_init) } } - if ((!link->wa_flags.dp_keep_receiver_powered) || hw_init) + if (((!link->wa_flags.dp_keep_receiver_powered) || hw_init) && + (link->type != dc_connection_none)) dpcd_write_rx_power_ctrl(link, false); } } @@ -842,14 +843,14 @@ void link_set_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable) dsc_cfg.dc_dsc_cfg.num_slices_h /= opp_cnt; if (should_use_dto_dscclk) - dccg->funcs->set_dto_dscclk(dccg, dsc->inst); + dccg->funcs->set_dto_dscclk(dccg, dsc->inst, dsc_cfg.dc_dsc_cfg.num_slices_h); dsc->funcs->dsc_set_config(dsc, &dsc_cfg, &dsc_optc_cfg); dsc->funcs->dsc_enable(dsc, pipe_ctx->stream_res.opp->inst); for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { struct display_stream_compressor *odm_dsc = odm_pipe->stream_res.dsc; if (should_use_dto_dscclk) - dccg->funcs->set_dto_dscclk(dccg, odm_dsc->inst); + dccg->funcs->set_dto_dscclk(dccg, odm_dsc->inst, dsc_cfg.dc_dsc_cfg.num_slices_h); odm_dsc->funcs->dsc_set_config(odm_dsc, &dsc_cfg, &dsc_optc_cfg); odm_dsc->funcs->dsc_enable(odm_dsc, odm_pipe->stream_res.opp->inst); } @@ -2296,8 +2297,7 @@ static bool allocate_usb4_bandwidth_for_stream(struct dc_stream_state *stream, i link->dpia_bw_alloc_config.remote_sink_req_bw[sink_index] = bw; } - /* get dp overhead for dp tunneling */ - link->dpia_bw_alloc_config.dp_overhead = link_dp_dpia_get_dp_overhead_in_dp_tunneling(link); + link->dpia_bw_alloc_config.dp_overhead = link_dpia_get_dp_overhead(link); req_bw += link->dpia_bw_alloc_config.dp_overhead; link_dp_dpia_allocate_usb4_bandwidth_for_stream(link, req_bw); diff --git a/drivers/gpu/drm/amd/display/dc/link/link_factory.c b/drivers/gpu/drm/amd/display/dc/link/link_factory.c index 1a04f4b7458593ed5560b143cd4fabc3c3cec425..c5f4e803be842aada38506cead200237ecb69e1b 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_factory.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_factory.c @@ -100,7 +100,7 @@ static void construct_link_service_validation(struct link_service *link_srv) { link_srv->validate_mode_timing = link_validate_mode_timing; link_srv->dp_link_bandwidth_kbps = dp_link_bandwidth_kbps; - link_srv->validate_dpia_bandwidth = link_validate_dpia_bandwidth; + link_srv->validate_dp_tunnel_bandwidth = link_validate_dp_tunnel_bandwidth; link_srv->dp_required_hblank_size_bytes = dp_required_hblank_size_bytes; } diff --git a/drivers/gpu/drm/amd/display/dc/link/link_validation.c b/drivers/gpu/drm/amd/display/dc/link/link_validation.c index 29606fda029d6bb6777aa46ec45a06fbc1594afe..aecaf37eee3520d8b30edbfb5b575df510a5b4a0 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_validation.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_validation.c @@ -86,6 +86,10 @@ static bool dp_active_dongle_validate_timing( if (!dongle_caps->is_dp_hdmi_ycbcr420_pass_through) return false; break; + case PIXEL_ENCODING_UNDEFINED: + /* These color depths are currently not supported */ + ASSERT(false); + break; default: /* Invalid Pixel Encoding*/ return false; @@ -104,6 +108,10 @@ static bool dp_active_dongle_validate_timing( if (dongle_caps->dp_hdmi_max_bpc < 12) return false; break; + case COLOR_DEPTH_UNDEFINED: + /* These color depths are currently not supported */ + ASSERT(false); + break; case COLOR_DEPTH_141414: case COLOR_DEPTH_161616: default: @@ -255,6 +263,14 @@ uint32_t dp_link_bandwidth_kbps( return link_rate_per_lane_kbps * link_settings->lane_count / 10000 * total_data_bw_efficiency_x10000; } +static uint32_t dp_get_timing_bandwidth_kbps( + const struct dc_crtc_timing *timing, + const struct dc_link *link) +{ + return dc_bandwidth_in_kbps_from_timing(timing, + dc_link_get_highest_encoding_format(link)); +} + static bool dp_validate_mode_timing( struct dc_link *link, const struct dc_crtc_timing *timing) @@ -351,63 +367,81 @@ enum dc_status link_validate_mode_timing( return DC_OK; } +static const struct dc_tunnel_settings *get_dp_tunnel_settings(const struct dc_state *context, + const struct dc_stream_state *stream) +{ + int i; + const struct dc_tunnel_settings *dp_tunnel_settings = NULL; + + for (i = 0; i < MAX_PIPES; i++) { + if (context->res_ctx.pipe_ctx[i].stream && (context->res_ctx.pipe_ctx[i].stream == stream)) { + dp_tunnel_settings = &context->res_ctx.pipe_ctx[i].link_config.dp_tunnel_settings; + break; + } + } + + return dp_tunnel_settings; +} + /* - * This function calculates the bandwidth required for the stream timing - * and aggregates the stream bandwidth for the respective dpia link - * - * @stream: pointer to the dc_stream_state struct instance - * @num_streams: number of streams to be validated + * Calculates the DP tunneling bandwidth required for the stream timing + * and aggregates the stream bandwidth for the respective DP tunneling link * - * return: true if validation is succeeded + * return: dc_status */ -bool link_validate_dpia_bandwidth(const struct dc_stream_state *stream, const unsigned int num_streams) +enum dc_status link_validate_dp_tunnel_bandwidth(const struct dc *dc, const struct dc_state *new_ctx) { - int bw_needed[MAX_DPIA_NUM] = {0}; - struct dc_link *dpia_link[MAX_DPIA_NUM] = {0}; - int num_dpias = 0; - - for (unsigned int i = 0; i < num_streams; ++i) { - if (stream[i].signal == SIGNAL_TYPE_DISPLAY_PORT) { - /* new dpia sst stream, check whether it exceeds max dpia */ - if (num_dpias >= MAX_DPIA_NUM) - return false; + struct dc_validation_dpia_set dpia_link_sets[MAX_DPIA_NUM] = { 0 }; + uint8_t link_count = 0; + enum dc_status result = DC_OK; - dpia_link[num_dpias] = stream[i].link; - bw_needed[num_dpias] = dc_bandwidth_in_kbps_from_timing(&stream[i].timing, - dc_link_get_highest_encoding_format(dpia_link[num_dpias])); - num_dpias++; - } else if (stream[i].signal == SIGNAL_TYPE_DISPLAY_PORT_MST) { - uint8_t j = 0; - /* check whether its a known dpia link */ - for (; j < num_dpias; ++j) { - if (dpia_link[j] == stream[i].link) - break; - } + // Iterate through streams in the new context + for (uint8_t i = 0; (i < MAX_PIPES && i < new_ctx->stream_count); i++) { + const struct dc_stream_state *stream = new_ctx->streams[i]; + const struct dc_link *link; + const struct dc_tunnel_settings *dp_tunnel_settings; + uint32_t timing_bw; + + if (stream == NULL) + continue; + + link = stream->link; + + if (!(link && (stream->signal == SIGNAL_TYPE_DISPLAY_PORT + || stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) + && link->hpd_status)) + continue; - if (j == num_dpias) { - /* new dpia mst stream, check whether it exceeds max dpia */ - if (num_dpias >= MAX_DPIA_NUM) - return false; - else { - dpia_link[j] = stream[i].link; - num_dpias++; - } + dp_tunnel_settings = get_dp_tunnel_settings(new_ctx, stream); + + if ((dp_tunnel_settings == NULL) || (dp_tunnel_settings->should_use_dp_bw_allocation == false)) + continue; + + timing_bw = dp_get_timing_bandwidth_kbps(&stream->timing, link); + + // Find an existing entry for this 'link' in 'dpia_link_sets' + for (uint8_t j = 0; j < MAX_DPIA_NUM; j++) { + bool is_new_slot = false; + + if (dpia_link_sets[j].link == NULL) { + is_new_slot = true; + link_count++; + dpia_link_sets[j].required_bw = 0; + dpia_link_sets[j].link = link; } - bw_needed[j] += dc_bandwidth_in_kbps_from_timing(&stream[i].timing, - dc_link_get_highest_encoding_format(dpia_link[j])); + if (is_new_slot || (dpia_link_sets[j].link == link)) { + dpia_link_sets[j].tunnel_settings = dp_tunnel_settings; + dpia_link_sets[j].required_bw += timing_bw; + break; + } } } - /* Include dp overheads */ - for (uint8_t i = 0; i < num_dpias; ++i) { - int dp_overhead = 0; - - dp_overhead = link_dp_dpia_get_dp_overhead_in_dp_tunneling(dpia_link[i]); - bw_needed[i] += dp_overhead; - } + if (link_count && link_dpia_validate_dp_tunnel_bandwidth(dpia_link_sets, link_count) == false) + result = DC_FAIL_DP_TUNNEL_BW_VALIDATE; - return dpia_validate_usb4_bw(dpia_link, bw_needed, num_dpias); + return result; } struct dp_audio_layout_config { diff --git a/drivers/gpu/drm/amd/display/dc/link/link_validation.h b/drivers/gpu/drm/amd/display/dc/link/link_validation.h index bf398c49c3e8674d48b2aff732ad270e0546b331..9553c81053fe1aefc555800456887e42c4453d92 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_validation.h +++ b/drivers/gpu/drm/amd/display/dc/link/link_validation.h @@ -30,9 +30,9 @@ enum dc_status link_validate_mode_timing( const struct dc_stream_state *stream, struct dc_link *link, const struct dc_crtc_timing *timing); -bool link_validate_dpia_bandwidth( - const struct dc_stream_state *stream, - const unsigned int num_streams); +enum dc_status link_validate_dp_tunnel_bandwidth( + const struct dc *dc, + const struct dc_state *new_ctx); uint32_t dp_link_bandwidth_kbps( const struct dc_link *link, const struct dc_link_settings *link_settings); diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c index a5127c2d47efb81a0bf988340bc6649d00e72faa..0f965380a9b4b926d38d937db08cf77d9d5e4f04 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c @@ -385,9 +385,15 @@ bool dp_is_128b_132b_signal(struct pipe_ctx *pipe_ctx) bool dp_is_lttpr_present(struct dc_link *link) { /* Some sink devices report invalid LTTPR revision, so don't validate against that cap */ - return (dp_parse_lttpr_repeater_count(link->dpcd_caps.lttpr_caps.phy_repeater_cnt) != 0 && + uint32_t lttpr_count = dp_parse_lttpr_repeater_count(link->dpcd_caps.lttpr_caps.phy_repeater_cnt); + bool is_lttpr_present = (lttpr_count > 0 && link->dpcd_caps.lttpr_caps.max_lane_count > 0 && link->dpcd_caps.lttpr_caps.max_lane_count <= 4); + + if (lttpr_count > 0 && !is_lttpr_present) + DC_LOG_ERROR("LTTPR count is nonzero but invalid lane count reported. Assuming no LTTPR present.\n"); + + return is_lttpr_present; } /* in DP compliance test, DPR-120 may have @@ -1551,6 +1557,8 @@ enum dc_status dp_retrieve_lttpr_cap(struct dc_link *link) uint8_t lttpr_dpcd_data[10] = {0}; enum dc_status status; bool is_lttpr_present; + uint32_t lttpr_count; + uint32_t closest_lttpr_offset; /* Logic to determine LTTPR support*/ bool vbios_lttpr_interop = link->dc->caps.vbios_lttpr_aware; @@ -1602,20 +1610,22 @@ enum dc_status dp_retrieve_lttpr_cap(struct dc_link *link) lttpr_dpcd_data[DP_LTTPR_ALPM_CAPABILITIES - DP_LT_TUNABLE_PHY_REPEATER_FIELD_DATA_STRUCTURE_REV]; + lttpr_count = dp_parse_lttpr_repeater_count(link->dpcd_caps.lttpr_caps.phy_repeater_cnt); + /* If this chip cap is set, at least one retimer must exist in the chain * Override count to 1 if we receive a known bad count (0 or an invalid value) */ if (((link->chip_caps & AMD_EXT_DISPLAY_PATH_CAPS__EXT_CHIP_MASK) == AMD_EXT_DISPLAY_PATH_CAPS__DP_FIXED_VS_EN) && - (dp_parse_lttpr_repeater_count(link->dpcd_caps.lttpr_caps.phy_repeater_cnt) == 0)) { + lttpr_count == 0) { /* If you see this message consistently, either the host platform has FIXED_VS flag * incorrectly configured or the sink device is returning an invalid count. */ DC_LOG_ERROR("lttpr_caps phy_repeater_cnt is 0x%x, forcing it to 0x80.", link->dpcd_caps.lttpr_caps.phy_repeater_cnt); link->dpcd_caps.lttpr_caps.phy_repeater_cnt = 0x80; + lttpr_count = 1; DC_LOG_DC("lttpr_caps forced phy_repeater_cnt = %d\n", link->dpcd_caps.lttpr_caps.phy_repeater_cnt); } - /* Attempt to train in LTTPR transparent mode if repeater count exceeds 8. */ is_lttpr_present = dp_is_lttpr_present(link); DC_LOG_DC("is_lttpr_present = %d\n", is_lttpr_present); @@ -1623,11 +1633,25 @@ enum dc_status dp_retrieve_lttpr_cap(struct dc_link *link) if (is_lttpr_present) { CONN_DATA_DETECT(link, lttpr_dpcd_data, sizeof(lttpr_dpcd_data), "LTTPR Caps: "); - core_link_read_dpcd(link, DP_LTTPR_IEEE_OUI, link->dpcd_caps.lttpr_caps.lttpr_ieee_oui, sizeof(link->dpcd_caps.lttpr_caps.lttpr_ieee_oui)); - CONN_DATA_DETECT(link, link->dpcd_caps.lttpr_caps.lttpr_ieee_oui, sizeof(link->dpcd_caps.lttpr_caps.lttpr_ieee_oui), "LTTPR IEEE OUI: "); + // Identify closest LTTPR to determine if workarounds required for known embedded LTTPR + closest_lttpr_offset = dp_get_closest_lttpr_offset(lttpr_count); - core_link_read_dpcd(link, DP_LTTPR_DEVICE_ID, link->dpcd_caps.lttpr_caps.lttpr_device_id, sizeof(link->dpcd_caps.lttpr_caps.lttpr_device_id)); - CONN_DATA_DETECT(link, link->dpcd_caps.lttpr_caps.lttpr_device_id, sizeof(link->dpcd_caps.lttpr_caps.lttpr_device_id), "LTTPR Device ID: "); + core_link_read_dpcd(link, (DP_LTTPR_IEEE_OUI + closest_lttpr_offset), + link->dpcd_caps.lttpr_caps.lttpr_ieee_oui, sizeof(link->dpcd_caps.lttpr_caps.lttpr_ieee_oui)); + core_link_read_dpcd(link, (DP_LTTPR_DEVICE_ID + closest_lttpr_offset), + link->dpcd_caps.lttpr_caps.lttpr_device_id, sizeof(link->dpcd_caps.lttpr_caps.lttpr_device_id)); + + if (lttpr_count > 1) { + CONN_DATA_DETECT(link, link->dpcd_caps.lttpr_caps.lttpr_ieee_oui, sizeof(link->dpcd_caps.lttpr_caps.lttpr_ieee_oui), + "Closest LTTPR To Host's IEEE OUI: "); + CONN_DATA_DETECT(link, link->dpcd_caps.lttpr_caps.lttpr_device_id, sizeof(link->dpcd_caps.lttpr_caps.lttpr_device_id), + "Closest LTTPR To Host's LTTPR Device ID: "); + } else { + CONN_DATA_DETECT(link, link->dpcd_caps.lttpr_caps.lttpr_ieee_oui, sizeof(link->dpcd_caps.lttpr_caps.lttpr_ieee_oui), + "LTTPR IEEE OUI: "); + CONN_DATA_DETECT(link, link->dpcd_caps.lttpr_caps.lttpr_device_id, sizeof(link->dpcd_caps.lttpr_caps.lttpr_device_id), + "LTTPR Device ID: "); + } } return status; diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.c index 22bfdced64ab6afb82f82a143890b1be70e294dc..9b2f1a7da1d12f67c974f3a94e259d7d63cbdb96 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.c @@ -75,12 +75,15 @@ enum dc_status dpcd_get_tunneling_device_data(struct dc_link *link) if (link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.bits.dpia_bw_alloc) { status = core_link_read_dpcd(link, USB4_DRIVER_BW_CAPABILITY, - dpcd_dp_tun_data, 1); + dpcd_dp_tun_data, 2); if (status != DC_OK) goto err; - link->dpcd_caps.usb4_dp_tun_info.driver_bw_cap.raw = dpcd_dp_tun_data[0]; + link->dpcd_caps.usb4_dp_tun_info.driver_bw_cap.raw = + dpcd_dp_tun_data[USB4_DRIVER_BW_CAPABILITY - USB4_DRIVER_BW_CAPABILITY]; + link->dpcd_caps.usb4_dp_tun_info.dpia_tunnel_info.raw = + dpcd_dp_tun_data[DP_IN_ADAPTER_TUNNEL_INFO - USB4_DRIVER_BW_CAPABILITY]; } DC_LOG_DEBUG("%s: Link[%d] DP tunneling support (RouterId=%d AdapterId=%d) " @@ -155,8 +158,14 @@ void link_decide_dp_tunnel_settings(struct dc_stream_state *stream, link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.bits.dp_tunneling; if (link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.bits.dpia_bw_alloc - && link->dpcd_caps.usb4_dp_tun_info.driver_bw_cap.bits.driver_bw_alloc_support) + && link->dpcd_caps.usb4_dp_tun_info.driver_bw_cap.bits.driver_bw_alloc_support) { dp_tunnel_setting->should_use_dp_bw_allocation = true; + dp_tunnel_setting->cm_id = link->dpcd_caps.usb4_dp_tun_info.usb4_driver_id & 0x0F; + dp_tunnel_setting->group_id = link->dpcd_caps.usb4_dp_tun_info.dpia_tunnel_info.bits.group_id; + dp_tunnel_setting->estimated_bw = link->dpia_bw_alloc_config.estimated_bw; + dp_tunnel_setting->allocated_bw = link->dpia_bw_alloc_config.allocated_bw; + dp_tunnel_setting->bw_granularity = link->dpia_bw_alloc_config.bw_granularity; + } } } diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c index 3af7564a84f107a249459bc7834203899fd27e84..819bf2d8ba530bec22f7ab5a954de0d04d95bbfc 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c @@ -35,6 +35,8 @@ #define Kbps_TO_Gbps (1000 * 1000) +#define MST_TIME_SLOT_COUNT 64 + // ------------------------------------------------------------------ // PRIVATE FUNCTIONS // ------------------------------------------------------------------ @@ -160,78 +162,6 @@ static void retrieve_usb4_dp_bw_allocation_info(struct dc_link *link) link->dpia_bw_alloc_config.nrd_max_lane_count); } -static uint8_t get_lowest_dpia_index(struct dc_link *link) -{ - const struct dc *dc_struct = link->dc; - uint8_t idx = 0xFF; - int i; - - for (i = 0; i < MAX_LINKS; ++i) { - - if (!dc_struct->links[i] || - dc_struct->links[i]->ep_type != DISPLAY_ENDPOINT_USB4_DPIA) - continue; - - if (idx > dc_struct->links[i]->link_index) { - idx = dc_struct->links[i]->link_index; - break; - } - } - - return idx; -} - -/* - * Get the maximum dp tunnel banwidth of host router - * - * @dc: pointer to the dc struct instance - * @hr_index: host router index - * - * return: host router maximum dp tunnel bandwidth - */ -static int get_host_router_total_dp_tunnel_bw(const struct dc *dc, uint8_t hr_index) -{ - uint8_t lowest_dpia_index = get_lowest_dpia_index(dc->links[0]); - uint8_t hr_index_temp = 0; - struct dc_link *link_dpia_primary, *link_dpia_secondary; - int total_bw = 0; - - for (uint8_t i = 0; i < MAX_LINKS - 1; ++i) { - - if (!dc->links[i] || dc->links[i]->ep_type != DISPLAY_ENDPOINT_USB4_DPIA) - continue; - - hr_index_temp = (dc->links[i]->link_index - lowest_dpia_index) / 2; - - if (hr_index_temp == hr_index) { - link_dpia_primary = dc->links[i]; - link_dpia_secondary = dc->links[i + 1]; - - /** - * If BW allocation enabled on both DPIAs, then - * HR BW = Estimated(dpia_primary) + Allocated(dpia_secondary) - * otherwise HR BW = Estimated(bw alloc enabled dpia) - */ - if ((link_dpia_primary->hpd_status && - link_dpia_primary->dpia_bw_alloc_config.bw_alloc_enabled) && - (link_dpia_secondary->hpd_status && - link_dpia_secondary->dpia_bw_alloc_config.bw_alloc_enabled)) { - total_bw += link_dpia_primary->dpia_bw_alloc_config.estimated_bw + - link_dpia_secondary->dpia_bw_alloc_config.allocated_bw; - } else if (link_dpia_primary->hpd_status && - link_dpia_primary->dpia_bw_alloc_config.bw_alloc_enabled) { - total_bw = link_dpia_primary->dpia_bw_alloc_config.estimated_bw; - } else if (link_dpia_secondary->hpd_status && - link_dpia_secondary->dpia_bw_alloc_config.bw_alloc_enabled) { - total_bw += link_dpia_secondary->dpia_bw_alloc_config.estimated_bw; - } - break; - } - } - - return total_bw; -} - /* * Cleanup function for when the dpia is unplugged to reset struct * and perform any required clean up @@ -251,32 +181,40 @@ static void dpia_bw_alloc_unplug(struct dc_link *link) static void link_dpia_send_bw_alloc_request(struct dc_link *link, int req_bw) { - uint8_t requested_bw; - uint32_t temp; + uint8_t request_reg_val; + uint32_t temp, request_bw; - /* Error check whether request bw greater than allocated */ - if (req_bw > link->dpia_bw_alloc_config.estimated_bw) { - DC_LOG_ERROR("%s: Request BW greater than estimated BW for link(%d)\n", - __func__, link->link_index); - req_bw = link->dpia_bw_alloc_config.estimated_bw; + if (link->dpia_bw_alloc_config.bw_granularity == 0) { + DC_LOG_ERROR("%s: Link[%d]: bw_granularity is zero!", __func__, link->link_index); + return; } temp = req_bw * link->dpia_bw_alloc_config.bw_granularity; - requested_bw = temp / Kbps_TO_Gbps; + request_reg_val = temp / Kbps_TO_Gbps; /* Always make sure to add more to account for floating points */ if (temp % Kbps_TO_Gbps) - ++requested_bw; + ++request_reg_val; - /* Error check whether requested and allocated are equal */ - req_bw = requested_bw * (Kbps_TO_Gbps / link->dpia_bw_alloc_config.bw_granularity); - if (req_bw && (req_bw == link->dpia_bw_alloc_config.allocated_bw)) { - DC_LOG_ERROR("%s: Request BW equals to allocated BW for link(%d)\n", - __func__, link->link_index); + request_bw = request_reg_val * (Kbps_TO_Gbps / link->dpia_bw_alloc_config.bw_granularity); + + if (request_bw > link->dpia_bw_alloc_config.estimated_bw) { + DC_LOG_ERROR("%s: Link[%d]: Request BW (%d --> %d) > Estimated BW (%d)... Set to Estimated BW!", + __func__, link->link_index, + req_bw, request_bw, link->dpia_bw_alloc_config.estimated_bw); + req_bw = link->dpia_bw_alloc_config.estimated_bw; + + temp = req_bw * link->dpia_bw_alloc_config.bw_granularity; + request_reg_val = temp / Kbps_TO_Gbps; + if (temp % Kbps_TO_Gbps) + ++request_reg_val; } + link->dpia_bw_alloc_config.allocated_bw = request_bw; + DC_LOG_DC("%s: Link[%d]: Request BW: %d", __func__, link->link_index, request_bw); + core_link_write_dpcd(link, REQUESTED_BW, - &requested_bw, + &request_reg_val, sizeof(uint8_t)); } @@ -304,14 +242,16 @@ bool link_dpia_enable_usb4_dp_bw_alloc_mode(struct dc_link *link) link->dpia_bw_alloc_config.bw_alloc_enabled = true; ret = true; - /* - * During DP tunnel creation, CM preallocates BW and reduces estimated BW of other - * DPIA. CM release preallocation only when allocation is complete. Do zero alloc - * to make the CM to release preallocation and update estimated BW correctly for - * all DPIAs per host router - */ - // TODO: Zero allocation can be removed once the MSFT CM fix has been released - link_dp_dpia_allocate_usb4_bandwidth_for_stream(link, 0); + if (link->dc->debug.dpia_debug.bits.enable_usb4_bw_zero_alloc_patch) { + /* + * During DP tunnel creation, the CM preallocates BW + * and reduces the estimated BW of other DPIAs. + * The CM releases the preallocation only when the allocation is complete. + * Perform a zero allocation to make the CM release the preallocation + * and correctly update the estimated BW for all DPIAs per host router. + */ + link_dp_dpia_allocate_usb4_bandwidth_for_stream(link, 0); + } } else DC_LOG_DEBUG("%s: link[%d] failed to enable DPTX BW allocation mode", __func__, link->link_index); } @@ -329,19 +269,17 @@ bool link_dpia_enable_usb4_dp_bw_alloc_mode(struct dc_link *link) */ void link_dp_dpia_handle_bw_alloc_status(struct dc_link *link, uint8_t status) { + link->dpia_bw_alloc_config.estimated_bw = get_estimated_bw(link); + if (status & DP_TUNNELING_BW_REQUEST_SUCCEEDED) { DC_LOG_DEBUG("%s: BW Allocation request succeeded on link(%d)", __func__, link->link_index); } else if (status & DP_TUNNELING_BW_REQUEST_FAILED) { - link->dpia_bw_alloc_config.estimated_bw = get_estimated_bw(link); - DC_LOG_DEBUG("%s: BW Allocation request failed on link(%d) allocated/estimated BW=%d", __func__, link->link_index, link->dpia_bw_alloc_config.estimated_bw); link_dpia_send_bw_alloc_request(link, link->dpia_bw_alloc_config.estimated_bw); } else if (status & DP_TUNNELING_ESTIMATED_BW_CHANGED) { - link->dpia_bw_alloc_config.estimated_bw = get_estimated_bw(link); - DC_LOG_DEBUG("%s: Estimated BW changed on link(%d) new estimated BW=%d", __func__, link->link_index, link->dpia_bw_alloc_config.estimated_bw); } @@ -374,9 +312,13 @@ void dpia_handle_usb4_bandwidth_allocation_for_link(struct dc_link *link, int pe void link_dp_dpia_allocate_usb4_bandwidth_for_stream(struct dc_link *link, int req_bw) { - DC_LOG_DEBUG("%s: ENTER: link(%d), hpd_status(%d), current allocated_bw(%d), req_bw(%d)\n", + link->dpia_bw_alloc_config.estimated_bw = get_estimated_bw(link); + + DC_LOG_DEBUG("%s: ENTER: link[%d] hpd(%d) Allocated_BW: %d Estimated_BW: %d Req_BW: %d", __func__, link->link_index, link->hpd_status, - link->dpia_bw_alloc_config.allocated_bw, req_bw); + link->dpia_bw_alloc_config.allocated_bw, + link->dpia_bw_alloc_config.estimated_bw, + req_bw); if (link_dp_is_bw_alloc_available(link)) link_dpia_send_bw_alloc_request(link, req_bw); @@ -384,73 +326,116 @@ void link_dp_dpia_allocate_usb4_bandwidth_for_stream(struct dc_link *link, int r DC_LOG_DEBUG("%s: BW Allocation mode not available", __func__); } -bool dpia_validate_usb4_bw(struct dc_link **link, int *bw_needed_per_dpia, const unsigned int num_dpias) +uint32_t link_dpia_get_dp_overhead(const struct dc_link *link) { - bool ret = true; - int bw_needed_per_hr[MAX_HR_NUM] = { 0, 0 }, host_router_total_dp_bw = 0; - uint8_t lowest_dpia_index, i, hr_index; + uint32_t link_dp_overhead = 0; - if (!num_dpias || num_dpias > MAX_DPIA_NUM) - return ret; + if ((link->type == dc_connection_mst_branch) && + !link->dpcd_caps.channel_coding_cap.bits.DP_128b_132b_SUPPORTED) { + /* For 8b/10b encoding: MTP is 64 time slots long, slot 0 is used for MTPH + * MST overhead is 1/64 of link bandwidth (excluding any overhead) + */ + const struct dc_link_settings *link_cap = dc_link_get_link_cap(link); - lowest_dpia_index = get_lowest_dpia_index(link[0]); + if (link_cap) { + uint32_t link_bw_in_kbps = (uint32_t)link_cap->link_rate * + (uint32_t)link_cap->lane_count * + LINK_RATE_REF_FREQ_IN_KHZ * 8; + link_dp_overhead = (link_bw_in_kbps / MST_TIME_SLOT_COUNT) + + ((link_bw_in_kbps % MST_TIME_SLOT_COUNT) ? 1 : 0); + } + } - /* get total Host Router BW with granularity for the given modes */ - for (i = 0; i < num_dpias; ++i) { - int granularity_Gbps = 0; - int bw_granularity = 0; + return link_dp_overhead; +} - if (!link[i]->dpia_bw_alloc_config.bw_alloc_enabled) - continue; +/* + * Aggregates the DPIA bandwidth usage for the respective USB4 Router. + * And then validate if the required bandwidth is within the router's capacity. + * + * @dc_validation_dpia_set: pointer to the dc_validation_dpia_set + * @count: number of DPIA validation sets + * + * return: true if validation is succeeded + */ +bool link_dpia_validate_dp_tunnel_bandwidth(const struct dc_validation_dpia_set *dpia_link_sets, uint8_t count) +{ + uint32_t granularity_Gbps; + const struct dc_link *link; + uint32_t link_bw_granularity; + uint32_t link_required_bw; + struct usb4_router_validation_set router_sets[MAX_HOST_ROUTERS_NUM] = { 0 }; + uint8_t i; + bool is_success = true; + uint8_t router_count = 0; + + if ((dpia_link_sets == NULL) || (count == 0)) + return is_success; + + // Iterate through each DP tunneling link (DPIA). + // Aggregate its bandwidth requirements onto the respective USB4 router. + for (i = 0; i < count; i++) { + link = dpia_link_sets[i].link; + link_required_bw = dpia_link_sets[i].required_bw; + const struct dc_tunnel_settings *dp_tunnel_settings = dpia_link_sets[i].tunnel_settings; + + if ((link == NULL) || (dp_tunnel_settings == NULL) || dp_tunnel_settings->bw_granularity == 0) + break; - if (link[i]->link_index < lowest_dpia_index) - continue; + if (link->type == dc_connection_mst_branch) + link_required_bw += link_dpia_get_dp_overhead(link); - granularity_Gbps = (Kbps_TO_Gbps / link[i]->dpia_bw_alloc_config.bw_granularity); - bw_granularity = (bw_needed_per_dpia[i] / granularity_Gbps) * granularity_Gbps + - ((bw_needed_per_dpia[i] % granularity_Gbps) ? granularity_Gbps : 0); + granularity_Gbps = (Kbps_TO_Gbps / dp_tunnel_settings->bw_granularity); + link_bw_granularity = (link_required_bw / granularity_Gbps) * granularity_Gbps + + ((link_required_bw % granularity_Gbps) ? granularity_Gbps : 0); - hr_index = (link[i]->link_index - lowest_dpia_index) / 2; - bw_needed_per_hr[hr_index] += bw_granularity; - } + // Find or add the USB4 router associated with the current DPIA link + for (uint8_t j = 0; j < MAX_HOST_ROUTERS_NUM; j++) { + if (router_sets[j].is_valid == false) { + router_sets[j].is_valid = true; + router_sets[j].cm_id = dp_tunnel_settings->cm_id; + router_count++; + } - /* validate against each Host Router max BW */ - for (hr_index = 0; hr_index < MAX_HR_NUM; ++hr_index) { - if (bw_needed_per_hr[hr_index]) { - host_router_total_dp_bw = get_host_router_total_dp_tunnel_bw(link[0]->dc, hr_index); - if (bw_needed_per_hr[hr_index] > host_router_total_dp_bw) { - ret = false; + if (router_sets[j].cm_id == dp_tunnel_settings->cm_id) { + uint32_t remaining_bw = + dp_tunnel_settings->estimated_bw - dp_tunnel_settings->allocated_bw; + + router_sets[j].allocated_bw += dp_tunnel_settings->allocated_bw; + + if (remaining_bw > router_sets[j].remaining_bw) + router_sets[j].remaining_bw = remaining_bw; + + // Get the max estimated BW within the same CM_ID + if (dp_tunnel_settings->estimated_bw > router_sets[j].estimated_bw) + router_sets[j].estimated_bw = dp_tunnel_settings->estimated_bw; + + router_sets[j].required_bw += link_bw_granularity; + router_sets[j].dpia_count++; break; } } } - return ret; -} + // Validate bandwidth for each unique router found. + for (i = 0; i < router_count; i++) { + uint32_t total_bw = 0; -int link_dp_dpia_get_dp_overhead_in_dp_tunneling(struct dc_link *link) -{ - int dp_overhead = 0, link_mst_overhead = 0; + if (router_sets[i].is_valid == false) + break; - if (!link_dp_is_bw_alloc_available(link)) - return dp_overhead; + // Determine the total available bandwidth for the current router based on aggregated data + if ((router_sets[i].dpia_count == 1) || (router_sets[i].allocated_bw == 0)) + total_bw = router_sets[i].estimated_bw; + else + total_bw = router_sets[i].allocated_bw + router_sets[i].remaining_bw; - /* if its mst link, add MTPH overhead */ - if ((link->type == dc_connection_mst_branch) && - !link->dpcd_caps.channel_coding_cap.bits.DP_128b_132b_SUPPORTED) { - /* For 8b/10b encoding: MTP is 64 time slots long, slot 0 is used for MTPH - * MST overhead is 1/64 of link bandwidth (excluding any overhead) - */ - const struct dc_link_settings *link_cap = - dc_link_get_link_cap(link); - uint32_t link_bw_in_kbps = (uint32_t)link_cap->link_rate * - (uint32_t)link_cap->lane_count * - LINK_RATE_REF_FREQ_IN_KHZ * 8; - link_mst_overhead = (link_bw_in_kbps / 64) + ((link_bw_in_kbps % 64) ? 1 : 0); + if (router_sets[i].required_bw > total_bw) { + is_success = false; + break; + } } - /* add all the overheads */ - dp_overhead = link_mst_overhead; - - return dp_overhead; + return is_success; } + diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.h b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.h index 801965b5f9a4addb174a93b23c190386c838e9b8..41efcb3e44e2e8d3ce58da86122ad69b9ce769d2 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.h +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.h @@ -28,10 +28,6 @@ #include "link.h" -/* Number of Host Routers per motherboard is 2 */ -#define MAX_HR_NUM 2 -/* Number of DPIA per host router is 2 */ -#define MAX_DPIA_NUM (MAX_HR_NUM * 2) /* * Host Router BW type @@ -42,6 +38,16 @@ enum bw_type { HOST_ROUTER_BW_INVALID, }; +struct usb4_router_validation_set { + bool is_valid; + uint8_t cm_id; + uint8_t dpia_count; + uint32_t required_bw; + uint32_t allocated_bw; + uint32_t estimated_bw; + uint32_t remaining_bw; +}; + /* * Enable USB4 DP BW allocation mode * @@ -73,18 +79,6 @@ void link_dp_dpia_allocate_usb4_bandwidth_for_stream(struct dc_link *link, int r */ void dpia_handle_usb4_bandwidth_allocation_for_link(struct dc_link *link, int peak_bw); -/* - * Handle the validation of total BW here and confirm that the bw used by each - * DPIA doesn't exceed available BW for each host router (HR) - * - * @link[]: array of link pointer to all possible DPIA links - * @bw_needed[]: bw needed for each DPIA link based on timing - * @num_dpias: Number of DPIAs for the above 2 arrays. Should always be <= MAX_DPIA_NUM - * - * return: TRUE if bw used by DPIAs doesn't exceed available BW else return FALSE - */ -bool dpia_validate_usb4_bw(struct dc_link **link, int *bw_needed, const unsigned int num_dpias); - /* * Obtain all the DP overheads in dp tunneling for the dpia link * @@ -92,7 +86,7 @@ bool dpia_validate_usb4_bw(struct dc_link **link, int *bw_needed, const unsigned * * return: DP overheads in DP tunneling */ -int link_dp_dpia_get_dp_overhead_in_dp_tunneling(struct dc_link *link); +uint32_t link_dpia_get_dp_overhead(const struct dc_link *link); /* * Handle DP BW allocation status register @@ -104,4 +98,15 @@ int link_dp_dpia_get_dp_overhead_in_dp_tunneling(struct dc_link *link); */ void link_dp_dpia_handle_bw_alloc_status(struct dc_link *link, uint8_t status); +/* + * Aggregates the DPIA bandwidth usage for the respective USB4 Router. + * + * @dc_validation_dpia_set: pointer to the dc_validation_dpia_set + * @count: number of DPIA validation sets + * + * return: true if validation is succeeded + */ +bool link_dpia_validate_dp_tunnel_bandwidth(const struct dc_validation_dpia_set *dpia_link_sets, uint8_t count); + #endif /* DC_INC_LINK_DP_DPIA_BW_H_ */ + diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c index da74c2b5854f3abdedba41235dd5ee713d7c5117..035795042a01d73295a3e008f30a9fee8c061898 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c @@ -173,6 +173,15 @@ bool edp_set_backlight_level_nits(struct dc_link *link, target_luminance = (struct target_luminance_value *)&backlight_millinits; + //make sure we disable AMD ABC first. + core_link_read_dpcd(link, DP_SOURCE_BACKLIGHT_CONTROL, + &backlight_enable, sizeof(uint8_t)); + if (backlight_enable) { + backlight_enable = 0; + core_link_write_dpcd(link, DP_SOURCE_BACKLIGHT_CONTROL, + &backlight_enable, 1); + } + core_link_read_dpcd(link, DP_EDP_BACKLIGHT_MODE_SET_REGISTER, &backlight_enable, sizeof(uint8_t)); @@ -193,10 +202,22 @@ bool edp_set_backlight_level_nits(struct dc_link *link, *(uint16_t *)&dpcd_backlight_set.backlight_transition_time_ms = (uint16_t)transition_time_in_ms; uint8_t backlight_control = isHDR ? 1 : 0; + uint8_t backlight_enable = 0; + // OLEDs have no PWM, they can only use AUX if (link->dpcd_sink_ext_caps.bits.oled == 1) backlight_control = 1; + //make sure we disable VESA ABC first. + core_link_read_dpcd(link, DP_EDP_BACKLIGHT_MODE_SET_REGISTER, + &backlight_enable, sizeof(uint8_t)); + + if (backlight_enable & DP_EDP_PANEL_LUMINANCE_CONTROL_ENABLE) { + backlight_enable &= ~DP_EDP_PANEL_LUMINANCE_CONTROL_ENABLE; + core_link_write_dpcd(link, DP_EDP_BACKLIGHT_MODE_SET_REGISTER, + &backlight_enable, sizeof(backlight_enable)); + } + if (core_link_write_dpcd(link, DP_SOURCE_BACKLIGHT_LEVEL, (uint8_t *)(&dpcd_backlight_set), sizeof(dpcd_backlight_set)) != DC_OK) diff --git a/drivers/gpu/drm/amd/display/dc/mpc/Makefile b/drivers/gpu/drm/amd/display/dc/mpc/Makefile index 1e2e66508192d4033d8a831b1f138b27815d6519..5402c3529f5eee106e516814a89f9d2060b6513b 100644 --- a/drivers/gpu/drm/amd/display/dc/mpc/Makefile +++ b/drivers/gpu/drm/amd/display/dc/mpc/Makefile @@ -68,5 +68,5 @@ MPC_DCN401 = dcn401_mpc.o AMD_DAL_MPC_DCN401 = $(addprefix $(AMDDALPATH)/dc/mpc/dcn401/,$(MPC_DCN401)) AMD_DISPLAY_FILES += $(AMD_DAL_MPC_DCN401) -endif +endif diff --git a/drivers/gpu/drm/amd/display/dc/mpc/dcn32/dcn32_mpc.c b/drivers/gpu/drm/amd/display/dc/mpc/dcn32/dcn32_mpc.c index b4cea2b8cb2a8aa33ff60ea78c078fa64c5e731d..6f0e017a8ae29c76fbf3791361c96a2fdc4b74ff 100644 --- a/drivers/gpu/drm/amd/display/dc/mpc/dcn32/dcn32_mpc.c +++ b/drivers/gpu/drm/amd/display/dc/mpc/dcn32/dcn32_mpc.c @@ -30,7 +30,6 @@ #include "basics/conversion.h" #include "dcn10/dcn10_cm_common.h" #include "dc.h" -#include "dcn401/dcn401_mpc.h" #define REG(reg)\ mpc30->mpc_regs->reg @@ -879,7 +878,7 @@ void mpc32_set3dlut_ram10( } -static void mpc32_set_3dlut_mode( +void mpc32_set_3dlut_mode( struct mpc *mpc, enum dc_lut_mode mode, bool is_color_channel_12bits, @@ -1022,8 +1021,6 @@ static const struct mpc_funcs dcn32_mpc_funcs = { .power_on_mpc_mem_pwr = mpc3_power_on_ogam_lut, .get_mpc_out_mux = mpc1_get_mpc_out_mux, .set_bg_color = mpc1_set_bg_color, - .set_movable_cm_location = mpc401_set_movable_cm_location, - .populate_lut = mpc401_populate_lut, }; diff --git a/drivers/gpu/drm/amd/display/dc/mpc/dcn32/dcn32_mpc.h b/drivers/gpu/drm/amd/display/dc/mpc/dcn32/dcn32_mpc.h index 9622518826c94ba576c7fcfa4be93afa539d8ded..8c9b20bcca85430f2beda9ab78f90d67cff8e602 100644 --- a/drivers/gpu/drm/amd/display/dc/mpc/dcn32/dcn32_mpc.h +++ b/drivers/gpu/drm/amd/display/dc/mpc/dcn32/dcn32_mpc.h @@ -391,4 +391,12 @@ void mpc32_select_3dlut_ram( enum dc_lut_mode mode, bool is_color_channel_12bits, uint32_t mpcc_id); + +void mpc32_set_3dlut_mode( + struct mpc *mpc, + enum dc_lut_mode mode, + bool is_color_channel_12bits, + bool is_lut_size17x17x17, + uint32_t mpcc_id); + #endif //__DC_MPCC_DCN32_H__ diff --git a/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.c b/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.c index 98cf0cbd59ba0ea05638e6724bbc55d6cdef2685..f3fb3fe137577b78a85859111be198e0e8f3622c 100644 --- a/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.c +++ b/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.c @@ -294,7 +294,7 @@ void mpc401_program_3dlut_size(struct mpc *mpc, bool is_17x17x17, int mpcc_id) REG_UPDATE(MPCC_MCM_3DLUT_MODE[mpcc_id], MPCC_MCM_3DLUT_SIZE, is_17x17x17 ? 0 : 1); } -static void program_gamut_remap( +void mpc_program_gamut_remap( struct mpc *mpc, unsigned int mpcc_id, const uint16_t *regval, @@ -426,7 +426,7 @@ void mpc401_set_gamut_remap( if (adjust->gamut_adjust_type != GRAPHICS_GAMUT_ADJUST_TYPE_SW) { /* Bypass / Disable if type is bypass or hw */ - program_gamut_remap(mpc, mpcc_id, NULL, + mpc_program_gamut_remap(mpc, mpcc_id, NULL, adjust->mpcc_gamut_remap_block_id, MPCC_GAMUT_REMAP_MODE_SELECT_0); } else { struct fixed31_32 arr_matrix[12]; @@ -460,12 +460,12 @@ void mpc401_set_gamut_remap( else mode_select = MPCC_GAMUT_REMAP_MODE_SELECT_2; - program_gamut_remap(mpc, mpcc_id, arr_reg_val, + mpc_program_gamut_remap(mpc, mpcc_id, arr_reg_val, adjust->mpcc_gamut_remap_block_id, mode_select); } } -static void read_gamut_remap(struct mpc *mpc, +void mpc_read_gamut_remap(struct mpc *mpc, int mpcc_id, uint16_t *regval, enum mpcc_gamut_remap_id gamut_remap_block_id, @@ -561,9 +561,9 @@ void mpc401_get_gamut_remap(struct mpc *mpc, struct mpc_grph_gamut_adjustment *adjust) { uint16_t arr_reg_val[12] = {0}; - uint32_t mode_select; + uint32_t mode_select = MPCC_GAMUT_REMAP_MODE_SELECT_0; - read_gamut_remap(mpc, mpcc_id, arr_reg_val, adjust->mpcc_gamut_remap_block_id, &mode_select); + mpc_read_gamut_remap(mpc, mpcc_id, arr_reg_val, adjust->mpcc_gamut_remap_block_id, &mode_select); if (mode_select == MPCC_GAMUT_REMAP_MODE_SELECT_0) { adjust->gamut_adjust_type = GRAPHICS_GAMUT_ADJUST_TYPE_BYPASS; diff --git a/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.h b/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.h index 8e35ebc603a9dd10a3efa8fb2d9a7ba996f9d8bc..eb0c68d0b0c76569e1f0714b6efd218a7bcfa6bf 100644 --- a/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.h +++ b/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.h @@ -241,6 +241,19 @@ void mpc401_update_3dlut_fast_load_select( int mpcc_id, int hubp_idx); +void mpc_program_gamut_remap( + struct mpc *mpc, + unsigned int mpcc_id, + const uint16_t *regval, + enum mpcc_gamut_remap_id gamut_remap_block_id, + enum mpcc_gamut_remap_mode_select mode_select); + +void mpc_read_gamut_remap(struct mpc *mpc, + int mpcc_id, + uint16_t *regval, + enum mpcc_gamut_remap_id gamut_remap_block_id, + uint32_t *mode_select); + void mpc401_update_3dlut_fast_load_select( struct mpc *mpc, int mpcc_id, diff --git a/drivers/gpu/drm/amd/display/dc/os_types.h b/drivers/gpu/drm/amd/display/dc/os_types.h index f2ba76c1e0c092cb276aa3ddf1bf31a3cb086573..782316348941b378d8e38ccacb3e1ce3b4d6e40d 100644 --- a/drivers/gpu/drm/amd/display/dc/os_types.h +++ b/drivers/gpu/drm/amd/display/dc/os_types.h @@ -31,6 +31,7 @@ #include #include #include +#include #include diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c index 84f73fdb0f952599685c4df62282d319a65a569c..3a51be63f020836c6cb94937caec203c280a9d11 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c @@ -839,7 +839,7 @@ static enum dc_status build_mapped_resource( static enum dc_status dce100_validate_bandwidth( struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { int i; bool at_least_one_pipe = false; diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce110/dce110_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce110/dce110_resource.c index f3d5baac11bfd055c70ad0102a30424a93268444..cccde5a6f3cdf6d29d429749044fc6104457d316 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dce110/dce110_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dce110/dce110_resource.c @@ -963,7 +963,7 @@ static enum dc_status build_mapped_resource( static enum dc_status dce110_validate_bandwidth( struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { bool result = false; diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.c index 4225cae68c109cba19573386ce39bc3a497c7ad5..164ba796f64ce1360f86eda954fb700ccd3d3665 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.c @@ -886,7 +886,7 @@ static enum dc_status build_mapped_resource( enum dc_status dce112_validate_bandwidth( struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { bool result = false; diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.h index 6221d749246d85e7d0380a23a4fb93534f72896b..3efc4c55d2d20a4f916ff9abc9f9035688dd3553 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.h +++ b/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.h @@ -45,7 +45,7 @@ enum dc_status dce112_validate_with_context( enum dc_status dce112_validate_bandwidth( struct dc *dc, struct dc_state *context, - bool fast_validate); + enum dc_validate_mode validate_mode); enum dc_status dce112_add_stream_to_ctx( struct dc *dc, diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c index d9ffdded5ce1e176678b2570bc2ea0dc32888ebf..58b59d52dc9d3e386f2c2c7d6c1c6e7ab0aae329 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c @@ -866,7 +866,7 @@ static void dce60_resource_destruct(struct dce110_resource_pool *pool) static enum dc_status dce60_validate_bandwidth( struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { int i; bool at_least_one_pipe = false; diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c index bd5811f975318223b4d29d18d18b6e46b4669453..3e8b0ac11d906db127089fe5fea0f47075c8520a 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c @@ -872,7 +872,7 @@ static void dce80_resource_destruct(struct dce110_resource_pool *pool) static enum dc_status dce80_validate_bandwidth( struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { int i; bool at_least_one_pipe = false; diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn10/dcn10_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn10/dcn10_resource.c index be4ade0853e9adf2008d551a998b61db00a80d3c..652c05c354947c68b9c631808ccee2803746134a 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn10/dcn10_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn10/dcn10_resource.c @@ -1129,12 +1129,12 @@ static void dcn10_destroy_resource_pool(struct resource_pool **pool) static enum dc_status dcn10_validate_bandwidth( struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { bool voltage_supported; DC_FP_START(); - voltage_supported = dcn_validate_bandwidth(dc, context, fast_validate); + voltage_supported = dcn_validate_bandwidth(dc, context, validate_mode); DC_FP_END(); return voltage_supported ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c index 3405be07f5e3abe22cd6920847550ab9323bed12..067a93420a2373b8ef3325ebae20d9c9575f5575 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c @@ -2007,7 +2007,7 @@ bool dcn20_fast_validate_bw( int *pipe_cnt_out, int *pipe_split_from, int *vlevel_out, - bool fast_validate) + enum dc_validate_mode validate_mode) { bool out = false; int split[MAX_PIPES] = { 0 }; @@ -2021,7 +2021,7 @@ bool dcn20_fast_validate_bw( dcn20_merge_pipes_for_validate(dc, context); DC_FP_START(); - pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, fast_validate); + pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, validate_mode); DC_FP_END(); *pipe_cnt_out = pipe_cnt; @@ -2125,7 +2125,7 @@ validate_out: } enum dc_status dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { bool voltage_supported; display_e2e_pipe_params_st *pipes; @@ -2135,7 +2135,7 @@ enum dc_status dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, return DC_FAIL_BANDWIDTH_VALIDATE; DC_FP_START(); - voltage_supported = dcn20_validate_bandwidth_fp(dc, context, fast_validate, pipes); + voltage_supported = dcn20_validate_bandwidth_fp(dc, context, validate_mode, pipes); DC_FP_END(); kfree(pipes); diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.h index c0e062c7407dc346b1f7893ba1aebb0387653d2c..e997d35a8b86ecc935dc550c933cbc84d5bbc3f5 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.h +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.h @@ -119,7 +119,7 @@ void dcn20_set_mcif_arb_params( struct dc_state *context, display_e2e_pipe_params_st *pipes, int pipe_cnt); -enum dc_status dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, bool fast_validate); +enum dc_status dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, enum dc_validate_mode validate_mode); void dcn20_merge_pipes_for_validate( struct dc *dc, struct dc_state *context); @@ -158,7 +158,7 @@ bool dcn20_fast_validate_bw( int *pipe_cnt_out, int *pipe_split_from, int *vlevel_out, - bool fast_validate); + enum dc_validate_mode validate_mode); enum dc_status dcn20_build_mapped_resource(const struct dc *dc, struct dc_state *context, struct dc_stream_state *stream); enum dc_status dcn20_add_stream_to_ctx(struct dc *dc, struct dc_state *new_ctx, struct dc_stream_state *dc_stream); diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.c index 9ab01b65b177be5c02e3c222b592db2cefd81abd..238d7f8beb7c79fe7e0fb645f36333224ce30950 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.c @@ -769,7 +769,7 @@ bool dcn21_fast_validate_bw(struct dc *dc, int *pipe_cnt_out, int *pipe_split_from, int *vlevel_out, - bool fast_validate) + enum dc_validate_mode validate_mode) { bool out = false; int split[MAX_PIPES] = { 0 }; @@ -783,7 +783,7 @@ bool dcn21_fast_validate_bw(struct dc *dc, dcn20_merge_pipes_for_validate(dc, context); DC_FP_START(); - pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, fast_validate); + pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, validate_mode); DC_FP_END(); *pipe_cnt_out = pipe_cnt; @@ -924,7 +924,7 @@ validate_out: * dcn20_validate_bandwidth in dcn20_resource.c. */ static enum dc_status dcn21_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { bool voltage_supported; display_e2e_pipe_params_st *pipes; @@ -934,7 +934,7 @@ static enum dc_status dcn21_validate_bandwidth(struct dc *dc, struct dc_state *c return DC_FAIL_BANDWIDTH_VALIDATE; DC_FP_START(); - voltage_supported = dcn21_validate_bandwidth_fp(dc, context, fast_validate, pipes); + voltage_supported = dcn21_validate_bandwidth_fp(dc, context, validate_mode, pipes); DC_FP_END(); kfree(pipes); diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.h index f7ecc002c2f7f61fa3c182a8fcddff3deccfff09..a017fd9854d14cfd5cdbb34ba9bd752a7d677b12 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.h +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.h @@ -51,6 +51,6 @@ bool dcn21_fast_validate_bw( int *pipe_cnt_out, int *pipe_split_from, int *vlevel_out, - bool fast_validate); + enum dc_validate_mode validate_mode); #endif /* _DCN21_RESOURCE_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c index f631ae34e3200b2736f8516e5747c6141f5510f3..4d4635e01eb644f0062aa932c922d7a37520a542 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c @@ -1319,13 +1319,13 @@ static struct clock_source *dcn30_clock_source_create( int dcn30_populate_dml_pipes_from_context( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate) + enum dc_validate_mode validate_mode) { int i, pipe_cnt; struct resource_context *res_ctx = &context->res_ctx; DC_FP_START(); - dcn20_populate_dml_pipes_from_context(dc, context, pipes, fast_validate); + dcn20_populate_dml_pipes_from_context(dc, context, pipes, validate_mode); DC_FP_END(); for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { @@ -1627,7 +1627,7 @@ noinline bool dcn30_internal_validate_bw( display_e2e_pipe_params_st *pipes, int *pipe_cnt_out, int *vlevel_out, - bool fast_validate, + enum dc_validate_mode validate_mode, bool allow_self_refresh_only) { bool out = false; @@ -1646,7 +1646,7 @@ noinline bool dcn30_internal_validate_bw( context->bw_ctx.dml.vba.VoltageLevel = 0; context->bw_ctx.dml.vba.DRAMClockChangeSupport[0][0] = dm_dram_clock_change_vactive; dc->res_pool->funcs->update_soc_for_wm_a(dc, context); - pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, fast_validate); + pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, validate_mode); if (!pipe_cnt) { out = true; @@ -1655,7 +1655,7 @@ noinline bool dcn30_internal_validate_bw( dml_log_pipe_params(&context->bw_ctx.dml, pipes, pipe_cnt); - if (!fast_validate || !allow_self_refresh_only) { + if (validate_mode == DC_VALIDATE_MODE_AND_PROGRAMMING || !allow_self_refresh_only) { /* * DML favors voltage over p-state, but we're more interested in * supporting p-state over voltage. We can't support p-state in @@ -1669,7 +1669,7 @@ noinline bool dcn30_internal_validate_bw( vlevel = dcn20_validate_apply_pipe_split_flags(dc, context, vlevel, split, merge); } if (allow_self_refresh_only && - (fast_validate || vlevel == context->bw_ctx.dml.soc.num_states || + (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING || vlevel == context->bw_ctx.dml.soc.num_states || vba->DRAMClockChangeSupport[vlevel][vba->maxMpcComb] == dm_dram_clock_change_unsupported)) { /* * If mode is unsupported or there's still no p-state support @@ -1678,7 +1678,7 @@ noinline bool dcn30_internal_validate_bw( * We don't actually support prefetch mode 2, so require that we * at least support prefetch mode 1. */ - context->bw_ctx.dml.validate_max_state = fast_validate; + context->bw_ctx.dml.validate_max_state = (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING); context->bw_ctx.dml.soc.allow_dram_self_refresh_or_dram_clock_change_in_vblank = dm_allow_self_refresh; @@ -1865,7 +1865,7 @@ noinline bool dcn30_internal_validate_bw( } if (repopulate_pipes) - pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, fast_validate); + pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, validate_mode); context->bw_ctx.dml.vba.VoltageLevel = vlevel; *vlevel_out = vlevel; *pipe_cnt_out = pipe_cnt; @@ -2037,7 +2037,7 @@ void dcn30_calculate_wm_and_dlg( enum dc_status dcn30_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { bool out = false; @@ -2055,7 +2055,7 @@ enum dc_status dcn30_validate_bandwidth(struct dc *dc, goto validate_fail; DC_FP_START(); - out = dcn30_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, fast_validate, true); + out = dcn30_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, validate_mode, true); DC_FP_END(); if (pipe_cnt == 0) @@ -2066,7 +2066,7 @@ enum dc_status dcn30_validate_bandwidth(struct dc *dc, BW_VAL_TRACE_END_VOLTAGE_LEVEL(); - if (fast_validate) { + if (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING) { BW_VAL_TRACE_SKIP(fast); goto validate_out; } diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.h index 689d9bdace81e94ad39a6b8598f47fca0b9d5a79..2c967fe55712167580ad2df74a66af33e9c53d19 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.h +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.h @@ -57,14 +57,14 @@ unsigned int dcn30_calc_max_scaled_time( unsigned int urgent_watermark); enum dc_status dcn30_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate); + enum dc_validate_mode validate_mode); bool dcn30_internal_validate_bw( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, int *pipe_cnt_out, int *vlevel_out, - bool fast_validate, + enum dc_validate_mode validate_mode, bool allow_self_refresh_only); void dcn30_calculate_wm_and_dlg( struct dc *dc, struct dc_state *context, @@ -78,7 +78,7 @@ void dcn30_populate_dml_writeback_from_context( int dcn30_populate_dml_pipes_from_context( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate); + enum dc_validate_mode validate_mode); bool dcn30_acquire_post_bldn_3dlut( struct resource_context *res_ctx, diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c index 7e0af5297dc425d880f5ef015e879a522dd22748..88afa59d17b0601e6d7c29d3c7d11b5af9c683d0 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c @@ -1616,14 +1616,14 @@ static bool is_dual_plane(enum surface_pixel_format format) int dcn31x_populate_dml_pipes_from_context(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate) + enum dc_validate_mode validate_mode) { uint32_t pipe_cnt; int i; dc_assert_fp_enabled(); - pipe_cnt = dcn20_populate_dml_pipes_from_context(dc, context, pipes, fast_validate); + pipe_cnt = dcn20_populate_dml_pipes_from_context(dc, context, pipes, validate_mode); for (i = 0; i < pipe_cnt; i++) { pipes[i].pipe.src.gpuvm = 1; @@ -1641,7 +1641,7 @@ int dcn31x_populate_dml_pipes_from_context(struct dc *dc, int dcn31_populate_dml_pipes_from_context( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate) + enum dc_validate_mode validate_mode) { int i, pipe_cnt; struct resource_context *res_ctx = &context->res_ctx; @@ -1649,7 +1649,7 @@ int dcn31_populate_dml_pipes_from_context( bool upscaled = false; DC_FP_START(); - dcn31x_populate_dml_pipes_from_context(dc, context, pipes, fast_validate); + dcn31x_populate_dml_pipes_from_context(dc, context, pipes, validate_mode); DC_FP_END(); for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { @@ -1760,7 +1760,7 @@ dcn31_set_mcif_arb_params(struct dc *dc, enum dc_status dcn31_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { bool out = false; @@ -1778,19 +1778,19 @@ enum dc_status dcn31_validate_bandwidth(struct dc *dc, goto validate_fail; DC_FP_START(); - out = dcn30_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, fast_validate, true); + out = dcn30_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, validate_mode, true); DC_FP_END(); - // Disable fast_validate to set min dcfclk in calculate_wm_and_dlg + // Disable DC_VALIDATE_MODE_ONLY and DC_VALIDATE_MODE_AND_STATE_INDEX to set min dcfclk in calculate_wm_and_dlg if (pipe_cnt == 0) - fast_validate = false; + validate_mode = DC_VALIDATE_MODE_AND_PROGRAMMING; if (!out) goto validate_fail; BW_VAL_TRACE_END_VOLTAGE_LEVEL(); - if (fast_validate) { + if (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING) { BW_VAL_TRACE_SKIP(fast); goto validate_out; } @@ -1850,7 +1850,9 @@ static struct resource_funcs dcn31_res_pool_funcs = { .patch_unknown_plane_state = dcn20_patch_unknown_plane_state, .get_panel_config_defaults = dcn31_get_panel_config_defaults, .get_det_buffer_size = dcn31_get_det_buffer_size, - .get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe + .get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe, + .update_dc_state_for_encoder_switch = dcn31_update_dc_state_for_encoder_switch, + .build_pipe_pix_clk_params = dcn20_build_pipe_pix_clk_params }; static struct clock_source *dcn30_clock_source_create( @@ -1954,6 +1956,9 @@ static bool dcn31_resource_construct( dc->caps.color.mpc.ogam_rom_caps.hlg = 0; dc->caps.color.mpc.ocsc = 1; + dc->caps.num_of_host_routers = 2; + dc->caps.num_of_dpias_per_host_router = 2; + /* Use pipe context based otg sync logic */ dc->config.use_pipe_ctx_sync_logic = true; dc->config.disable_hbr_audio_dp2 = true; @@ -2228,3 +2233,35 @@ struct resource_pool *dcn31_create_resource_pool( kfree(pool); return NULL; } + +enum dc_status dcn31_update_dc_state_for_encoder_switch(struct dc_link *link, + struct dc_link_settings *link_setting, + uint8_t pipe_count, + struct pipe_ctx *pipes, + struct audio_output *audio_output) +{ + struct dc_state *state = link->dc->current_state; + int i; + +#if defined(CONFIG_DRM_AMD_DC_FP) + for (i = 0; i < state->stream_count; i++) + if (state->streams[i] && state->streams[i]->link && state->streams[i]->link == link) + link->dc->hwss.calculate_pix_rate_divider((struct dc *)link->dc, state, state->streams[i]); + + for (i = 0; i < pipe_count; i++) { + link->dc->res_pool->funcs->build_pipe_pix_clk_params(&pipes[i]); + + // Setup audio + if (pipes[i].stream_res.audio != NULL) + build_audio_output(state, &pipes[i], &audio_output[i]); + } +#else + /* This DCN requires rate divider updates and audio reprogramming to allow DP1<-->DP2 link rate switching, + * but the above will not compile on architectures without an FPU. + */ + DC_LOG_WARNING("%s: DP1<-->DP2 link retraining will not work on this DCN on non-FPU platforms", __func__); + ASSERT(0); +#endif + + return DC_OK; +} diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.h index dd82815d7efeceba44b931ff6165557eb47766b5..c32c85ef0ba4774579cb2645c25ab0c3c0c544b4 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.h +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.h @@ -39,7 +39,7 @@ struct dcn31_resource_pool { enum dc_status dcn31_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate); + enum dc_validate_mode validate_mode); void dcn31_calculate_wm_and_dlg( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, @@ -48,7 +48,7 @@ void dcn31_calculate_wm_and_dlg( int dcn31_populate_dml_pipes_from_context( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate); + enum dc_validate_mode validate_mode); void dcn31_populate_dml_writeback_from_context(struct dc *dc, struct resource_context *res_ctx, @@ -66,6 +66,12 @@ struct resource_pool *dcn31_create_resource_pool( unsigned int dcn31_get_det_buffer_size( const struct dc_state *context); +enum dc_status dcn31_update_dc_state_for_encoder_switch(struct dc_link *link, + struct dc_link_settings *link_setting, + uint8_t pipe_count, + struct pipe_ctx *pipes, + struct audio_output *audio_output); + /*temp: B0 specific before switch to dcn313 headers*/ #ifndef regPHYPLLF_PIXCLK_RESYNC_CNTL #define regPHYPLLF_PIXCLK_RESYNC_CNTL 0x007e diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.c index d96bc6cb73adcde383632ad634c0ad985a4e60c2..dedf7bce6ecea03041aadcc7e609834cc6b81df7 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.c @@ -1667,12 +1667,12 @@ static struct clock_source *dcn31_clock_source_create( static int dcn314_populate_dml_pipes_from_context( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate) + enum dc_validate_mode validate_mode) { int pipe_cnt; DC_FP_START(); - pipe_cnt = dcn314_populate_dml_pipes_from_context_fpu(dc, context, pipes, fast_validate); + pipe_cnt = dcn314_populate_dml_pipes_from_context_fpu(dc, context, pipes, validate_mode); DC_FP_END(); return pipe_cnt; @@ -1696,7 +1696,7 @@ static void dcn314_get_panel_config_defaults(struct dc_panel_config *panel_confi enum dc_status dcn314_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { bool out = false; @@ -1715,19 +1715,19 @@ enum dc_status dcn314_validate_bandwidth(struct dc *dc, DC_FP_START(); // do not support self refresh only - out = dcn30_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, fast_validate, false); + out = dcn30_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, validate_mode, false); DC_FP_END(); - // Disable fast_validate to set min dcfclk in calculate_wm_and_dlg + // Disable DC_VALIDATE_MODE_ONLY and DC_VALIDATE_MODE_AND_STATE_INDEX to set min dcfclk in calculate_wm_and_dlg if (pipe_cnt == 0) - fast_validate = false; + validate_mode = DC_VALIDATE_MODE_AND_PROGRAMMING; if (!out) goto validate_fail; BW_VAL_TRACE_END_VOLTAGE_LEVEL(); - if (fast_validate) { + if (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING) { BW_VAL_TRACE_SKIP(fast); goto validate_out; } @@ -1779,7 +1779,9 @@ static struct resource_funcs dcn314_res_pool_funcs = { .get_panel_config_defaults = dcn314_get_panel_config_defaults, .get_preferred_eng_id_dpia = dcn314_get_preferred_eng_id_dpia, .get_det_buffer_size = dcn31_get_det_buffer_size, - .get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe + .get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe, + .update_dc_state_for_encoder_switch = dcn31_update_dc_state_for_encoder_switch, + .build_pipe_pix_clk_params = dcn20_build_pipe_pix_clk_params }; static struct clock_source *dcn30_clock_source_create( @@ -1885,6 +1887,9 @@ static bool dcn314_resource_construct( dc->caps.max_disp_clock_khz_at_vmin = 650000; + dc->caps.num_of_host_routers = 2; + dc->caps.num_of_dpias_per_host_router = 2; + /* Use pipe context based otg sync logic */ dc->config.use_pipe_ctx_sync_logic = true; diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.h index f8ba531d6342f2d38b6b92ed84312e4f719c19a1..ac9bb7f097d5c35ae0b425b285db9600accef320 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.h +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.h @@ -41,7 +41,7 @@ struct dcn314_resource_pool { enum dc_status dcn314_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate); + enum dc_validate_mode validate_mode); struct resource_pool *dcn314_create_resource_pool( const struct dc_init_data *init_data, diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn315/dcn315_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn315/dcn315_resource.c index 6c2bb3f63be15ead8a8c37341d4ae6ef3d0cdbf7..d110be626bc26f3dc6a1c432c4c580c23d322087 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn315/dcn315_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn315/dcn315_resource.c @@ -1664,7 +1664,7 @@ static bool allow_pixel_rate_crb(struct dc *dc, struct dc_state *context) static int dcn315_populate_dml_pipes_from_context( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate) + enum dc_validate_mode validate_mode) { int i, pipe_cnt, crb_idx, crb_pipes; struct resource_context *res_ctx = &context->res_ctx; @@ -1674,7 +1674,7 @@ static int dcn315_populate_dml_pipes_from_context( bool pixel_rate_crb = allow_pixel_rate_crb(dc, context); DC_FP_START(); - dcn31x_populate_dml_pipes_from_context(dc, context, pipes, fast_validate); + dcn31x_populate_dml_pipes_from_context(dc, context, pipes, validate_mode); DC_FP_END(); for (i = 0, pipe_cnt = 0, crb_pipes = 0; i < dc->res_pool->pipe_count; i++) { @@ -1844,7 +1844,9 @@ static struct resource_funcs dcn315_res_pool_funcs = { .get_panel_config_defaults = dcn315_get_panel_config_defaults, .get_power_profile = dcn315_get_power_profile, .get_det_buffer_size = dcn31_get_det_buffer_size, - .get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe + .get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe, + .update_dc_state_for_encoder_switch = dcn31_update_dc_state_for_encoder_switch, + .build_pipe_pix_clk_params = dcn20_build_pipe_pix_clk_params }; static bool dcn315_resource_construct( diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn316/dcn316_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn316/dcn316_resource.c index 568094827212dfa60e6b0eeb25934b84c2f3898d..939811858ff79db1d69568350f32b80e88e2a703 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn316/dcn316_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn316/dcn316_resource.c @@ -1610,7 +1610,7 @@ static bool is_dual_plane(enum surface_pixel_format format) static int dcn316_populate_dml_pipes_from_context( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate) + enum dc_validate_mode validate_mode) { int i, pipe_cnt; struct resource_context *res_ctx = &context->res_ctx; @@ -1618,7 +1618,7 @@ static int dcn316_populate_dml_pipes_from_context( const int max_usable_det = context->bw_ctx.dml.ip.config_return_buffer_size_in_kbytes - DCN3_16_MIN_COMPBUF_SIZE_KB; DC_FP_START(); - dcn31x_populate_dml_pipes_from_context(dc, context, pipes, fast_validate); + dcn31x_populate_dml_pipes_from_context(dc, context, pipes, validate_mode); DC_FP_END(); for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { @@ -1720,7 +1720,9 @@ static struct resource_funcs dcn316_res_pool_funcs = { .patch_unknown_plane_state = dcn20_patch_unknown_plane_state, .get_panel_config_defaults = dcn316_get_panel_config_defaults, .get_det_buffer_size = dcn31_get_det_buffer_size, - .get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe + .get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe, + .update_dc_state_for_encoder_switch = dcn31_update_dc_state_for_encoder_switch, + .build_pipe_pix_clk_params = dcn20_build_pipe_pix_clk_params }; static bool dcn316_resource_construct( diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c index bb0dae0be5b87a1919ebfe623a08e49eb1167365..9ffa10189eee60d6396f1bf7853eb7fd75361f0c 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c @@ -1742,7 +1742,7 @@ void dcn32_add_phantom_pipes(struct dc *dc, struct dc_state *context, } } -static bool dml1_validate(struct dc *dc, struct dc_state *context, bool fast_validate) +static bool dml1_validate(struct dc *dc, struct dc_state *context, enum dc_validate_mode validate_mode) { bool out = false; @@ -1767,7 +1767,7 @@ static bool dml1_validate(struct dc *dc, struct dc_state *context, bool fast_val goto validate_fail; DC_FP_START(); - out = dcn32_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, fast_validate); + out = dcn32_internal_validate_bw(dc, context, pipes, &pipe_cnt, &vlevel, validate_mode); DC_FP_END(); if (pipe_cnt == 0) @@ -1778,7 +1778,7 @@ static bool dml1_validate(struct dc *dc, struct dc_state *context, bool fast_val BW_VAL_TRACE_END_VOLTAGE_LEVEL(); - if (fast_validate) { + if (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING) { BW_VAL_TRACE_SKIP(fast); goto validate_out; } @@ -1809,7 +1809,7 @@ validate_out: enum dc_status dcn32_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { unsigned int i; enum dc_status status; @@ -1827,11 +1827,11 @@ enum dc_status dcn32_validate_bandwidth(struct dc *dc, if (dc->debug.using_dml2) status = dml2_validate(dc, context, context->power_source == DC_POWER_SOURCE_DC ? context->bw_ctx.dml2_dc_power_source : context->bw_ctx.dml2, - fast_validate) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; + validate_mode) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; else - status = dml1_validate(dc, context, fast_validate) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; + status = dml1_validate(dc, context, validate_mode) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; - if (!fast_validate && status == DC_OK && dc_state_is_subvp_in_use(context)) { + if (validate_mode == DC_VALIDATE_MODE_AND_PROGRAMMING && status == DC_OK && dc_state_is_subvp_in_use(context)) { /* check new stream configuration still supports cursor if subvp used */ for (i = 0; i < context->stream_count; i++) { stream = context->streams[i]; @@ -1846,14 +1846,14 @@ enum dc_status dcn32_validate_bandwidth(struct dc *dc, }; } - if (!fast_validate && status == DC_FAIL_HW_CURSOR_SUPPORT) { + if (validate_mode == DC_VALIDATE_MODE_AND_PROGRAMMING && status == DC_FAIL_HW_CURSOR_SUPPORT) { /* attempt to validate again with subvp disabled due to cursor */ if (dc->debug.using_dml2) status = dml2_validate(dc, context, context->power_source == DC_POWER_SOURCE_DC ? context->bw_ctx.dml2_dc_power_source : context->bw_ctx.dml2, - fast_validate) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; + validate_mode) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; else - status = dml1_validate(dc, context, fast_validate) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; + status = dml1_validate(dc, context, validate_mode) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; } return status; @@ -1862,7 +1862,7 @@ enum dc_status dcn32_validate_bandwidth(struct dc *dc, int dcn32_populate_dml_pipes_from_context( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate) + enum dc_validate_mode validate_mode) { int i, pipe_cnt; struct resource_context *res_ctx = &context->res_ctx; @@ -1878,7 +1878,7 @@ int dcn32_populate_dml_pipes_from_context( int num_subvp_none = 0; int odm_slice_count; - dcn20_populate_dml_pipes_from_context(dc, context, pipes, fast_validate); + dcn20_populate_dml_pipes_from_context(dc, context, pipes, validate_mode); /* For single display subvp, look for subvp main so if we have phantom * pipe, we can set odm policy to match main pipe @@ -1960,7 +1960,7 @@ int dcn32_populate_dml_pipes_from_context( /* Only populate DML input with subvp info for full updates. * This is just a workaround -- needs a proper fix. */ - if (!fast_validate) { + if (validate_mode == DC_VALIDATE_MODE_AND_PROGRAMMING) { switch (dc_state_get_pipe_subvp_type(context, pipe)) { case SUBVP_MAIN: pipes[pipe_cnt].pipe.src.use_mall_for_pstate_change = dm_use_mall_pstate_change_sub_viewport; @@ -2061,21 +2061,15 @@ void dcn32_calculate_wm_and_dlg(struct dc *dc, struct dc_state *context, static void dcn32_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params) { - struct dml2_configuration_options *dml2_opt = &dc->dml2_tmp; - - memcpy(dml2_opt, &dc->dml2_options, sizeof(dc->dml2_options)); - DC_FP_START(); dcn32_update_bw_bounding_box_fpu(dc, bw_params); - dml2_opt->use_clock_dc_limits = false; if (dc->debug.using_dml2 && dc->current_state && dc->current_state->bw_ctx.dml2) - dml2_reinit(dc, dml2_opt, &dc->current_state->bw_ctx.dml2); + dml2_reinit(dc, &dc->dml2_options, &dc->current_state->bw_ctx.dml2); - dml2_opt->use_clock_dc_limits = true; if (dc->debug.using_dml2 && dc->current_state && dc->current_state->bw_ctx.dml2_dc_power_source) - dml2_reinit(dc, dml2_opt, &dc->current_state->bw_ctx.dml2_dc_power_source); + dml2_reinit(dc, &dc->dml2_dc_power_options, &dc->current_state->bw_ctx.dml2_dc_power_source); DC_FP_END(); } @@ -2257,7 +2251,7 @@ static bool dcn32_resource_construct( dc->caps.color.dpp.gamma_corr = 1; dc->caps.color.dpp.dgam_rom_for_yuv = 0; - dc->caps.color.dpp.hw_3d_lut = 1; + dc->caps.color.dpp.hw_3d_lut = 0; dc->caps.color.dpp.ogam_ram = 0; // no OGAM in DPP since DCN1 // no OGAM ROM on DCN2 and later ASICs dc->caps.color.dpp.ogam_rom_caps.srgb = 0; @@ -2276,6 +2270,7 @@ static bool dcn32_resource_construct( dc->caps.color.mpc.ogam_rom_caps.pq = 0; dc->caps.color.mpc.ogam_rom_caps.hlg = 0; dc->caps.color.mpc.ocsc = 1; + dc->caps.color.mpc.preblend = true; /* Use pipe context based otg sync logic */ dc->config.use_pipe_ctx_sync_logic = true; @@ -2519,7 +2514,6 @@ static bool dcn32_resource_construct( } dc->dml2_options.dcn_pipe_count = pool->base.pipe_count; - dc->dml2_options.use_native_pstate_optimization = false; dc->dml2_options.use_native_soc_bb_construction = true; dc->dml2_options.minimize_dispclk_using_odm = true; @@ -2551,6 +2545,10 @@ static bool dcn32_resource_construct( if (ASICREV_IS_GC_11_0_3(dc->ctx->asic_id.hw_internal_rev) && (dc->config.sdpif_request_limit_words_per_umc == 0)) dc->config.sdpif_request_limit_words_per_umc = 16; + /* init DC limited DML2 options */ + memcpy(&dc->dml2_dc_power_options, &dc->dml2_options, sizeof(struct dml2_configuration_options)); + dc->dml2_dc_power_options.use_clock_dc_limits = true; + return true; create_fail: diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h index d60ed77eda80ba3bd00a3482cdac1abdf0500ec8..82f966cf4ed271ca86c7c0e3d66d8621bad56385 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h @@ -100,12 +100,12 @@ void dcn32_add_phantom_pipes(struct dc *dc, enum dc_status dcn32_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate); + enum dc_validate_mode validate_mode); int dcn32_populate_dml_pipes_from_context( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, - bool fast_validate); + enum dc_validate_mode validate_mode); void dcn32_calculate_wm_and_dlg( struct dc *dc, struct dc_state *context, diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c index 7db1f7a5613fe02749d6377a5381422f146c57b6..c53266e16c5823e974db5db38639d8eccd9b765d 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c @@ -1580,21 +1580,15 @@ static struct dc_cap_funcs cap_funcs = { static void dcn321_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params) { - struct dml2_configuration_options *dml2_opt = &dc->dml2_tmp; - - memcpy(dml2_opt, &dc->dml2_options, sizeof(dc->dml2_options)); - DC_FP_START(); dcn321_update_bw_bounding_box_fpu(dc, bw_params); - dml2_opt->use_clock_dc_limits = false; if (dc->debug.using_dml2 && dc->current_state && dc->current_state->bw_ctx.dml2) - dml2_reinit(dc, dml2_opt, &dc->current_state->bw_ctx.dml2); + dml2_reinit(dc, &dc->dml2_options, &dc->current_state->bw_ctx.dml2); - dml2_opt->use_clock_dc_limits = true; if (dc->debug.using_dml2 && dc->current_state && dc->current_state->bw_ctx.dml2_dc_power_source) - dml2_reinit(dc, dml2_opt, &dc->current_state->bw_ctx.dml2_dc_power_source); + dml2_reinit(dc, &dc->dml2_dc_power_options, &dc->current_state->bw_ctx.dml2_dc_power_source); DC_FP_END(); } @@ -1761,8 +1755,8 @@ static bool dcn321_resource_construct( dc->caps.color.dpp.gamma_corr = 1; dc->caps.color.dpp.dgam_rom_for_yuv = 0; - dc->caps.color.dpp.hw_3d_lut = 1; - dc->caps.color.dpp.ogam_ram = 1; + dc->caps.color.dpp.hw_3d_lut = 0; + dc->caps.color.dpp.ogam_ram = 0; // no OGAM ROM on DCN2 and later ASICs dc->caps.color.dpp.ogam_rom_caps.srgb = 0; dc->caps.color.dpp.ogam_rom_caps.bt2020 = 0; @@ -1780,6 +1774,7 @@ static bool dcn321_resource_construct( dc->caps.color.mpc.ogam_rom_caps.pq = 0; dc->caps.color.mpc.ogam_rom_caps.hlg = 0; dc->caps.color.mpc.ocsc = 1; + dc->caps.color.mpc.preblend = true; /* Use pipe context based otg sync logic */ dc->config.use_pipe_ctx_sync_logic = true; @@ -2018,7 +2013,6 @@ static bool dcn321_resource_construct( } dc->dml2_options.dcn_pipe_count = pool->base.pipe_count; - dc->dml2_options.use_native_pstate_optimization = false; dc->dml2_options.use_native_soc_bb_construction = true; dc->dml2_options.minimize_dispclk_using_odm = true; @@ -2046,6 +2040,10 @@ static bool dcn321_resource_construct( dc->dml2_options.max_segments_per_hubp = 18; dc->dml2_options.det_segment_size = DCN3_2_DET_SEG_SIZE; + /* init DC limited DML2 options */ + memcpy(&dc->dml2_dc_power_options, &dc->dml2_options, sizeof(struct dml2_configuration_options)); + dc->dml2_dc_power_options.use_clock_dc_limits = true; + return true; create_fail: diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c index 72c6cf047db06f4c9fc6bf0f9a7308d582a2f6b9..e327dee9be21f956fbcfec1ef8fe2b906aed827b 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c @@ -1734,15 +1734,15 @@ static void dcn35_get_panel_config_defaults(struct dc_panel_config *panel_config static enum dc_status dcn35_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { bool out = false; out = dml2_validate(dc, context, context->power_source == DC_POWER_SOURCE_DC ? context->bw_ctx.dml2_dc_power_source : context->bw_ctx.dml2, - fast_validate); + validate_mode); - if (fast_validate) + if (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING) return out ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; DC_FP_START(); @@ -1786,7 +1786,9 @@ static struct resource_funcs dcn35_res_pool_funcs = { .get_panel_config_defaults = dcn35_get_panel_config_defaults, .get_preferred_eng_id_dpia = dcn35_get_preferred_eng_id_dpia, .get_det_buffer_size = dcn31_get_det_buffer_size, - .get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe + .get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe, + .update_dc_state_for_encoder_switch = dcn31_update_dc_state_for_encoder_switch, + .build_pipe_pix_clk_params = dcn20_build_pipe_pix_clk_params }; static bool dcn35_resource_construct( @@ -1874,7 +1876,7 @@ static bool dcn35_resource_construct( dc->caps.color.dpp.gamma_corr = 1; dc->caps.color.dpp.dgam_rom_for_yuv = 0; - dc->caps.color.dpp.hw_3d_lut = 1; + dc->caps.color.dpp.hw_3d_lut = 0; dc->caps.color.dpp.ogam_ram = 0; // no OGAM in DPP since DCN1 // no OGAM ROM on DCN301 dc->caps.color.dpp.ogam_rom_caps.srgb = 0; @@ -1893,6 +1895,10 @@ static bool dcn35_resource_construct( dc->caps.color.mpc.ogam_rom_caps.pq = 0; dc->caps.color.mpc.ogam_rom_caps.hlg = 0; dc->caps.color.mpc.ocsc = 1; + dc->caps.color.mpc.preblend = true; + + dc->caps.num_of_host_routers = 2; + dc->caps.num_of_dpias_per_host_router = 2; /* max_disp_clock_khz_at_vmin is slightly lower than the STA value in order * to provide some margin. @@ -2153,7 +2159,6 @@ static bool dcn35_resource_construct( dc->dcn_ip->max_num_dpp = pool->base.pipe_count; dc->dml2_options.dcn_pipe_count = pool->base.pipe_count; - dc->dml2_options.use_native_pstate_optimization = true; dc->dml2_options.use_native_soc_bb_construction = true; dc->dml2_options.minimize_dispclk_using_odm = false; if (dc->config.EnableMinDispClkODM) diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c index 989a270f7deaa02b2b23cd78afad675a17b1afae..2f2976afc2296a1e766c9ae1b3cbb6924679efc0 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c @@ -1714,15 +1714,15 @@ static void dcn35_get_panel_config_defaults(struct dc_panel_config *panel_config static enum dc_status dcn351_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { bool out = false; out = dml2_validate(dc, context, context->power_source == DC_POWER_SOURCE_DC ? context->bw_ctx.dml2_dc_power_source : context->bw_ctx.dml2, - fast_validate); + validate_mode); - if (fast_validate) + if (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING) return out ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; DC_FP_START(); @@ -1758,7 +1758,9 @@ static struct resource_funcs dcn351_res_pool_funcs = { .get_panel_config_defaults = dcn35_get_panel_config_defaults, .get_preferred_eng_id_dpia = dcn351_get_preferred_eng_id_dpia, .get_det_buffer_size = dcn31_get_det_buffer_size, - .get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe + .get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe, + .update_dc_state_for_encoder_switch = dcn31_update_dc_state_for_encoder_switch, + .build_pipe_pix_clk_params = dcn20_build_pipe_pix_clk_params }; static bool dcn351_resource_construct( @@ -1846,7 +1848,7 @@ static bool dcn351_resource_construct( dc->caps.color.dpp.gamma_corr = 1; dc->caps.color.dpp.dgam_rom_for_yuv = 0; - dc->caps.color.dpp.hw_3d_lut = 1; + dc->caps.color.dpp.hw_3d_lut = 0; dc->caps.color.dpp.ogam_ram = 0; // no OGAM in DPP since DCN1 // no OGAM ROM on DCN301 dc->caps.color.dpp.ogam_rom_caps.srgb = 0; @@ -1865,6 +1867,10 @@ static bool dcn351_resource_construct( dc->caps.color.mpc.ogam_rom_caps.pq = 0; dc->caps.color.mpc.ogam_rom_caps.hlg = 0; dc->caps.color.mpc.ocsc = 1; + dc->caps.color.mpc.preblend = true; + + dc->caps.num_of_host_routers = 2; + dc->caps.num_of_dpias_per_host_router = 2; /* max_disp_clock_khz_at_vmin is slightly lower than the STA value in order * to provide some margin. @@ -2125,7 +2131,6 @@ static bool dcn351_resource_construct( dc->dcn_ip->max_num_dpp = pool->base.pipe_count; dc->dml2_options.dcn_pipe_count = pool->base.pipe_count; - dc->dml2_options.use_native_pstate_optimization = true; dc->dml2_options.use_native_soc_bb_construction = true; dc->dml2_options.minimize_dispclk_using_odm = false; if (dc->config.EnableMinDispClkODM) diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn36/dcn36_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn36/dcn36_resource.c index 48e1f234185ff96bab0b75469c791d1b2509494e..5b7848496a70a3b27cdfef4eb7731e79952b8c75 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn36/dcn36_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn36/dcn36_resource.c @@ -1715,15 +1715,15 @@ static void dcn35_get_panel_config_defaults(struct dc_panel_config *panel_config static enum dc_status dcn35_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { bool out = false; out = dml2_validate(dc, context, context->power_source == DC_POWER_SOURCE_DC ? context->bw_ctx.dml2_dc_power_source : context->bw_ctx.dml2, - fast_validate); + validate_mode); - if (fast_validate) + if (validate_mode != DC_VALIDATE_MODE_AND_PROGRAMMING) return out ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; DC_FP_START(); @@ -1759,7 +1759,9 @@ static struct resource_funcs dcn36_res_pool_funcs = { .patch_unknown_plane_state = dcn20_patch_unknown_plane_state, .get_panel_config_defaults = dcn35_get_panel_config_defaults, .get_preferred_eng_id_dpia = dcn36_get_preferred_eng_id_dpia, - .get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe + .get_vstartup_for_pipe = dcn10_get_vstartup_for_pipe, + .update_dc_state_for_encoder_switch = dcn31_update_dc_state_for_encoder_switch, + .build_pipe_pix_clk_params = dcn20_build_pipe_pix_clk_params }; static bool dcn36_resource_construct( @@ -1847,7 +1849,7 @@ static bool dcn36_resource_construct( dc->caps.color.dpp.gamma_corr = 1; dc->caps.color.dpp.dgam_rom_for_yuv = 0; - dc->caps.color.dpp.hw_3d_lut = 1; + dc->caps.color.dpp.hw_3d_lut = 0; dc->caps.color.dpp.ogam_ram = 0; // no OGAM in DPP since DCN1 // no OGAM ROM on DCN301 dc->caps.color.dpp.ogam_rom_caps.srgb = 0; @@ -1866,6 +1868,10 @@ static bool dcn36_resource_construct( dc->caps.color.mpc.ogam_rom_caps.pq = 0; dc->caps.color.mpc.ogam_rom_caps.hlg = 0; dc->caps.color.mpc.ocsc = 1; + dc->caps.color.mpc.preblend = true; + + dc->caps.num_of_host_routers = 2; + dc->caps.num_of_dpias_per_host_router = 2; /* max_disp_clock_khz_at_vmin is slightly lower than the STA value in order * to provide some margin. @@ -2126,7 +2132,6 @@ static bool dcn36_resource_construct( dc->dcn_ip->max_num_dpp = pool->base.pipe_count; dc->dml2_options.dcn_pipe_count = pool->base.pipe_count; - dc->dml2_options.use_native_pstate_optimization = true; dc->dml2_options.use_native_soc_bb_construction = true; dc->dml2_options.minimize_dispclk_using_odm = false; if (dc->config.EnableMinDispClkODM) diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c index f420c4dafa03c42afe410e125f2efad4a0a28428..a9d989f204056fbe93268efcbfbaa913ebddccb4 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c @@ -70,7 +70,6 @@ #include "dml/dcn30/display_mode_vba_30.h" #include "vm_helper.h" #include "dcn20/dcn20_vmid.h" -#include "dml/dcn401/dcn401_fpu.h" #include "dc_state_priv.h" @@ -1608,10 +1607,6 @@ static struct dc_cap_funcs cap_funcs = { static void dcn401_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params) { - struct dml2_configuration_options *dml2_opt = &dc->dml2_tmp; - - memcpy(dml2_opt, &dc->dml2_options, sizeof(dc->dml2_options)); - /* re-calculate the available MALL size if required */ if (bw_params->num_channels > 0) { dc->caps.max_cab_allocation_bytes = dcn401_calc_num_avail_chans_for_mall( @@ -1622,15 +1617,11 @@ static void dcn401_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *b DC_FP_START(); - dcn401_update_bw_bounding_box_fpu(dc, bw_params); - - dml2_opt->use_clock_dc_limits = false; if (dc->debug.using_dml2 && dc->current_state && dc->current_state->bw_ctx.dml2) - dml2_reinit(dc, dml2_opt, &dc->current_state->bw_ctx.dml2); + dml2_reinit(dc, &dc->dml2_options, &dc->current_state->bw_ctx.dml2); - dml2_opt->use_clock_dc_limits = true; if (dc->debug.using_dml2 && dc->current_state && dc->current_state->bw_ctx.dml2_dc_power_source) - dml2_reinit(dc, dml2_opt, &dc->current_state->bw_ctx.dml2_dc_power_source); + dml2_reinit(dc, &dc->dml2_dc_power_options, &dc->current_state->bw_ctx.dml2_dc_power_source); DC_FP_END(); } @@ -1644,7 +1635,7 @@ enum dc_status dcn401_patch_unknown_plane_state(struct dc_plane_state *plane_sta enum dc_status dcn401_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate) + enum dc_validate_mode validate_mode) { unsigned int i; enum dc_status status = DC_OK; @@ -1662,9 +1653,9 @@ enum dc_status dcn401_validate_bandwidth(struct dc *dc, if (dc->debug.using_dml2) status = dml2_validate(dc, context, context->power_source == DC_POWER_SOURCE_DC ? context->bw_ctx.dml2_dc_power_source : context->bw_ctx.dml2, - fast_validate) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; + validate_mode) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; - if (!fast_validate && status == DC_OK && dc_state_is_subvp_in_use(context)) { + if (validate_mode == DC_VALIDATE_MODE_AND_PROGRAMMING && status == DC_OK && dc_state_is_subvp_in_use(context)) { /* check new stream configuration still supports cursor if subvp used */ for (i = 0; i < context->stream_count; i++) { stream = context->streams[i]; @@ -1679,12 +1670,12 @@ enum dc_status dcn401_validate_bandwidth(struct dc *dc, }; } - if (!fast_validate && status == DC_FAIL_HW_CURSOR_SUPPORT) { + if (validate_mode == DC_VALIDATE_MODE_AND_PROGRAMMING && status == DC_FAIL_HW_CURSOR_SUPPORT) { /* attempt to validate again with subvp disabled due to cursor */ if (dc->debug.using_dml2) status = dml2_validate(dc, context, context->power_source == DC_POWER_SOURCE_DC ? context->bw_ctx.dml2_dc_power_source : context->bw_ctx.dml2, - fast_validate) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; + validate_mode) ? DC_OK : DC_FAIL_BANDWIDTH_VALIDATE; } return status; @@ -1957,8 +1948,30 @@ static bool dcn401_resource_construct( dc->caps.color.mpc.ogam_rom_caps.pq = 0; dc->caps.color.mpc.ogam_rom_caps.hlg = 0; dc->caps.color.mpc.ocsc = 1; + dc->caps.color.mpc.preblend = true; dc->config.use_spl = true; dc->config.prefer_easf = true; + + dc->config.dcn_sharpness_range.sdr_rgb_min = 0; + dc->config.dcn_sharpness_range.sdr_rgb_max = 1750; + dc->config.dcn_sharpness_range.sdr_rgb_mid = 750; + dc->config.dcn_sharpness_range.sdr_yuv_min = 0; + dc->config.dcn_sharpness_range.sdr_yuv_max = 3500; + dc->config.dcn_sharpness_range.sdr_yuv_mid = 1500; + dc->config.dcn_sharpness_range.hdr_rgb_min = 0; + dc->config.dcn_sharpness_range.hdr_rgb_max = 2750; + dc->config.dcn_sharpness_range.hdr_rgb_mid = 1500; + + dc->config.dcn_override_sharpness_range.sdr_rgb_min = 0; + dc->config.dcn_override_sharpness_range.sdr_rgb_max = 3250; + dc->config.dcn_override_sharpness_range.sdr_rgb_mid = 1250; + dc->config.dcn_override_sharpness_range.sdr_yuv_min = 0; + dc->config.dcn_override_sharpness_range.sdr_yuv_max = 3500; + dc->config.dcn_override_sharpness_range.sdr_yuv_mid = 1500; + dc->config.dcn_override_sharpness_range.hdr_rgb_min = 0; + dc->config.dcn_override_sharpness_range.hdr_rgb_max = 2750; + dc->config.dcn_override_sharpness_range.hdr_rgb_mid = 1500; + dc->config.dc_mode_clk_limit_support = true; dc->config.enable_windowed_mpo_odm = true; dc->config.set_pipe_unlock_order = true; /* Need to ensure DET gets freed before allocating */ @@ -2195,7 +2208,6 @@ static bool dcn401_resource_construct( dc->config.sdpif_request_limit_words_per_umc = 16; dc->dml2_options.dcn_pipe_count = pool->base.pipe_count; - dc->dml2_options.use_native_pstate_optimization = false; dc->dml2_options.use_native_soc_bb_construction = true; dc->dml2_options.minimize_dispclk_using_odm = true; dc->dml2_options.map_dc_pipes_with_callbacks = true; @@ -2228,6 +2240,10 @@ static bool dcn401_resource_construct( /* SPL */ dc->caps.scl_caps.sharpener_support = true; + /* init DC limited DML2 options */ + memcpy(&dc->dml2_dc_power_options, &dc->dml2_options, sizeof(struct dml2_configuration_options)); + dc->dml2_dc_power_options.use_clock_dc_limits = true; + return true; create_fail: diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.h index dc52a30991afde1f069439bbee8f2549c629858e..2ae6831c31effff849f671966ababf3182791a7b 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.h +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.h @@ -24,7 +24,7 @@ enum dc_status dcn401_patch_unknown_plane_state(struct dc_plane_state *plane_sta enum dc_status dcn401_validate_bandwidth(struct dc *dc, struct dc_state *context, - bool fast_validate); + enum dc_validate_mode validate_mode); void dcn401_prepare_mcache_programming(struct dc *dc, struct dc_state *context); diff --git a/drivers/gpu/drm/amd/display/dc/sspl/dc_spl.c b/drivers/gpu/drm/amd/display/dc/sspl/dc_spl.c index e0008c5f08ad234b3bb19ac39ddbcebaabafe29a..55b929ca798298a6d9f994436770a962c984c0a6 100644 --- a/drivers/gpu/drm/amd/display/dc/sspl/dc_spl.c +++ b/drivers/gpu/drm/amd/display/dc/sspl/dc_spl.c @@ -196,7 +196,12 @@ static struct spl_rect calculate_mpc_slice_in_timing_active( int epimo = mpc_slice_count - plane_clip_rec->width % mpc_slice_count - 1; struct spl_rect mpc_rec; - if (use_recout_width_aligned) { + if (spl_in->basic_in.custom_width != 0) { + mpc_rec.width = spl_in->basic_in.custom_width; + mpc_rec.x = spl_in->basic_in.custom_x; + mpc_rec.height = plane_clip_rec->height; + mpc_rec.y = plane_clip_rec->y; + } else if (use_recout_width_aligned) { mpc_rec.width = recout_width_align; if ((mpc_rec.width * (mpc_slice_idx + 1)) > plane_clip_rec->width) { mpc_rec.width = plane_clip_rec->width % recout_width_align; @@ -219,7 +224,7 @@ static struct spl_rect calculate_mpc_slice_in_timing_active( /* extra pixels in the division remainder need to go to pipes after * the extra pixel index minus one(epimo) defined here as: */ - if (mpc_slice_idx > epimo) { + if (mpc_slice_idx > epimo && spl_in->basic_in.custom_width == 0) { mpc_rec.x += mpc_slice_idx - epimo - 1; mpc_rec.width += 1; } @@ -252,10 +257,10 @@ static struct spl_rect calculate_odm_slice_in_timing_active(struct spl_in *spl_i odm_rec.x = odm_slice_width * odm_slice_idx; odm_rec.width = is_last_odm_slice ? - /* last slice width is the reminder of h_active */ - h_active - odm_slice_width * (odm_slice_count - 1) : - /* odm slice width is the floor of h_active / count */ - odm_slice_width; + /* last slice width is the reminder of h_active */ + h_active - odm_slice_width * (odm_slice_count - 1) : + /* odm slice width is the floor of h_active / count */ + odm_slice_width; odm_rec.y = 0; odm_rec.height = v_active; @@ -884,7 +889,9 @@ static bool spl_get_isharp_en(struct spl_in *spl_in, /* Calculate number of tap with adaptive scaling off */ static void spl_get_taps_non_adaptive_scaler( - struct spl_scratch *spl_scratch, const struct spl_taps *in_taps, bool always_scale) + struct spl_scratch *spl_scratch, + const struct spl_taps *in_taps, + bool is_subsampled) { bool check_max_downscale = false; @@ -945,14 +952,15 @@ static void spl_get_taps_non_adaptive_scaler( SPL_ASSERT(check_max_downscale); - if (IDENTITY_RATIO(spl_scratch->scl_data.ratios.horz) && !always_scale) + if (IDENTITY_RATIO(spl_scratch->scl_data.ratios.horz)) spl_scratch->scl_data.taps.h_taps = 1; - if (IDENTITY_RATIO(spl_scratch->scl_data.ratios.vert) && !always_scale) + if (IDENTITY_RATIO(spl_scratch->scl_data.ratios.vert)) spl_scratch->scl_data.taps.v_taps = 1; - if (IDENTITY_RATIO(spl_scratch->scl_data.ratios.horz_c) && !always_scale) + if (IDENTITY_RATIO(spl_scratch->scl_data.ratios.horz_c) && !is_subsampled) spl_scratch->scl_data.taps.h_taps_c = 1; - if (IDENTITY_RATIO(spl_scratch->scl_data.ratios.vert_c) && !always_scale) + if (IDENTITY_RATIO(spl_scratch->scl_data.ratios.vert_c) && !is_subsampled) spl_scratch->scl_data.taps.v_taps_c = 1; + } /* Calculate optimal number of taps */ @@ -965,15 +973,13 @@ static bool spl_get_optimal_number_of_taps( unsigned int max_taps_y, max_taps_c; unsigned int min_taps_y, min_taps_c; enum lb_memory_config lb_config; - bool skip_easf = false; - bool always_scale = spl_in->basic_out.always_scale; + bool skip_easf = false; bool is_subsampled = spl_is_subsampled_format(spl_in->basic_in.format); - if (spl_scratch->scl_data.viewport.width > spl_scratch->scl_data.h_active && max_downscale_src_width != 0 && spl_scratch->scl_data.viewport.width > max_downscale_src_width) { - spl_get_taps_non_adaptive_scaler(spl_scratch, in_taps, always_scale); + spl_get_taps_non_adaptive_scaler(spl_scratch, in_taps, is_subsampled); *enable_easf_v = false; *enable_easf_h = false; *enable_isharp = false; @@ -982,7 +988,7 @@ static bool spl_get_optimal_number_of_taps( /* Disable adaptive scaler and sharpener when integer scaling is enabled */ if (spl_in->scaling_quality.integer_scaling) { - spl_get_taps_non_adaptive_scaler(spl_scratch, in_taps, always_scale); + spl_get_taps_non_adaptive_scaler(spl_scratch, in_taps, is_subsampled); *enable_easf_v = false; *enable_easf_h = false; *enable_isharp = false; @@ -997,8 +1003,9 @@ static bool spl_get_optimal_number_of_taps( * From programming guide: taps = min{ ceil(2*H_RATIO,1), 8} for downscaling * taps = 4 for upscaling */ - if (skip_easf) - spl_get_taps_non_adaptive_scaler(spl_scratch, in_taps, always_scale); + if (skip_easf) { + spl_get_taps_non_adaptive_scaler(spl_scratch, in_taps, is_subsampled); + } else { if (spl_is_video_format(spl_in->basic_in.format)) { spl_scratch->scl_data.taps.h_taps = 6; @@ -1124,7 +1131,6 @@ static bool spl_get_optimal_number_of_taps( (IDENTITY_RATIO(spl_scratch->scl_data.ratios.vert))) { spl_scratch->scl_data.taps.h_taps = 1; spl_scratch->scl_data.taps.v_taps = 1; - if (IDENTITY_RATIO(spl_scratch->scl_data.ratios.horz_c) && !is_subsampled) spl_scratch->scl_data.taps.h_taps_c = 1; @@ -1149,6 +1155,7 @@ static bool spl_get_optimal_number_of_taps( if ((!*enable_easf_v) && !is_subsampled && (IDENTITY_RATIO(spl_scratch->scl_data.ratios.vert_c))) spl_scratch->scl_data.taps.v_taps_c = 1; + } } return true; diff --git a/drivers/gpu/drm/amd/display/dc/sspl/dc_spl_types.h b/drivers/gpu/drm/amd/display/dc/sspl/dc_spl_types.h index 36a284305a70f9cad9b89d672d235b66c7cfac85..23d254dea18f2410b4259088233f33140126dd21 100644 --- a/drivers/gpu/drm/amd/display/dc/sspl/dc_spl_types.h +++ b/drivers/gpu/drm/amd/display/dc/sspl/dc_spl_types.h @@ -460,6 +460,8 @@ struct basic_in { enum spl_color_space color_space; // Color Space unsigned int max_luminance; // Max Luminance TODO: Is determined in dc_hw_sequencer.c is_sdr bool film_grain_applied; // Film Grain Applied // TODO: To check from where to get this? + int custom_width; // Width for non-standard segmentation - used when != 0 + int custom_x; // Start x for non-standard segmentation - used when custom_width != 0 }; // Basic output information diff --git a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h index 3f3fa1b6a69e8ea7c20c1ee5798ce25cdd0557a4..0bafb6710761841d99c3d176366a33e4fca8a29e 100644 --- a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h +++ b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h @@ -129,7 +129,9 @@ enum dmub_window_id { DMUB_WINDOW_5_TRACEBUFF, DMUB_WINDOW_6_FW_STATE, DMUB_WINDOW_7_SCRATCH_MEM, + DMUB_WINDOW_IB_MEM, DMUB_WINDOW_SHARED_STATE, + DMUB_WINDOW_LSDMA_BUFFER, DMUB_WINDOW_TOTAL, }; @@ -355,6 +357,7 @@ struct dmub_diagnostic_data { uint8_t is_traceport_en : 1; uint8_t is_cw0_enabled : 1; uint8_t is_cw6_enabled : 1; + uint8_t is_pwait : 1; }; struct dmub_srv_inbox { @@ -539,6 +542,7 @@ struct dmub_srv { uint32_t fw_version; bool is_virtual; struct dmub_fb scratch_mem_fb; + struct dmub_fb ib_mem_gart; volatile struct dmub_shared_state_feature_block *shared_state; volatile const struct dmub_fw_state *fw_state; @@ -576,6 +580,7 @@ struct dmub_srv { enum dmub_srv_power_state_type power_state; struct dmub_diagnostic_data debug; + struct dmub_fb lsdma_rb_fb; }; /** @@ -602,14 +607,6 @@ struct dmub_notification { }; }; -/* enum dmub_ips_mode - IPS mode identifier */ -enum dmub_ips_mode { - DMUB_IPS_MODE_IPS1_MAX = 0, - DMUB_IPS_MODE_IPS2, - DMUB_IPS_MODE_IPS1_RCG, - DMUB_IPS_MODE_IPS1_ONO2_ON -}; - /** * DMUB firmware version helper macro - useful for checking if the version * of a firmware to know if feature or functionality is supported or present. diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h index b66bd10cdc9b836bfcf0d4e7dba0ac552caaa88c..5cf5dd5831fcce3623051c031344070edda2a785 100644 --- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h @@ -104,6 +104,14 @@ */ #define DMUB_MAX_FPO_STREAMS 4 +/* Define to ensure that the "common" members always appear in the same + * order in different structs for back compat purposes + */ +#define COMMON_STREAM_STATIC_SUB_STATE \ + struct dmub_fams2_cmd_legacy_stream_static_state legacy; \ + struct dmub_fams2_cmd_subvp_stream_static_state subvp; \ + struct dmub_fams2_cmd_drr_stream_static_state drr; + /* Maximum number of streams on any ASIC. */ #define DMUB_MAX_STREAMS 6 @@ -291,6 +299,31 @@ union dmub_addr { } u; /*<< Low/high bit access */ uint64_t quad_part; /*<< 64 bit address */ }; + +/* Flattened structure containing SOC BB parameters stored in the VBIOS + * It is not practical to store the entire bounding box in VBIOS since the bounding box struct can gain new parameters. + * This also prevents alighment issues when new parameters are added to the SoC BB. + * The following parameters should be added since these values can't be obtained elsewhere: + * -dml2_soc_power_management_parameters + * -dml2_soc_vmin_clock_limits + */ +struct dmub_soc_bb_params { + uint32_t dram_clk_change_blackout_ns; + uint32_t dram_clk_change_read_only_ns; + uint32_t dram_clk_change_write_only_ns; + uint32_t fclk_change_blackout_ns; + uint32_t g7_ppt_blackout_ns; + uint32_t stutter_enter_plus_exit_latency_ns; + uint32_t stutter_exit_latency_ns; + uint32_t z8_stutter_enter_plus_exit_latency_ns; + uint32_t z8_stutter_exit_latency_ns; + uint32_t z8_min_idle_time_ns; + uint32_t type_b_dram_clk_change_blackout_ns; + uint32_t type_b_ppt_blackout_ns; + uint32_t vmin_limit_dispclk_khz; + uint32_t vmin_limit_dcfclk_khz; + uint32_t g7_temperature_read_blackout_ns; +}; #pragma pack(pop) /** @@ -762,6 +795,17 @@ enum dmub_ips_rcg_disable_type { #define DMUB_IPS1_COMMIT_MASK 0x00000004 #define DMUB_IPS2_COMMIT_MASK 0x00000008 +enum dmub_ips_comand_type { + /** + * Start/stop IPS residency measurements for a given IPS mode + */ + DMUB_CMD__IPS_RESIDENCY_CNTL = 0, + /** + * Query IPS residency information for a given IPS mode + */ + DMUB_CMD__IPS_QUERY_RESIDENCY_INFO = 1, +}; + /** * union dmub_fw_boot_options - Boot option definitions for SCRATCH14 */ @@ -856,7 +900,9 @@ union dmub_shared_state_ips_driver_signals { uint32_t allow_ips0_rcg : 1; /**< 1 is IPS0 RCG is allowed */ uint32_t allow_ips1_rcg : 1; /**< 1 is IPS1 RCG is allowed */ uint32_t allow_ips1z8 : 1; /**< 1 is IPS1 Z8 Retention is allowed */ - uint32_t reserved_bits : 24; /**< Reversed bits */ + uint32_t allow_dynamic_ips1 : 1; /**< 1 if IPS1 is allowed in dynamic use cases such as VPB */ + uint32_t allow_dynamic_ips1_z8: 1; /**< 1 if IPS1 z8 ret is allowed in dynamic use cases such as VPB */ + uint32_t reserved_bits : 22; /**< Reversed bits */ } bits; uint32_t all; }; @@ -1508,6 +1554,16 @@ enum dmub_cmd_type { */ DMUB_CMD__FUSED_IO = 89, + /** + * Command type used for all LSDMA commands. + */ + DMUB_CMD__LSDMA = 90, + + /** + * Command type use for all IPS commands. + */ + DMUB_CMD__IPS = 91, + DMUB_CMD__VBIOS = 128, }; @@ -1918,6 +1974,121 @@ struct dmub_rb_cmd_fams2_flip { struct dmub_fams2_flip_info flip_info; }; +struct dmub_cmd_lsdma_data { + union { + struct lsdma_init_data { + union dmub_addr gpu_addr_base; + uint32_t ring_size; + } init_data; + struct lsdma_tiled_copy_data { + uint32_t src_addr_lo; + uint32_t src_addr_hi; + uint32_t dst_addr_lo; + uint32_t dst_addr_hi; + + uint32_t src_x : 16; + uint32_t src_y : 16; + + uint32_t src_width : 16; + uint32_t src_height : 16; + + uint32_t dst_x : 16; + uint32_t dst_y : 16; + + uint32_t dst_width : 16; + uint32_t dst_height : 16; + + uint32_t rect_x : 16; + uint32_t rect_y : 16; + + uint32_t src_swizzle_mode : 5; + uint32_t src_mip_max : 5; + uint32_t src_mip_id : 5; + uint32_t dst_mip_max : 5; + uint32_t dst_swizzle_mode : 5; + uint32_t dst_mip_id : 5; + uint32_t tmz : 1; + uint32_t dcc : 1; + + uint32_t data_format : 6; + uint32_t padding1 : 4; + uint32_t dst_element_size : 3; + uint32_t num_type : 3; + uint32_t src_element_size : 3; + uint32_t write_compress : 2; + uint32_t cache_policy_dst : 2; + uint32_t cache_policy_src : 2; + uint32_t read_compress : 2; + uint32_t src_dim : 2; + uint32_t dst_dim : 2; + uint32_t max_uncom : 1; + + uint32_t max_com : 2; + uint32_t padding : 30; + } tiled_copy_data; + struct lsdma_linear_copy_data { + uint32_t count : 30; + uint32_t cache_policy_dst : 2; + + uint32_t tmz : 1; + uint32_t cache_policy_src : 2; + uint32_t padding : 29; + + uint32_t src_lo; + uint32_t src_hi; + uint32_t dst_lo; + uint32_t dst_hi; + } linear_copy_data; + struct lsdma_reg_write_data { + uint32_t reg_addr; + uint32_t reg_data; + } reg_write_data; + struct lsdma_pio_copy_data { + union { + struct { + uint32_t byte_count : 26; + uint32_t src_loc : 1; + uint32_t dst_loc : 1; + uint32_t src_addr_inc : 1; + uint32_t dst_addr_inc : 1; + uint32_t overlap_disable : 1; + uint32_t constant_fill : 1; + } fields; + uint32_t raw; + } packet; + uint32_t src_lo; + uint32_t src_hi; + uint32_t dst_lo; + uint32_t dst_hi; + } pio_copy_data; + struct lsdma_pio_constfill_data { + union { + struct { + uint32_t byte_count : 26; + uint32_t src_loc : 1; + uint32_t dst_loc : 1; + uint32_t src_addr_inc : 1; + uint32_t dst_addr_inc : 1; + uint32_t overlap_disable : 1; + uint32_t constant_fill : 1; + } fields; + uint32_t raw; + } packet; + uint32_t dst_lo; + uint32_t dst_hi; + uint32_t data; + } pio_constfill_data; + + uint32_t all[14]; + } u; + +}; + +struct dmub_rb_cmd_lsdma { + struct dmub_cmd_header header; + struct dmub_cmd_lsdma_data lsdma_data; +}; + struct dmub_optc_state_v2 { uint32_t v_total_min; uint32_t v_total_max; @@ -1949,6 +2120,28 @@ enum fams2_stream_type { FAMS2_STREAM_TYPE_SUBVP = 4, }; +struct dmub_rect16 { + /** + * Dirty rect x offset. + */ + uint16_t x; + + /** + * Dirty rect y offset. + */ + uint16_t y; + + /** + * Dirty rect width. + */ + uint16_t width; + + /** + * Dirty rect height. + */ + uint16_t height; +}; + /* static stream state */ struct dmub_fams2_legacy_stream_static_state { uint8_t vactive_det_fill_delay_otg_vlines; @@ -2021,11 +2214,13 @@ union dmub_fams2_stream_static_sub_state { }; //v0 union dmub_fams2_cmd_stream_static_sub_state { - struct dmub_fams2_cmd_legacy_stream_static_state legacy; - struct dmub_fams2_cmd_subvp_stream_static_state subvp; - struct dmub_fams2_cmd_drr_stream_static_state drr; + COMMON_STREAM_STATIC_SUB_STATE }; //v1 +union dmub_fams2_stream_static_sub_state_v2 { + COMMON_STREAM_STATIC_SUB_STATE +}; //v2 + struct dmub_fams2_stream_static_state { enum fams2_stream_type type; uint32_t otg_vline_time_ns; @@ -2091,7 +2286,7 @@ struct dmub_fams2_cmd_stream_static_base_state { struct dmub_fams2_stream_static_state_v1 { struct dmub_fams2_cmd_stream_static_base_state base; - union dmub_fams2_cmd_stream_static_sub_state sub_state; + union dmub_fams2_stream_static_sub_state_v2 sub_state; }; //v1 /** @@ -2139,6 +2334,11 @@ union dmub_cmd_fams2_config { } stream_v1; //v1 }; +struct dmub_fams2_config_v2 { + struct dmub_cmd_fams2_global_config global; + struct dmub_fams2_stream_static_state_v1 stream_v1[DMUB_MAX_STREAMS]; //v1 +}; + /** * DMUB rb command definition for FAMS2 (merged SubVP, FPO, Legacy) */ @@ -2147,6 +2347,22 @@ struct dmub_rb_cmd_fams2 { union dmub_cmd_fams2_config config; }; +/** + * Indirect buffer descriptor + */ +struct dmub_ib_data { + union dmub_addr src; // location of indirect buffer in memory + uint16_t size; // indirect buffer size in bytes +}; + +/** + * DMUB rb command definition for commands passed over indirect buffer + */ +struct dmub_rb_cmd_ib { + struct dmub_cmd_header header; + struct dmub_ib_data ib_data; +}; + /** * enum dmub_cmd_idle_opt_type - Idle optimization command type. */ @@ -2170,6 +2386,11 @@ enum dmub_cmd_idle_opt_type { * DCN hardware notify power state. */ DMUB_CMD__IDLE_OPT_SET_DC_POWER_STATE = 3, + + /** + * DCN notify to release HW. + */ + DMUB_CMD__IDLE_OPT_RELEASE_HW = 4, }; /** @@ -2315,7 +2536,8 @@ struct dmub_dig_transmitter_control_data_v1_7 { uint8_t connobj_id; /**< Connector Object Id defined in ObjectId.h */ uint8_t HPO_instance; /**< HPO instance (0: inst0, 1: inst1) */ uint8_t reserved1; /**< For future use */ - uint8_t reserved2[3]; /**< For future use */ + uint8_t skip_phy_ssc_reduction; + uint8_t reserved2[2]; /**< For future use */ uint32_t reserved3[11]; /**< For future use */ }; @@ -2933,6 +3155,7 @@ enum dmub_cmd_fams_type { DMUB_CMD__FAMS2_CONFIG = 4, DMUB_CMD__FAMS2_DRR_UPDATE = 5, DMUB_CMD__FAMS2_FLIP = 6, + DMUB_CMD__FAMS2_IB_CONFIG = 7, }; /** @@ -4416,6 +4639,37 @@ enum dmub_cmd_abm_type { DMUB_CMD__ABM_GET_HISTOGRAM_DATA = 11, }; +/** + * LSDMA command sub-types. + */ +enum dmub_cmd_lsdma_type { + /** + * Initialize parameters for LSDMA. + * Ring buffer is mapped to the ring buffer + */ + DMUB_CMD__LSDMA_INIT_CONFIG = 0, + /** + * LSDMA copies data from source to destination linearly + */ + DMUB_CMD__LSDMA_LINEAR_COPY = 1, + /** + * Send the tiled-to-tiled copy command + */ + DMUB_CMD__LSDMA_TILED_TO_TILED_COPY = 2, + /** + * Send the poll reg write command + */ + DMUB_CMD__LSDMA_POLL_REG_WRITE = 3, + /** + * Send the pio copy command + */ + DMUB_CMD__LSDMA_PIO_COPY = 4, + /** + * Send the pio constfill command + */ + DMUB_CMD__LSDMA_PIO_CONSTFILL = 5, +}; + struct abm_ace_curve { /** * @offsets: ACE curve offsets. @@ -5620,6 +5874,59 @@ struct dmub_rb_cmd_assr_enable { uint32_t reserved[3]; }; +/** + * Current definition of "ips_mode" from driver + */ +enum ips_residency_mode { + IPS_RESIDENCY__IPS1_MAX, + IPS_RESIDENCY__IPS2, + IPS_RESIDENCY__IPS1_RCG, + IPS_RESIDENCY__IPS1_ONO2_ON, +}; + +#define NUM_IPS_HISTOGRAM_BUCKETS 16 + +/** + * IPS residency statistics to be sent to driver - subset of struct dmub_ips_residency_stats + */ +struct dmub_ips_residency_info { + uint32_t residency_millipercent; + uint32_t entry_counter; + uint32_t histogram[NUM_IPS_HISTOGRAM_BUCKETS]; + uint64_t total_time_us; + uint64_t total_inactive_time_us; +}; + +/** + * Data passed from driver to FW in a DMUB_CMD__IPS_RESIDENCY_CNTL command. + */ +struct dmub_cmd_ips_residency_cntl_data { + uint8_t panel_inst; + uint8_t start_measurement; + uint8_t padding[2]; // align to 4-byte boundary +}; + +struct dmub_rb_cmd_ips_residency_cntl { + struct dmub_cmd_header header; + struct dmub_cmd_ips_residency_cntl_data cntl_data; +}; + +/** + * Data passed from FW to driver in a DMUB_CMD__IPS_QUERY_RESIDENCY_INFO command. + */ +struct dmub_cmd_ips_query_residency_info_data { + union dmub_addr dest; + uint32_t size; + uint32_t ips_mode; + uint8_t panel_inst; + uint8_t padding[3]; // align to 4-byte boundary +}; + +struct dmub_rb_cmd_ips_query_residency_info { + struct dmub_cmd_header header; + struct dmub_cmd_ips_query_residency_info_data info_data; +}; + /** * union dmub_rb_cmd - DMUB inbox command. */ @@ -5926,13 +6233,25 @@ union dmub_rb_cmd { * Definition of a DMUB_CMD__PSP_ASSR_ENABLE command. */ struct dmub_rb_cmd_assr_enable assr_enable; + struct dmub_rb_cmd_fams2 fams2_config; + struct dmub_rb_cmd_ib ib_fams2_config; + struct dmub_rb_cmd_fams2_drr_update fams2_drr_update; struct dmub_rb_cmd_fams2_flip fams2_flip; struct dmub_rb_cmd_fused_io fused_io; + + /** + * Definition of a DMUB_CMD__LSDMA command. + */ + struct dmub_rb_cmd_lsdma lsdma; + + struct dmub_rb_cmd_ips_residency_cntl ips_residency_cntl; + + struct dmub_rb_cmd_ips_query_residency_info ips_query_residency_info; }; /** diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c index a308bd6046778368d0f12e2887bf95ddb10f44d9..3f38db752b84462c8319b90e581aa6b6be8f5606 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c @@ -416,7 +416,7 @@ uint32_t dmub_dcn31_get_current_time(struct dmub_srv *dmub) void dmub_dcn31_get_diagnostic_data(struct dmub_srv *dmub) { - uint32_t is_dmub_enabled, is_soft_reset, is_sec_reset; + uint32_t is_dmub_enabled, is_soft_reset, is_sec_reset, is_pwait; uint32_t is_traceport_enabled, is_cw0_enabled, is_cw6_enabled; struct dmub_timeout_info timeout = {0}; @@ -466,6 +466,9 @@ void dmub_dcn31_get_diagnostic_data(struct dmub_srv *dmub) REG_GET(DMCUB_CNTL, DMCUB_ENABLE, &is_dmub_enabled); dmub->debug.is_dmcub_enabled = is_dmub_enabled; + REG_GET(DMCUB_CNTL, DMCUB_PWAIT_MODE_STATUS, &is_pwait); + dmub->debug.is_pwait = is_pwait; + REG_GET(DMCUB_CNTL2, DMCUB_SOFT_RESET, &is_soft_reset); dmub->debug.is_dmcub_soft_reset = is_soft_reset; diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn35.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn35.c index 72a0f078cd1a589f86c512a51082cde31474a391..2228d62adc7e465c9c317bd1c0fb6024644a13f7 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn35.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn35.c @@ -92,19 +92,15 @@ void dmub_dcn35_reset(struct dmub_srv *dmub) uint32_t in_reset, is_enabled, scratch, i, pwait_mode; REG_GET(DMCUB_CNTL2, DMCUB_SOFT_RESET, &in_reset); + REG_GET(DMCUB_CNTL, DMCUB_ENABLE, &is_enabled); - if (in_reset == 0) { + if (in_reset == 0 && is_enabled != 0) { cmd.bits.status = 1; cmd.bits.command_code = DMUB_GPINT__STOP_FW; cmd.bits.param = 0; dmub->hw_funcs.set_gpint(dmub, cmd); - /** - * Timeout covers both the ACK and the wait - * for remaining work to finish. - */ - for (i = 0; i < timeout; ++i) { if (dmub->hw_funcs.is_gpint_acked(dmub, cmd)) break; @@ -130,11 +126,9 @@ void dmub_dcn35_reset(struct dmub_srv *dmub) /* Force reset in case we timed out, DMCUB is likely hung. */ } - REG_GET(DMCUB_CNTL, DMCUB_ENABLE, &is_enabled); - if (is_enabled) { REG_UPDATE(DMCUB_CNTL2, DMCUB_SOFT_RESET, 1); - REG_UPDATE(MMHUBBUB_SOFT_RESET, DMUIF_SOFT_RESET, 1); + udelay(1); REG_UPDATE(DMCUB_CNTL, DMCUB_ENABLE, 0); } @@ -160,11 +154,7 @@ void dmub_dcn35_reset_release(struct dmub_srv *dmub) LONO_SOCCLK_GATE_DISABLE, 1, LONO_DMCUBCLK_GATE_DISABLE, 1); - REG_UPDATE(MMHUBBUB_SOFT_RESET, DMUIF_SOFT_RESET, 1); - udelay(1); REG_UPDATE_2(DMCUB_CNTL, DMCUB_ENABLE, 1, DMCUB_TRACEPORT_EN, 1); - REG_UPDATE(DMCUB_CNTL2, DMCUB_SOFT_RESET, 1); - udelay(1); REG_UPDATE(MMHUBBUB_SOFT_RESET, DMUIF_SOFT_RESET, 0); REG_UPDATE(DMCUB_CNTL2, DMCUB_SOFT_RESET, 0); } @@ -464,7 +454,7 @@ uint32_t dmub_dcn35_get_current_time(struct dmub_srv *dmub) void dmub_dcn35_get_diagnostic_data(struct dmub_srv *dmub) { - uint32_t is_dmub_enabled, is_soft_reset; + uint32_t is_dmub_enabled, is_soft_reset, is_pwait; uint32_t is_traceport_enabled, is_cw6_enabled; struct dmub_timeout_info timeout = {0}; @@ -515,6 +505,9 @@ void dmub_dcn35_get_diagnostic_data(struct dmub_srv *dmub) REG_GET(DMCUB_CNTL, DMCUB_ENABLE, &is_dmub_enabled); dmub->debug.is_dmcub_enabled = is_dmub_enabled; + REG_GET(DMCUB_CNTL, DMCUB_PWAIT_MODE_STATUS, &is_pwait); + dmub->debug.is_pwait = is_pwait; + REG_GET(DMCUB_CNTL2, DMCUB_SOFT_RESET, &is_soft_reset); dmub->debug.is_dmcub_soft_reset = is_soft_reset; diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn401.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn401.c index 2575dbc448f70953f449afbceba401f8d86b6487..b31adbd0d685f16e4c9a4117ded9165b84c9536b 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn401.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn401.c @@ -413,7 +413,7 @@ uint32_t dmub_dcn401_get_current_time(struct dmub_srv *dmub) void dmub_dcn401_get_diagnostic_data(struct dmub_srv *dmub) { - uint32_t is_dmub_enabled, is_soft_reset, is_sec_reset; + uint32_t is_dmub_enabled, is_soft_reset, is_sec_reset, is_pwait; uint32_t is_traceport_enabled, is_cw0_enabled, is_cw6_enabled; struct dmub_timeout_info timeout = {0}; @@ -464,6 +464,9 @@ void dmub_dcn401_get_diagnostic_data(struct dmub_srv *dmub) REG_GET(DMCUB_CNTL, DMCUB_ENABLE, &is_dmub_enabled); dmub->debug.is_dmcub_enabled = is_dmub_enabled; + REG_GET(DMCUB_CNTL, DMCUB_PWAIT_MODE_STATUS, &is_pwait); + dmub->debug.is_pwait = is_pwait; + REG_GET(DMCUB_CNTL2, DMCUB_SOFT_RESET, &is_soft_reset); dmub->debug.is_dmcub_soft_reset = is_soft_reset; diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c index acca7943a8c8289cf6f9e76e4d63b23601a891b6..b17a19400c067eede5b370252f0bcd1acf0d9a17 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c @@ -65,6 +65,12 @@ /* Default scratch mem size. */ #define DMUB_SCRATCH_MEM_SIZE (1024) +/* Default indirect buffer size. */ +#define DMUB_IB_MEM_SIZE (1280) + +/* Default LSDMA ring buffer size. */ +#define DMUB_LSDMA_RB_SIZE (64 * 1024) + /* Number of windows in use. */ #define DMUB_NUM_WINDOWS (DMUB_WINDOW_TOTAL) /* Base addresses. */ @@ -559,7 +565,9 @@ enum dmub_status window_sizes[DMUB_WINDOW_5_TRACEBUFF] = trace_buffer_size; window_sizes[DMUB_WINDOW_6_FW_STATE] = fw_state_size; window_sizes[DMUB_WINDOW_7_SCRATCH_MEM] = DMUB_SCRATCH_MEM_SIZE; + window_sizes[DMUB_WINDOW_IB_MEM] = DMUB_IB_MEM_SIZE; window_sizes[DMUB_WINDOW_SHARED_STATE] = max(DMUB_FW_HEADER_SHARED_STATE_SIZE, shared_state_size); + window_sizes[DMUB_WINDOW_LSDMA_BUFFER] = DMUB_LSDMA_RB_SIZE; out->fb_size = dmub_srv_calc_regions_for_memory_type(params, out, window_sizes, DMUB_WINDOW_MEMORY_TYPE_FB); @@ -645,6 +653,7 @@ enum dmub_status dmub_srv_hw_init(struct dmub_srv *dmub, struct dmub_fb *tracebuff_fb = params->fb[DMUB_WINDOW_5_TRACEBUFF]; struct dmub_fb *fw_state_fb = params->fb[DMUB_WINDOW_6_FW_STATE]; struct dmub_fb *scratch_mem_fb = params->fb[DMUB_WINDOW_7_SCRATCH_MEM]; + struct dmub_fb *ib_mem_gart = params->fb[DMUB_WINDOW_IB_MEM]; struct dmub_fb *shared_state_fb = params->fb[DMUB_WINDOW_SHARED_STATE]; struct dmub_rb_init_params rb_params, outbox0_rb_params; @@ -655,7 +664,7 @@ enum dmub_status dmub_srv_hw_init(struct dmub_srv *dmub, return DMUB_STATUS_INVALID; if (!inst_fb || !stack_fb || !data_fb || !bios_fb || !mail_fb || - !tracebuff_fb || !fw_state_fb || !scratch_mem_fb) { + !tracebuff_fb || !fw_state_fb || !scratch_mem_fb || !ib_mem_gart) { ASSERT(0); return DMUB_STATUS_INVALID; } @@ -741,6 +750,8 @@ enum dmub_status dmub_srv_hw_init(struct dmub_srv *dmub, dmub->scratch_mem_fb = *scratch_mem_fb; + dmub->ib_mem_gart = *ib_mem_gart; + if (dmub->hw_funcs.setup_windows) dmub->hw_funcs.setup_windows(dmub, &cw2, &cw3, &cw4, &cw5, &cw6, ®ion6); diff --git a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c index 3ba9b62ba70b8d512e9c555a751a593f6d217739..71efd2770c99c2c48cbbb14fffde7713b6e88449 100644 --- a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c +++ b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c @@ -147,7 +147,7 @@ unsigned int mod_freesync_calc_v_total_from_refresh( ((unsigned int)(div64_u64((1000000000ULL * 1000000), refresh_in_uhz))); - if (MICRO_HZ_TO_HZ(refresh_in_uhz) <= stream->timing.min_refresh_in_uhz) { + if (refresh_in_uhz <= stream->timing.min_refresh_in_uhz) { /* When the target refresh rate is the minimum panel refresh rate, * round down the vtotal value to avoid stretching vblank over * panel's vtotal boundary. @@ -155,6 +155,14 @@ unsigned int mod_freesync_calc_v_total_from_refresh( v_total = div64_u64(div64_u64(((unsigned long long)( frame_duration_in_ns) * (stream->timing.pix_clk_100hz / 10)), stream->timing.h_total), 1000000); + } else if (refresh_in_uhz >= stream->timing.max_refresh_in_uhz) { + /* When the target refresh rate is the maximum panel refresh rate + * round up the vtotal value to prevent off-by-one error causing + * v_total_min to be below the panel's lower bound + */ + v_total = div64_u64(div64_u64(((unsigned long long)( + frame_duration_in_ns) * (stream->timing.pix_clk_100hz / 10)), + stream->timing.h_total) + (1000000 - 1), 1000000); } else { v_total = div64_u64(div64_u64(((unsigned long long)( frame_duration_in_ns) * (stream->timing.pix_clk_100hz / 10)), diff --git a/drivers/gpu/drm/amd/include/amd_shared.h b/drivers/gpu/drm/amd/include/amd_shared.h index 11374a2cbab87eb06329b573cbf7e0bfe8a76389..bfb446736ca8c4a5c24b415b52eedf16ade89f79 100644 --- a/drivers/gpu/drm/amd/include/amd_shared.h +++ b/drivers/gpu/drm/amd/include/amd_shared.h @@ -396,6 +396,7 @@ enum amd_dpm_forced_level; * (such as allocating any required memory) * @suspend: handles IP specific hw/sw changes for suspend * @resume: handles IP specific hw/sw changes for resume + * @complete: handles IP specific changes after resume * @is_idle: returns current IP block idle status * @wait_for_idle: poll for idle * @check_soft_reset: check soft reset the IP block @@ -427,6 +428,7 @@ struct amd_ip_funcs { int (*prepare_suspend)(struct amdgpu_ip_block *ip_block); int (*suspend)(struct amdgpu_ip_block *ip_block); int (*resume)(struct amdgpu_ip_block *ip_block); + void (*complete)(struct amdgpu_ip_block *ip_block); bool (*is_idle)(struct amdgpu_ip_block *ip_block); int (*wait_for_idle)(struct amdgpu_ip_block *ip_block); bool (*check_soft_reset)(struct amdgpu_ip_block *ip_block); diff --git a/drivers/gpu/drm/amd/include/kgd_pp_interface.h b/drivers/gpu/drm/amd/include/kgd_pp_interface.h index f4d914dc731f8e078b72086ed3c824e18b49acfc..e2b1ea7467b099de493d963e05b8a17301df5dd5 100644 --- a/drivers/gpu/drm/amd/include/kgd_pp_interface.h +++ b/drivers/gpu/drm/amd/include/kgd_pp_interface.h @@ -108,6 +108,8 @@ enum pp_clock_type { PP_VCLK1, PP_DCLK, PP_DCLK1, + PP_ISPICLK, + PP_ISPXCLK, OD_SCLK, OD_MCLK, OD_VDDC_CURVE, diff --git a/drivers/gpu/drm/amd/pm/amdgpu_dpm.c b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c index 5c1cbdc122d24a2134cab411e5adb2b33ee26a93..71d986dd7a6e84d3e01cfa83d81b870a60385d0e 100644 --- a/drivers/gpu/drm/amd/pm/amdgpu_dpm.c +++ b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c @@ -98,6 +98,7 @@ int amdgpu_dpm_set_powergating_by_smu(struct amdgpu_device *adev, case AMD_IP_BLOCK_TYPE_GMC: case AMD_IP_BLOCK_TYPE_ACP: case AMD_IP_BLOCK_TYPE_VPE: + case AMD_IP_BLOCK_TYPE_ISP: if (pp_funcs && pp_funcs->set_powergating_by_smu) ret = (pp_funcs->set_powergating_by_smu( (adev)->powerplay.pp_handle, block_type, gate, 0)); @@ -852,22 +853,16 @@ int amdgpu_dpm_set_soft_freq_range(struct amdgpu_device *adev, uint32_t max) { struct smu_context *smu = adev->powerplay.pp_handle; - int ret = 0; - - if (type != PP_SCLK) - return -EINVAL; if (!is_support_sw_smu(adev)) return -EOPNOTSUPP; - mutex_lock(&adev->pm.mutex); - ret = smu_set_soft_freq_range(smu, - SMU_SCLK, + guard(mutex)(&adev->pm.mutex); + + return smu_set_soft_freq_range(smu, + type, min, max); - mutex_unlock(&adev->pm.mutex); - - return ret; } int amdgpu_dpm_write_watermarks_table(struct amdgpu_device *adev) diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c index 34e71727b27d76effb229c5d92e07734ee2b132a..307ebf7e32267b79dadf3d59d4a5568ad290fad0 100644 --- a/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c @@ -1242,7 +1242,7 @@ static void kv_dpm_enable_bapm(void *handle, bool enable) if (pi->bapm_enable) { ret = amdgpu_kv_smc_bapm_enable(adev, enable); if (ret) - DRM_ERROR("amdgpu_kv_smc_bapm_enable failed\n"); + drm_err(adev_to_drm(adev), "amdgpu_kv_smc_bapm_enable failed\n"); } } @@ -1266,40 +1266,40 @@ static int kv_dpm_enable(struct amdgpu_device *adev) ret = kv_process_firmware_header(adev); if (ret) { - DRM_ERROR("kv_process_firmware_header failed\n"); + drm_err(adev_to_drm(adev), "kv_process_firmware_header failed\n"); return ret; } kv_init_fps_limits(adev); kv_init_graphics_levels(adev); ret = kv_program_bootup_state(adev); if (ret) { - DRM_ERROR("kv_program_bootup_state failed\n"); + drm_err(adev_to_drm(adev), "kv_program_bootup_state failed\n"); return ret; } kv_calculate_dfs_bypass_settings(adev); ret = kv_upload_dpm_settings(adev); if (ret) { - DRM_ERROR("kv_upload_dpm_settings failed\n"); + drm_err(adev_to_drm(adev), "kv_upload_dpm_settings failed\n"); return ret; } ret = kv_populate_uvd_table(adev); if (ret) { - DRM_ERROR("kv_populate_uvd_table failed\n"); + drm_err(adev_to_drm(adev), "kv_populate_uvd_table failed\n"); return ret; } ret = kv_populate_vce_table(adev); if (ret) { - DRM_ERROR("kv_populate_vce_table failed\n"); + drm_err(adev_to_drm(adev), "kv_populate_vce_table failed\n"); return ret; } ret = kv_populate_samu_table(adev); if (ret) { - DRM_ERROR("kv_populate_samu_table failed\n"); + drm_err(adev_to_drm(adev), "kv_populate_samu_table failed\n"); return ret; } ret = kv_populate_acp_table(adev); if (ret) { - DRM_ERROR("kv_populate_acp_table failed\n"); + drm_err(adev_to_drm(adev), "kv_populate_acp_table failed\n"); return ret; } kv_program_vc(adev); @@ -1310,39 +1310,39 @@ static int kv_dpm_enable(struct amdgpu_device *adev) if (pi->enable_auto_thermal_throttling) { ret = kv_enable_auto_thermal_throttling(adev); if (ret) { - DRM_ERROR("kv_enable_auto_thermal_throttling failed\n"); + drm_err(adev_to_drm(adev), "kv_enable_auto_thermal_throttling failed\n"); return ret; } } ret = kv_enable_dpm_voltage_scaling(adev); if (ret) { - DRM_ERROR("kv_enable_dpm_voltage_scaling failed\n"); + drm_err(adev_to_drm(adev), "kv_enable_dpm_voltage_scaling failed\n"); return ret; } ret = kv_set_dpm_interval(adev); if (ret) { - DRM_ERROR("kv_set_dpm_interval failed\n"); + drm_err(adev_to_drm(adev), "kv_set_dpm_interval failed\n"); return ret; } ret = kv_set_dpm_boot_state(adev); if (ret) { - DRM_ERROR("kv_set_dpm_boot_state failed\n"); + drm_err(adev_to_drm(adev), "kv_set_dpm_boot_state failed\n"); return ret; } ret = kv_enable_ulv(adev, true); if (ret) { - DRM_ERROR("kv_enable_ulv failed\n"); + drm_err(adev_to_drm(adev), "kv_enable_ulv failed\n"); return ret; } kv_start_dpm(adev); ret = kv_enable_didt(adev, true); if (ret) { - DRM_ERROR("kv_enable_didt failed\n"); + drm_err(adev_to_drm(adev), "kv_enable_didt failed\n"); return ret; } ret = kv_enable_smc_cac(adev, true); if (ret) { - DRM_ERROR("kv_enable_smc_cac failed\n"); + drm_err(adev_to_drm(adev), "kv_enable_smc_cac failed\n"); return ret; } @@ -1350,7 +1350,7 @@ static int kv_dpm_enable(struct amdgpu_device *adev) ret = amdgpu_kv_smc_bapm_enable(adev, false); if (ret) { - DRM_ERROR("amdgpu_kv_smc_bapm_enable failed\n"); + drm_err(adev_to_drm(adev), "amdgpu_kv_smc_bapm_enable failed\n"); return ret; } @@ -1358,7 +1358,7 @@ static int kv_dpm_enable(struct amdgpu_device *adev) kv_is_internal_thermal_sensor(adev->pm.int_thermal_type)) { ret = kv_set_thermal_temperature_range(adev, KV_TEMP_RANGE_MIN, KV_TEMP_RANGE_MAX); if (ret) { - DRM_ERROR("kv_set_thermal_temperature_range failed\n"); + drm_err(adev_to_drm(adev), "kv_set_thermal_temperature_range failed\n"); return ret; } amdgpu_irq_get(adev, &adev->pm.dpm.thermal.irq, @@ -1382,7 +1382,7 @@ static void kv_dpm_disable(struct amdgpu_device *adev) err = amdgpu_kv_smc_bapm_enable(adev, false); if (err) - DRM_ERROR("amdgpu_kv_smc_bapm_enable failed\n"); + drm_err(adev_to_drm(adev), "amdgpu_kv_smc_bapm_enable failed\n"); if (adev->asic_type == CHIP_MULLINS) kv_enable_nb_dpm(adev, false); @@ -1920,7 +1920,7 @@ static int kv_dpm_set_power_state(void *handle) if (pi->bapm_enable) { ret = amdgpu_kv_smc_bapm_enable(adev, adev->pm.ac_power); if (ret) { - DRM_ERROR("amdgpu_kv_smc_bapm_enable failed\n"); + drm_err(adev_to_drm(adev), "amdgpu_kv_smc_bapm_enable failed\n"); return ret; } } @@ -1931,7 +1931,7 @@ static int kv_dpm_set_power_state(void *handle) kv_update_dfs_bypass_settings(adev, new_ps); ret = kv_calculate_ds_divider(adev); if (ret) { - DRM_ERROR("kv_calculate_ds_divider failed\n"); + drm_err(adev_to_drm(adev), "kv_calculate_ds_divider failed\n"); return ret; } kv_calculate_nbps_level_settings(adev); @@ -1947,7 +1947,7 @@ static int kv_dpm_set_power_state(void *handle) ret = kv_update_vce_dpm(adev, new_ps, old_ps); if (ret) { - DRM_ERROR("kv_update_vce_dpm failed\n"); + drm_err(adev_to_drm(adev), "kv_update_vce_dpm failed\n"); return ret; } kv_update_sclk_t(adev); @@ -1960,7 +1960,7 @@ static int kv_dpm_set_power_state(void *handle) kv_update_dfs_bypass_settings(adev, new_ps); ret = kv_calculate_ds_divider(adev); if (ret) { - DRM_ERROR("kv_calculate_ds_divider failed\n"); + drm_err(adev_to_drm(adev), "kv_calculate_ds_divider failed\n"); return ret; } kv_calculate_nbps_level_settings(adev); @@ -1972,7 +1972,7 @@ static int kv_dpm_set_power_state(void *handle) kv_set_enabled_levels(adev); ret = kv_update_vce_dpm(adev, new_ps, old_ps); if (ret) { - DRM_ERROR("kv_update_vce_dpm failed\n"); + drm_err(adev_to_drm(adev), "kv_update_vce_dpm failed\n"); return ret; } kv_update_acp_boot_level(adev); @@ -2521,7 +2521,7 @@ static int kv_set_thermal_temperature_range(struct amdgpu_device *adev, if (high_temp > max_temp) high_temp = max_temp; if (high_temp < low_temp) { - DRM_ERROR("invalid thermal range: %d - %d\n", low_temp, high_temp); + drm_err(adev_to_drm(adev), "invalid thermal range: %d - %d\n", low_temp, high_temp); return -EINVAL; } @@ -2563,7 +2563,7 @@ static int kv_parse_sys_info_table(struct amdgpu_device *adev) data_offset); if (crev != 8) { - DRM_ERROR("Unsupported IGP table: %d %d\n", frev, crev); + drm_err(adev_to_drm(adev), "Unsupported IGP table: %d %d\n", frev, crev); return -EINVAL; } pi->sys_info.bootup_sclk = le32_to_cpu(igp_info->info_8.ulBootUpEngineClock); @@ -2579,7 +2579,7 @@ static int kv_parse_sys_info_table(struct amdgpu_device *adev) else pi->sys_info.htc_hyst_lmt = igp_info->info_8.ucHtcHystLmt; if (pi->sys_info.htc_tmp_lmt <= pi->sys_info.htc_hyst_lmt) { - DRM_ERROR("The htcTmpLmt should be larger than htcHystLmt.\n"); + drm_err(adev_to_drm(adev), "The htcTmpLmt should be larger than htcHystLmt.\n"); } if (le32_to_cpu(igp_info->info_8.ulSystemConfig) & (1 << 3)) @@ -2886,16 +2886,18 @@ kv_dpm_print_power_state(void *handle, void *request_ps) struct kv_ps *ps = kv_get_ps(rps); struct amdgpu_device *adev = (struct amdgpu_device *)handle; - amdgpu_dpm_print_class_info(rps->class, rps->class2); - amdgpu_dpm_print_cap_info(rps->caps); - printk("\tuvd vclk: %d dclk: %d\n", rps->vclk, rps->dclk); + amdgpu_dpm_dbg_print_class_info(adev, rps->class, rps->class2); + amdgpu_dpm_dbg_print_cap_info(adev, rps->caps); + drm_dbg(adev_to_drm(adev), "vclk: %d, dclk: %d\n", + rps->vclk, rps->dclk); for (i = 0; i < ps->num_levels; i++) { struct kv_pl *pl = &ps->levels[i]; - printk("\t\tpower level %d sclk: %u vddc: %u\n", - i, pl->sclk, - kv_convert_8bit_index_to_voltage(adev, pl->vddc_index)); + drm_dbg(adev_to_drm(adev), + "power level %d sclk: %u vddc: %u\n", + i, pl->sclk, + kv_convert_8bit_index_to_voltage(adev, pl->vddc_index)); } - amdgpu_dpm_print_ps_status(adev, rps); + amdgpu_dpm_dbg_print_ps_status(adev, rps); } static void kv_dpm_fini(struct amdgpu_device *adev) @@ -3013,13 +3015,13 @@ static int kv_dpm_sw_init(struct amdgpu_ip_block *ip_block) adev->pm.dpm.current_ps = adev->pm.dpm.requested_ps = adev->pm.dpm.boot_ps; if (amdgpu_dpm == 1) amdgpu_pm_print_power_states(adev); - DRM_INFO("amdgpu: dpm initialized\n"); + drm_info(adev_to_drm(adev), "dpm initialized\n"); return 0; dpm_failed: kv_dpm_fini(adev); - DRM_ERROR("amdgpu: dpm initialization failed\n"); + drm_err(adev_to_drm(adev), "dpm initialization failed: %d\n", ret); return ret; } diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.c index c7518b13e787955759f645d18c815452d96f7d77..ea3ace882a10ab7fd542320aa00fc54b62ec868a 100644 --- a/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.c +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.c @@ -47,7 +47,7 @@ #define amdgpu_dpm_check_state_equal(adev, cps, rps, equal) \ ((adev)->powerplay.pp_funcs->check_state_equal((adev)->powerplay.pp_handle, (cps), (rps), (equal))) -void amdgpu_dpm_print_class_info(u32 class, u32 class2) +void amdgpu_dpm_dbg_print_class_info(struct amdgpu_device *adev, u32 class, u32 class2) { const char *s; @@ -66,71 +66,45 @@ void amdgpu_dpm_print_class_info(u32 class, u32 class2) s = "performance"; break; } - printk("\tui class: %s\n", s); - printk("\tinternal class:"); + drm_dbg(adev_to_drm(adev), "\tui class: %s\n", s); if (((class & ~ATOM_PPLIB_CLASSIFICATION_UI_MASK) == 0) && (class2 == 0)) - pr_cont(" none"); - else { - if (class & ATOM_PPLIB_CLASSIFICATION_BOOT) - pr_cont(" boot"); - if (class & ATOM_PPLIB_CLASSIFICATION_THERMAL) - pr_cont(" thermal"); - if (class & ATOM_PPLIB_CLASSIFICATION_LIMITEDPOWERSOURCE) - pr_cont(" limited_pwr"); - if (class & ATOM_PPLIB_CLASSIFICATION_REST) - pr_cont(" rest"); - if (class & ATOM_PPLIB_CLASSIFICATION_FORCED) - pr_cont(" forced"); - if (class & ATOM_PPLIB_CLASSIFICATION_3DPERFORMANCE) - pr_cont(" 3d_perf"); - if (class & ATOM_PPLIB_CLASSIFICATION_OVERDRIVETEMPLATE) - pr_cont(" ovrdrv"); - if (class & ATOM_PPLIB_CLASSIFICATION_UVDSTATE) - pr_cont(" uvd"); - if (class & ATOM_PPLIB_CLASSIFICATION_3DLOW) - pr_cont(" 3d_low"); - if (class & ATOM_PPLIB_CLASSIFICATION_ACPI) - pr_cont(" acpi"); - if (class & ATOM_PPLIB_CLASSIFICATION_HD2STATE) - pr_cont(" uvd_hd2"); - if (class & ATOM_PPLIB_CLASSIFICATION_HDSTATE) - pr_cont(" uvd_hd"); - if (class & ATOM_PPLIB_CLASSIFICATION_SDSTATE) - pr_cont(" uvd_sd"); - if (class2 & ATOM_PPLIB_CLASSIFICATION2_LIMITEDPOWERSOURCE_2) - pr_cont(" limited_pwr2"); - if (class2 & ATOM_PPLIB_CLASSIFICATION2_ULV) - pr_cont(" ulv"); - if (class2 & ATOM_PPLIB_CLASSIFICATION2_MVC) - pr_cont(" uvd_mvc"); - } - pr_cont("\n"); + drm_dbg(adev_to_drm(adev), "\tinternal class: none\n"); + else + drm_dbg(adev_to_drm(adev), "\tinternal class: %s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s\n", + (class & ATOM_PPLIB_CLASSIFICATION_BOOT) ? " boot" : "", + (class & ATOM_PPLIB_CLASSIFICATION_THERMAL) ? " thermal" : "", + (class & ATOM_PPLIB_CLASSIFICATION_LIMITEDPOWERSOURCE) ? " limited_pwr" : "", + (class & ATOM_PPLIB_CLASSIFICATION_REST) ? " rest" : "", + (class & ATOM_PPLIB_CLASSIFICATION_FORCED) ? " forced" : "", + (class & ATOM_PPLIB_CLASSIFICATION_3DPERFORMANCE) ? " 3d_perf" : "", + (class & ATOM_PPLIB_CLASSIFICATION_OVERDRIVETEMPLATE) ? " ovrdrv" : "", + (class & ATOM_PPLIB_CLASSIFICATION_UVDSTATE) ? " uvd" : "", + (class & ATOM_PPLIB_CLASSIFICATION_3DLOW) ? " 3d_low" : "", + (class & ATOM_PPLIB_CLASSIFICATION_ACPI) ? " acpi" : "", + (class & ATOM_PPLIB_CLASSIFICATION_HD2STATE) ? " uvd_hd2" : "", + (class & ATOM_PPLIB_CLASSIFICATION_HDSTATE) ? " uvd_hd" : "", + (class & ATOM_PPLIB_CLASSIFICATION_SDSTATE) ? " uvd_sd" : "", + (class2 & ATOM_PPLIB_CLASSIFICATION2_LIMITEDPOWERSOURCE_2) ? " limited_pwr2" : "", + (class2 & ATOM_PPLIB_CLASSIFICATION2_ULV) ? " ulv" : "", + (class2 & ATOM_PPLIB_CLASSIFICATION2_MVC) ? " uvd_mvc" : ""); } -void amdgpu_dpm_print_cap_info(u32 caps) +void amdgpu_dpm_dbg_print_cap_info(struct amdgpu_device *adev, u32 caps) { - printk("\tcaps:"); - if (caps & ATOM_PPLIB_SINGLE_DISPLAY_ONLY) - pr_cont(" single_disp"); - if (caps & ATOM_PPLIB_SUPPORTS_VIDEO_PLAYBACK) - pr_cont(" video"); - if (caps & ATOM_PPLIB_DISALLOW_ON_DC) - pr_cont(" no_dc"); - pr_cont("\n"); + drm_dbg(adev_to_drm(adev), "\tcaps: %s%s%s\n", + (caps & ATOM_PPLIB_SINGLE_DISPLAY_ONLY) ? " single_disp" : "", + (caps & ATOM_PPLIB_SUPPORTS_VIDEO_PLAYBACK) ? " video" : "", + (caps & ATOM_PPLIB_DISALLOW_ON_DC) ? " no_dc" : ""); } -void amdgpu_dpm_print_ps_status(struct amdgpu_device *adev, +void amdgpu_dpm_dbg_print_ps_status(struct amdgpu_device *adev, struct amdgpu_ps *rps) { - printk("\tstatus:"); - if (rps == adev->pm.dpm.current_ps) - pr_cont(" c"); - if (rps == adev->pm.dpm.requested_ps) - pr_cont(" r"); - if (rps == adev->pm.dpm.boot_ps) - pr_cont(" b"); - pr_cont("\n"); + drm_dbg(adev_to_drm(adev), "\tstatus:%s%s%s\n", + rps == adev->pm.dpm.current_ps ? " c" : "", + rps == adev->pm.dpm.requested_ps ? " r" : "", + rps == adev->pm.dpm.boot_ps ? " b" : ""); } void amdgpu_pm_print_power_states(struct amdgpu_device *adev) @@ -699,64 +673,64 @@ void amdgpu_add_thermal_controller(struct amdgpu_device *adev) adev->pm.fan_max_rpm = controller->ucFanMaxRPM; } if (controller->ucType == ATOM_PP_THERMALCONTROLLER_RV6xx) { - DRM_INFO("Internal thermal controller %s fan control\n", + drm_info(adev_to_drm(adev), "Internal thermal controller %s fan control\n", (controller->ucFanParameters & ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); adev->pm.int_thermal_type = THERMAL_TYPE_RV6XX; } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_RV770) { - DRM_INFO("Internal thermal controller %s fan control\n", + drm_info(adev_to_drm(adev), "Internal thermal controller %s fan control\n", (controller->ucFanParameters & ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); adev->pm.int_thermal_type = THERMAL_TYPE_RV770; } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_EVERGREEN) { - DRM_INFO("Internal thermal controller %s fan control\n", + drm_info(adev_to_drm(adev), "Internal thermal controller %s fan control\n", (controller->ucFanParameters & ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); adev->pm.int_thermal_type = THERMAL_TYPE_EVERGREEN; } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_SUMO) { - DRM_INFO("Internal thermal controller %s fan control\n", + drm_info(adev_to_drm(adev), "Internal thermal controller %s fan control\n", (controller->ucFanParameters & ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); adev->pm.int_thermal_type = THERMAL_TYPE_SUMO; } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_NISLANDS) { - DRM_INFO("Internal thermal controller %s fan control\n", + drm_info(adev_to_drm(adev), "Internal thermal controller %s fan control\n", (controller->ucFanParameters & ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); adev->pm.int_thermal_type = THERMAL_TYPE_NI; } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_SISLANDS) { - DRM_INFO("Internal thermal controller %s fan control\n", + drm_info(adev_to_drm(adev), "Internal thermal controller %s fan control\n", (controller->ucFanParameters & ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); adev->pm.int_thermal_type = THERMAL_TYPE_SI; } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_CISLANDS) { - DRM_INFO("Internal thermal controller %s fan control\n", + drm_info(adev_to_drm(adev), "Internal thermal controller %s fan control\n", (controller->ucFanParameters & ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); adev->pm.int_thermal_type = THERMAL_TYPE_CI; } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_KAVERI) { - DRM_INFO("Internal thermal controller %s fan control\n", + drm_info(adev_to_drm(adev), "Internal thermal controller %s fan control\n", (controller->ucFanParameters & ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); adev->pm.int_thermal_type = THERMAL_TYPE_KV; } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_EXTERNAL_GPIO) { - DRM_INFO("External GPIO thermal controller %s fan control\n", + drm_info(adev_to_drm(adev), "External GPIO thermal controller %s fan control\n", (controller->ucFanParameters & ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); adev->pm.int_thermal_type = THERMAL_TYPE_EXTERNAL_GPIO; } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_ADT7473_WITH_INTERNAL) { - DRM_INFO("ADT7473 with internal thermal controller %s fan control\n", + drm_info(adev_to_drm(adev), "ADT7473 with internal thermal controller %s fan control\n", (controller->ucFanParameters & ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); adev->pm.int_thermal_type = THERMAL_TYPE_ADT7473_WITH_INTERNAL; } else if (controller->ucType == ATOM_PP_THERMALCONTROLLER_EMC2103_WITH_INTERNAL) { - DRM_INFO("EMC2103 with internal thermal controller %s fan control\n", + drm_info(adev_to_drm(adev), "EMC2103 with internal thermal controller %s fan control\n", (controller->ucFanParameters & ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); adev->pm.int_thermal_type = THERMAL_TYPE_EMC2103_WITH_INTERNAL; } else if (controller->ucType < ARRAY_SIZE(pp_lib_thermal_controller_names)) { - DRM_INFO("Possible %s thermal controller at 0x%02x %s fan control\n", + drm_info(adev_to_drm(adev), "Possible %s thermal controller at 0x%02x %s fan control\n", pp_lib_thermal_controller_names[controller->ucType], controller->ucI2cAddress >> 1, (controller->ucFanParameters & @@ -772,7 +746,7 @@ void amdgpu_add_thermal_controller(struct amdgpu_device *adev) i2c_new_client_device(&adev->pm.i2c_bus->adapter, &info); } } else { - DRM_INFO("Unknown thermal controller type %d at 0x%02x %s fan control\n", + drm_info(adev_to_drm(adev), "Unknown thermal controller type %d at 0x%02x %s fan control\n", controller->ucType, controller->ucI2cAddress >> 1, (controller->ucFanParameters & @@ -943,9 +917,9 @@ static int amdgpu_dpm_change_power_state_locked(struct amdgpu_device *adev) return -EINVAL; if (amdgpu_dpm == 1 && pp_funcs->print_power_state) { - printk("switching from power state:\n"); + drm_dbg(adev_to_drm(adev), "switching from power state\n"); amdgpu_dpm_print_power_state(adev, adev->pm.dpm.current_ps); - printk("switching to power state:\n"); + drm_dbg(adev_to_drm(adev), "switching to power state\n"); amdgpu_dpm_print_power_state(adev, adev->pm.dpm.requested_ps); } diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.h b/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.h index 93bd3973330cda3270293264c876e72feaf7ee42..7120eef30509296611189ac12f7191ddc8170cfe 100644 --- a/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.h +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.h @@ -23,10 +23,9 @@ #ifndef __LEGACY_DPM_H__ #define __LEGACY_DPM_H__ -void amdgpu_dpm_print_class_info(u32 class, u32 class2); -void amdgpu_dpm_print_cap_info(u32 caps); -void amdgpu_dpm_print_ps_status(struct amdgpu_device *adev, - struct amdgpu_ps *rps); +void amdgpu_dpm_dbg_print_class_info(struct amdgpu_device *adev, u32 class, u32 class2); +void amdgpu_dpm_dbg_print_cap_info(struct amdgpu_device *adev, u32 caps); +void amdgpu_dpm_dbg_print_ps_status(struct amdgpu_device *adev, struct amdgpu_ps *rps); int amdgpu_get_platform_caps(struct amdgpu_device *adev); int amdgpu_parse_extended_power_table(struct amdgpu_device *adev); void amdgpu_free_extended_power_table(struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c index 4c0e976004ba43548e9a3d631760068a40652460..52e732be59e36b8fec8c87387e92369dc211b39e 100644 --- a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c @@ -7951,15 +7951,15 @@ static void si_dpm_print_power_state(void *handle, struct rv7xx_pl *pl; int i; - amdgpu_dpm_print_class_info(rps->class, rps->class2); - amdgpu_dpm_print_cap_info(rps->caps); - DRM_INFO("\tuvd vclk: %d dclk: %d\n", rps->vclk, rps->dclk); + amdgpu_dpm_dbg_print_class_info(adev, rps->class, rps->class2); + amdgpu_dpm_dbg_print_cap_info(adev, rps->caps); + drm_dbg(adev_to_drm(adev), "\tuvd vclk: %d dclk: %d\n", rps->vclk, rps->dclk); for (i = 0; i < ps->performance_level_count; i++) { pl = &ps->performance_levels[i]; - DRM_INFO("\t\tpower level %d sclk: %u mclk: %u vddc: %u vddci: %u pcie gen: %u\n", + drm_dbg(adev_to_drm(adev), "\t\tpower level %d sclk: %u mclk: %u vddc: %u vddci: %u pcie gen: %u\n", i, pl->sclk, pl->mclk, pl->vddc, pl->vddci, pl->pcie_gen + 1); } - amdgpu_dpm_print_ps_status(adev, rps); + amdgpu_dpm_dbg_print_ps_status(adev, rps); } static int si_dpm_early_init(struct amdgpu_ip_block *ip_block) diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c index d79a1d94661a5407b36caa75132bb1b77d278c2a..756afe78a6e5ace9913d2142a07c375c30e458a1 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c +++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c @@ -76,6 +76,7 @@ static void smu_power_profile_mode_get(struct smu_context *smu, enum PP_SMC_POWER_PROFILE profile_mode); static void smu_power_profile_mode_put(struct smu_context *smu, enum PP_SMC_POWER_PROFILE profile_mode); +static enum smu_clk_type smu_convert_to_smuclk(enum pp_clock_type type); static int smu_sys_get_pp_feature_mask(void *handle, char *buf) @@ -134,12 +135,17 @@ int smu_get_status_gfxoff(struct smu_context *smu, uint32_t *value) } int smu_set_soft_freq_range(struct smu_context *smu, - enum smu_clk_type clk_type, + enum pp_clock_type type, uint32_t min, uint32_t max) { + enum smu_clk_type clk_type; int ret = 0; + clk_type = smu_convert_to_smuclk(type); + if (clk_type == SMU_CLK_COUNT) + return -EINVAL; + if (smu->ppt_funcs->set_soft_freq_limited_range) ret = smu->ppt_funcs->set_soft_freq_limited_range(smu, clk_type, @@ -307,6 +313,26 @@ static int smu_dpm_set_vpe_enable(struct smu_context *smu, return ret; } +static int smu_dpm_set_isp_enable(struct smu_context *smu, + bool enable) +{ + struct smu_power_context *smu_power = &smu->smu_power; + struct smu_power_gate *power_gate = &smu_power->power_gate; + int ret; + + if (!smu->ppt_funcs->dpm_set_isp_enable) + return 0; + + if (atomic_read(&power_gate->isp_gated) ^ enable) + return 0; + + ret = smu->ppt_funcs->dpm_set_isp_enable(smu, enable); + if (!ret) + atomic_set(&power_gate->isp_gated, !enable); + + return ret; +} + static int smu_dpm_set_umsch_mm_enable(struct smu_context *smu, bool enable) { @@ -408,6 +434,12 @@ static int smu_dpm_set_power_gate(void *handle, dev_err(smu->adev->dev, "Failed to power %s VPE!\n", gate ? "gate" : "ungate"); break; + case AMD_IP_BLOCK_TYPE_ISP: + ret = smu_dpm_set_isp_enable(smu, !gate); + if (ret) + dev_err(smu->adev->dev, "Failed to power %s ISP!\n", + gate ? "gate" : "ungate"); + break; default: dev_err(smu->adev->dev, "Unsupported block type!\n"); return -EINVAL; @@ -1004,6 +1036,21 @@ static int smu_fini_fb_allocations(struct smu_context *smu) return 0; } +static void smu_update_gpu_addresses(struct smu_context *smu) +{ + struct smu_table_context *smu_table = &smu->smu_table; + struct smu_table *pm_status_table = smu_table->tables + SMU_TABLE_PMSTATUSLOG; + struct smu_table *driver_table = &(smu_table->driver_table); + struct smu_table *dummy_read_1_table = &smu_table->dummy_read_1_table; + + if (pm_status_table->bo) + pm_status_table->mc_address = amdgpu_bo_fb_aper_addr(pm_status_table->bo); + if (driver_table->bo) + driver_table->mc_address = amdgpu_bo_fb_aper_addr(driver_table->bo); + if (dummy_read_1_table->bo) + dummy_read_1_table->mc_address = amdgpu_bo_fb_aper_addr(dummy_read_1_table->bo); +} + /** * smu_alloc_memory_pool - allocate memory pool in the system memory * @@ -1285,6 +1332,7 @@ static int smu_sw_init(struct amdgpu_ip_block *ip_block) atomic_set(&smu->smu_power.power_gate.vcn_gated[i], 1); atomic_set(&smu->smu_power.power_gate.jpeg_gated, 1); atomic_set(&smu->smu_power.power_gate.vpe_gated, 1); + atomic_set(&smu->smu_power.power_gate.isp_gated, 1); atomic_set(&smu->smu_power.power_gate.umsch_mm_gated, 1); smu_init_power_profile(smu); @@ -1672,37 +1720,6 @@ static int smu_smc_hw_setup(struct smu_context *smu) } } - ret = smu_system_features_control(smu, true); - if (ret) { - dev_err(adev->dev, "Failed to enable requested dpm features!\n"); - return ret; - } - - smu_init_xgmi_plpd_mode(smu); - - ret = smu_feature_get_enabled_mask(smu, &features_supported); - if (ret) { - dev_err(adev->dev, "Failed to retrieve supported dpm features!\n"); - return ret; - } - bitmap_copy(feature->supported, - (unsigned long *)&features_supported, - feature->feature_num); - - if (!smu_is_dpm_running(smu)) - dev_info(adev->dev, "dpm has been disabled\n"); - - /* - * Set initialized values (get from vbios) to dpm tables context such as - * gfxclk, memclk, dcefclk, and etc. And enable the DPM feature for each - * type of clks. - */ - ret = smu_set_default_dpm_table(smu); - if (ret) { - dev_err(adev->dev, "Failed to setup default dpm clock tables!\n"); - return ret; - } - if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN5) pcie_gen = 4; else if (adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_GEN4) @@ -1738,6 +1755,37 @@ static int smu_smc_hw_setup(struct smu_context *smu) return ret; } + ret = smu_system_features_control(smu, true); + if (ret) { + dev_err(adev->dev, "Failed to enable requested dpm features!\n"); + return ret; + } + + smu_init_xgmi_plpd_mode(smu); + + ret = smu_feature_get_enabled_mask(smu, &features_supported); + if (ret) { + dev_err(adev->dev, "Failed to retrieve supported dpm features!\n"); + return ret; + } + bitmap_copy(feature->supported, + (unsigned long *)&features_supported, + feature->feature_num); + + if (!smu_is_dpm_running(smu)) + dev_info(adev->dev, "dpm has been disabled\n"); + + /* + * Set initialized values (get from vbios) to dpm tables context such as + * gfxclk, memclk, dcefclk, and etc. And enable the DPM feature for each + * type of clks. + */ + ret = smu_set_default_dpm_table(smu); + if (ret) { + dev_err(adev->dev, "Failed to setup default dpm clock tables!\n"); + return ret; + } + ret = smu_get_thermal_temperature_range(smu); if (ret) { dev_err(adev->dev, "Failed to get thermal temperature ranges!\n"); @@ -1780,6 +1828,9 @@ static int smu_start_smc_engine(struct smu_context *smu) struct amdgpu_device *adev = smu->adev; int ret = 0; + if (amdgpu_virt_xgmi_migrate_enabled(adev)) + smu_update_gpu_addresses(smu); + smu->smc_fw_state = SMU_FW_INIT; if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) { @@ -2935,6 +2986,12 @@ static enum smu_clk_type smu_convert_to_smuclk(enum pp_clock_type type) clk_type = SMU_DCLK; break; case PP_DCLK1: clk_type = SMU_DCLK1; break; + case PP_ISPICLK: + clk_type = SMU_ISPICLK; + break; + case PP_ISPXCLK: + clk_type = SMU_ISPXCLK; + break; case OD_SCLK: clk_type = SMU_OD_SCLK; break; case OD_MCLK: diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h index 9aacc7bc1c6978fd3e1bf9d2311e214426f2a2bd..b52e194397e2e34506e25498658e369991237244 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h @@ -402,6 +402,7 @@ struct smu_power_gate { atomic_t vcn_gated[AMDGPU_MAX_VCN_INSTANCES]; atomic_t jpeg_gated; atomic_t vpe_gated; + atomic_t isp_gated; atomic_t umsch_mm_gated; }; @@ -1435,6 +1436,12 @@ struct pptable_funcs { */ int (*dpm_set_vpe_enable)(struct smu_context *smu, bool enable); + /** + * @dpm_set_isp_enable: Enable/disable ISP engine dynamic power + * management. + */ + int (*dpm_set_isp_enable)(struct smu_context *smu, bool enable); + /** * @dpm_set_umsch_mm_enable: Enable/disable UMSCH engine dynamic power * management. @@ -1635,7 +1642,7 @@ int smu_write_watermarks_table(struct smu_context *smu); int smu_get_dpm_freq_range(struct smu_context *smu, enum smu_clk_type clk_type, uint32_t *min, uint32_t *max); -int smu_set_soft_freq_range(struct smu_context *smu, enum smu_clk_type clk_type, +int smu_set_soft_freq_range(struct smu_context *smu, enum pp_clock_type clk_type, uint32_t min, uint32_t max); int smu_set_gfx_power_up_by_imu(struct smu_context *smu); diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu14_driver_if_v14_0_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu14_driver_if_v14_0_0.h index 1bc30db22f9c980a56f525c9ebb95be20f906b1a..cd44f4254134d526b7945d116c226b0e5ea0a65c 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu14_driver_if_v14_0_0.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu14_driver_if_v14_0_0.h @@ -106,6 +106,7 @@ typedef struct { #define NUM_FCLK_DPM_LEVELS 8 #define NUM_MEM_PSTATE_LEVELS 4 +#define ISP_ALL_TILES_MASK 0x7FF typedef struct { uint32_t UClk; diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_pmfw.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_pmfw.h index d7505cfc433a68914d1a4a3159551a23f763291a..0a2ca544f4e382350e882230ded6b86df006d813 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_pmfw.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_pmfw.h @@ -86,8 +86,10 @@ typedef enum { /*36*/ FEATURE_PIT = 36, /*37*/ FEATURE_DVO = 37, /*38*/ FEATURE_XVMINORPSM_CLKSTOP_DS = 38, +/*39*/ FEATURE_GLOBAL_DPM = 39, +/*40*/ FEATURE_NODE_POWER_MANAGER = 40, -/*39*/ NUM_FEATURES = 39 +/*41*/ NUM_FEATURES = 41 } FEATURE_LIST_e; //enum for MPIO PCIe gen speed msgs @@ -133,7 +135,7 @@ typedef enum { GFX_DVM_MARGIN_COUNT } GFX_DVM_MARGIN_e; -#define SMU_METRICS_TABLE_VERSION 0x12 +#define SMU_METRICS_TABLE_VERSION 0x13 typedef struct __attribute__((packed, aligned(4))) { uint64_t AccumulationCounter; @@ -275,6 +277,16 @@ typedef struct { //PSNs uint64_t PublicSerialNumber_AID[4]; uint64_t PublicSerialNumber_XCD[8]; + + //XGMI + uint32_t MaxXgmiWidth; + uint32_t MaxXgmiBitrate; + + // Telemetry + uint32_t InputTelemetryVoltageInmV; + + // General info + uint32_t pldmVersion[2]; } StaticMetricsTable_t; #pragma pack(pop) diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h index eefdaa0b5df65ef3a916f1c7b6dfc31e72438063..d7a9e41820fae0f3878090ab5791a7e167df7c34 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h @@ -305,6 +305,8 @@ enum smu_clk_type { SMU_MCLK, SMU_PCIE, SMU_LCLK, + SMU_ISPICLK, + SMU_ISPXCLK, SMU_OD_CCLK, SMU_OD_SCLK, SMU_OD_MCLK, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c index 7fad5dfb39c44087cfef43559e07d3f6f2e56e68..aac202d0c30e5649dacf6207fccfe095f4990120 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c @@ -2444,7 +2444,8 @@ static int navi10_update_pcie_parameters(struct smu_context *smu, struct smu_11_0_dpm_context *dpm_context = smu->smu_dpm.dpm_context; PPTable_t *pptable = smu->smu_table.driver_pptable; uint32_t smu_pcie_arg; - int ret, i; + int ret = 0; + int i; /* lclk dpm table setup */ for (i = 0; i < MAX_PCIE_CONF; i++) { @@ -2453,25 +2454,27 @@ static int navi10_update_pcie_parameters(struct smu_context *smu, } for (i = 0; i < NUM_LINK_LEVELS; i++) { - smu_pcie_arg = (i << 16) | - ((pptable->PcieGenSpeed[i] <= pcie_gen_cap) ? (pptable->PcieGenSpeed[i] << 8) : - (pcie_gen_cap << 8)) | ((pptable->PcieLaneCount[i] <= pcie_width_cap) ? - pptable->PcieLaneCount[i] : pcie_width_cap); - ret = smu_cmn_send_smc_msg_with_param(smu, - SMU_MSG_OverridePcieParameters, - smu_pcie_arg, - NULL); - - if (ret) - return ret; - - if (pptable->PcieGenSpeed[i] > pcie_gen_cap) - dpm_context->dpm_tables.pcie_table.pcie_gen[i] = pcie_gen_cap; - if (pptable->PcieLaneCount[i] > pcie_width_cap) - dpm_context->dpm_tables.pcie_table.pcie_lane[i] = pcie_width_cap; + if (pptable->PcieGenSpeed[i] > pcie_gen_cap || + pptable->PcieLaneCount[i] > pcie_width_cap) { + dpm_context->dpm_tables.pcie_table.pcie_gen[i] = + pptable->PcieGenSpeed[i] > pcie_gen_cap ? + pcie_gen_cap : pptable->PcieGenSpeed[i]; + dpm_context->dpm_tables.pcie_table.pcie_lane[i] = + pptable->PcieLaneCount[i] > pcie_width_cap ? + pcie_width_cap : pptable->PcieLaneCount[i]; + smu_pcie_arg = i << 16; + smu_pcie_arg |= pcie_gen_cap << 8; + smu_pcie_arg |= pcie_width_cap; + ret = smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_OverridePcieParameters, + smu_pcie_arg, + NULL); + if (ret) + break; + } } - return 0; + return ret; } static inline void navi10_dump_od_table(struct smu_context *smu, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c index 115e3fa456bcac045f12115e32039ca2ab0aa6f6..d57591509aed181978abb8221f6feba66703273e 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c @@ -2145,7 +2145,8 @@ static int sienna_cichlid_update_pcie_parameters(struct smu_context *smu, uint8_t min_gen_speed, max_gen_speed; uint8_t min_lane_width, max_lane_width; uint32_t smu_pcie_arg; - int ret, i; + int ret = 0; + int i; GET_PPTABLE_MEMBER(PcieGenSpeed, &table_member1); GET_PPTABLE_MEMBER(PcieLaneCount, &table_member2); @@ -2170,19 +2171,22 @@ static int sienna_cichlid_update_pcie_parameters(struct smu_context *smu, pcie_table->pcie_lane[1] = max_lane_width; for (i = 0; i < NUM_LINK_LEVELS; i++) { - smu_pcie_arg = (i << 16 | + if (!(smu->adev->pm.pp_feature & PP_PCIE_DPM_MASK) || + table_member1[i] > pcie_gen_cap || table_member2[i] > pcie_width_cap) { + smu_pcie_arg = (i << 16 | pcie_table->pcie_gen[i] << 8 | pcie_table->pcie_lane[i]); - ret = smu_cmn_send_smc_msg_with_param(smu, - SMU_MSG_OverridePcieParameters, - smu_pcie_arg, - NULL); - if (ret) - return ret; + ret = smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_OverridePcieParameters, + smu_pcie_arg, + NULL); + if (ret) + break; + } } - return 0; + return ret; } static int sienna_cichlid_get_dpm_ultimate_freq(struct smu_context *smu, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c index a55ea76d7399692ce39d9d0903906612f8fb791a..2c9869feba610f08b0e17f3f299024f4d039ed52 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c @@ -666,7 +666,6 @@ static int vangogh_print_clk_levels(struct smu_context *smu, { DpmClocks_t *clk_table = smu->smu_table.clocks_table; SmuMetrics_t metrics; - struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm); int i, idx, size = 0, ret = 0; uint32_t cur_value = 0, value = 0, count = 0; bool cur_value_match_level = false; @@ -682,31 +681,25 @@ static int vangogh_print_clk_levels(struct smu_context *smu, switch (clk_type) { case SMU_OD_SCLK: - if (smu_dpm_ctx->dpm_level == AMD_DPM_FORCED_LEVEL_MANUAL) { - size += sysfs_emit_at(buf, size, "%s:\n", "OD_SCLK"); - size += sysfs_emit_at(buf, size, "0: %10uMhz\n", - (smu->gfx_actual_hard_min_freq > 0) ? smu->gfx_actual_hard_min_freq : smu->gfx_default_hard_min_freq); - size += sysfs_emit_at(buf, size, "1: %10uMhz\n", - (smu->gfx_actual_soft_max_freq > 0) ? smu->gfx_actual_soft_max_freq : smu->gfx_default_soft_max_freq); - } + size += sysfs_emit_at(buf, size, "%s:\n", "OD_SCLK"); + size += sysfs_emit_at(buf, size, "0: %10uMhz\n", + (smu->gfx_actual_hard_min_freq > 0) ? smu->gfx_actual_hard_min_freq : smu->gfx_default_hard_min_freq); + size += sysfs_emit_at(buf, size, "1: %10uMhz\n", + (smu->gfx_actual_soft_max_freq > 0) ? smu->gfx_actual_soft_max_freq : smu->gfx_default_soft_max_freq); break; case SMU_OD_CCLK: - if (smu_dpm_ctx->dpm_level == AMD_DPM_FORCED_LEVEL_MANUAL) { - size += sysfs_emit_at(buf, size, "CCLK_RANGE in Core%d:\n", smu->cpu_core_id_select); - size += sysfs_emit_at(buf, size, "0: %10uMhz\n", - (smu->cpu_actual_soft_min_freq > 0) ? smu->cpu_actual_soft_min_freq : smu->cpu_default_soft_min_freq); - size += sysfs_emit_at(buf, size, "1: %10uMhz\n", - (smu->cpu_actual_soft_max_freq > 0) ? smu->cpu_actual_soft_max_freq : smu->cpu_default_soft_max_freq); - } + size += sysfs_emit_at(buf, size, "CCLK_RANGE in Core%d:\n", smu->cpu_core_id_select); + size += sysfs_emit_at(buf, size, "0: %10uMhz\n", + (smu->cpu_actual_soft_min_freq > 0) ? smu->cpu_actual_soft_min_freq : smu->cpu_default_soft_min_freq); + size += sysfs_emit_at(buf, size, "1: %10uMhz\n", + (smu->cpu_actual_soft_max_freq > 0) ? smu->cpu_actual_soft_max_freq : smu->cpu_default_soft_max_freq); break; case SMU_OD_RANGE: - if (smu_dpm_ctx->dpm_level == AMD_DPM_FORCED_LEVEL_MANUAL) { - size += sysfs_emit_at(buf, size, "%s:\n", "OD_RANGE"); - size += sysfs_emit_at(buf, size, "SCLK: %7uMhz %10uMhz\n", - smu->gfx_default_hard_min_freq, smu->gfx_default_soft_max_freq); - size += sysfs_emit_at(buf, size, "CCLK: %7uMhz %10uMhz\n", - smu->cpu_default_soft_min_freq, smu->cpu_default_soft_max_freq); - } + size += sysfs_emit_at(buf, size, "%s:\n", "OD_RANGE"); + size += sysfs_emit_at(buf, size, "SCLK: %7uMhz %10uMhz\n", + smu->gfx_default_hard_min_freq, smu->gfx_default_soft_max_freq); + size += sysfs_emit_at(buf, size, "CCLK: %7uMhz %10uMhz\n", + smu->cpu_default_soft_min_freq, smu->cpu_default_soft_max_freq); break; case SMU_SOCCLK: /* the level 3 ~ 6 of socclk use the same frequency for vangogh */ diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c index 9481f897432d771b517e9b69e4dd8d21f77dc90a..e97b0cf19197e9e51e29ac1b3d3e29606994a636 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c @@ -497,7 +497,6 @@ static int renoir_print_clk_levels(struct smu_context *smu, int i, idx, size = 0, ret = 0; uint32_t cur_value = 0, value = 0, count = 0, min = 0, max = 0; SmuMetrics_t metrics; - struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm); bool cur_value_match_level = false; memset(&metrics, 0, sizeof(metrics)); @@ -510,28 +509,24 @@ static int renoir_print_clk_levels(struct smu_context *smu, switch (clk_type) { case SMU_OD_RANGE: - if (smu_dpm_ctx->dpm_level == AMD_DPM_FORCED_LEVEL_MANUAL) { - ret = smu_cmn_send_smc_msg_with_param(smu, - SMU_MSG_GetMinGfxclkFrequency, - 0, &min); - if (ret) - return ret; - ret = smu_cmn_send_smc_msg_with_param(smu, - SMU_MSG_GetMaxGfxclkFrequency, - 0, &max); - if (ret) - return ret; - size += sysfs_emit_at(buf, size, "OD_RANGE\nSCLK: %10uMhz %10uMhz\n", min, max); - } + ret = smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_GetMinGfxclkFrequency, + 0, &min); + if (ret) + return ret; + ret = smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_GetMaxGfxclkFrequency, + 0, &max); + if (ret) + return ret; + size += sysfs_emit_at(buf, size, "OD_RANGE\nSCLK: %10uMhz %10uMhz\n", min, max); break; case SMU_OD_SCLK: - if (smu_dpm_ctx->dpm_level == AMD_DPM_FORCED_LEVEL_MANUAL) { - min = (smu->gfx_actual_hard_min_freq > 0) ? smu->gfx_actual_hard_min_freq : smu->gfx_default_hard_min_freq; - max = (smu->gfx_actual_soft_max_freq > 0) ? smu->gfx_actual_soft_max_freq : smu->gfx_default_soft_max_freq; - size += sysfs_emit_at(buf, size, "OD_SCLK\n"); - size += sysfs_emit_at(buf, size, "0:%10uMhz\n", min); - size += sysfs_emit_at(buf, size, "1:%10uMhz\n", max); - } + min = (smu->gfx_actual_hard_min_freq > 0) ? smu->gfx_actual_hard_min_freq : smu->gfx_default_hard_min_freq; + max = (smu->gfx_actual_soft_max_freq > 0) ? smu->gfx_actual_soft_max_freq : smu->gfx_default_soft_max_freq; + size += sysfs_emit_at(buf, size, "OD_SCLK\n"); + size += sysfs_emit_at(buf, size, "0:%10uMhz\n", min); + size += sysfs_emit_at(buf, size, "1:%10uMhz\n", max); break; case SMU_GFXCLK: case SMU_SCLK: diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c index a7167668d1894484f7898cbd85fccb71dbbd20f3..1a1f2a6b2e52128828e820451792da2b34384db3 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c @@ -58,6 +58,7 @@ MODULE_FIRMWARE("amdgpu/aldebaran_smc.bin"); MODULE_FIRMWARE("amdgpu/smu_13_0_0.bin"); +MODULE_FIRMWARE("amdgpu/smu_13_0_0_kicker.bin"); MODULE_FIRMWARE("amdgpu/smu_13_0_7.bin"); MODULE_FIRMWARE("amdgpu/smu_13_0_10.bin"); @@ -92,7 +93,7 @@ const int pmfw_decoded_link_width[7] = {0, 1, 2, 4, 8, 12, 16}; int smu_v13_0_init_microcode(struct smu_context *smu) { struct amdgpu_device *adev = smu->adev; - char ucode_prefix[15]; + char ucode_prefix[30]; int err = 0; const struct smc_firmware_header_v1_0 *hdr; const struct common_firmware_header *header; @@ -103,8 +104,13 @@ int smu_v13_0_init_microcode(struct smu_context *smu) return 0; amdgpu_ucode_ip_version_decode(adev, MP1_HWIP, ucode_prefix, sizeof(ucode_prefix)); - err = amdgpu_ucode_request(adev, &adev->pm.fw, AMDGPU_UCODE_REQUIRED, - "amdgpu/%s.bin", ucode_prefix); + + if (amdgpu_is_kicker_fw(adev)) + err = amdgpu_ucode_request(adev, &adev->pm.fw, AMDGPU_UCODE_REQUIRED, + "amdgpu/%s_kicker.bin", ucode_prefix); + else + err = amdgpu_ucode_request(adev, &adev->pm.fw, AMDGPU_UCODE_REQUIRED, + "amdgpu/%s.bin", ucode_prefix); if (err) goto out; @@ -2380,7 +2386,8 @@ int smu_v13_0_update_pcie_parameters(struct smu_context *smu, &dpm_context->dpm_tables.pcie_table; int num_of_levels = pcie_table->num_of_link_levels; uint32_t smu_pcie_arg; - int ret, i; + int ret = 0; + int i; if (!num_of_levels) return 0; @@ -2396,30 +2403,38 @@ int smu_v13_0_update_pcie_parameters(struct smu_context *smu, for (i = 0; i < num_of_levels; i++) { pcie_table->pcie_gen[i] = pcie_gen_cap; pcie_table->pcie_lane[i] = pcie_width_cap; + smu_pcie_arg = i << 16; + smu_pcie_arg |= pcie_table->pcie_gen[i] << 8; + smu_pcie_arg |= pcie_table->pcie_lane[i]; + + ret = smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_OverridePcieParameters, + smu_pcie_arg, + NULL); + if (ret) + break; } } else { for (i = 0; i < num_of_levels; i++) { - if (pcie_table->pcie_gen[i] > pcie_gen_cap) + if (pcie_table->pcie_gen[i] > pcie_gen_cap || + pcie_table->pcie_lane[i] > pcie_width_cap) { pcie_table->pcie_gen[i] = pcie_gen_cap; - if (pcie_table->pcie_lane[i] > pcie_width_cap) pcie_table->pcie_lane[i] = pcie_width_cap; + smu_pcie_arg = i << 16; + smu_pcie_arg |= pcie_table->pcie_gen[i] << 8; + smu_pcie_arg |= pcie_table->pcie_lane[i]; + + ret = smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_OverridePcieParameters, + smu_pcie_arg, + NULL); + if (ret) + break; + } } } - for (i = 0; i < num_of_levels; i++) { - smu_pcie_arg = i << 16; - smu_pcie_arg |= pcie_table->pcie_gen[i] << 8; - smu_pcie_arg |= pcie_table->pcie_lane[i]; - - ret = smu_cmn_send_smc_msg_with_param(smu, - SMU_MSG_OverridePcieParameters, - smu_pcie_arg, - NULL); - if (ret) - return ret; - } - - return 0; + return ret; } int smu_v13_0_disable_pmfw_state(struct smu_context *smu) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c index 5a9711e8cf689569a13b7f1e9041d1c66ea70ef6..e084ed99ec0e913934674ccb55f0345b003a74b1 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c @@ -572,8 +572,6 @@ static int smu_v13_0_0_set_default_dpm_table(struct smu_context *smu) PPTable_t *pptable = table_context->driver_pptable; SkuTable_t *skutable = &pptable->SkuTable; struct smu_13_0_dpm_table *dpm_table; - struct smu_13_0_pcie_table *pcie_table; - uint32_t link_level; int ret = 0; /* socclk dpm table setup */ @@ -689,24 +687,6 @@ static int smu_v13_0_0_set_default_dpm_table(struct smu_context *smu) dpm_table->max = dpm_table->dpm_levels[0].value; } - /* lclk dpm table setup */ - pcie_table = &dpm_context->dpm_tables.pcie_table; - pcie_table->num_of_link_levels = 0; - for (link_level = 0; link_level < NUM_LINK_LEVELS; link_level++) { - if (!skutable->PcieGenSpeed[link_level] && - !skutable->PcieLaneCount[link_level] && - !skutable->LclkFreq[link_level]) - continue; - - pcie_table->pcie_gen[pcie_table->num_of_link_levels] = - skutable->PcieGenSpeed[link_level]; - pcie_table->pcie_lane[pcie_table->num_of_link_levels] = - skutable->PcieLaneCount[link_level]; - pcie_table->clk_freq[pcie_table->num_of_link_levels] = - skutable->LclkFreq[link_level]; - pcie_table->num_of_link_levels++; - } - /* dcefclk dpm table setup */ dpm_table = &dpm_context->dpm_tables.dcef_table; if (smu_cmn_feature_is_enabled(smu, SMU_FEATURE_DPM_DCN_BIT)) { @@ -3150,6 +3130,90 @@ static int smu_v13_0_0_set_power_limit(struct smu_context *smu, return 0; } +static int smu_v13_0_0_update_pcie_parameters(struct smu_context *smu, + uint8_t pcie_gen_cap, + uint8_t pcie_width_cap) +{ + struct smu_13_0_dpm_context *dpm_context = smu->smu_dpm.dpm_context; + struct smu_13_0_pcie_table *pcie_table = + &dpm_context->dpm_tables.pcie_table; + int num_of_levels; + uint32_t smu_pcie_arg; + uint32_t link_level; + struct smu_table_context *table_context = &smu->smu_table; + PPTable_t *pptable = table_context->driver_pptable; + SkuTable_t *skutable = &pptable->SkuTable; + int ret = 0; + int i; + + pcie_table->num_of_link_levels = 0; + + for (link_level = 0; link_level < NUM_LINK_LEVELS; link_level++) { + if (!skutable->PcieGenSpeed[link_level] && + !skutable->PcieLaneCount[link_level] && + !skutable->LclkFreq[link_level]) + continue; + + pcie_table->pcie_gen[pcie_table->num_of_link_levels] = + skutable->PcieGenSpeed[link_level]; + pcie_table->pcie_lane[pcie_table->num_of_link_levels] = + skutable->PcieLaneCount[link_level]; + pcie_table->clk_freq[pcie_table->num_of_link_levels] = + skutable->LclkFreq[link_level]; + pcie_table->num_of_link_levels++; + } + + num_of_levels = pcie_table->num_of_link_levels; + if (!num_of_levels) + return 0; + + if (!(smu->adev->pm.pp_feature & PP_PCIE_DPM_MASK)) { + if (pcie_table->pcie_gen[num_of_levels - 1] < pcie_gen_cap) + pcie_gen_cap = pcie_table->pcie_gen[num_of_levels - 1]; + + if (pcie_table->pcie_lane[num_of_levels - 1] < pcie_width_cap) + pcie_width_cap = pcie_table->pcie_lane[num_of_levels - 1]; + + /* Force all levels to use the same settings */ + for (i = 0; i < num_of_levels; i++) { + pcie_table->pcie_gen[i] = pcie_gen_cap; + pcie_table->pcie_lane[i] = pcie_width_cap; + smu_pcie_arg = i << 16; + smu_pcie_arg |= pcie_table->pcie_gen[i] << 8; + smu_pcie_arg |= pcie_table->pcie_lane[i]; + + ret = smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_OverridePcieParameters, + smu_pcie_arg, + NULL); + if (ret) + break; + } + } else { + for (i = 0; i < num_of_levels; i++) { + if (pcie_table->pcie_gen[i] > pcie_gen_cap || + pcie_table->pcie_lane[i] > pcie_width_cap) { + pcie_table->pcie_gen[i] = pcie_table->pcie_gen[i] > pcie_gen_cap ? + pcie_gen_cap : pcie_table->pcie_gen[i]; + pcie_table->pcie_lane[i] = pcie_table->pcie_lane[i] > pcie_width_cap ? + pcie_width_cap : pcie_table->pcie_lane[i]; + smu_pcie_arg = i << 16; + smu_pcie_arg |= pcie_table->pcie_gen[i] << 8; + smu_pcie_arg |= pcie_table->pcie_lane[i]; + + ret = smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_OverridePcieParameters, + smu_pcie_arg, + NULL); + if (ret) + break; + } + } + } + + return ret; +} + static const struct pptable_funcs smu_v13_0_0_ppt_funcs = { .get_allowed_feature_mask = smu_v13_0_0_get_allowed_feature_mask, .set_default_dpm_table = smu_v13_0_0_set_default_dpm_table, @@ -3179,7 +3243,7 @@ static const struct pptable_funcs smu_v13_0_0_ppt_funcs = { .feature_is_enabled = smu_cmn_feature_is_enabled, .print_clk_levels = smu_v13_0_0_print_clk_levels, .force_clk_levels = smu_v13_0_0_force_clk_levels, - .update_pcie_parameters = smu_v13_0_update_pcie_parameters, + .update_pcie_parameters = smu_v13_0_0_update_pcie_parameters, .get_thermal_temperature_range = smu_v13_0_0_get_thermal_temperature_range, .register_irq_handler = smu_v13_0_register_irq_handler, .enable_thermal_alert = smu_v13_0_enable_thermal_alert, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c index e0d356f93ab064058cba3573d0776c705d8b6466..b3adeb6e43a81d4402181385b128b11900214390 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c @@ -187,8 +187,34 @@ int smu_v13_0_12_get_max_metrics_size(void) return max(sizeof(StaticMetricsTable_t), sizeof(MetricsTable_t)); } +static void smu_v13_0_12_init_xgmi_data(struct smu_context *smu, + StaticMetricsTable_t *static_metrics) +{ + struct smu_table_context *smu_table = &smu->smu_table; + uint16_t max_speed; + uint8_t max_width; + int ret; + + if (smu_table->tables[SMU_TABLE_SMU_METRICS].version >= 0x13) { + max_width = (uint8_t)static_metrics->MaxXgmiWidth; + max_speed = (uint16_t)static_metrics->MaxXgmiBitrate; + ret = 0; + } else { + MetricsTable_t *metrics = (MetricsTable_t *)smu_table->metrics_table; + + ret = smu_v13_0_6_get_metrics_table(smu, NULL, true); + if (!ret) { + max_width = (uint8_t)metrics->XgmiWidth; + max_speed = (uint16_t)metrics->XgmiBitrate; + } + } + if (!ret) + amgpu_xgmi_set_max_speed_width(smu->adev, max_speed, max_width); +} + int smu_v13_0_12_setup_driver_pptable(struct smu_context *smu) { + struct smu_13_0_dpm_context *dpm_context = smu->smu_dpm.dpm_context; struct smu_table_context *smu_table = &smu->smu_table; StaticMetricsTable_t *static_metrics = (StaticMetricsTable_t *)smu_table->metrics_table; struct PPTable_t *pptable = @@ -237,6 +263,18 @@ int smu_v13_0_12_setup_driver_pptable(struct smu_context *smu) if (ret) return ret; + if (smu_v13_0_6_cap_supported(smu, SMU_CAP(BOARD_VOLTAGE))) { + if (!static_metrics->InputTelemetryVoltageInmV) { + dev_warn(smu->adev->dev, "Invalid board voltage %d\n", + static_metrics->InputTelemetryVoltageInmV); + } + dpm_context->board_volt = static_metrics->InputTelemetryVoltageInmV; + } + if (smu_v13_0_6_cap_supported(smu, SMU_CAP(PLDM_VERSION)) && + static_metrics->pldmVersion[0] != 0xFFFFFFFF) + smu->adev->firmware.pldm_version = + static_metrics->pldmVersion[0]; + smu_v13_0_12_init_xgmi_data(smu, static_metrics); pptable->Init = true; } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c index f00ef7f3f3551a58e3e76aeb0ed004707c456659..68624afe7d836b2a407401c9970db551077fb26b 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c @@ -345,6 +345,11 @@ static void smu_v13_0_12_init_caps(struct smu_context *smu) if (fw_ver >= 0x00562500) smu_v13_0_6_cap_set(smu, SMU_CAP(HST_LIMIT_METRICS)); + + if (fw_ver >= 0x04560100) { + smu_v13_0_6_cap_set(smu, SMU_CAP(BOARD_VOLTAGE)); + smu_v13_0_6_cap_set(smu, SMU_CAP(PLDM_VERSION)); + } } static void smu_v13_0_6_init_caps(struct smu_context *smu) @@ -685,8 +690,8 @@ static int smu_v13_0_6_get_allowed_feature_mask(struct smu_context *smu, return 0; } -static int smu_v13_0_6_get_metrics_table(struct smu_context *smu, - void *metrics_table, bool bypass_cache) +int smu_v13_0_6_get_metrics_table(struct smu_context *smu, void *metrics_table, + bool bypass_cache) { struct smu_table_context *smu_table = &smu->smu_table; uint32_t table_size = smu_table->tables[SMU_TABLE_SMU_METRICS].size; @@ -800,6 +805,8 @@ static int smu_v13_0_6_setup_driver_pptable(struct smu_context *smu) int version = smu_v13_0_6_get_metrics_version(smu); int ret, i, retry = 100; uint32_t table_version; + uint16_t max_speed; + uint8_t max_width; if (amdgpu_ip_version(smu->adev, MP1_HWIP, 0) == IP_VERSION(13, 0, 12) && smu_v13_0_6_cap_supported(smu, SMU_CAP(STATIC_METRICS))) @@ -835,6 +842,9 @@ static int smu_v13_0_6_setup_driver_pptable(struct smu_context *smu) SMUQ10_ROUND(GET_METRIC_FIELD(MaxGfxclkFrequency, version)); pptable->MinGfxclkFrequency = SMUQ10_ROUND(GET_METRIC_FIELD(MinGfxclkFrequency, version)); + max_width = (uint8_t)GET_METRIC_FIELD(XgmiWidth, version); + max_speed = (uint16_t)GET_METRIC_FIELD(XgmiBitrate, version); + amgpu_xgmi_set_max_speed_width(smu->adev, max_speed, max_width); for (i = 0; i < 4; ++i) { pptable->FclkFrequencyTable[i] = @@ -1377,8 +1387,9 @@ static int smu_v13_0_6_print_clk_levels(struct smu_context *smu, return ret; } - min_clk = pstate_table->gfxclk_pstate.curr.min; - max_clk = pstate_table->gfxclk_pstate.curr.max; + single_dpm_table = &(dpm_context->dpm_tables.gfx_table); + min_clk = single_dpm_table->min; + max_clk = single_dpm_table->max; if (now < SMU_13_0_6_DSCLK_THRESHOLD) { size += sysfs_emit_at(buf, size, "S: %uMhz *\n", diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h index d38d6d76b1e79861345810f467131f2b2a61eaff..67b30674fd319904b5ce5d88e74c492f12457ceb 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h @@ -74,6 +74,8 @@ enum smu_v13_0_6_caps { extern void smu_v13_0_6_set_ppt_funcs(struct smu_context *smu); bool smu_v13_0_6_cap_supported(struct smu_context *smu, enum smu_v13_0_6_caps cap); int smu_v13_0_6_get_static_metrics_table(struct smu_context *smu); +int smu_v13_0_6_get_metrics_table(struct smu_context *smu, void *metrics_table, + bool bypass_cache); bool smu_v13_0_12_is_dpm_running(struct smu_context *smu); int smu_v13_0_12_get_max_metrics_size(void); diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c index c8f4f6fb408345a641b0c4832b143c344f0cbf8d..c96fa5e49ed655af4452da9eca27e6eb390b3a32 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c @@ -579,8 +579,6 @@ static int smu_v13_0_7_set_default_dpm_table(struct smu_context *smu) PPTable_t *driver_ppt = smu->smu_table.driver_pptable; SkuTable_t *skutable = &driver_ppt->SkuTable; struct smu_13_0_dpm_table *dpm_table; - struct smu_13_0_pcie_table *pcie_table; - uint32_t link_level; int ret = 0; /* socclk dpm table setup */ @@ -687,24 +685,6 @@ static int smu_v13_0_7_set_default_dpm_table(struct smu_context *smu) dpm_table->max = dpm_table->dpm_levels[0].value; } - /* lclk dpm table setup */ - pcie_table = &dpm_context->dpm_tables.pcie_table; - pcie_table->num_of_link_levels = 0; - for (link_level = 0; link_level < NUM_LINK_LEVELS; link_level++) { - if (!skutable->PcieGenSpeed[link_level] && - !skutable->PcieLaneCount[link_level] && - !skutable->LclkFreq[link_level]) - continue; - - pcie_table->pcie_gen[pcie_table->num_of_link_levels] = - skutable->PcieGenSpeed[link_level]; - pcie_table->pcie_lane[pcie_table->num_of_link_levels] = - skutable->PcieLaneCount[link_level]; - pcie_table->clk_freq[pcie_table->num_of_link_levels] = - skutable->LclkFreq[link_level]; - pcie_table->num_of_link_levels++; - } - /* dcefclk dpm table setup */ dpm_table = &dpm_context->dpm_tables.dcef_table; if (smu_cmn_feature_is_enabled(smu, SMU_FEATURE_DPM_DCN_BIT)) { @@ -2739,6 +2719,89 @@ static int smu_v13_0_7_set_power_limit(struct smu_context *smu, return 0; } +static int smu_v13_0_7_update_pcie_parameters(struct smu_context *smu, + uint8_t pcie_gen_cap, + uint8_t pcie_width_cap) +{ + struct smu_13_0_dpm_context *dpm_context = smu->smu_dpm.dpm_context; + struct smu_13_0_pcie_table *pcie_table = + &dpm_context->dpm_tables.pcie_table; + int num_of_levels; + int link_level; + uint32_t smu_pcie_arg; + struct smu_table_context *table_context = &smu->smu_table; + PPTable_t *pptable = table_context->driver_pptable; + SkuTable_t *skutable = &pptable->SkuTable; + int ret = 0; + int i; + + pcie_table->num_of_link_levels = 0; + for (link_level = 0; link_level < NUM_LINK_LEVELS; link_level++) { + if (!skutable->PcieGenSpeed[link_level] && + !skutable->PcieLaneCount[link_level] && + !skutable->LclkFreq[link_level]) + continue; + + pcie_table->pcie_gen[pcie_table->num_of_link_levels] = + skutable->PcieGenSpeed[link_level]; + pcie_table->pcie_lane[pcie_table->num_of_link_levels] = + skutable->PcieLaneCount[link_level]; + pcie_table->clk_freq[pcie_table->num_of_link_levels] = + skutable->LclkFreq[link_level]; + pcie_table->num_of_link_levels++; + } + + num_of_levels = pcie_table->num_of_link_levels; + if (!num_of_levels) + return 0; + + if (!(smu->adev->pm.pp_feature & PP_PCIE_DPM_MASK)) { + if (pcie_table->pcie_gen[num_of_levels - 1] < pcie_gen_cap) + pcie_gen_cap = pcie_table->pcie_gen[num_of_levels - 1]; + + if (pcie_table->pcie_lane[num_of_levels - 1] < pcie_width_cap) + pcie_width_cap = pcie_table->pcie_lane[num_of_levels - 1]; + + /* Force all levels to use the same settings */ + for (i = 0; i < num_of_levels; i++) { + pcie_table->pcie_gen[i] = pcie_gen_cap; + pcie_table->pcie_lane[i] = pcie_width_cap; + smu_pcie_arg = i << 16; + smu_pcie_arg |= pcie_table->pcie_gen[i] << 8; + smu_pcie_arg |= pcie_table->pcie_lane[i]; + + ret = smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_OverridePcieParameters, + smu_pcie_arg, + NULL); + if (ret) + break; + } + } else { + for (i = 0; i < num_of_levels; i++) { + if (pcie_table->pcie_gen[i] > pcie_gen_cap || + pcie_table->pcie_lane[i] > pcie_width_cap) { + pcie_table->pcie_gen[i] = pcie_table->pcie_gen[i] > pcie_gen_cap ? + pcie_gen_cap : pcie_table->pcie_gen[i]; + pcie_table->pcie_lane[i] = pcie_table->pcie_lane[i] > pcie_width_cap ? + pcie_width_cap : pcie_table->pcie_lane[i]; + smu_pcie_arg = i << 16; + smu_pcie_arg |= pcie_table->pcie_gen[i] << 8; + smu_pcie_arg |= pcie_table->pcie_lane[i]; + + ret = smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_OverridePcieParameters, + smu_pcie_arg, + NULL); + if (ret) + break; + } + } + } + + return ret; +} + static const struct pptable_funcs smu_v13_0_7_ppt_funcs = { .get_allowed_feature_mask = smu_v13_0_7_get_allowed_feature_mask, .set_default_dpm_table = smu_v13_0_7_set_default_dpm_table, @@ -2768,7 +2831,7 @@ static const struct pptable_funcs smu_v13_0_7_ppt_funcs = { .feature_is_enabled = smu_cmn_feature_is_enabled, .print_clk_levels = smu_v13_0_7_print_clk_levels, .force_clk_levels = smu_v13_0_7_force_clk_levels, - .update_pcie_parameters = smu_v13_0_update_pcie_parameters, + .update_pcie_parameters = smu_v13_0_7_update_pcie_parameters, .get_thermal_temperature_range = smu_v13_0_7_get_thermal_temperature_range, .register_irq_handler = smu_v13_0_register_irq_handler, .enable_thermal_alert = smu_v13_0_enable_thermal_alert, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_0_ppt.c index 84f9b007b59f2eab19604f1204d6cd9d8f9e656f..fe00c84b1cc66b911877a3e9dc6144758753a401 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_0_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_0_ppt.c @@ -1207,11 +1207,13 @@ static int smu_v14_0_0_print_clk_levels(struct smu_context *smu, static int smu_v14_0_0_set_soft_freq_limited_range(struct smu_context *smu, enum smu_clk_type clk_type, - uint32_t min, - uint32_t max) + u32 min, + u32 max, + bool __always_unused automatic) { - enum smu_message_type msg_set_min, msg_set_max; - int ret = 0; + enum smu_message_type msg_set_min = SMU_MSG_MAX_COUNT; + enum smu_message_type msg_set_max = SMU_MSG_MAX_COUNT; + int ret = -EINVAL; if (!smu_v14_0_0_clk_dpm_is_enabled(smu, clk_type)) return -EINVAL; @@ -1240,16 +1242,23 @@ static int smu_v14_0_0_set_soft_freq_limited_range(struct smu_context *smu, msg_set_min = SMU_MSG_SetHardMinVcn1; msg_set_max = SMU_MSG_SetSoftMaxVcn1; break; + case SMU_ISPICLK: + msg_set_min = SMU_MSG_SetHardMinIspiclkByFreq; + break; + case SMU_ISPXCLK: + msg_set_min = SMU_MSG_SetHardMinIspxclkByFreq; + break; default: return -EINVAL; } - ret = smu_cmn_send_smc_msg_with_param(smu, msg_set_min, min, NULL); - if (ret) - return ret; + if (min && msg_set_min != SMU_MSG_MAX_COUNT) + ret = smu_cmn_send_smc_msg_with_param(smu, msg_set_min, min, NULL); + + if (max && msg_set_max != SMU_MSG_MAX_COUNT) + ret = smu_cmn_send_smc_msg_with_param(smu, msg_set_max, max, NULL); - return smu_cmn_send_smc_msg_with_param(smu, msg_set_max, - max, NULL); + return ret; } static int smu_v14_0_0_force_clk_levels(struct smu_context *smu, @@ -1278,7 +1287,7 @@ static int smu_v14_0_0_force_clk_levels(struct smu_context *smu, if (ret) break; - ret = smu_v14_0_0_set_soft_freq_limited_range(smu, clk_type, min_freq, max_freq); + ret = smu_v14_0_0_set_soft_freq_limited_range(smu, clk_type, min_freq, max_freq, false); break; default: ret = -EINVAL; @@ -1426,7 +1435,8 @@ static int smu_v14_0_common_set_performance_level(struct smu_context *smu, ret = smu_v14_0_0_set_soft_freq_limited_range(smu, SMU_SCLK, sclk_min, - sclk_max); + sclk_max, + false); if (ret) return ret; @@ -1438,7 +1448,8 @@ static int smu_v14_0_common_set_performance_level(struct smu_context *smu, ret = smu_v14_0_0_set_soft_freq_limited_range(smu, SMU_FCLK, fclk_min, - fclk_max); + fclk_max, + false); if (ret) return ret; } @@ -1447,7 +1458,8 @@ static int smu_v14_0_common_set_performance_level(struct smu_context *smu, ret = smu_v14_0_0_set_soft_freq_limited_range(smu, SMU_SOCCLK, socclk_min, - socclk_max); + socclk_max, + false); if (ret) return ret; } @@ -1456,7 +1468,8 @@ static int smu_v14_0_common_set_performance_level(struct smu_context *smu, ret = smu_v14_0_0_set_soft_freq_limited_range(smu, SMU_VCLK, vclk_min, - vclk_max); + vclk_max, + false); if (ret) return ret; } @@ -1465,7 +1478,8 @@ static int smu_v14_0_common_set_performance_level(struct smu_context *smu, ret = smu_v14_0_0_set_soft_freq_limited_range(smu, SMU_VCLK1, vclk1_min, - vclk1_max); + vclk1_max, + false); if (ret) return ret; } @@ -1474,7 +1488,8 @@ static int smu_v14_0_common_set_performance_level(struct smu_context *smu, ret = smu_v14_0_0_set_soft_freq_limited_range(smu, SMU_DCLK, dclk_min, - dclk_max); + dclk_max, + false); if (ret) return ret; } @@ -1483,7 +1498,8 @@ static int smu_v14_0_common_set_performance_level(struct smu_context *smu, ret = smu_v14_0_0_set_soft_freq_limited_range(smu, SMU_DCLK1, dclk1_min, - dclk1_max); + dclk1_max, + false); if (ret) return ret; } @@ -1533,6 +1549,14 @@ static int smu_v14_0_0_set_vpe_enable(struct smu_context *smu, 0, NULL); } +static int smu_v14_0_0_set_isp_enable(struct smu_context *smu, + bool enable) +{ + return smu_cmn_send_smc_msg_with_param(smu, enable ? + SMU_MSG_PowerUpIspByTile : SMU_MSG_PowerDownIspByTile, + ISP_ALL_TILES_MASK, NULL); +} + static int smu_v14_0_0_set_umsch_mm_enable(struct smu_context *smu, bool enable) { @@ -1662,6 +1686,7 @@ static const struct pptable_funcs smu_v14_0_0_ppt_funcs = { .gfx_off_control = smu_v14_0_gfx_off_control, .mode2_reset = smu_v14_0_0_mode2_reset, .get_dpm_ultimate_freq = smu_v14_0_common_get_dpm_ultimate_freq, + .set_soft_freq_limited_range = smu_v14_0_0_set_soft_freq_limited_range, .od_edit_dpm_table = smu_v14_0_od_edit_dpm_table, .print_clk_levels = smu_v14_0_0_print_clk_levels, .force_clk_levels = smu_v14_0_0_force_clk_levels, @@ -1669,6 +1694,7 @@ static const struct pptable_funcs smu_v14_0_0_ppt_funcs = { .set_fine_grain_gfx_freq_parameters = smu_v14_0_common_set_fine_grain_gfx_freq_parameters, .set_gfx_power_up_by_imu = smu_v14_0_set_gfx_power_up_by_imu, .dpm_set_vpe_enable = smu_v14_0_0_set_vpe_enable, + .dpm_set_isp_enable = smu_v14_0_0_set_isp_enable, .dpm_set_umsch_mm_enable = smu_v14_0_0_set_umsch_mm_enable, .get_dpm_clock_table = smu_v14_0_common_get_dpm_table, .set_mall_enable = smu_v14_0_common_set_mall_enable, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c index 82c2db972491d46752961245b0fa343174b068aa..3aea32baea3da244d0d47088f726b969cd0f2b97 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c @@ -502,8 +502,6 @@ static int smu_v14_0_2_set_default_dpm_table(struct smu_context *smu) PPTable_t *pptable = table_context->driver_pptable; SkuTable_t *skutable = &pptable->SkuTable; struct smu_14_0_dpm_table *dpm_table; - struct smu_14_0_pcie_table *pcie_table; - uint32_t link_level; int ret = 0; /* socclk dpm table setup */ @@ -619,27 +617,6 @@ static int smu_v14_0_2_set_default_dpm_table(struct smu_context *smu) dpm_table->max = dpm_table->dpm_levels[0].value; } - /* lclk dpm table setup */ - pcie_table = &dpm_context->dpm_tables.pcie_table; - pcie_table->num_of_link_levels = 0; - for (link_level = 0; link_level < NUM_LINK_LEVELS; link_level++) { - if (!skutable->PcieGenSpeed[link_level] && - !skutable->PcieLaneCount[link_level] && - !skutable->LclkFreq[link_level]) - continue; - - pcie_table->pcie_gen[pcie_table->num_of_link_levels] = - skutable->PcieGenSpeed[link_level]; - pcie_table->pcie_lane[pcie_table->num_of_link_levels] = - skutable->PcieLaneCount[link_level]; - pcie_table->clk_freq[pcie_table->num_of_link_levels] = - skutable->LclkFreq[link_level]; - pcie_table->num_of_link_levels++; - - if (link_level == 0) - link_level++; - } - /* dcefclk dpm table setup */ dpm_table = &dpm_context->dpm_tables.dcef_table; if (smu_cmn_feature_is_enabled(smu, SMU_FEATURE_DPM_DCN_BIT)) { @@ -1487,10 +1464,31 @@ static int smu_v14_0_2_update_pcie_parameters(struct smu_context *smu, struct smu_14_0_dpm_context *dpm_context = smu->smu_dpm.dpm_context; struct smu_14_0_pcie_table *pcie_table = &dpm_context->dpm_tables.pcie_table; - int num_of_levels = pcie_table->num_of_link_levels; + int num_of_levels; uint32_t smu_pcie_arg; - int ret, i; + uint32_t link_level; + struct smu_table_context *table_context = &smu->smu_table; + PPTable_t *pptable = table_context->driver_pptable; + SkuTable_t *skutable = &pptable->SkuTable; + int ret = 0; + int i; + + pcie_table->num_of_link_levels = 0; + for (link_level = 0; link_level < NUM_LINK_LEVELS; link_level++) { + if (!skutable->PcieGenSpeed[link_level] && + !skutable->PcieLaneCount[link_level] && + !skutable->LclkFreq[link_level]) + continue; + pcie_table->pcie_gen[pcie_table->num_of_link_levels] = + skutable->PcieGenSpeed[link_level]; + pcie_table->pcie_lane[pcie_table->num_of_link_levels] = + skutable->PcieLaneCount[link_level]; + pcie_table->clk_freq[pcie_table->num_of_link_levels] = + skutable->LclkFreq[link_level]; + pcie_table->num_of_link_levels++; + } + num_of_levels = pcie_table->num_of_link_levels; if (!num_of_levels) return 0; @@ -1505,30 +1503,40 @@ static int smu_v14_0_2_update_pcie_parameters(struct smu_context *smu, for (i = 0; i < num_of_levels; i++) { pcie_table->pcie_gen[i] = pcie_gen_cap; pcie_table->pcie_lane[i] = pcie_width_cap; + smu_pcie_arg = i << 16; + smu_pcie_arg |= pcie_table->pcie_gen[i] << 8; + smu_pcie_arg |= pcie_table->pcie_lane[i]; + + ret = smu_cmn_send_smc_msg_with_param(smu, + SMU_MSG_OverridePcieParameters, + smu_pcie_arg, + NULL); + if (ret) + break; } } else { for (i = 0; i < num_of_levels; i++) { - if (pcie_table->pcie_gen[i] > pcie_gen_cap) - pcie_table->pcie_gen[i] = pcie_gen_cap; - if (pcie_table->pcie_lane[i] > pcie_width_cap) - pcie_table->pcie_lane[i] = pcie_width_cap; - } - } - - for (i = 0; i < num_of_levels; i++) { - smu_pcie_arg = i << 16; - smu_pcie_arg |= pcie_table->pcie_gen[i] << 8; - smu_pcie_arg |= pcie_table->pcie_lane[i]; - - ret = smu_cmn_send_smc_msg_with_param(smu, + if (pcie_table->pcie_gen[i] > pcie_gen_cap || + pcie_table->pcie_lane[i] > pcie_width_cap) { + pcie_table->pcie_gen[i] = pcie_table->pcie_gen[i] > pcie_gen_cap ? + pcie_gen_cap : pcie_table->pcie_gen[i]; + pcie_table->pcie_lane[i] = pcie_table->pcie_lane[i] > pcie_width_cap ? + pcie_width_cap : pcie_table->pcie_lane[i]; + smu_pcie_arg = i << 16; + smu_pcie_arg |= pcie_table->pcie_gen[i] << 8; + smu_pcie_arg |= pcie_table->pcie_lane[i]; + + ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_OverridePcieParameters, smu_pcie_arg, NULL); - if (ret) - return ret; + if (ret) + break; + } + } } - return 0; + return ret; } static const struct smu_temperature_range smu14_thermal_policy[] = { diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c index 7eaf58fd7f9ae509280b9b2c087b1d938190b0e1..59f9abd0f7b8cefa0c6bea45cd38decd425bf067 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c @@ -86,6 +86,7 @@ static void smu_cmn_read_arg(struct smu_context *smu, #define SMU_RESP_BUSY_OTHER 0xFC #define SMU_RESP_DEBUG_END 0xFB +#define SMU_RESP_UNEXP (~0U) /** * __smu_cmn_poll_stat -- poll for a status from the SMU * @smu: a pointer to SMU context @@ -171,6 +172,15 @@ static void __smu_cmn_reg_print_error(struct smu_context *smu, dev_err_ratelimited(adev->dev, "SMU: I'm debugging!"); break; + case SMU_RESP_UNEXP: + if (amdgpu_device_bus_status_check(smu->adev)) { + /* print error immediately if device is off the bus */ + dev_err(adev->dev, + "SMU: response:0x%08X for index:%d param:0x%08X message:%s?", + reg_c2pmsg_90, msg_index, param, message); + break; + } + fallthrough; default: dev_err_ratelimited(adev->dev, "SMU: response:0x%08X for index:%d param:0x%08X message:%s?", diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h index 7473672abd2a91c8a4d278643b3b81a3d000d351..a608cdbdada4cbf968c260b7d2506915f2a133c0 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h +++ b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h @@ -40,28 +40,29 @@ #define SMU_IH_INTERRUPT_CONTEXT_ID_FAN_ABNORMAL 0x8 #define SMU_IH_INTERRUPT_CONTEXT_ID_FAN_RECOVERY 0x9 -#define smu_cmn_init_soft_gpu_metrics(ptr, frev, crev) \ - do { \ - typecheck(struct gpu_metrics_v##frev##_##crev, \ - typeof(*(ptr))); \ - struct metrics_table_header *header = \ - (struct metrics_table_header *)(ptr); \ - memset(header, 0xFF, sizeof(*(ptr))); \ - header->format_revision = frev; \ - header->content_revision = crev; \ - header->structure_size = sizeof(*(ptr)); \ +#define smu_cmn_init_soft_gpu_metrics(ptr, frev, crev) \ + do { \ + typecheck(struct gpu_metrics_v##frev##_##crev *, (ptr)); \ + struct gpu_metrics_v##frev##_##crev *tmp = (ptr); \ + struct metrics_table_header *header = \ + (struct metrics_table_header *)tmp; \ + memset(header, 0xFF, sizeof(*tmp)); \ + header->format_revision = frev; \ + header->content_revision = crev; \ + header->structure_size = sizeof(*tmp); \ } while (0) -#define smu_cmn_init_partition_metrics(ptr, frev, crev) \ - do { \ - typecheck(struct amdgpu_partition_metrics_v##frev##_##crev, \ - typeof(*(ptr))); \ - struct metrics_table_header *header = \ - (struct metrics_table_header *)(ptr); \ - memset(header, 0xFF, sizeof(*(ptr))); \ - header->format_revision = frev; \ - header->content_revision = crev; \ - header->structure_size = sizeof(*(ptr)); \ +#define smu_cmn_init_partition_metrics(ptr, fr, cr) \ + do { \ + typecheck(struct amdgpu_partition_metrics_v##fr##_##cr *, \ + (ptr)); \ + struct amdgpu_partition_metrics_v##fr##_##cr *tmp = (ptr); \ + struct metrics_table_header *header = \ + (struct metrics_table_header *)tmp; \ + memset(header, 0xFF, sizeof(*tmp)); \ + header->format_revision = fr; \ + header->content_revision = cr; \ + header->structure_size = sizeof(*tmp); \ } while (0) extern const int link_speed[]; diff --git a/drivers/gpu/drm/ci/build-igt.sh b/drivers/gpu/drm/ci/build-igt.sh index caa2f4804ed5313147c2f35252ce282d11ffc496..eddb5f782a5e7531290752aab2a42a29aff1f9a5 100644 --- a/drivers/gpu/drm/ci/build-igt.sh +++ b/drivers/gpu/drm/ci/build-igt.sh @@ -71,4 +71,4 @@ tar -cf artifacts/igt.tar /igt # Pass needed files to the test stage S3_ARTIFACT_NAME="igt.tar.gz" gzip -c artifacts/igt.tar > ${S3_ARTIFACT_NAME} -s3_upload ${S3_ARTIFACT_NAME} https://${PIPELINE_ARTIFACTS_BASE}/${KERNEL_ARCH}/ +ci-fairy s3cp --token-file "${S3_JWT_FILE}" ${S3_ARTIFACT_NAME} https://${PIPELINE_ARTIFACTS_BASE}/${KERNEL_ARCH}/${S3_ARTIFACT_NAME} diff --git a/drivers/gpu/drm/ci/build.sh b/drivers/gpu/drm/ci/build.sh index 6fb74c51abe27e10e7c9f24342a7e4f0fadfd9e3..ac5e7ed195cf4b26d6b652071feb7a322ce62a72 100644 --- a/drivers/gpu/drm/ci/build.sh +++ b/drivers/gpu/drm/ci/build.sh @@ -113,17 +113,6 @@ mkdir -p install/modules/ INSTALL_MOD_PATH=install/modules/ make modules_install if [[ ${DEBIAN_ARCH} = "arm64" ]]; then - make Image.lzma - mkimage \ - -f auto \ - -A arm \ - -O linux \ - -d arch/arm64/boot/Image.lzma \ - -C lzma\ - -b arch/arm64/boot/dts/qcom/sdm845-cheza-r3.dtb \ - /kernel/cheza-kernel - KERNEL_IMAGE_NAME+=" cheza-kernel" - # Make a gzipped copy of the Image for db410c. gzip -k /kernel/Image KERNEL_IMAGE_NAME+=" Image.gz" @@ -148,13 +137,13 @@ if [[ "$UPLOAD_TO_MINIO" = "1" ]]; then ls -l "${S3_JWT_FILE}" for f in $FILES_TO_UPLOAD; do - s3_upload /kernel/$f \ - https://${PIPELINE_ARTIFACTS_BASE}/${DEBIAN_ARCH}/ + ci-fairy s3cp --token-file "${S3_JWT_FILE}" /kernel/$f \ + https://${PIPELINE_ARTIFACTS_BASE}/${DEBIAN_ARCH}/$f done S3_ARTIFACT_NAME="kernel-files.tar.zst" tar --zstd -cf $S3_ARTIFACT_NAME install - s3_upload ${S3_ARTIFACT_NAME} https://${PIPELINE_ARTIFACTS_BASE}/${DEBIAN_ARCH}/ + ci-fairy s3cp --token-file "${S3_JWT_FILE}" ${S3_ARTIFACT_NAME} https://${PIPELINE_ARTIFACTS_BASE}/${DEBIAN_ARCH}/${S3_ARTIFACT_NAME} echo "Download vmlinux.xz from https://${PIPELINE_ARTIFACTS_BASE}/${DEBIAN_ARCH}/vmlinux.xz" fi diff --git a/drivers/gpu/drm/ci/build.yml b/drivers/gpu/drm/ci/build.yml index 8eb56ebcf4aa728874dcb086baf0f82faddc9180..af27ff5de369d4389cb00fec51af4c305614570f 100644 --- a/drivers/gpu/drm/ci/build.yml +++ b/drivers/gpu/drm/ci/build.yml @@ -13,7 +13,7 @@ - .build - .use-debian/arm64_build tags: - - aarch64 + - $FDO_RUNNER_JOB_PRIORITY_TAG_AARCH64 variables: DEFCONFIG: "arch/arm/configs/multi_v7_defconfig" KERNEL_IMAGE_NAME: "zImage" @@ -24,7 +24,7 @@ - .build - .use-debian/arm64_build tags: - - aarch64 + - $FDO_RUNNER_JOB_PRIORITY_TAG_AARCH64 variables: DEFCONFIG: "arch/arm64/configs/defconfig" KERNEL_IMAGE_NAME: "Image" @@ -44,16 +44,22 @@ igt:arm32: extends: .build:arm32 + variables: + GIT_DEPTH: 10 script: - FDO_CI_CONCURRENT=${FDO_CI_CONCURRENT} bash drivers/gpu/drm/ci/build-igt.sh igt:arm64: extends: .build:arm64 + variables: + GIT_DEPTH: 10 script: - FDO_CI_CONCURRENT=${FDO_CI_CONCURRENT} bash drivers/gpu/drm/ci/build-igt.sh igt:x86_64: extends: .build:x86_64 + variables: + GIT_DEPTH: 10 script: - FDO_CI_CONCURRENT=${FDO_CI_CONCURRENT} bash drivers/gpu/drm/ci/build-igt.sh diff --git a/drivers/gpu/drm/ci/container.yml b/drivers/gpu/drm/ci/container.yml index 56c95c2f91ae588e8bcfe399ecc973200a3309fc..5f90508578a36ec59a74d26a9cb6318c58baacd1 100644 --- a/drivers/gpu/drm/ci/container.yml +++ b/drivers/gpu/drm/ci/container.yml @@ -20,31 +20,15 @@ debian/arm64_build: EXTRA_LOCAL_PACKAGES: "jq libasound2 libcairo2 libdw1 libglib2.0-0 libjson-c5" # Disable container jobs that we won't use -alpine/x86_64_build: - rules: - - when: never - -debian/arm32_test-base: - rules: - - when: never - -debian/arm32_test-gl: - rules: - - when: never - -debian/arm32_test-vk: - rules: - - when: never - -debian/arm64_test-gl: +debian/arm64_test-vk: rules: - when: never -debian/arm64_test-vk: +debian/baremetal_arm32_test-gl: rules: - when: never -debian/baremetal_arm32_test: +debian/baremetal_arm64_test-vk: rules: - when: never @@ -64,19 +48,19 @@ debian/x86_64_test-android: rules: - when: never -debian/x86_64_test-vk: +debian/x86_64_test-video: rules: - when: never -fedora/x86_64_build: +debian/x86_64_test-vk: rules: - when: never -debian/android_build: +fedora/x86_64_build: rules: - when: never -.debian/x86_64_test-android: +debian/android_build: rules: - when: never diff --git a/drivers/gpu/drm/ci/gitlab-ci.yml b/drivers/gpu/drm/ci/gitlab-ci.yml index 9a32cd736f2fef43bae291f91e1a2048bf4c597d..d502d146b17733905a6eac1ecd32c8a6e600ecd8 100644 --- a/drivers/gpu/drm/ci/gitlab-ci.yml +++ b/drivers/gpu/drm/ci/gitlab-ci.yml @@ -1,17 +1,17 @@ variables: DRM_CI_PROJECT_PATH: &drm-ci-project-path mesa/mesa - DRM_CI_COMMIT_SHA: &drm-ci-commit-sha f73132f1215a37ce8ffc711a0136c90649aaf128 + DRM_CI_COMMIT_SHA: &drm-ci-commit-sha 02337aec715c25dae7ff2479d986f831c77fe536 UPSTREAM_REPO: https://gitlab.freedesktop.org/drm/kernel.git TARGET_BRANCH: drm-next - IGT_VERSION: 04bedb9238586b81d4d4ca62b02e584f6cfc77af + IGT_VERSION: 129d5b10baaadea1d6cd6377341c4cb42e7ee6fd DEQP_RUNNER_GIT_URL: https://gitlab.freedesktop.org/mesa/deqp-runner.git DEQP_RUNNER_GIT_TAG: v0.20.0 FDO_UPSTREAM_REPO: helen.fornazier/linux # The repo where the git-archive daily runs - MESA_TEMPLATES_COMMIT: &ci-templates-commit d5aa3941aa03c2f716595116354fb81eb8012acb + MESA_TEMPLATES_COMMIT: &ci-templates-commit c6aeb16f86e32525fa630fb99c66c4f3e62fc3cb DRM_CI_PROJECT_URL: https://gitlab.freedesktop.org/${DRM_CI_PROJECT_PATH} CI_PRE_CLONE_SCRIPT: |- set -o xtrace @@ -20,10 +20,8 @@ variables: rm download-git-cache.sh set +o xtrace S3_JWT_FILE: /s3_jwt - S3_JWT_HEADER_FILE: /s3_jwt_header S3_JWT_FILE_SCRIPT: |- echo -n '${S3_JWT}' > '${S3_JWT_FILE}' && - echo -n "Authorization: Bearer ${S3_JWT}" > '${S3_JWT_HEADER_FILE}' && unset CI_JOB_JWT S3_JWT # Unsetting vulnerable env variables S3_HOST: s3.freedesktop.org # This bucket is used to fetch the kernel image @@ -38,7 +36,11 @@ variables: JOB_ARTIFACTS_BASE: ${PIPELINE_ARTIFACTS_BASE}/${CI_JOB_ID} KERNEL_IMAGE_BASE: https://${S3_HOST}/${S3_KERNEL_BUCKET}/${KERNEL_REPO}/${KERNEL_TAG} LAVA_TAGS: subset-1-gfx - LAVA_JOB_PRIORITY: 30 + # Default priority for non-merge pipelines + FDO_RUNNER_JOB_PRIORITY_TAG_X86_64: "" # Empty tags are ignored by gitlab + FDO_RUNNER_JOB_PRIORITY_TAG_X86_64_KVM: kvm + FDO_RUNNER_JOB_PRIORITY_TAG_AARCH64: aarch64 + JOB_PRIORITY: 30 ARTIFACTS_BASE_URL: https://${CI_PROJECT_ROOT_NAMESPACE}.${CI_PAGES_DOMAIN}/-/${CI_PROJECT_NAME}/-/jobs/${CI_JOB_ID}/artifacts # Python scripts for structured logger PYTHONPATH: "$PYTHONPATH:$CI_PROJECT_DIR/install" @@ -72,16 +74,13 @@ default: include: - - project: 'freedesktop/ci-templates' - ref: 16bc29078de5e0a067ff84a1a199a3760d3b3811 - file: - - '/templates/ci-fairy.yml' - project: 'freedesktop/ci-templates' ref: *ci-templates-commit file: - '/templates/alpine.yml' - '/templates/debian.yml' - '/templates/fedora.yml' + - '/templates/ci-fairy.yml' - project: *drm-ci-project-path ref: *drm-ci-commit-sha file: @@ -105,6 +104,7 @@ include: - '/src/microsoft/ci/gitlab-ci-inc.yml' - '/src/nouveau/ci/gitlab-ci-inc.yml' - '/src/virtio/ci/gitlab-ci-inc.yml' + - 'docs/gitlab-ci.yml' - drivers/gpu/drm/ci/image-tags.yml - drivers/gpu/drm/ci/container.yml - drivers/gpu/drm/ci/static-checks.yml @@ -118,6 +118,7 @@ include: stages: - sanity - container + - deploy - git-archive - build-for-tests - build-only @@ -236,16 +237,20 @@ stages: - _build/meson-logs/strace +python-artifacts: + variables: + GIT_DEPTH: 10 + + # Git archive -make git archive: +make-git-archive: extends: - .fdo.ci-fairy stage: git-archive rules: - !reference [.scheduled_pipeline-rules, rules] - # ensure we are running on packet tags: - - packet.net + - $FDO_RUNNER_JOB_PRIORITY_TAG_X86_64 script: # Remove drm-ci files we just added - rm -rf .gitlab-ci.* @@ -257,7 +262,7 @@ make git archive: - tar -cvzf ../$CI_PROJECT_NAME.tar.gz . # Use id_tokens for JWT auth - - s3_upload ../$CI_PROJECT_NAME.tar.gz https://$S3_HOST/${S3_GITCACHE_BUCKET}/$CI_PROJECT_NAMESPACE/$CI_PROJECT_NAME/ + - ci-fairy s3cp --token-file "${S3_JWT_FILE}" ../$CI_PROJECT_NAME.tar.gz https://$S3_HOST/${S3_GITCACHE_BUCKET}/$CI_PROJECT_NAMESPACE/$CI_PROJECT_NAME/$CI_PROJECT_NAME.tar.gz # Sanity checks of MR settings and commit logs @@ -265,6 +270,8 @@ sanity: extends: - .fdo.ci-fairy stage: sanity + tags: + - $FDO_RUNNER_JOB_PRIORITY_TAG_X86_64 rules: - if: *is-pre-merge when: on_success @@ -283,7 +290,6 @@ sanity: DEBIAN_BUILD_TAG DEBIAN_PYUTILS_TAG DEBIAN_TEST_GL_TAG - KERNEL_ROOTFS_TAG KERNEL_TAG PKG_REPO_REV ) @@ -299,14 +305,14 @@ sanity: when: on_failure reports: junit: check-*.xml - tags: - - placeholder-job mr-label-maker-test: extends: - .fdo.ci-fairy stage: sanity + tags: + - $FDO_RUNNER_JOB_PRIORITY_TAG_X86_64 rules: - !reference [.mr-label-maker-rules, rules] variables: @@ -329,3 +335,15 @@ mr-label-maker-test: optional: true - job: toml-lint optional: true + +deploy-docs: + rules: + - when: never + +linkcheck-docs: + rules: + - when: never + +test-docs: + rules: + - when: never diff --git a/drivers/gpu/drm/ci/igt_runner.sh b/drivers/gpu/drm/ci/igt_runner.sh index 2a0599f12c588a3c2ffebe42daeb16010ea658a7..b24d4bc53cda34c4f6e9c913a03005d54b206102 100755 --- a/drivers/gpu/drm/ci/igt_runner.sh +++ b/drivers/gpu/drm/ci/igt_runner.sh @@ -19,6 +19,7 @@ set +e cat /sys/kernel/debug/dri/*/state set -e +mkdir -p /lib/modules case "$DRIVER_NAME" in amdgpu|vkms) # Cannot use HWCI_KERNEL_MODULES as at that point we don't have the module in /lib diff --git a/drivers/gpu/drm/ci/image-tags.yml b/drivers/gpu/drm/ci/image-tags.yml index 53fe34b865788cf470108104643a2c6bc9586813..7acc2e2a8eaac9d9a16a8de4eb94963752b55a4e 100644 --- a/drivers/gpu/drm/ci/image-tags.yml +++ b/drivers/gpu/drm/ci/image-tags.yml @@ -1,24 +1,18 @@ variables: - CONTAINER_TAG: "20250328-mesa-uprev" - DEBIAN_X86_64_BUILD_BASE_IMAGE: "debian/x86_64_build-base" + CONTAINER_TAG: "20250502-mesa-uprev" DEBIAN_BASE_TAG: "${CONTAINER_TAG}" - - DEBIAN_X86_64_BUILD_IMAGE_PATH: "debian/x86_64_build" DEBIAN_BUILD_TAG: "${CONTAINER_TAG}" - KERNEL_ROOTFS_TAG: "${CONTAINER_TAG}" + DEBIAN_TEST_GL_TAG: "${CONTAINER_TAG}" # default kernel for rootfs before injecting the current kernel tree - KERNEL_TAG: "v6.13-rc4-mesa-5e77" + KERNEL_TAG: "v6.14-mesa-0bdd" KERNEL_REPO: "gfx-ci/linux" - PKG_REPO_REV: "bca9635d" - - DEBIAN_X86_64_TEST_BASE_IMAGE: "debian/x86_64_test-base" - DEBIAN_X86_64_TEST_IMAGE_GL_PATH: "debian/x86_64_test-gl" - DEBIAN_TEST_GL_TAG: "${CONTAINER_TAG}" + PKG_REPO_REV: "95bf62c" - DEBIAN_PYUTILS_IMAGE: "debian/x86_64_pyutils" DEBIAN_PYUTILS_TAG: "${CONTAINER_TAG}" + ALPINE_X86_64_BUILD_TAG: "${CONTAINER_TAG}" ALPINE_X86_64_LAVA_SSH_TAG: "${CONTAINER_TAG}" - CONDITIONAL_BUILD_ANGLE_TAG: fec96cc945650c5fe9f7188cabe80d8a + CONDITIONAL_BUILD_ANGLE_TAG: 384145a4023315dae658259bee07c43a + CONDITIONAL_BUILD_PIGLIT_TAG: a19e424b8a3f020dbf1b9dd29f220a4f diff --git a/drivers/gpu/drm/ci/lava-submit.sh b/drivers/gpu/drm/ci/lava-submit.sh index a1e8b34fb2d491b5b2d02965b0f6453b3a01aab0..a295102c3468ee3eefb754581930c6d2271e9ec1 100755 --- a/drivers/gpu/drm/ci/lava-submit.sh +++ b/drivers/gpu/drm/ci/lava-submit.sh @@ -41,7 +41,6 @@ section_start prepare_rootfs "Preparing root filesystem" set -ex -section_switch rootfs "Assembling root filesystem" ROOTFS_URL="$(get_path_to_artifact lava-rootfs.tar.zst)" [ $? != 1 ] || exit 1 @@ -54,7 +53,7 @@ cp artifacts/ci-common/init-*.sh results/job-rootfs-overlay/ cp "$SCRIPTS_DIR"/setup-test-env.sh results/job-rootfs-overlay/ tar zcf job-rootfs-overlay.tar.gz -C results/job-rootfs-overlay/ . -s3_upload job-rootfs-overlay.tar.gz "https://${JOB_ARTIFACTS_BASE}" +ci-fairy s3cp --token-file "${S3_JWT_FILE}" job-rootfs-overlay.tar.gz "https://${JOB_ROOTFS_OVERLAY_PATH}" # Prepare env vars for upload. section_switch variables "Environment variables passed through to device:" @@ -64,6 +63,9 @@ section_switch lava_submit "Submitting job for scheduling" touch results/lava.log tail -f results/lava.log & +# Ensure that we are printing the commands that are being executed, +# making it easier to debug the job in case it fails. +set -x PYTHONPATH=artifacts/ artifacts/lava/lava_job_submitter.py \ --farm "${FARM}" \ --device-type "${DEVICE_TYPE}" \ diff --git a/drivers/gpu/drm/ci/test.yml b/drivers/gpu/drm/ci/test.yml index 84a25f0e783b6ca7f8a993c709d5a0fc86bf18d3..81147e86bfd0898bf665d802bf6cbf2a98927455 100644 --- a/drivers/gpu/drm/ci/test.yml +++ b/drivers/gpu/drm/ci/test.yml @@ -24,7 +24,7 @@ .lava-igt:arm32: extends: - - .lava-test:arm32 + - .lava-arm32-test-gl variables: HWCI_TEST_SCRIPT: "/install/igt_runner.sh" DEBIAN_ARCH: "armhf" @@ -33,15 +33,14 @@ - testing:arm32 needs: - alpine/x86_64_lava_ssh_client - - kernel+rootfs_arm32 - - debian/x86_64_build + - debian/arm32_test-gl - python-artifacts - testing:arm32 - igt:arm32 .lava-igt:arm64: extends: - - .lava-test:arm64 + - .lava-arm64-test-gl variables: HWCI_TEST_SCRIPT: "/install/igt_runner.sh" DEBIAN_ARCH: "arm64" @@ -50,15 +49,14 @@ - testing:arm64 needs: - alpine/x86_64_lava_ssh_client - - kernel+rootfs_arm64 - - debian/x86_64_build + - debian/arm64_test-gl - python-artifacts - testing:arm64 - igt:arm64 .lava-igt:x86_64: extends: - - .lava-test:x86_64 + - .lava-x86_64-test-gl variables: HWCI_TEST_SCRIPT: "/install/igt_runner.sh" DEBIAN_ARCH: "amd64" @@ -67,16 +65,15 @@ - testing:x86_64 needs: - alpine/x86_64_lava_ssh_client - - kernel+rootfs_x86_64 - - debian/x86_64_build + - debian/x86_64_test-gl - python-artifacts - testing:x86_64 - igt:x86_64 .baremetal-igt-arm64: extends: - - .baremetal-test-arm64 - - .use-debian/baremetal_arm64_test + - .baremetal-test-arm64-gl + - .use-debian/baremetal_arm64_test-gl - .allow_failure_lockdep timeout: "1h30m" rules: @@ -91,7 +88,7 @@ BM_CMDLINE: "ip=dhcp console=ttyMSM0,115200n8 $BM_KERNEL_EXTRA_ARGS root=/dev/nfs rw nfsrootdebug nfsroot=,tcp,nfsvers=4.2 init=/init $BM_KERNELARGS" FARM: google needs: - - debian/baremetal_arm64_test + - debian/baremetal_arm64_test-gl - job: testing:arm64 artifacts: false - igt:arm64 @@ -101,19 +98,21 @@ .software-driver: stage: software-driver extends: + - .test-gl - .allow_failure_lockdep timeout: "1h30m" rules: - !reference [.scheduled_pipeline-rules, rules] - when: on_success - extends: - - .test-gl tags: - - kvm + - $FDO_RUNNER_JOB_PRIORITY_TAG_X86_64_KVM + before_script: + - !reference [default, before_script] + - rm -rf install + - tar -xf artifacts/install.tar script: - ln -sf $CI_PROJECT_DIR/install /install - mv install/bzImage /kernel/bzImage - - mkdir -p /lib/modules - install/crosvm-runner.sh install/igt_runner.sh needs: - debian/x86_64_test-gl @@ -180,20 +179,6 @@ msm:apq8096: script: - ./install/bare-metal/fastboot.sh || exit $? -msm:sdm845: - extends: - - .baremetal-igt-arm64 - stage: msm - parallel: 6 - variables: - DEVICE_TYPE: sdm845-cheza-r3 - DRIVER_NAME: msm - BM_KERNEL: https://${PIPELINE_ARTIFACTS_BASE}/arm64/cheza-kernel - GPU_VERSION: sdm845 - RUNNER_TAG: google-freedreno-cheza - script: - - ./install/bare-metal/cros-servo.sh || exit $? - msm:sm8350-hdk: extends: - .lava-igt:arm64 @@ -323,7 +308,7 @@ i915:cml: variables: DEVICE_TYPE: asus-C436FA-Flip-hatch GPU_VERSION: cml - RUNNER_TAG: mesa-ci-x86-64-lava-asus-C436FA-flip-hatch + RUNNER_TAG: mesa-ci-x86-64-lava-asus-C436FA-Flip-hatch i915:tgl: extends: diff --git a/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-kingoftown-skips.txt b/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-kingoftown-skips.txt index d4b8ba3a54a977002f60594ed42071dbc710ec93..154b047787b2415496ef1d0718ab4e7b05dbc9c0 100644 --- a/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-kingoftown-skips.txt +++ b/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-kingoftown-skips.txt @@ -32,3 +32,8 @@ kms_display_modes@mst-extended-mode-negative # It causes other tests to fail, so skip it. kms_invalid_mode@overflow-vrefresh + +# sc7180 does not have APRIV, so memptrs is not protected. +# (Preemption is not supported on devices that do not have +# APRIV, so this is ok) +msm/msm_mapping@memptrs diff --git a/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-lazor-limozeen-skips.txt b/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-lazor-limozeen-skips.txt index 022db559cc7d946080a8ba419f659e8e5be3e23b..a9bb3e1ad75ca57a758beaa88c4f9087c762f341 100644 --- a/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-lazor-limozeen-skips.txt +++ b/drivers/gpu/drm/ci/xfails/msm-sc7180-trogdor-lazor-limozeen-skips.txt @@ -23,3 +23,8 @@ core_hotunplug.* # https://gitlab.freedesktop.org/drm/igt-gpu-tools/-/issues/162 kms_display_modes@extended-mode-basic kms_display_modes@mst-extended-mode-negative + +# sc7180 does not have APRIV, so memptrs is not protected. +# (Preemption is not supported on devices that do not have +# APRIV, so this is ok) +msm/msm_mapping@memptrs diff --git a/drivers/gpu/drm/ci/xfails/msm-sdm845-fails.txt b/drivers/gpu/drm/ci/xfails/msm-sdm845-fails.txt deleted file mode 100644 index 7a2ab58b706f9ccaaec02be870bc589453d4927e..0000000000000000000000000000000000000000 --- a/drivers/gpu/drm/ci/xfails/msm-sdm845-fails.txt +++ /dev/null @@ -1,29 +0,0 @@ -kms_color@ctm-0-25,Fail -kms_color@ctm-0-50,Fail -kms_color@ctm-0-75,Fail -kms_color@ctm-blue-to-red,Fail -kms_color@ctm-green-to-red,Fail -kms_color@ctm-negative,Fail -kms_color@ctm-red-to-blue,Fail -kms_color@ctm-signed,Fail -kms_cursor_legacy@basic-flip-before-cursor-atomic,Fail -kms_cursor_legacy@basic-flip-before-cursor-legacy,Fail -kms_cursor_legacy@cursor-vs-flip-atomic,Fail -kms_cursor_legacy@cursor-vs-flip-atomic-transitions,Fail -kms_cursor_legacy@cursor-vs-flip-atomic-transitions-varying-size,Fail -kms_cursor_legacy@cursor-vs-flip-legacy,Fail -kms_cursor_legacy@cursor-vs-flip-toggle,Fail -kms_cursor_legacy@cursor-vs-flip-varying-size,Fail -kms_cursor_legacy@flip-vs-cursor-atomic,Fail -kms_cursor_legacy@flip-vs-cursor-crc-atomic,Fail -kms_cursor_legacy@flip-vs-cursor-crc-legacy,Fail -kms_cursor_legacy@flip-vs-cursor-legacy,Fail -kms_flip@flip-vs-modeset-vs-hang,Fail -kms_flip@flip-vs-panning-vs-hang,Fail -kms_lease@lease-uevent,Fail -kms_pipe_crc_basic@compare-crc-sanitycheck-nv12,Fail -kms_plane_alpha_blend@alpha-7efc,Fail -kms_plane_alpha_blend@coverage-7efc,Fail -kms_plane_alpha_blend@coverage-vs-premult-vs-constant,Fail -kms_plane_cursor@overlay,Fail -kms_plane_cursor@viewport,Fail diff --git a/drivers/gpu/drm/ci/xfails/msm-sdm845-flakes.txt b/drivers/gpu/drm/ci/xfails/msm-sdm845-flakes.txt deleted file mode 100644 index e32d73c6c98ea56bddde1472302a6b985d2fe9f6..0000000000000000000000000000000000000000 --- a/drivers/gpu/drm/ci/xfails/msm-sdm845-flakes.txt +++ /dev/null @@ -1,139 +0,0 @@ -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -kms_cursor_legacy@basic-flip-after-cursor-atomic - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -kms_cursor_legacy@basic-flip-after-cursor-legacy - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -kms_cursor_legacy@basic-flip-after-cursor-varying-size - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -kms_cursor_legacy@basic-flip-before-cursor-varying-size - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -kms_cursor_legacy@flip-vs-cursor-atomic-transitions - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -kms_cursor_legacy@flip-vs-cursor-varying-size - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -kms_cursor_legacy@short-flip-after-cursor-atomic-transitions - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -kms_cursor_legacy@short-flip-after-cursor-atomic-transitions-varying-size - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -kms_cursor_legacy@short-flip-after-cursor-toggle - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -kms_cursor_legacy@short-flip-before-cursor-atomic-transitions - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -kms_cursor_legacy@short-flip-before-cursor-atomic-transitions-varying-size - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -msm/msm_shrink@copy-gpu-32 - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-g0df7b9b97 -# Linux Version: 6.9.0-rc7 -msm/msm_shrink@copy-gpu-oom-32 - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-gf13702b8e -# Linux Version: 6.10.0-rc5 -kms_cursor_legacy@short-flip-before-cursor-toggle - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-gf13702b8e -# Linux Version: 6.10.0-rc5 -kms_cursor_legacy@flip-vs-cursor-toggle - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/661483c8-ad82-400d-bcd8-e94986d20d7d@collabora.com/T/#u -# Failure Rate: 50 -# IGT Version: 1.28-gf13702b8e -# Linux Version: 6.10.0-rc5 -msm/msm_shrink@copy-mmap-oom-8 - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/64bc4bcf-de51-4e60-a9f7-1295a1e64c65@collabora.com/T/#t -# Failure Rate: 50 -# IGT Version: 1.28-ga73311079 -# Linux Version: 6.11.0-rc2 -kms_lease@page-flip-implicit-plane - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://lore.kernel.org/linux-arm-msm/64bc4bcf-de51-4e60-a9f7-1295a1e64c65@collabora.com/T/#t -# Failure Rate: 50 -# IGT Version: 1.28-ga73311079 -# Linux Version: 6.11.0-rc5 -kms_flip@flip-vs-expired-vblank - -# Board Name: sdm845-cheza-r3 -# Bug Report: https://gitlab.freedesktop.org/drm/msm/-/issues/75 -# Failure Rate: 20 -# IGT Version: 1.30-g04bedb923 -# Linux Version: 6.14.0-rc4 -kms_flip@plain-flip-ts-check-interruptible diff --git a/drivers/gpu/drm/ci/xfails/msm-sdm845-skips.txt b/drivers/gpu/drm/ci/xfails/msm-sdm845-skips.txt deleted file mode 100644 index 6c86d1953e11f96237e6df5f389075c8524028b6..0000000000000000000000000000000000000000 --- a/drivers/gpu/drm/ci/xfails/msm-sdm845-skips.txt +++ /dev/null @@ -1,350 +0,0 @@ -# Hangs machine -kms_bw.* - -# Failing due to a bootloader/fw issue. The workaround in mesa CI involves these two patches -# https://gitlab.freedesktop.org/gfx-ci/linux/-/commit/4b49f902ec6f2bb382cbbf489870573f4b43371e -# https://gitlab.freedesktop.org/gfx-ci/linux/-/commit/38cdf4c5559771e2474ae0fecef8469f65147bc1 -msm/msm_mapping@* - -# Skip driver specific tests -^amdgpu.* -nouveau_.* -^panfrost.* -^v3d.* -^vc4.* -^vmwgfx* - -# Skip intel specific tests -gem_.* -i915_.* -tools_test.* -kms_dp_link_training.* - -# Currently fails and causes coverage loss for other tests -# since core_getversion also fails. -core_hotunplug.* - -# Whole machine hangs -kms_cursor_crc.* - -# IGT test crash -# IGT Version: 1.28-ga73311079 -# Linux Version: 6.11.0-rc2 -kms_content_protection@uevent - -# IGT issue. is_joiner_mode() should return false for non-Intel hardware. -# https://gitlab.freedesktop.org/drm/igt-gpu-tools/-/issues/162 -kms_display_modes@extended-mode-basic -kms_display_modes@mst-extended-mode-negative - -# Kernel panic -msm/msm_recovery@hangcheck -# DEBUG - Begin test msm/msm_recovery@hangcheck -# Console: switching to colour dummy device 80x25 -# [ 489.526286] [IGT] msm_recovery: executing -# [ 489.531926] [IGT] msm_recovery: starting subtest hangcheck -# [ 492.808574] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 492.820358] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 492.831154] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 493.832570] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 493.844177] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 493.854971] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 494.824633] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 494.836237] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 494.847034] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 495.816570] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 495.828170] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 495.838966] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 496.804643] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 496.816246] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 496.827041] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 497.832570] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 497.844170] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 497.854963] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 498.820636] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 498.832232] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 498.843024] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 499.816568] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 499.828163] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 499.838958] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 500.808570] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 500.820165] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 500.830960] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 501.832570] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 501.844175] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 501.854965] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 502.824568] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 502.836171] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 502.846965] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 503.816570] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 503.828176] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 503.838969] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 504.804640] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 504.816237] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 504.827033] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 505.828643] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 505.840247] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 505.851043] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 506.820637] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 506.832233] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 506.843026] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 507.816567] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 507.828171] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 507.838965] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 508.808568] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 508.820173] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 508.830969] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 509.832568] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 509.844173] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 509.854967] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 510.824568] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 510.836162] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 510.846954] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 511.816569] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 511.828173] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 511.838968] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 512.804641] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 512.816246] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 512.827040] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 513.828641] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 513.840239] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 513.851035] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 514.824568] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 514.836164] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 514.846959] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 515.812640] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 515.824235] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 515.835030] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 515.912427] rcu: INFO: rcu_preempt self-detected stall on CPU -# [ 515.918398] rcu: 0-....: (6452 ticks this GP) idle=6afc/1/0x4000000000000000 softirq=12492/12697 fqs=3179 -# [ 515.929296] rcu: (t=6505 jiffies g=36205 q=58 ncpus=8) -# [ 515.934709] CPU: 0 UID: 0 PID: 126 Comm: sugov:0 Tainted: G W 6.14.0-rc4-gdddf15cff632 #1 -# [ 515.934727] Tainted: [W]=WARN -# [ 515.934732] Hardware name: Google Cheza (rev3+) (DT) -# [ 515.934739] pstate: 00400009 (nzcv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) -# [ 515.934751] pc : rcu_core+0x59c/0xe68 -# [ 515.934769] lr : rcu_core+0x74/0xe68 -# [ 515.934781] sp : ffff800080003e50 -# [ 515.934785] x29: ffff800080003e50 x28: ffff225d038e9bc0 x27: 0000000000000002 -# [ 515.934805] x26: ffffc171a8ee6108 x25: ffffc171a85bc2c0 x24: ffff60ecd691e000 -# [ 515.934820] x23: ffffc171a85d15c0 x22: ffffc171a8f8d780 x21: ffff225e7eeef5c0 -# [ 515.934835] x20: ffffc171a8ef0e80 x19: ffffc171a85d15d1 x18: ffffc171a9461e70 -# [ 515.934850] x17: ffff60ecd691e000 x16: ffff800080000000 x15: 0000000000000000 -# [ 515.934866] x14: ffffc171a85d0780 x13: 0000000000000400 x12: 0000000000000000 -# [ 515.934880] x11: ffffc171a85ce900 x10: ffffc171a8ef5000 x9 : ffffc171a8ef0000 -# [ 515.934894] x8 : ffff800080003d88 x7 : ffffc171a8ee6100 x6 : ffff800080003de0 -# [ 515.934909] x5 : ffff800080003dc8 x4 : 0000000000000003 x3 : 0000000000000000 -# [ 515.934923] x2 : 0000000000000101 x1 : 0000000000000000 x0 : ffff225d038e9bc0 -# [ 515.934939] Call trace: -# [ 515.934945] rcu_core+0x59c/0xe68 (P) -# [ 515.934962] rcu_core_si+0x10/0x1c -# [ 515.934976] handle_softirqs+0x118/0x4b8 -# [ 515.934994] __do_softirq+0x14/0x20 -# [ 515.935007] ____do_softirq+0x10/0x1c -# [ 515.935021] call_on_irq_stack+0x24/0x4c -# [ 515.935034] do_softirq_own_stack+0x1c/0x28 -# [ 515.935048] __irq_exit_rcu+0x174/0x1b4 -# [ 515.935063] irq_exit_rcu+0x10/0x38 -# [ 515.935077] el1_interrupt+0x38/0x64 -# [ 515.935092] el1h_64_irq_handler+0x18/0x24 -# [ 515.935104] el1h_64_irq+0x6c/0x70 -# [ 515.935115] lock_acquire+0x1e0/0x338 (P) -# [ 515.935129] __mutex_lock+0xa8/0x4b8 -# [ 515.935144] mutex_lock_nested+0x24/0x30 -# [ 515.935159] _find_opp_table_unlocked+0x40/0xfc -# [ 515.935174] _find_key+0x64/0x16c -# [ 515.935184] dev_pm_opp_find_freq_exact+0x4c/0x74 -# [ 515.935197] qcom_cpufreq_hw_target_index+0xe8/0x128 -# [ 515.935211] __cpufreq_driver_target+0x144/0x29c -# [ 515.935227] sugov_work+0x58/0x74 -# [ 515.935239] kthread_worker_fn+0xf4/0x324 -# [ 515.935254] kthread+0x12c/0x208 -# [ 515.935266] ret_from_fork+0x10/0x20 -# [ 516.808569] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 516.820174] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 516.830968] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 517.828641] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 517.840236] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 517.851032] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 518.820642] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 518.832237] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 518.843030] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 519.812636] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 519.824231] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 519.835026] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 520.808570] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 520.820165] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 520.830959] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 521.828643] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 521.840238] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 521.851033] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 522.820636] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 522.832232] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 522.843027] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 523.812639] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 523.824239] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 523.835034] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 524.804640] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 524.816235] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 524.827026] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 525.828641] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 525.840236] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 525.851031] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 526.820641] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 526.832244] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 526.843041] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 527.812642] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 527.824242] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 527.835038] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 528.804639] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 528.816234] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 528.827027] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 529.832634] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 529.844231] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 529.855017] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 530.820646] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 530.832270] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 530.843065] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 531.812640] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 531.824238] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 531.835030] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 532.804640] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 532.816237] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 532.827031] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 533.828640] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 533.840243] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 533.851037] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 534.820640] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 534.832245] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 534.843038] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 535.812641] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 535.824238] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 535.835033] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 536.804639] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 536.816235] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 536.827030] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 537.828640] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 537.840234] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 537.851020] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 538.820640] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 538.832235] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 538.843027] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 539.812644] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: hangcheck detected gpu lockup rb 0! -# [ 539.824247] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: completed fence: 45605 -# [ 539.835040] msm_dpu ae01000.display-controller: [drm:hangcheck_handler] *ERROR* 6.3.0.2: submitted fence: 45611 -# [ 540.124426] watchdog: BUG: soft lockup - CPU#0 stuck for 49s! [sugov:0:126] -# [ 540.124439] Modules linked in: -# [ 540.124448] irq event stamp: 9912389 -# [ 540.124453] hardirqs last enabled at (9912388): [] exit_to_kernel_mode+0x38/0x130 -# [ 540.124473] hardirqs last disabled at (9912389): [] el1_interrupt+0x24/0x64 -# [ 540.124486] softirqs last enabled at (9898068): [] handle_softirqs+0x4a0/0x4b8 -# [ 540.124505] softirqs last disabled at (9898071): [] __do_softirq+0x14/0x20 -# [ 540.124525] CPU: 0 UID: 0 PID: 126 Comm: sugov:0 Tainted: G W 6.14.0-rc4-gdddf15cff632 #1 -# [ 540.124540] Tainted: [W]=WARN -# [ 540.124544] Hardware name: Google Cheza (rev3+) (DT) -# [ 540.124549] pstate: 60400009 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) -# [ 540.124560] pc : xhci_urb_enqueue+0xbc/0x32c -# [ 540.124573] lr : xhci_urb_enqueue+0xb4/0x32c -# [ 540.124581] sp : ffff800080003c20 -# [ 540.124586] x29: ffff800080003c20 x28: 0000000000000000 x27: ffff225d00b1e6a0 -# [ 540.124602] x26: ffff225d01c3d800 x25: 0000000000000001 x24: 0000000000000006 -# [ 540.124617] x23: ffff225d044dc000 x22: ffff225d044dc000 x21: 0000000000000001 -# [ 540.124632] x20: ffff225d002d7280 x19: ffff225d0573a780 x18: ffff225e7eff0f50 -# [ 540.124647] x17: 000000000000cab0 x16: 0000000000000000 x15: ffff225d0353a000 -# [ 540.124661] x14: 0000000000000000 x13: 0000000000000820 x12: 0000000000000000 -# [ 540.124674] x11: ffff800080003a30 x10: 0000000000000001 x9 : 0000000000000000 -# [ 540.124689] x8 : ffff225d002d7300 x7 : 0000000000000000 x6 : 000000000000003f -# [ 540.124702] x5 : 00000000ffffffff x4 : 0000000000000920 x3 : 0000000000000080 -# [ 540.124716] x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff225d002d7280 -# [ 540.124731] Call trace: -# [ 540.124736] xhci_urb_enqueue+0xbc/0x32c (P) -# [ 540.124751] usb_hcd_submit_urb+0x98/0x7fc -# [ 540.124766] usb_submit_urb+0x294/0x560 -# [ 540.124780] intr_callback+0x78/0x1fc -# [ 540.124798] __usb_hcd_giveback_urb+0x68/0x128 -# [ 540.124812] usb_giveback_urb_bh+0xa8/0x140 -# [ 540.124825] process_one_work+0x208/0x5e8 -# [ 540.124840] bh_worker+0x1a8/0x20c -# [ 540.124853] workqueue_softirq_action+0x78/0x88 -# [ 540.124868] tasklet_hi_action+0x14/0x3c -# [ 540.124883] handle_softirqs+0x118/0x4b8 -# [ 540.124897] __do_softirq+0x14/0x20 -# [ 540.124908] ____do_softirq+0x10/0x1c -# [ 540.124922] call_on_irq_stack+0x24/0x4c -# [ 540.124934] do_softirq_own_stack+0x1c/0x28 -# [ 540.124947] __irq_exit_rcu+0x174/0x1b4 -# [ 540.124961] irq_exit_rcu+0x10/0x38 -# [ 540.124976] el1_interrupt+0x38/0x64 -# [ 540.124987] el1h_64_irq_handler+0x18/0x24 -# [ 540.124998] el1h_64_irq+0x6c/0x70 -# [ 540.125009] lock_acquire+0x1e0/0x338 (P) -# [ 540.125023] __mutex_lock+0xa8/0x4b8 -# [ 540.125038] mutex_lock_nested+0x24/0x30 -# [ 540.125052] _find_opp_table_unlocked+0x40/0xfc -# [ 540.125067] _find_key+0x64/0x16c -# [ 540.125078] dev_pm_opp_find_freq_exact+0x4c/0x74 -# [ 540.125090] qcom_cpufreq_hw_target_index+0xe8/0x128 -# [ 540.125105] __cpufreq_driver_target+0x144/0x29c -# [ 540.125121] sugov_work+0x58/0x74 -# [ 540.125133] kthread_worker_fn+0xf4/0x324 -# [ 540.125148] kthread+0x12c/0x208 -# [ 540.125160] ret_from_fork+0x10/0x20 -# [ 540.125176] Kernel panic - not syncing: softlockup: hung tasks -# [ 540.423567] CPU: 0 UID: 0 PID: 126 Comm: sugov:0 Tainted: G W L 6.14.0-rc4-gdddf15cff632 #1 -# [ 540.433411] Tainted: [W]=WARN, [L]=SOFTLOCKUP -# [ 540.437901] Hardware name: Google Cheza (rev3+) (DT) -# [ 540.443022] Call trace: -# [ 540.445559] show_stack+0x18/0x24 (C) -# [ 540.449357] dump_stack_lvl+0x38/0xd0 -# [ 540.453157] dump_stack+0x18/0x24 -# [ 540.456599] panic+0x3bc/0x41c -# [ 540.459767] watchdog_timer_fn+0x254/0x2e4 -# [ 540.464005] __hrtimer_run_queues+0x3c4/0x440 -# [ 540.468508] hrtimer_interrupt+0xe4/0x244 -# [ 540.472662] arch_timer_handler_phys+0x2c/0x44 -# [ 540.477256] handle_percpu_devid_irq+0x90/0x1f0 -# [ 540.481943] handle_irq_desc+0x40/0x58 -# [ 540.485829] generic_handle_domain_irq+0x1c/0x28 -# [ 540.490604] gic_handle_irq+0x4c/0x11c -# [ 540.494483] do_interrupt_handler+0x50/0x84 -# [ 540.498811] el1_interrupt+0x34/0x64 -# [ 540.502518] el1h_64_irq_handler+0x18/0x24 -# [ 540.506758] el1h_64_irq+0x6c/0x70 -# [ 540.510279] xhci_urb_enqueue+0xbc/0x32c (P) -# [ 540.514693] usb_hcd_submit_urb+0x98/0x7fc -# [ 540.518932] usb_submit_urb+0x294/0x560 -# [ 540.522901] intr_callback+0x78/0x1fc -# [ 540.526700] __usb_hcd_giveback_urb+0x68/0x128 -# [ 540.531288] usb_giveback_urb_bh+0xa8/0x140 -# [ 540.535614] process_one_work+0x208/0x5e8 -# [ 540.539769] bh_worker+0x1a8/0x20c -# [ 540.543293] workqueue_softirq_action+0x78/0x88 -# [ 540.547980] tasklet_hi_action+0x14/0x3c -# [ 540.552038] handle_softirqs+0x118/0x4b8 -# [ 540.556096] __do_softirq+0x14/0x20 -# [ 540.559705] ____do_softirq+0x10/0x1c -# [ 540.563500] call_on_irq_stack+0x24/0x4c -# [ 540.567554] do_softirq_own_stack+0x1c/0x28 -# [ 540.571878] __irq_exit_rcu+0x174/0x1b4 -# [ 540.575849] irq_exit_rcu+0x10/0x38 -# [ 540.579462] el1_interrupt+0x38/0x64 -# [ 540.583158] el1h_64_irq_handler+0x18/0x24 -# [ 540.587397] el1h_64_irq+0x6c/0x70 -# [ 540.590918] lock_acquire+0x1e0/0x338 (P) -# [ 540.595060] __mutex_lock+0xa8/0x4b8 -# [ 540.598760] mutex_lock_nested+0x24/0x30 -# [ 540.602818] _find_opp_table_unlocked+0x40/0xfc -# [ 540.607503] _find_key+0x64/0x16c -# [ 540.610940] dev_pm_opp_find_freq_exact+0x4c/0x74 -# [ 540.615798] qcom_cpufreq_hw_target_index+0xe8/0x128 -# [ 540.620924] __cpufreq_driver_target+0x144/0x29c -# [ 540.625698] sugov_work+0x58/0x74 -# [ 540.629134] kthread_worker_fn+0xf4/0x324 -# [ 540.633278] kthread+0x12c/0x208 -# [ 540.636619] ret_from_fork+0x10/0x20 -# [ 540.640321] SMP: stopping secondary CPUs -# [ 540.644518] Kernel Offset: 0x417126200000 from 0xffff800080000000 -# [ 540.650848] PHYS_OFFSET: 0xfff0dda400000000 -# [ 540.655170] CPU features: 0x000,00000100,00901250,8200721b -# [ 540.660829] Memory Limit: none -# [ 540.663999] ---[ end Kernel panic - not syncing: softlockup: hung tasks ]--- diff --git a/drivers/gpu/drm/display/drm_dp_helper.c b/drivers/gpu/drm/display/drm_dp_helper.c index 5d8a4691af2c8dfa164d4ad0c072f95ca8314622..1c3920297906b8229c18723807a2afd0dea5b767 100644 --- a/drivers/gpu/drm/display/drm_dp_helper.c +++ b/drivers/gpu/drm/display/drm_dp_helper.c @@ -692,6 +692,34 @@ void drm_dp_dpcd_set_powered(struct drm_dp_aux *aux, bool powered) } EXPORT_SYMBOL(drm_dp_dpcd_set_powered); +/** + * drm_dp_dpcd_set_probe() - Set whether a probing before DPCD access is done + * @aux: DisplayPort AUX channel + * @enable: Enable the probing if required + */ +void drm_dp_dpcd_set_probe(struct drm_dp_aux *aux, bool enable) +{ + WRITE_ONCE(aux->dpcd_probe_disabled, !enable); +} +EXPORT_SYMBOL(drm_dp_dpcd_set_probe); + +static bool dpcd_access_needs_probe(struct drm_dp_aux *aux) +{ + /* + * HP ZR24w corrupts the first DPCD access after entering power save + * mode. Eg. on a read, the entire buffer will be filled with the same + * byte. Do a throw away read to avoid corrupting anything we care + * about. Afterwards things will work correctly until the monitor + * gets woken up and subsequently re-enters power save mode. + * + * The user pressing any button on the monitor is enough to wake it + * up, so there is no particularly good place to do the workaround. + * We just have to do it before any DPCD access and hope that the + * monitor doesn't power down exactly after the throw away read. + */ + return !aux->is_remote && !READ_ONCE(aux->dpcd_probe_disabled); +} + /** * drm_dp_dpcd_read() - read a series of bytes from the DPCD * @aux: DisplayPort AUX channel (SST or MST) @@ -713,20 +741,8 @@ ssize_t drm_dp_dpcd_read(struct drm_dp_aux *aux, unsigned int offset, { int ret; - /* - * HP ZR24w corrupts the first DPCD access after entering power save - * mode. Eg. on a read, the entire buffer will be filled with the same - * byte. Do a throw away read to avoid corrupting anything we care - * about. Afterwards things will work correctly until the monitor - * gets woken up and subsequently re-enters power save mode. - * - * The user pressing any button on the monitor is enough to wake it - * up, so there is no particularly good place to do the workaround. - * We just have to do it before any DPCD access and hope that the - * monitor doesn't power down exactly after the throw away read. - */ - if (!aux->is_remote) { - ret = drm_dp_dpcd_probe(aux, DP_DPCD_REV); + if (dpcd_access_needs_probe(aux)) { + ret = drm_dp_dpcd_probe(aux, DP_LANE0_1_STATUS); if (ret < 0) return ret; } diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c index c21b5e570013633155fb8152504a95ae9423ef76..e2e85345aa9a4081004eb44f19fa50dd41196593 100644 --- a/drivers/gpu/drm/drm_edid.c +++ b/drivers/gpu/drm/drm_edid.c @@ -67,34 +67,36 @@ static int oui(u8 first, u8 second, u8 third) * on as many displays as possible). */ -/* First detailed mode wrong, use largest 60Hz mode */ -#define EDID_QUIRK_PREFER_LARGE_60 (1 << 0) -/* Reported 135MHz pixel clock is too high, needs adjustment */ -#define EDID_QUIRK_135_CLOCK_TOO_HIGH (1 << 1) -/* Prefer the largest mode at 75 Hz */ -#define EDID_QUIRK_PREFER_LARGE_75 (1 << 2) -/* Detail timing is in cm not mm */ -#define EDID_QUIRK_DETAILED_IN_CM (1 << 3) -/* Detailed timing descriptors have bogus size values, so just take the - * maximum size and use that. - */ -#define EDID_QUIRK_DETAILED_USE_MAXIMUM_SIZE (1 << 4) -/* use +hsync +vsync for detailed mode */ -#define EDID_QUIRK_DETAILED_SYNC_PP (1 << 6) -/* Force reduced-blanking timings for detailed modes */ -#define EDID_QUIRK_FORCE_REDUCED_BLANKING (1 << 7) -/* Force 8bpc */ -#define EDID_QUIRK_FORCE_8BPC (1 << 8) -/* Force 12bpc */ -#define EDID_QUIRK_FORCE_12BPC (1 << 9) -/* Force 6bpc */ -#define EDID_QUIRK_FORCE_6BPC (1 << 10) -/* Force 10bpc */ -#define EDID_QUIRK_FORCE_10BPC (1 << 11) -/* Non desktop display (i.e. HMD) */ -#define EDID_QUIRK_NON_DESKTOP (1 << 12) -/* Cap the DSC target bitrate to 15bpp */ -#define EDID_QUIRK_CAP_DSC_15BPP (1 << 13) +enum drm_edid_internal_quirk { + /* First detailed mode wrong, use largest 60Hz mode */ + EDID_QUIRK_PREFER_LARGE_60 = DRM_EDID_QUIRK_NUM, + /* Reported 135MHz pixel clock is too high, needs adjustment */ + EDID_QUIRK_135_CLOCK_TOO_HIGH, + /* Prefer the largest mode at 75 Hz */ + EDID_QUIRK_PREFER_LARGE_75, + /* Detail timing is in cm not mm */ + EDID_QUIRK_DETAILED_IN_CM, + /* Detailed timing descriptors have bogus size values, so just take the + * maximum size and use that. + */ + EDID_QUIRK_DETAILED_USE_MAXIMUM_SIZE, + /* use +hsync +vsync for detailed mode */ + EDID_QUIRK_DETAILED_SYNC_PP, + /* Force reduced-blanking timings for detailed modes */ + EDID_QUIRK_FORCE_REDUCED_BLANKING, + /* Force 8bpc */ + EDID_QUIRK_FORCE_8BPC, + /* Force 12bpc */ + EDID_QUIRK_FORCE_12BPC, + /* Force 6bpc */ + EDID_QUIRK_FORCE_6BPC, + /* Force 10bpc */ + EDID_QUIRK_FORCE_10BPC, + /* Non desktop display (i.e. HMD) */ + EDID_QUIRK_NON_DESKTOP, + /* Cap the DSC target bitrate to 15bpp */ + EDID_QUIRK_CAP_DSC_15BPP, +}; #define MICROSOFT_IEEE_OUI 0xca125c @@ -129,124 +131,132 @@ static const struct edid_quirk { u32 quirks; } edid_quirk_list[] = { /* Acer AL1706 */ - EDID_QUIRK('A', 'C', 'R', 44358, EDID_QUIRK_PREFER_LARGE_60), + EDID_QUIRK('A', 'C', 'R', 44358, BIT(EDID_QUIRK_PREFER_LARGE_60)), /* Acer F51 */ - EDID_QUIRK('A', 'P', 'I', 0x7602, EDID_QUIRK_PREFER_LARGE_60), + EDID_QUIRK('A', 'P', 'I', 0x7602, BIT(EDID_QUIRK_PREFER_LARGE_60)), /* AEO model 0 reports 8 bpc, but is a 6 bpc panel */ - EDID_QUIRK('A', 'E', 'O', 0, EDID_QUIRK_FORCE_6BPC), + EDID_QUIRK('A', 'E', 'O', 0, BIT(EDID_QUIRK_FORCE_6BPC)), /* BenQ GW2765 */ - EDID_QUIRK('B', 'N', 'Q', 0x78d6, EDID_QUIRK_FORCE_8BPC), + EDID_QUIRK('B', 'N', 'Q', 0x78d6, BIT(EDID_QUIRK_FORCE_8BPC)), /* BOE model on HP Pavilion 15-n233sl reports 8 bpc, but is a 6 bpc panel */ - EDID_QUIRK('B', 'O', 'E', 0x78b, EDID_QUIRK_FORCE_6BPC), + EDID_QUIRK('B', 'O', 'E', 0x78b, BIT(EDID_QUIRK_FORCE_6BPC)), /* CPT panel of Asus UX303LA reports 8 bpc, but is a 6 bpc panel */ - EDID_QUIRK('C', 'P', 'T', 0x17df, EDID_QUIRK_FORCE_6BPC), + EDID_QUIRK('C', 'P', 'T', 0x17df, BIT(EDID_QUIRK_FORCE_6BPC)), /* SDC panel of Lenovo B50-80 reports 8 bpc, but is a 6 bpc panel */ - EDID_QUIRK('S', 'D', 'C', 0x3652, EDID_QUIRK_FORCE_6BPC), + EDID_QUIRK('S', 'D', 'C', 0x3652, BIT(EDID_QUIRK_FORCE_6BPC)), /* BOE model 0x0771 reports 8 bpc, but is a 6 bpc panel */ - EDID_QUIRK('B', 'O', 'E', 0x0771, EDID_QUIRK_FORCE_6BPC), + EDID_QUIRK('B', 'O', 'E', 0x0771, BIT(EDID_QUIRK_FORCE_6BPC)), /* Belinea 10 15 55 */ - EDID_QUIRK('M', 'A', 'X', 1516, EDID_QUIRK_PREFER_LARGE_60), - EDID_QUIRK('M', 'A', 'X', 0x77e, EDID_QUIRK_PREFER_LARGE_60), + EDID_QUIRK('M', 'A', 'X', 1516, BIT(EDID_QUIRK_PREFER_LARGE_60)), + EDID_QUIRK('M', 'A', 'X', 0x77e, BIT(EDID_QUIRK_PREFER_LARGE_60)), /* Envision Peripherals, Inc. EN-7100e */ - EDID_QUIRK('E', 'P', 'I', 59264, EDID_QUIRK_135_CLOCK_TOO_HIGH), + EDID_QUIRK('E', 'P', 'I', 59264, BIT(EDID_QUIRK_135_CLOCK_TOO_HIGH)), /* Envision EN2028 */ - EDID_QUIRK('E', 'P', 'I', 8232, EDID_QUIRK_PREFER_LARGE_60), + EDID_QUIRK('E', 'P', 'I', 8232, BIT(EDID_QUIRK_PREFER_LARGE_60)), /* Funai Electronics PM36B */ - EDID_QUIRK('F', 'C', 'M', 13600, EDID_QUIRK_PREFER_LARGE_75 | - EDID_QUIRK_DETAILED_IN_CM), + EDID_QUIRK('F', 'C', 'M', 13600, BIT(EDID_QUIRK_PREFER_LARGE_75) | + BIT(EDID_QUIRK_DETAILED_IN_CM)), /* LG 27GP950 */ - EDID_QUIRK('G', 'S', 'M', 0x5bbf, EDID_QUIRK_CAP_DSC_15BPP), + EDID_QUIRK('G', 'S', 'M', 0x5bbf, BIT(EDID_QUIRK_CAP_DSC_15BPP)), /* LG 27GN950 */ - EDID_QUIRK('G', 'S', 'M', 0x5b9a, EDID_QUIRK_CAP_DSC_15BPP), + EDID_QUIRK('G', 'S', 'M', 0x5b9a, BIT(EDID_QUIRK_CAP_DSC_15BPP)), /* LGD panel of HP zBook 17 G2, eDP 10 bpc, but reports unknown bpc */ - EDID_QUIRK('L', 'G', 'D', 764, EDID_QUIRK_FORCE_10BPC), + EDID_QUIRK('L', 'G', 'D', 764, BIT(EDID_QUIRK_FORCE_10BPC)), /* LG Philips LCD LP154W01-A5 */ - EDID_QUIRK('L', 'P', 'L', 0, EDID_QUIRK_DETAILED_USE_MAXIMUM_SIZE), - EDID_QUIRK('L', 'P', 'L', 0x2a00, EDID_QUIRK_DETAILED_USE_MAXIMUM_SIZE), + EDID_QUIRK('L', 'P', 'L', 0, BIT(EDID_QUIRK_DETAILED_USE_MAXIMUM_SIZE)), + EDID_QUIRK('L', 'P', 'L', 0x2a00, BIT(EDID_QUIRK_DETAILED_USE_MAXIMUM_SIZE)), /* Samsung SyncMaster 205BW. Note: irony */ - EDID_QUIRK('S', 'A', 'M', 541, EDID_QUIRK_DETAILED_SYNC_PP), + EDID_QUIRK('S', 'A', 'M', 541, BIT(EDID_QUIRK_DETAILED_SYNC_PP)), /* Samsung SyncMaster 22[5-6]BW */ - EDID_QUIRK('S', 'A', 'M', 596, EDID_QUIRK_PREFER_LARGE_60), - EDID_QUIRK('S', 'A', 'M', 638, EDID_QUIRK_PREFER_LARGE_60), + EDID_QUIRK('S', 'A', 'M', 596, BIT(EDID_QUIRK_PREFER_LARGE_60)), + EDID_QUIRK('S', 'A', 'M', 638, BIT(EDID_QUIRK_PREFER_LARGE_60)), /* Sony PVM-2541A does up to 12 bpc, but only reports max 8 bpc */ - EDID_QUIRK('S', 'N', 'Y', 0x2541, EDID_QUIRK_FORCE_12BPC), + EDID_QUIRK('S', 'N', 'Y', 0x2541, BIT(EDID_QUIRK_FORCE_12BPC)), /* ViewSonic VA2026w */ - EDID_QUIRK('V', 'S', 'C', 5020, EDID_QUIRK_FORCE_REDUCED_BLANKING), + EDID_QUIRK('V', 'S', 'C', 5020, BIT(EDID_QUIRK_FORCE_REDUCED_BLANKING)), /* Medion MD 30217 PG */ - EDID_QUIRK('M', 'E', 'D', 0x7b8, EDID_QUIRK_PREFER_LARGE_75), + EDID_QUIRK('M', 'E', 'D', 0x7b8, BIT(EDID_QUIRK_PREFER_LARGE_75)), /* Lenovo G50 */ - EDID_QUIRK('S', 'D', 'C', 18514, EDID_QUIRK_FORCE_6BPC), + EDID_QUIRK('S', 'D', 'C', 18514, BIT(EDID_QUIRK_FORCE_6BPC)), /* Panel in Samsung NP700G7A-S01PL notebook reports 6bpc */ - EDID_QUIRK('S', 'E', 'C', 0xd033, EDID_QUIRK_FORCE_8BPC), + EDID_QUIRK('S', 'E', 'C', 0xd033, BIT(EDID_QUIRK_FORCE_8BPC)), /* Rotel RSX-1058 forwards sink's EDID but only does HDMI 1.1*/ - EDID_QUIRK('E', 'T', 'R', 13896, EDID_QUIRK_FORCE_8BPC), + EDID_QUIRK('E', 'T', 'R', 13896, BIT(EDID_QUIRK_FORCE_8BPC)), /* Valve Index Headset */ - EDID_QUIRK('V', 'L', 'V', 0x91a8, EDID_QUIRK_NON_DESKTOP), - EDID_QUIRK('V', 'L', 'V', 0x91b0, EDID_QUIRK_NON_DESKTOP), - EDID_QUIRK('V', 'L', 'V', 0x91b1, EDID_QUIRK_NON_DESKTOP), - EDID_QUIRK('V', 'L', 'V', 0x91b2, EDID_QUIRK_NON_DESKTOP), - EDID_QUIRK('V', 'L', 'V', 0x91b3, EDID_QUIRK_NON_DESKTOP), - EDID_QUIRK('V', 'L', 'V', 0x91b4, EDID_QUIRK_NON_DESKTOP), - EDID_QUIRK('V', 'L', 'V', 0x91b5, EDID_QUIRK_NON_DESKTOP), - EDID_QUIRK('V', 'L', 'V', 0x91b6, EDID_QUIRK_NON_DESKTOP), - EDID_QUIRK('V', 'L', 'V', 0x91b7, EDID_QUIRK_NON_DESKTOP), - EDID_QUIRK('V', 'L', 'V', 0x91b8, EDID_QUIRK_NON_DESKTOP), - EDID_QUIRK('V', 'L', 'V', 0x91b9, EDID_QUIRK_NON_DESKTOP), - EDID_QUIRK('V', 'L', 'V', 0x91ba, EDID_QUIRK_NON_DESKTOP), - EDID_QUIRK('V', 'L', 'V', 0x91bb, EDID_QUIRK_NON_DESKTOP), - EDID_QUIRK('V', 'L', 'V', 0x91bc, EDID_QUIRK_NON_DESKTOP), - EDID_QUIRK('V', 'L', 'V', 0x91bd, EDID_QUIRK_NON_DESKTOP), - EDID_QUIRK('V', 'L', 'V', 0x91be, EDID_QUIRK_NON_DESKTOP), - EDID_QUIRK('V', 'L', 'V', 0x91bf, EDID_QUIRK_NON_DESKTOP), + EDID_QUIRK('V', 'L', 'V', 0x91a8, BIT(EDID_QUIRK_NON_DESKTOP)), + EDID_QUIRK('V', 'L', 'V', 0x91b0, BIT(EDID_QUIRK_NON_DESKTOP)), + EDID_QUIRK('V', 'L', 'V', 0x91b1, BIT(EDID_QUIRK_NON_DESKTOP)), + EDID_QUIRK('V', 'L', 'V', 0x91b2, BIT(EDID_QUIRK_NON_DESKTOP)), + EDID_QUIRK('V', 'L', 'V', 0x91b3, BIT(EDID_QUIRK_NON_DESKTOP)), + EDID_QUIRK('V', 'L', 'V', 0x91b4, BIT(EDID_QUIRK_NON_DESKTOP)), + EDID_QUIRK('V', 'L', 'V', 0x91b5, BIT(EDID_QUIRK_NON_DESKTOP)), + EDID_QUIRK('V', 'L', 'V', 0x91b6, BIT(EDID_QUIRK_NON_DESKTOP)), + EDID_QUIRK('V', 'L', 'V', 0x91b7, BIT(EDID_QUIRK_NON_DESKTOP)), + EDID_QUIRK('V', 'L', 'V', 0x91b8, BIT(EDID_QUIRK_NON_DESKTOP)), + EDID_QUIRK('V', 'L', 'V', 0x91b9, BIT(EDID_QUIRK_NON_DESKTOP)), + EDID_QUIRK('V', 'L', 'V', 0x91ba, BIT(EDID_QUIRK_NON_DESKTOP)), + EDID_QUIRK('V', 'L', 'V', 0x91bb, BIT(EDID_QUIRK_NON_DESKTOP)), + EDID_QUIRK('V', 'L', 'V', 0x91bc, BIT(EDID_QUIRK_NON_DESKTOP)), + EDID_QUIRK('V', 'L', 'V', 0x91bd, BIT(EDID_QUIRK_NON_DESKTOP)), + EDID_QUIRK('V', 'L', 'V', 0x91be, BIT(EDID_QUIRK_NON_DESKTOP)), + EDID_QUIRK('V', 'L', 'V', 0x91bf, BIT(EDID_QUIRK_NON_DESKTOP)), /* HTC Vive and Vive Pro VR Headsets */ - EDID_QUIRK('H', 'V', 'R', 0xaa01, EDID_QUIRK_NON_DESKTOP), - EDID_QUIRK('H', 'V', 'R', 0xaa02, EDID_QUIRK_NON_DESKTOP), + EDID_QUIRK('H', 'V', 'R', 0xaa01, BIT(EDID_QUIRK_NON_DESKTOP)), + EDID_QUIRK('H', 'V', 'R', 0xaa02, BIT(EDID_QUIRK_NON_DESKTOP)), /* Oculus Rift DK1, DK2, CV1 and Rift S VR Headsets */ - EDID_QUIRK('O', 'V', 'R', 0x0001, EDID_QUIRK_NON_DESKTOP), - EDID_QUIRK('O', 'V', 'R', 0x0003, EDID_QUIRK_NON_DESKTOP), - EDID_QUIRK('O', 'V', 'R', 0x0004, EDID_QUIRK_NON_DESKTOP), - EDID_QUIRK('O', 'V', 'R', 0x0012, EDID_QUIRK_NON_DESKTOP), + EDID_QUIRK('O', 'V', 'R', 0x0001, BIT(EDID_QUIRK_NON_DESKTOP)), + EDID_QUIRK('O', 'V', 'R', 0x0003, BIT(EDID_QUIRK_NON_DESKTOP)), + EDID_QUIRK('O', 'V', 'R', 0x0004, BIT(EDID_QUIRK_NON_DESKTOP)), + EDID_QUIRK('O', 'V', 'R', 0x0012, BIT(EDID_QUIRK_NON_DESKTOP)), /* Windows Mixed Reality Headsets */ - EDID_QUIRK('A', 'C', 'R', 0x7fce, EDID_QUIRK_NON_DESKTOP), - EDID_QUIRK('L', 'E', 'N', 0x0408, EDID_QUIRK_NON_DESKTOP), - EDID_QUIRK('F', 'U', 'J', 0x1970, EDID_QUIRK_NON_DESKTOP), - EDID_QUIRK('D', 'E', 'L', 0x7fce, EDID_QUIRK_NON_DESKTOP), - EDID_QUIRK('S', 'E', 'C', 0x144a, EDID_QUIRK_NON_DESKTOP), - EDID_QUIRK('A', 'U', 'S', 0xc102, EDID_QUIRK_NON_DESKTOP), + EDID_QUIRK('A', 'C', 'R', 0x7fce, BIT(EDID_QUIRK_NON_DESKTOP)), + EDID_QUIRK('L', 'E', 'N', 0x0408, BIT(EDID_QUIRK_NON_DESKTOP)), + EDID_QUIRK('F', 'U', 'J', 0x1970, BIT(EDID_QUIRK_NON_DESKTOP)), + EDID_QUIRK('D', 'E', 'L', 0x7fce, BIT(EDID_QUIRK_NON_DESKTOP)), + EDID_QUIRK('S', 'E', 'C', 0x144a, BIT(EDID_QUIRK_NON_DESKTOP)), + EDID_QUIRK('A', 'U', 'S', 0xc102, BIT(EDID_QUIRK_NON_DESKTOP)), /* Sony PlayStation VR Headset */ - EDID_QUIRK('S', 'N', 'Y', 0x0704, EDID_QUIRK_NON_DESKTOP), + EDID_QUIRK('S', 'N', 'Y', 0x0704, BIT(EDID_QUIRK_NON_DESKTOP)), /* Sensics VR Headsets */ - EDID_QUIRK('S', 'E', 'N', 0x1019, EDID_QUIRK_NON_DESKTOP), + EDID_QUIRK('S', 'E', 'N', 0x1019, BIT(EDID_QUIRK_NON_DESKTOP)), /* OSVR HDK and HDK2 VR Headsets */ - EDID_QUIRK('S', 'V', 'R', 0x1019, EDID_QUIRK_NON_DESKTOP), - EDID_QUIRK('A', 'U', 'O', 0x1111, EDID_QUIRK_NON_DESKTOP), + EDID_QUIRK('S', 'V', 'R', 0x1019, BIT(EDID_QUIRK_NON_DESKTOP)), + EDID_QUIRK('A', 'U', 'O', 0x1111, BIT(EDID_QUIRK_NON_DESKTOP)), + + /* + * @drm_edid_internal_quirk entries end here, following with the + * @drm_edid_quirk entries. + */ + + /* HP ZR24w DP AUX DPCD access requires probing to prevent corruption. */ + EDID_QUIRK('H', 'W', 'P', 0x2869, BIT(DRM_EDID_QUIRK_DP_DPCD_PROBE)), }; /* @@ -2952,6 +2962,18 @@ static u32 edid_get_quirks(const struct drm_edid *drm_edid) return 0; } +static bool drm_edid_has_internal_quirk(struct drm_connector *connector, + enum drm_edid_internal_quirk quirk) +{ + return connector->display_info.quirks & BIT(quirk); +} + +bool drm_edid_has_quirk(struct drm_connector *connector, enum drm_edid_quirk quirk) +{ + return connector->display_info.quirks & BIT(quirk); +} +EXPORT_SYMBOL(drm_edid_has_quirk); + #define MODE_SIZE(m) ((m)->hdisplay * (m)->vdisplay) #define MODE_REFRESH_DIFF(c,t) (abs((c) - (t))) @@ -2961,7 +2983,6 @@ static u32 edid_get_quirks(const struct drm_edid *drm_edid) */ static void edid_fixup_preferred(struct drm_connector *connector) { - const struct drm_display_info *info = &connector->display_info; struct drm_display_mode *t, *cur_mode, *preferred_mode; int target_refresh = 0; int cur_vrefresh, preferred_vrefresh; @@ -2969,9 +2990,9 @@ static void edid_fixup_preferred(struct drm_connector *connector) if (list_empty(&connector->probed_modes)) return; - if (info->quirks & EDID_QUIRK_PREFER_LARGE_60) + if (drm_edid_has_internal_quirk(connector, EDID_QUIRK_PREFER_LARGE_60)) target_refresh = 60; - if (info->quirks & EDID_QUIRK_PREFER_LARGE_75) + if (drm_edid_has_internal_quirk(connector, EDID_QUIRK_PREFER_LARGE_75)) target_refresh = 75; preferred_mode = list_first_entry(&connector->probed_modes, @@ -3475,7 +3496,6 @@ static struct drm_display_mode *drm_mode_detailed(struct drm_connector *connecto const struct drm_edid *drm_edid, const struct detailed_timing *timing) { - const struct drm_display_info *info = &connector->display_info; struct drm_device *dev = connector->dev; struct drm_display_mode *mode; const struct detailed_pixel_timing *pt = &timing->data.pixel_data; @@ -3509,7 +3529,7 @@ static struct drm_display_mode *drm_mode_detailed(struct drm_connector *connecto return NULL; } - if (info->quirks & EDID_QUIRK_FORCE_REDUCED_BLANKING) { + if (drm_edid_has_internal_quirk(connector, EDID_QUIRK_FORCE_REDUCED_BLANKING)) { mode = drm_cvt_mode(dev, hactive, vactive, 60, true, false, false); if (!mode) return NULL; @@ -3521,7 +3541,7 @@ static struct drm_display_mode *drm_mode_detailed(struct drm_connector *connecto if (!mode) return NULL; - if (info->quirks & EDID_QUIRK_135_CLOCK_TOO_HIGH) + if (drm_edid_has_internal_quirk(connector, EDID_QUIRK_135_CLOCK_TOO_HIGH)) mode->clock = 1088 * 10; else mode->clock = le16_to_cpu(timing->pixel_clock) * 10; @@ -3552,7 +3572,7 @@ static struct drm_display_mode *drm_mode_detailed(struct drm_connector *connecto drm_mode_do_interlace_quirk(mode, pt); - if (info->quirks & EDID_QUIRK_DETAILED_SYNC_PP) { + if (drm_edid_has_internal_quirk(connector, EDID_QUIRK_DETAILED_SYNC_PP)) { mode->flags |= DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC; } else { mode->flags |= (pt->misc & DRM_EDID_PT_HSYNC_POSITIVE) ? @@ -3565,12 +3585,12 @@ set_size: mode->width_mm = pt->width_mm_lo | (pt->width_height_mm_hi & 0xf0) << 4; mode->height_mm = pt->height_mm_lo | (pt->width_height_mm_hi & 0xf) << 8; - if (info->quirks & EDID_QUIRK_DETAILED_IN_CM) { + if (drm_edid_has_internal_quirk(connector, EDID_QUIRK_DETAILED_IN_CM)) { mode->width_mm *= 10; mode->height_mm *= 10; } - if (info->quirks & EDID_QUIRK_DETAILED_USE_MAXIMUM_SIZE) { + if (drm_edid_has_internal_quirk(connector, EDID_QUIRK_DETAILED_USE_MAXIMUM_SIZE)) { mode->width_mm = drm_edid->edid->width_cm * 10; mode->height_mm = drm_edid->edid->height_cm * 10; } @@ -6736,26 +6756,26 @@ static void update_display_info(struct drm_connector *connector, drm_update_mso(connector, drm_edid); out: - if (info->quirks & EDID_QUIRK_NON_DESKTOP) { + if (drm_edid_has_internal_quirk(connector, EDID_QUIRK_NON_DESKTOP)) { drm_dbg_kms(connector->dev, "[CONNECTOR:%d:%s] Non-desktop display%s\n", connector->base.id, connector->name, info->non_desktop ? " (redundant quirk)" : ""); info->non_desktop = true; } - if (info->quirks & EDID_QUIRK_CAP_DSC_15BPP) + if (drm_edid_has_internal_quirk(connector, EDID_QUIRK_CAP_DSC_15BPP)) info->max_dsc_bpp = 15; - if (info->quirks & EDID_QUIRK_FORCE_6BPC) + if (drm_edid_has_internal_quirk(connector, EDID_QUIRK_FORCE_6BPC)) info->bpc = 6; - if (info->quirks & EDID_QUIRK_FORCE_8BPC) + if (drm_edid_has_internal_quirk(connector, EDID_QUIRK_FORCE_8BPC)) info->bpc = 8; - if (info->quirks & EDID_QUIRK_FORCE_10BPC) + if (drm_edid_has_internal_quirk(connector, EDID_QUIRK_FORCE_10BPC)) info->bpc = 10; - if (info->quirks & EDID_QUIRK_FORCE_12BPC) + if (drm_edid_has_internal_quirk(connector, EDID_QUIRK_FORCE_12BPC)) info->bpc = 12; /* Depends on info->cea_rev set by drm_parse_cea_ext() above */ @@ -6920,7 +6940,6 @@ static int add_displayid_detailed_modes(struct drm_connector *connector, static int _drm_edid_connector_add_modes(struct drm_connector *connector, const struct drm_edid *drm_edid) { - const struct drm_display_info *info = &connector->display_info; int num_modes = 0; if (!drm_edid) @@ -6950,7 +6969,8 @@ static int _drm_edid_connector_add_modes(struct drm_connector *connector, if (drm_edid->edid->features & DRM_EDID_FEATURE_CONTINUOUS_FREQ) num_modes += add_inferred_modes(connector, drm_edid); - if (info->quirks & (EDID_QUIRK_PREFER_LARGE_60 | EDID_QUIRK_PREFER_LARGE_75)) + if (drm_edid_has_internal_quirk(connector, EDID_QUIRK_PREFER_LARGE_60) || + drm_edid_has_internal_quirk(connector, EDID_QUIRK_PREFER_LARGE_75)) edid_fixup_preferred(connector); return num_modes; diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c index 19d50d254fe691b6be129a1061cc8f05786a58d7..0905ef6786e9688956d3e69d84c2737d92fd3d59 100644 --- a/drivers/gpu/drm/drm_gem.c +++ b/drivers/gpu/drm/drm_gem.c @@ -1429,12 +1429,14 @@ EXPORT_SYMBOL(drm_gem_lru_move_tail); * @nr_to_scan: The number of pages to try to reclaim * @remaining: The number of pages left to reclaim, should be initialized by caller * @shrink: Callback to try to shrink/reclaim the object. + * @ticket: Optional ww_acquire_ctx context to use for locking */ unsigned long drm_gem_lru_scan(struct drm_gem_lru *lru, unsigned int nr_to_scan, unsigned long *remaining, - bool (*shrink)(struct drm_gem_object *obj)) + bool (*shrink)(struct drm_gem_object *obj, struct ww_acquire_ctx *ticket), + struct ww_acquire_ctx *ticket) { struct drm_gem_lru still_in_lru; struct drm_gem_object *obj; @@ -1467,17 +1469,20 @@ drm_gem_lru_scan(struct drm_gem_lru *lru, */ mutex_unlock(lru->lock); + if (ticket) + ww_acquire_init(ticket, &reservation_ww_class); + /* * Note that this still needs to be trylock, since we can * hit shrinker in response to trying to get backing pages * for this obj (ie. while it's lock is already held) */ - if (!dma_resv_trylock(obj->resv)) { + if (!ww_mutex_trylock(&obj->resv->lock, ticket)) { *remaining += obj->size >> PAGE_SHIFT; goto tail; } - if (shrink(obj)) { + if (shrink(obj, ticket)) { freed += obj->size >> PAGE_SHIFT; /* @@ -1491,6 +1496,9 @@ drm_gem_lru_scan(struct drm_gem_lru *lru, dma_resv_unlock(obj->resv); + if (ticket) + ww_acquire_fini(ticket); + tail: drm_gem_object_put(obj); mutex_lock(lru->lock); diff --git a/drivers/gpu/drm/drm_gpusvm.c b/drivers/gpu/drm/drm_gpusvm.c index 3b0c3886a2d0b60fe5fa14ae87116f638c353b29..ab198645d90f01ce6a87674dc5371c9126bfc566 100644 --- a/drivers/gpu/drm/drm_gpusvm.c +++ b/drivers/gpu/drm/drm_gpusvm.c @@ -981,6 +981,40 @@ static void drm_gpusvm_driver_lock_held(struct drm_gpusvm *gpusvm) } #endif +/** + * drm_gpusvm_find_vma_start() - Find start address for first VMA in range + * @gpusvm: Pointer to the GPU SVM structure + * @start: The inclusive start user address. + * @end: The exclusive end user address. + * + * Returns: The start address of first VMA within the provided range, + * ULONG_MAX otherwise. Assumes start_addr < end_addr. + */ +unsigned long +drm_gpusvm_find_vma_start(struct drm_gpusvm *gpusvm, + unsigned long start, + unsigned long end) +{ + struct mm_struct *mm = gpusvm->mm; + struct vm_area_struct *vma; + unsigned long addr = ULONG_MAX; + + if (!mmget_not_zero(mm)) + return addr; + + mmap_read_lock(mm); + + vma = find_vma_intersection(mm, start, end); + if (vma) + addr = vma->vm_start; + + mmap_read_unlock(mm); + mmput(mm); + + return addr; +} +EXPORT_SYMBOL_GPL(drm_gpusvm_find_vma_start); + /** * drm_gpusvm_range_find_or_insert() - Find or insert GPU SVM range * @gpusvm: Pointer to the GPU SVM structure diff --git a/drivers/gpu/drm/drm_gpuvm.c b/drivers/gpu/drm/drm_gpuvm.c index e89b932e987c759f924a52eca0c2ad9ff5405ecf..bbc7fecb6f4ab4c7208f7d88a660b837ec2de340 100644 --- a/drivers/gpu/drm/drm_gpuvm.c +++ b/drivers/gpu/drm/drm_gpuvm.c @@ -2300,13 +2300,13 @@ __drm_gpuvm_sm_unmap(struct drm_gpuvm *gpuvm, } /** - * drm_gpuvm_sm_map() - creates the &drm_gpuva_op split/merge steps + * drm_gpuvm_sm_map() - calls the &drm_gpuva_op split/merge steps * @gpuvm: the &drm_gpuvm representing the GPU VA space + * @priv: pointer to a driver private data structure * @req_addr: the start address of the new mapping * @req_range: the range of the new mapping * @req_obj: the &drm_gem_object to map * @req_offset: the offset within the &drm_gem_object - * @priv: pointer to a driver private data structure * * This function iterates the given range of the GPU VA space. It utilizes the * &drm_gpuvm_ops to call back into the driver providing the split and merge @@ -2350,7 +2350,7 @@ drm_gpuvm_sm_map(struct drm_gpuvm *gpuvm, void *priv, EXPORT_SYMBOL_GPL(drm_gpuvm_sm_map); /** - * drm_gpuvm_sm_unmap() - creates the &drm_gpuva_ops to split on unmap + * drm_gpuvm_sm_unmap() - calls the &drm_gpuva_ops to split on unmap * @gpuvm: the &drm_gpuvm representing the GPU VA space * @priv: pointer to a driver private data structure * @req_addr: the start address of the range to unmap @@ -2391,6 +2391,132 @@ drm_gpuvm_sm_unmap(struct drm_gpuvm *gpuvm, void *priv, } EXPORT_SYMBOL_GPL(drm_gpuvm_sm_unmap); +static int +drm_gpuva_sm_step_lock(struct drm_gpuva_op *op, void *priv) +{ + struct drm_exec *exec = priv; + + switch (op->op) { + case DRM_GPUVA_OP_REMAP: + if (op->remap.unmap->va->gem.obj) + return drm_exec_lock_obj(exec, op->remap.unmap->va->gem.obj); + return 0; + case DRM_GPUVA_OP_UNMAP: + if (op->unmap.va->gem.obj) + return drm_exec_lock_obj(exec, op->unmap.va->gem.obj); + return 0; + default: + return 0; + } +} + +static const struct drm_gpuvm_ops lock_ops = { + .sm_step_map = drm_gpuva_sm_step_lock, + .sm_step_remap = drm_gpuva_sm_step_lock, + .sm_step_unmap = drm_gpuva_sm_step_lock, +}; + +/** + * drm_gpuvm_sm_map_exec_lock() - locks the objects touched by a drm_gpuvm_sm_map() + * @gpuvm: the &drm_gpuvm representing the GPU VA space + * @exec: the &drm_exec locking context + * @num_fences: for newly mapped objects, the # of fences to reserve + * @req_addr: the start address of the range to unmap + * @req_range: the range of the mappings to unmap + * @req_obj: the &drm_gem_object to map + * @req_offset: the offset within the &drm_gem_object + * + * This function locks (drm_exec_lock_obj()) objects that will be unmapped/ + * remapped, and locks+prepares (drm_exec_prepare_object()) objects that + * will be newly mapped. + * + * The expected usage is: + * + * vm_bind { + * struct drm_exec exec; + * + * // IGNORE_DUPLICATES is required, INTERRUPTIBLE_WAIT is recommended: + * drm_exec_init(&exec, IGNORE_DUPLICATES | INTERRUPTIBLE_WAIT, 0); + * + * drm_exec_until_all_locked (&exec) { + * for_each_vm_bind_operation { + * switch (op->op) { + * case DRIVER_OP_UNMAP: + * ret = drm_gpuvm_sm_unmap_exec_lock(gpuvm, &exec, op->addr, op->range); + * break; + * case DRIVER_OP_MAP: + * ret = drm_gpuvm_sm_map_exec_lock(gpuvm, &exec, num_fences, + * op->addr, op->range, + * obj, op->obj_offset); + * break; + * } + * + * drm_exec_retry_on_contention(&exec); + * if (ret) + * return ret; + * } + * } + * } + * + * This enables all locking to be performed before the driver begins modifying + * the VM. This is safe to do in the case of overlapping DRIVER_VM_BIND_OPs, + * where an earlier op can alter the sequence of steps generated for a later + * op, because the later altered step will involve the same GEM object(s) + * already seen in the earlier locking step. For example: + * + * 1) An earlier driver DRIVER_OP_UNMAP op removes the need for a + * DRM_GPUVA_OP_REMAP/UNMAP step. This is safe because we've already + * locked the GEM object in the earlier DRIVER_OP_UNMAP op. + * + * 2) An earlier DRIVER_OP_MAP op overlaps with a later DRIVER_OP_MAP/UNMAP + * op, introducing a DRM_GPUVA_OP_REMAP/UNMAP that wouldn't have been + * required without the earlier DRIVER_OP_MAP. This is safe because we've + * already locked the GEM object in the earlier DRIVER_OP_MAP step. + * + * Returns: 0 on success or a negative error codec + */ +int +drm_gpuvm_sm_map_exec_lock(struct drm_gpuvm *gpuvm, + struct drm_exec *exec, unsigned int num_fences, + u64 req_addr, u64 req_range, + struct drm_gem_object *req_obj, u64 req_offset) +{ + if (req_obj) { + int ret = drm_exec_prepare_obj(exec, req_obj, num_fences); + if (ret) + return ret; + } + + return __drm_gpuvm_sm_map(gpuvm, &lock_ops, exec, + req_addr, req_range, + req_obj, req_offset); + +} +EXPORT_SYMBOL_GPL(drm_gpuvm_sm_map_exec_lock); + +/** + * drm_gpuvm_sm_unmap_exec_lock() - locks the objects touched by drm_gpuvm_sm_unmap() + * @gpuvm: the &drm_gpuvm representing the GPU VA space + * @exec: the &drm_exec locking context + * @req_addr: the start address of the range to unmap + * @req_range: the range of the mappings to unmap + * + * This function locks (drm_exec_lock_obj()) objects that will be unmapped/ + * remapped by drm_gpuvm_sm_unmap(). + * + * See drm_gpuvm_sm_map_exec_lock() for expected usage. + * + * Returns: 0 on success or a negative error code + */ +int +drm_gpuvm_sm_unmap_exec_lock(struct drm_gpuvm *gpuvm, struct drm_exec *exec, + u64 req_addr, u64 req_range) +{ + return __drm_gpuvm_sm_unmap(gpuvm, &lock_ops, exec, + req_addr, req_range); +} +EXPORT_SYMBOL_GPL(drm_gpuvm_sm_unmap_exec_lock); + static struct drm_gpuva_op * gpuva_op_alloc(struct drm_gpuvm *gpuvm) { diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile index e153686256c9ebac78a607b84364231d16194eb5..85354344307284ecff1f888e29c943412f355b3c 100644 --- a/drivers/gpu/drm/i915/Makefile +++ b/drivers/gpu/drm/i915/Makefile @@ -40,12 +40,11 @@ i915-y += \ intel_pcode.o \ intel_region_ttm.o \ intel_runtime_pm.o \ - intel_sbi.o \ intel_step.o \ intel_uncore.o \ intel_uncore_trace.o \ intel_wakeref.o \ - vlv_sideband.o \ + vlv_iosf_sb.o \ vlv_suspend.o # core peripheral code @@ -219,12 +218,11 @@ i915-$(CONFIG_HWMON) += \ # modesetting core code i915-y += \ display/hsw_ips.o \ - display/i9xx_plane.o \ display/i9xx_display_sr.o \ + display/i9xx_plane.o \ display/i9xx_wm.o \ display/intel_alpm.o \ display/intel_atomic.o \ - display/intel_atomic_plane.o \ display/intel_audio.o \ display/intel_bios.o \ display/intel_bo.o \ @@ -266,6 +264,7 @@ i915-y += \ display/intel_fbc.o \ display/intel_fdi.o \ display/intel_fifo_underrun.o \ + display/intel_flipq.o \ display/intel_frontbuffer.o \ display/intel_global_state.o \ display/intel_hdcp.o \ @@ -284,10 +283,12 @@ i915-y += \ display/intel_pch.o \ display/intel_pch_display.o \ display/intel_pch_refclk.o \ + display/intel_plane.o \ display/intel_plane_initial.o \ display/intel_pmdemand.o \ display/intel_psr.o \ display/intel_quirks.o \ + display/intel_sbi.o \ display/intel_sprite.o \ display/intel_sprite_uapi.o \ display/intel_tc.o \ @@ -296,7 +297,8 @@ i915-y += \ display/intel_wm.o \ display/skl_scaler.o \ display/skl_universal_plane.o \ - display/skl_watermark.o + display/skl_watermark.o \ + display/vlv_sideband.o i915-$(CONFIG_ACPI) += \ display/intel_acpi.o \ display/intel_opregion.o diff --git a/drivers/gpu/drm/i915/display/g4x_dp.c b/drivers/gpu/drm/i915/display/g4x_dp.c index e0a98e6fd6d1def6c4b6b7c520040440771836bf..87f6b9602b16a0833aa4e16fb44e3312fa281883 100644 --- a/drivers/gpu/drm/i915/display/g4x_dp.c +++ b/drivers/gpu/drm/i915/display/g4x_dp.c @@ -18,6 +18,7 @@ #include "intel_crtc.h" #include "intel_de.h" #include "intel_display_power.h" +#include "intel_display_regs.h" #include "intel_display_types.h" #include "intel_dp.h" #include "intel_dp_aux.h" diff --git a/drivers/gpu/drm/i915/display/g4x_hdmi.c b/drivers/gpu/drm/i915/display/g4x_hdmi.c index 1d252432d729a7a5639414115c1c1de3af32ae4f..2610f5702fb9a7b5591b70114b996cfed40748e7 100644 --- a/drivers/gpu/drm/i915/display/g4x_hdmi.c +++ b/drivers/gpu/drm/i915/display/g4x_hdmi.c @@ -15,6 +15,7 @@ #include "intel_crtc.h" #include "intel_de.h" #include "intel_display_power.h" +#include "intel_display_regs.h" #include "intel_display_types.h" #include "intel_dp_aux.h" #include "intel_dpio_phy.h" diff --git a/drivers/gpu/drm/i915/display/hsw_ips.c b/drivers/gpu/drm/i915/display/hsw_ips.c index 4307e2ed03d949aea70fd5d1263c349ef8dcc5bc..927fe56aec777416327db31987b631ffb3d4320f 100644 --- a/drivers/gpu/drm/i915/display/hsw_ips.c +++ b/drivers/gpu/drm/i915/display/hsw_ips.c @@ -5,11 +5,13 @@ #include +#include + #include "hsw_ips.h" -#include "i915_drv.h" #include "i915_reg.h" #include "intel_color_regs.h" #include "intel_de.h" +#include "intel_display_regs.h" #include "intel_display_rpm.h" #include "intel_display_types.h" #include "intel_pcode.h" @@ -17,8 +19,6 @@ static void hsw_ips_enable(const struct intel_crtc_state *crtc_state) { struct intel_display *display = to_intel_display(crtc_state); - struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); - struct drm_i915_private *i915 = to_i915(crtc->base.dev); u32 val; if (!crtc_state->ips_enabled) @@ -39,8 +39,8 @@ static void hsw_ips_enable(const struct intel_crtc_state *crtc_state) if (display->platform.broadwell) { drm_WARN_ON(display->drm, - snb_pcode_write(&i915->uncore, DISPLAY_IPS_CONTROL, - val | IPS_PCODE_CONTROL)); + intel_pcode_write(display->drm, DISPLAY_IPS_CONTROL, + val | IPS_PCODE_CONTROL)); /* * Quoting Art Runyan: "its not safe to expect any particular * value in IPS_CTL bit 31 after enabling IPS through the @@ -65,8 +65,6 @@ static void hsw_ips_enable(const struct intel_crtc_state *crtc_state) bool hsw_ips_disable(const struct intel_crtc_state *crtc_state) { struct intel_display *display = to_intel_display(crtc_state); - struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); - struct drm_i915_private *i915 = to_i915(crtc->base.dev); bool need_vblank_wait = false; if (!crtc_state->ips_enabled) @@ -74,7 +72,7 @@ bool hsw_ips_disable(const struct intel_crtc_state *crtc_state) if (display->platform.broadwell) { drm_WARN_ON(display->drm, - snb_pcode_write(&i915->uncore, DISPLAY_IPS_CONTROL, 0)); + intel_pcode_write(display->drm, DISPLAY_IPS_CONTROL, 0)); /* * Wait for PCODE to finish disabling IPS. The BSpec specified * 42ms timeout value leads to occasional timeouts so use 100ms @@ -267,7 +265,7 @@ int hsw_ips_compute_config(struct intel_atomic_state *state, return PTR_ERR(cdclk_state); /* pixel rate mustn't exceed 95% of cdclk with IPS on BDW */ - if (crtc_state->pixel_rate > cdclk_state->logical.cdclk * 95 / 100) + if (crtc_state->pixel_rate > intel_cdclk_logical(cdclk_state) * 95 / 100) return 0; } diff --git a/drivers/gpu/drm/i915/display/i9xx_display_sr.c b/drivers/gpu/drm/i915/display/i9xx_display_sr.c index 32abe9743014b8c8afd5dc530c62b0cf2712b2f9..935419441709db3f66394cbc82ea7915caa4ca50 100644 --- a/drivers/gpu/drm/i915/display/i9xx_display_sr.c +++ b/drivers/gpu/drm/i915/display/i9xx_display_sr.c @@ -5,10 +5,10 @@ #include -#include "i915_reg.h" #include "i9xx_display_sr.h" #include "i9xx_wm_regs.h" #include "intel_de.h" +#include "intel_display_regs.h" #include "intel_gmbus.h" #include "intel_pci_config.h" diff --git a/drivers/gpu/drm/i915/display/i9xx_plane.c b/drivers/gpu/drm/i915/display/i9xx_plane.c index a2a6d52be0a5ff064bc73a2bb0c893170684b9fe..f291ced989dc197add3c2707864d0451cb041aca 100644 --- a/drivers/gpu/drm/i915/display/i9xx_plane.c +++ b/drivers/gpu/drm/i915/display/i9xx_plane.c @@ -2,6 +2,7 @@ /* * Copyright © 2020 Intel Corporation */ + #include #include @@ -14,13 +15,15 @@ #include "i9xx_plane.h" #include "i9xx_plane_regs.h" #include "intel_atomic.h" -#include "intel_atomic_plane.h" +#include "intel_bo.h" #include "intel_de.h" #include "intel_display_irq.h" +#include "intel_display_regs.h" #include "intel_display_types.h" #include "intel_fb.h" #include "intel_fbc.h" #include "intel_frontbuffer.h" +#include "intel_plane.h" #include "intel_sprite.h" /* Primary plane formats for gen <= 3 */ @@ -334,10 +337,10 @@ i9xx_plane_check(struct intel_crtc_state *crtc_state, if (ret) return ret; - ret = intel_atomic_plane_check_clipping(plane_state, crtc_state, - DRM_PLANE_NO_SCALING, - DRM_PLANE_NO_SCALING, - i9xx_plane_has_windowing(plane)); + ret = intel_plane_check_clipping(plane_state, crtc_state, + DRM_PLANE_NO_SCALING, + DRM_PLANE_NO_SCALING, + i9xx_plane_has_windowing(plane)); if (ret) return ret; @@ -903,6 +906,27 @@ static const struct drm_plane_funcs i8xx_plane_funcs = { .format_mod_supported_async = intel_plane_format_mod_supported_async, }; +static void i9xx_disable_tiling(struct intel_plane *plane) +{ + struct intel_display *display = to_intel_display(plane); + enum i9xx_plane_id i9xx_plane = plane->i9xx_plane; + u32 dspcntr; + u32 reg; + + dspcntr = intel_de_read_fw(display, DSPCNTR(display, i9xx_plane)); + dspcntr &= ~DISP_TILED; + intel_de_write_fw(display, DSPCNTR(display, i9xx_plane), dspcntr); + + if (DISPLAY_VER(display) >= 4) { + reg = intel_de_read_fw(display, DSPSURF(display, i9xx_plane)); + intel_de_write_fw(display, DSPSURF(display, i9xx_plane), reg); + + } else { + reg = intel_de_read_fw(display, DSPADDR(display, i9xx_plane)); + intel_de_write_fw(display, DSPADDR(display, i9xx_plane), reg); + } +} + struct intel_plane * intel_primary_plane_create(struct intel_display *display, enum pipe pipe) { @@ -1045,6 +1069,8 @@ intel_primary_plane_create(struct intel_display *display, enum pipe pipe) } } + plane->disable_tiling = i9xx_disable_tiling; + modifiers = intel_fb_plane_get_modifiers(display, INTEL_PLANE_CAP_TILING_X); if (DISPLAY_VER(display) >= 5 || display->platform.g4x) @@ -1149,7 +1175,7 @@ i9xx_get_initial_plane_config(struct intel_crtc *crtc, drm_WARN_ON(display->drm, pipe != crtc->pipe); - intel_fb = kzalloc(sizeof(*intel_fb), GFP_KERNEL); + intel_fb = intel_bo_alloc_framebuffer(); if (!intel_fb) { drm_dbg_kms(display->drm, "failed to alloc fb\n"); return; diff --git a/drivers/gpu/drm/i915/display/i9xx_wm.c b/drivers/gpu/drm/i915/display/i9xx_wm.c index 77876ef735b746b486e2efcb492d8b4752522ba2..1f9db511877721bc3f7d7e04fd110c6fef763e12 100644 --- a/drivers/gpu/drm/i915/display/i9xx_wm.c +++ b/drivers/gpu/drm/i915/display/i9xx_wm.c @@ -11,6 +11,7 @@ #include "intel_bo.h" #include "intel_de.h" #include "intel_display.h" +#include "intel_display_regs.h" #include "intel_display_trace.h" #include "intel_fb.h" #include "intel_mchbar_regs.h" @@ -107,43 +108,41 @@ static const struct cxsr_latency *pnv_get_cxsr_latency(struct intel_display *dis static void chv_set_memory_dvfs(struct intel_display *display, bool enable) { - struct drm_i915_private *dev_priv = to_i915(display->drm); u32 val; - vlv_punit_get(dev_priv); + vlv_punit_get(display->drm); - val = vlv_punit_read(dev_priv, PUNIT_REG_DDR_SETUP2); + val = vlv_punit_read(display->drm, PUNIT_REG_DDR_SETUP2); if (enable) val &= ~FORCE_DDR_HIGH_FREQ; else val |= FORCE_DDR_HIGH_FREQ; val &= ~FORCE_DDR_LOW_FREQ; val |= FORCE_DDR_FREQ_REQ_ACK; - vlv_punit_write(dev_priv, PUNIT_REG_DDR_SETUP2, val); + vlv_punit_write(display->drm, PUNIT_REG_DDR_SETUP2, val); - if (wait_for((vlv_punit_read(dev_priv, PUNIT_REG_DDR_SETUP2) & + if (wait_for((vlv_punit_read(display->drm, PUNIT_REG_DDR_SETUP2) & FORCE_DDR_FREQ_REQ_ACK) == 0, 3)) drm_err(display->drm, "timed out waiting for Punit DDR DVFS request\n"); - vlv_punit_put(dev_priv); + vlv_punit_put(display->drm); } static void chv_set_memory_pm5(struct intel_display *display, bool enable) { - struct drm_i915_private *dev_priv = to_i915(display->drm); u32 val; - vlv_punit_get(dev_priv); + vlv_punit_get(display->drm); - val = vlv_punit_read(dev_priv, PUNIT_REG_DSPSSPM); + val = vlv_punit_read(display->drm, PUNIT_REG_DSPSSPM); if (enable) val |= DSP_MAXFIFO_PM5_ENABLE; else val &= ~DSP_MAXFIFO_PM5_ENABLE; - vlv_punit_write(dev_priv, PUNIT_REG_DSPSSPM, val); + vlv_punit_write(display->drm, PUNIT_REG_DSPSSPM, val); - vlv_punit_put(dev_priv); + vlv_punit_put(display->drm); } #define FW_WM(value, plane) \ @@ -3900,7 +3899,6 @@ static void g4x_wm_sanitize(struct intel_display *display) static void vlv_wm_get_hw_state(struct intel_display *display) { - struct drm_i915_private *dev_priv = to_i915(display->drm); struct vlv_wm_values *wm = &display->wm.vlv; struct intel_crtc *crtc; u32 val; @@ -3911,9 +3909,9 @@ static void vlv_wm_get_hw_state(struct intel_display *display) wm->level = VLV_WM_LEVEL_PM2; if (display->platform.cherryview) { - vlv_punit_get(dev_priv); + vlv_punit_get(display->drm); - val = vlv_punit_read(dev_priv, PUNIT_REG_DSPSSPM); + val = vlv_punit_read(display->drm, PUNIT_REG_DSPSSPM); if (val & DSP_MAXFIFO_PM5_ENABLE) wm->level = VLV_WM_LEVEL_PM5; @@ -3926,23 +3924,23 @@ static void vlv_wm_get_hw_state(struct intel_display *display) * HIGH/LOW bits so that we don't actually change * the current state. */ - val = vlv_punit_read(dev_priv, PUNIT_REG_DDR_SETUP2); + val = vlv_punit_read(display->drm, PUNIT_REG_DDR_SETUP2); val |= FORCE_DDR_FREQ_REQ_ACK; - vlv_punit_write(dev_priv, PUNIT_REG_DDR_SETUP2, val); + vlv_punit_write(display->drm, PUNIT_REG_DDR_SETUP2, val); - if (wait_for((vlv_punit_read(dev_priv, PUNIT_REG_DDR_SETUP2) & + if (wait_for((vlv_punit_read(display->drm, PUNIT_REG_DDR_SETUP2) & FORCE_DDR_FREQ_REQ_ACK) == 0, 3)) { drm_dbg_kms(display->drm, "Punit not acking DDR DVFS request, " "assuming DDR DVFS is disabled\n"); display->wm.num_levels = VLV_WM_LEVEL_PM5 + 1; } else { - val = vlv_punit_read(dev_priv, PUNIT_REG_DDR_SETUP2); + val = vlv_punit_read(display->drm, PUNIT_REG_DDR_SETUP2); if ((val & FORCE_DDR_HIGH_FREQ) == 0) wm->level = VLV_WM_LEVEL_DDR_DVFS; } - vlv_punit_put(dev_priv); + vlv_punit_put(display->drm); } for_each_intel_crtc(display->drm, crtc) { diff --git a/drivers/gpu/drm/i915/display/icl_dsi.c b/drivers/gpu/drm/i915/display/icl_dsi.c index aa566ad3f5cd72fb8f9440e2645da0047da222b1..8d9cb73a93a7d026e64553576a29f4932a47f91c 100644 --- a/drivers/gpu/drm/i915/display/icl_dsi.c +++ b/drivers/gpu/drm/i915/display/icl_dsi.c @@ -45,6 +45,7 @@ #include "intel_crtc.h" #include "intel_ddi.h" #include "intel_de.h" +#include "intel_display_regs.h" #include "intel_dsi.h" #include "intel_dsi_vbt.h" #include "intel_panel.h" @@ -192,12 +193,12 @@ static int dsi_send_pkt_hdr(struct intel_dsi_host *host, else tmp &= ~PAYLOAD_PRESENT; - tmp &= ~VBLANK_FENCE; + tmp &= ~(VBLANK_FENCE | LP_DATA_TRANSFER | PIPELINE_FLUSH); if (enable_lpdt) tmp |= LP_DATA_TRANSFER; else - tmp &= ~LP_DATA_TRANSFER; + tmp |= PIPELINE_FLUSH; tmp &= ~(PARAM_WC_MASK | VC_MASK | DT_MASK); tmp |= ((packet->header[0] & VC_MASK) << VC_SHIFT); @@ -658,7 +659,7 @@ static void gen11_dsi_map_pll(struct intel_encoder *encoder, { struct intel_display *display = to_intel_display(encoder); struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder); - struct intel_shared_dpll *pll = crtc_state->shared_dpll; + struct intel_dpll *pll = crtc_state->intel_dpll; enum phy phy; u32 val; diff --git a/drivers/gpu/drm/i915/display/icl_dsi_regs.h b/drivers/gpu/drm/i915/display/icl_dsi_regs.h index d4845ac65acce9b0af5384188aba003f445f6408..b601b7632339a54caa3e310fb9010708e605b1c7 100644 --- a/drivers/gpu/drm/i915/display/icl_dsi_regs.h +++ b/drivers/gpu/drm/i915/display/icl_dsi_regs.h @@ -272,6 +272,7 @@ #define PAYLOAD_PRESENT (1 << 31) #define LP_DATA_TRANSFER (1 << 30) #define VBLANK_FENCE (1 << 29) +#define PIPELINE_FLUSH (1 << 28) #define PARAM_WC_MASK (0xffff << 8) #define PARAM_WC_LOWER_SHIFT 8 #define PARAM_WC_UPPER_SHIFT 16 diff --git a/drivers/gpu/drm/i915/display/intel_alpm.c b/drivers/gpu/drm/i915/display/intel_alpm.c index c176bdbc19a34208ebc3495116d369dd9b8b5f9f..dfdde8e4eabe46dc6c4bc7e6b6176a5bd1f1333b 100644 --- a/drivers/gpu/drm/i915/display/intel_alpm.c +++ b/drivers/gpu/drm/i915/display/intel_alpm.c @@ -26,6 +26,13 @@ bool intel_alpm_aux_less_wake_supported(struct intel_dp *intel_dp) return intel_dp->alpm_dpcd & DP_ALPM_AUX_LESS_CAP; } +bool intel_alpm_is_alpm_aux_less(struct intel_dp *intel_dp, + const struct intel_crtc_state *crtc_state) +{ + return intel_psr_needs_alpm_aux_less(intel_dp, crtc_state) || + (crtc_state->has_lobf && intel_alpm_aux_less_wake_supported(intel_dp)); +} + void intel_alpm_init(struct intel_dp *intel_dp) { u8 dpcd; @@ -329,7 +336,6 @@ static void lnl_alpm_configure(struct intel_dp *intel_dp, { struct intel_display *display = to_intel_display(intel_dp); enum transcoder cpu_transcoder = crtc_state->cpu_transcoder; - enum port port = dp_to_dig_port(intel_dp)->base.port; u32 alpm_ctl; if (DISPLAY_VER(display) < 20 || (!intel_psr_needs_alpm(intel_dp, crtc_state) && @@ -341,30 +347,26 @@ static void lnl_alpm_configure(struct intel_dp *intel_dp, * Panel Replay on eDP is always using ALPM aux less. I.e. no need to * check panel support at this point. */ - if ((crtc_state->has_panel_replay && intel_dp_is_edp(intel_dp)) || - (crtc_state->has_lobf && intel_alpm_aux_less_wake_supported(intel_dp))) { + if (intel_alpm_is_alpm_aux_less(intel_dp, crtc_state)) { alpm_ctl = ALPM_CTL_ALPM_ENABLE | ALPM_CTL_ALPM_AUX_LESS_ENABLE | ALPM_CTL_AUX_LESS_SLEEP_HOLD_TIME_50_SYMBOLS | ALPM_CTL_AUX_LESS_WAKE_TIME(intel_dp->alpm_parameters.aux_less_wake_lines); - intel_de_write(display, - PORT_ALPM_CTL(port), - PORT_ALPM_CTL_ALPM_AUX_LESS_ENABLE | - PORT_ALPM_CTL_MAX_PHY_SWING_SETUP(15) | - PORT_ALPM_CTL_MAX_PHY_SWING_HOLD(0) | - PORT_ALPM_CTL_SILENCE_PERIOD( - intel_dp->alpm_parameters.silence_period_sym_clocks)); - - intel_de_write(display, - PORT_ALPM_LFPS_CTL(port), - PORT_ALPM_LFPS_CTL_LFPS_CYCLE_COUNT(10) | - PORT_ALPM_LFPS_CTL_LFPS_HALF_CYCLE_DURATION( - intel_dp->alpm_parameters.lfps_half_cycle_num_of_syms) | - PORT_ALPM_LFPS_CTL_FIRST_LFPS_HALF_CYCLE_DURATION( - intel_dp->alpm_parameters.lfps_half_cycle_num_of_syms) | - PORT_ALPM_LFPS_CTL_LAST_LFPS_HALF_CYCLE_DURATION( - intel_dp->alpm_parameters.lfps_half_cycle_num_of_syms)); + if (intel_dp->as_sdp_supported) { + u32 pr_alpm_ctl = PR_ALPM_CTL_ADAPTIVE_SYNC_SDP_POSITION_T1; + + if (intel_dp->pr_dpcd[INTEL_PR_DPCD_INDEX(DP_PANEL_REPLAY_CAP_CAPABILITY)] & + DP_PANEL_REPLAY_LINK_OFF_SUPPORTED_IN_PR_AFTER_ADAPTIVE_SYNC_SDP) + pr_alpm_ctl |= PR_ALPM_CTL_ALLOW_LINK_OFF_BETWEEN_AS_SDP_AND_SU; + if (!(intel_dp->pr_dpcd[INTEL_PR_DPCD_INDEX(DP_PANEL_REPLAY_CAP_CAPABILITY)] & + DP_PANEL_REPLAY_ASYNC_VIDEO_TIMING_NOT_SUPPORTED_IN_PR)) + pr_alpm_ctl |= PR_ALPM_CTL_AS_SDP_TRANSMISSION_IN_ACTIVE_DISABLE; + + intel_de_write(display, PR_ALPM_CTL(display, cpu_transcoder), + pr_alpm_ctl); + } + } else { alpm_ctl = ALPM_CTL_EXTENDED_FAST_WAKE_ENABLE | ALPM_CTL_EXTENDED_FAST_WAKE_TIME(intel_dp->alpm_parameters.fast_wake_lines); @@ -388,6 +390,36 @@ void intel_alpm_configure(struct intel_dp *intel_dp, intel_dp->alpm_parameters.transcoder = crtc_state->cpu_transcoder; } +void intel_alpm_port_configure(struct intel_dp *intel_dp, + const struct intel_crtc_state *crtc_state) +{ + struct intel_display *display = to_intel_display(intel_dp); + enum port port = dp_to_dig_port(intel_dp)->base.port; + u32 alpm_ctl_val = 0, lfps_ctl_val = 0; + + if (DISPLAY_VER(display) < 20) + return; + + if (intel_alpm_is_alpm_aux_less(intel_dp, crtc_state)) { + alpm_ctl_val = PORT_ALPM_CTL_ALPM_AUX_LESS_ENABLE | + PORT_ALPM_CTL_MAX_PHY_SWING_SETUP(15) | + PORT_ALPM_CTL_MAX_PHY_SWING_HOLD(0) | + PORT_ALPM_CTL_SILENCE_PERIOD( + intel_dp->alpm_parameters.silence_period_sym_clocks); + lfps_ctl_val = PORT_ALPM_LFPS_CTL_LFPS_CYCLE_COUNT(10) | + PORT_ALPM_LFPS_CTL_LFPS_HALF_CYCLE_DURATION( + intel_dp->alpm_parameters.lfps_half_cycle_num_of_syms) | + PORT_ALPM_LFPS_CTL_FIRST_LFPS_HALF_CYCLE_DURATION( + intel_dp->alpm_parameters.lfps_half_cycle_num_of_syms) | + PORT_ALPM_LFPS_CTL_LAST_LFPS_HALF_CYCLE_DURATION( + intel_dp->alpm_parameters.lfps_half_cycle_num_of_syms); + } + + intel_de_write(display, PORT_ALPM_CTL(port), alpm_ctl_val); + + intel_de_write(display, PORT_ALPM_LFPS_CTL(port), lfps_ctl_val); +} + void intel_alpm_pre_plane_update(struct intel_atomic_state *state, struct intel_crtc *crtc) { diff --git a/drivers/gpu/drm/i915/display/intel_alpm.h b/drivers/gpu/drm/i915/display/intel_alpm.h index c9fe21e3e72cee3561eccb30fab8f676161dbdd9..a861c20b5d79f152e464a30e3ff9e1bdcf08f69a 100644 --- a/drivers/gpu/drm/i915/display/intel_alpm.h +++ b/drivers/gpu/drm/i915/display/intel_alpm.h @@ -27,11 +27,15 @@ void intel_alpm_enable_sink(struct intel_dp *intel_dp, const struct intel_crtc_state *crtc_state); void intel_alpm_pre_plane_update(struct intel_atomic_state *state, struct intel_crtc *crtc); +void intel_alpm_port_configure(struct intel_dp *intel_dp, + const struct intel_crtc_state *crtc_state); void intel_alpm_post_plane_update(struct intel_atomic_state *state, struct intel_crtc *crtc); void intel_alpm_lobf_debugfs_add(struct intel_connector *connector); bool intel_alpm_aux_wake_supported(struct intel_dp *intel_dp); bool intel_alpm_aux_less_wake_supported(struct intel_dp *intel_dp); +bool intel_alpm_is_alpm_aux_less(struct intel_dp *intel_dp, + const struct intel_crtc_state *crtc_state); void intel_alpm_disable(struct intel_dp *intel_dp); bool intel_alpm_get_error(struct intel_dp *intel_dp); #endif diff --git a/drivers/gpu/drm/i915/display/intel_atomic.c b/drivers/gpu/drm/i915/display/intel_atomic.c index e83feca5c9c9b50fcde61a31b4f61194f0acd11f..348b1655435e1cc597d2adca3aa10555e670335d 100644 --- a/drivers/gpu/drm/i915/display/intel_atomic.c +++ b/drivers/gpu/drm/i915/display/intel_atomic.c @@ -26,7 +26,7 @@ * * The functions here implement the state management and hardware programming * dispatch required by the atomic modeset infrastructure. - * See intel_atomic_plane.c for the plane-specific atomic functionality. + * See intel_plane.c for the plane-specific atomic functionality. */ #include @@ -274,7 +274,7 @@ intel_crtc_duplicate_state(struct drm_crtc *crtc) crtc_state->do_async_flip = false; crtc_state->fb_bits = 0; crtc_state->update_planes = 0; - crtc_state->dsb_color_vblank = NULL; + crtc_state->dsb_color = NULL; crtc_state->dsb_commit = NULL; crtc_state->use_dsb = false; @@ -310,7 +310,7 @@ intel_crtc_destroy_state(struct drm_crtc *crtc, { struct intel_crtc_state *crtc_state = to_intel_crtc_state(state); - drm_WARN_ON(crtc->dev, crtc_state->dsb_color_vblank); + drm_WARN_ON(crtc->dev, crtc_state->dsb_color); drm_WARN_ON(crtc->dev, crtc_state->dsb_commit); __drm_atomic_helper_crtc_destroy_state(&crtc_state->uapi); diff --git a/drivers/gpu/drm/i915/display/intel_audio.c b/drivers/gpu/drm/i915/display/intel_audio.c index 55af3a553c58a1abe25bd5265a26d4654310793b..5bdaef38f13d68222c9e1fc17360810b3e81aefc 100644 --- a/drivers/gpu/drm/i915/display/intel_audio.c +++ b/drivers/gpu/drm/i915/display/intel_audio.c @@ -951,7 +951,7 @@ static int glk_force_audio_cdclk_commit(struct intel_atomic_state *state, if (IS_ERR(cdclk_state)) return PTR_ERR(cdclk_state); - cdclk_state->force_min_cdclk = enable ? 2 * 96000 : 0; + intel_cdclk_force_min_cdclk(cdclk_state, enable ? 2 * 96000 : 0); return drm_atomic_commit(&state->base); } diff --git a/drivers/gpu/drm/i915/display/intel_backlight.c b/drivers/gpu/drm/i915/display/intel_backlight.c index 5827da5860032adf7a824867d38aaa5f41b39fbc..e007380e9a6311f3762c75f1a81d9d0326287ae7 100644 --- a/drivers/gpu/drm/i915/display/intel_backlight.c +++ b/drivers/gpu/drm/i915/display/intel_backlight.c @@ -7,7 +7,6 @@ #include #include #include - #include #include @@ -19,6 +18,7 @@ #include "intel_backlight_regs.h" #include "intel_connector.h" #include "intel_de.h" +#include "intel_display_regs.h" #include "intel_display_rpm.h" #include "intel_display_types.h" #include "intel_dp_aux_backlight.h" diff --git a/drivers/gpu/drm/i915/display/intel_bios.c b/drivers/gpu/drm/i915/display/intel_bios.c index ba7b8938b17c31eda2be3a73eb199b4f1fb68960..9ce41e689d501301d55be5965984a55d5ff263f3 100644 --- a/drivers/gpu/drm/i915/display/intel_bios.c +++ b/drivers/gpu/drm/i915/display/intel_bios.c @@ -37,6 +37,7 @@ #include "i915_drv.h" #include "intel_display.h" +#include "intel_display_core.h" #include "intel_display_rpm.h" #include "intel_display_types.h" #include "intel_gmbus.h" diff --git a/drivers/gpu/drm/i915/display/intel_bo.c b/drivers/gpu/drm/i915/display/intel_bo.c index fbd16d7b58d954b08a691b4391f5ffa759445cf8..65d64f79a4bd1d41d7a48c251e77b98014f1a7dc 100644 --- a/drivers/gpu/drm/i915/display/intel_bo.c +++ b/drivers/gpu/drm/i915/display/intel_bo.c @@ -1,6 +1,8 @@ // SPDX-License-Identifier: MIT /* Copyright © 2024 Intel Corporation */ +#include +#include "display/intel_display_types.h" #include "gem/i915_gem_mman.h" #include "gem/i915_gem_object.h" #include "gem/i915_gem_object_frontbuffer.h" @@ -57,3 +59,18 @@ void intel_bo_describe(struct seq_file *m, struct drm_gem_object *obj) { i915_debugfs_describe_obj(m, to_intel_bo(obj)); } + +struct intel_framebuffer *intel_bo_alloc_framebuffer(void) +{ + return i915_gem_object_alloc_framebuffer(); +} + +int intel_bo_panic_setup(struct drm_scanout_buffer *sb) +{ + return i915_gem_object_panic_setup(sb); +} + +void intel_bo_panic_finish(struct intel_framebuffer *fb) +{ + return i915_gem_object_panic_finish(fb); +} diff --git a/drivers/gpu/drm/i915/display/intel_bo.h b/drivers/gpu/drm/i915/display/intel_bo.h index ea7a2253aaa57f85d393ff910339c1d4182272ff..97087a64d23b547453abf4d5b3c86d4cd18f060a 100644 --- a/drivers/gpu/drm/i915/display/intel_bo.h +++ b/drivers/gpu/drm/i915/display/intel_bo.h @@ -7,6 +7,8 @@ #include struct drm_gem_object; +struct drm_scanout_buffer; +struct intel_framebuffer; struct seq_file; struct vm_area_struct; @@ -23,5 +25,8 @@ struct intel_frontbuffer *intel_bo_set_frontbuffer(struct drm_gem_object *obj, struct intel_frontbuffer *front); void intel_bo_describe(struct seq_file *m, struct drm_gem_object *obj); +struct intel_framebuffer *intel_bo_alloc_framebuffer(void); +int intel_bo_panic_setup(struct drm_scanout_buffer *sb); +void intel_bo_panic_finish(struct intel_framebuffer *fb); #endif /* __INTEL_BO__ */ diff --git a/drivers/gpu/drm/i915/display/intel_bw.c b/drivers/gpu/drm/i915/display/intel_bw.c index a5dd2932b852d656a70794e91f1eb0dec5ec98e6..d29a755612deff2a95f50989d95b042cc7909cee 100644 --- a/drivers/gpu/drm/i915/display/intel_bw.c +++ b/drivers/gpu/drm/i915/display/intel_bw.c @@ -5,6 +5,8 @@ #include +#include "soc/intel_dram.h" + #include "i915_drv.h" #include "i915_reg.h" #include "i915_utils.h" @@ -12,10 +14,47 @@ #include "intel_bw.h" #include "intel_cdclk.h" #include "intel_display_core.h" +#include "intel_display_regs.h" #include "intel_display_types.h" -#include "skl_watermark.h" #include "intel_mchbar_regs.h" #include "intel_pcode.h" +#include "intel_uncore.h" +#include "skl_watermark.h" + +struct intel_dbuf_bw { + unsigned int max_bw[I915_MAX_DBUF_SLICES]; + u8 active_planes[I915_MAX_DBUF_SLICES]; +}; + +struct intel_bw_state { + struct intel_global_state base; + struct intel_dbuf_bw dbuf_bw[I915_MAX_PIPES]; + + /* + * Contains a bit mask, used to determine, whether correspondent + * pipe allows SAGV or not. + */ + u8 pipe_sagv_reject; + + /* bitmask of active pipes */ + u8 active_pipes; + + /* + * From MTL onwards, to lock a QGV point, punit expects the peak BW of + * the selected QGV point as the parameter in multiples of 100MB/s + */ + u16 qgv_point_peakbw; + + /* + * Current QGV points mask, which restricts + * some particular SAGV states, not to confuse + * with pipe_sagv_mask. + */ + u16 qgv_points_mask; + + unsigned int data_rate[I915_MAX_PIPES]; + u8 num_active_planes[I915_MAX_PIPES]; +}; /* Parameters for Qclk Geyserville (QGV) */ struct intel_qgv_point { @@ -79,14 +118,13 @@ static int icl_pcode_read_qgv_point_info(struct intel_display *display, struct intel_qgv_point *sp, int point) { - struct drm_i915_private *i915 = to_i915(display->drm); u32 val = 0, val2 = 0; u16 dclk; int ret; - ret = snb_pcode_read(&i915->uncore, ICL_PCODE_MEM_SUBSYSYSTEM_INFO | - ICL_PCODE_MEM_SS_READ_QGV_POINT_INFO(point), - &val, &val2); + ret = intel_pcode_read(display->drm, ICL_PCODE_MEM_SUBSYSYSTEM_INFO | + ICL_PCODE_MEM_SS_READ_QGV_POINT_INFO(point), + &val, &val2); if (ret) return ret; @@ -107,13 +145,12 @@ static int icl_pcode_read_qgv_point_info(struct intel_display *display, static int adls_pcode_read_psf_gv_point_info(struct intel_display *display, struct intel_psf_gv_point *points) { - struct drm_i915_private *i915 = to_i915(display->drm); u32 val = 0; int ret; int i; - ret = snb_pcode_read(&i915->uncore, ICL_PCODE_MEM_SUBSYSYSTEM_INFO | - ADL_PCODE_MEM_SS_READ_PSF_GV_INFO, &val, NULL); + ret = intel_pcode_read(display->drm, ICL_PCODE_MEM_SUBSYSYSTEM_INFO | + ADL_PCODE_MEM_SS_READ_PSF_GV_INFO, &val, NULL); if (ret) return ret; @@ -151,21 +188,20 @@ static bool is_sagv_enabled(struct intel_display *display, u16 points_mask) ICL_PCODE_REQ_QGV_PT_MASK); } -int icl_pcode_restrict_qgv_points(struct intel_display *display, - u32 points_mask) +static int icl_pcode_restrict_qgv_points(struct intel_display *display, + u32 points_mask) { - struct drm_i915_private *i915 = to_i915(display->drm); int ret; if (DISPLAY_VER(display) >= 14) return 0; /* bspec says to keep retrying for at least 1 ms */ - ret = skl_pcode_request(&i915->uncore, ICL_PCODE_SAGV_DE_MEM_SS_CONFIG, - points_mask, - ICL_PCODE_REP_QGV_MASK | ADLS_PCODE_REP_PSF_MASK, - ICL_PCODE_REP_QGV_SAFE | ADLS_PCODE_REP_PSF_SAFE, - 1); + ret = intel_pcode_request(display->drm, ICL_PCODE_SAGV_DE_MEM_SS_CONFIG, + points_mask, + ICL_PCODE_REP_QGV_MASK | ADLS_PCODE_REP_PSF_MASK, + ICL_PCODE_REP_QGV_SAFE | ADLS_PCODE_REP_PSF_SAFE, + 1); if (ret < 0) { drm_err(display->drm, @@ -218,11 +254,10 @@ intel_read_qgv_point_info(struct intel_display *display, } static int icl_get_qgv_points(struct intel_display *display, + const struct dram_info *dram_info, struct intel_qgv_info *qi, bool is_y_tile) { - struct drm_i915_private *i915 = to_i915(display->drm); - const struct dram_info *dram_info = &i915->dram_info; int i, ret; qi->num_points = dram_info->num_qgv_points; @@ -418,19 +453,27 @@ static const struct intel_sa_info xe3lpd_sa_info = { .derating = 10, }; -static int icl_get_bw_info(struct intel_display *display, const struct intel_sa_info *sa) +static const struct intel_sa_info xe3lpd_3002_sa_info = { + .deburst = 32, + .deprogbwlimit = 22, /* GB/s */ + .displayrtids = 256, + .derating = 10, +}; + +static int icl_get_bw_info(struct intel_display *display, + const struct dram_info *dram_info, + const struct intel_sa_info *sa) { - struct drm_i915_private *i915 = to_i915(display->drm); struct intel_qgv_info qi = {}; bool is_y_tile = true; /* assume y tile may be used */ - int num_channels = max_t(u8, 1, i915->dram_info.num_channels); + int num_channels = max_t(u8, 1, dram_info->num_channels); int ipqdepth, ipqdepthpch = 16; int dclk_max; int maxdebw; int num_groups = ARRAY_SIZE(display->bw.max); int i, ret; - ret = icl_get_qgv_points(display, &qi, is_y_tile); + ret = icl_get_qgv_points(display, dram_info, &qi, is_y_tile); if (ret) { drm_dbg_kms(display->drm, "Failed to get memory subsystem information, ignoring bandwidth limits"); @@ -488,11 +531,11 @@ static int icl_get_bw_info(struct intel_display *display, const struct intel_sa_ return 0; } -static int tgl_get_bw_info(struct intel_display *display, const struct intel_sa_info *sa) +static int tgl_get_bw_info(struct intel_display *display, + const struct dram_info *dram_info, + const struct intel_sa_info *sa) { - struct drm_i915_private *i915 = to_i915(display->drm); struct intel_qgv_info qi = {}; - const struct dram_info *dram_info = &i915->dram_info; bool is_y_tile = true; /* assume y tile may be used */ int num_channels = max_t(u8, 1, dram_info->num_channels); int ipqdepth, ipqdepthpch = 16; @@ -502,7 +545,7 @@ static int tgl_get_bw_info(struct intel_display *display, const struct intel_sa_ int num_groups = ARRAY_SIZE(display->bw.max); int i, ret; - ret = icl_get_qgv_points(display, &qi, is_y_tile); + ret = icl_get_qgv_points(display, dram_info, &qi, is_y_tile); if (ret) { drm_dbg_kms(display->drm, "Failed to get memory subsystem information, ignoring bandwidth limits"); @@ -632,15 +675,15 @@ static void dg2_get_bw_info(struct intel_display *display) } static int xe2_hpd_get_bw_info(struct intel_display *display, + const struct dram_info *dram_info, const struct intel_sa_info *sa) { - struct drm_i915_private *i915 = to_i915(display->drm); struct intel_qgv_info qi = {}; - int num_channels = i915->dram_info.num_channels; + int num_channels = dram_info->num_channels; int peakbw, maxdebw; int ret, i; - ret = icl_get_qgv_points(display, &qi, true); + ret = icl_get_qgv_points(display, dram_info, &qi, true); if (ret) { drm_dbg_kms(display->drm, "Failed to get memory subsystem information, ignoring bandwidth limits"); @@ -763,32 +806,34 @@ static unsigned int icl_qgv_bw(struct intel_display *display, void intel_bw_init_hw(struct intel_display *display) { - const struct dram_info *dram_info = &to_i915(display->drm)->dram_info; + const struct dram_info *dram_info = intel_dram_info(display->drm); if (!HAS_DISPLAY(display)) return; - if (DISPLAY_VER(display) >= 30) - tgl_get_bw_info(display, &xe3lpd_sa_info); + if (DISPLAY_VERx100(display) >= 3002) + tgl_get_bw_info(display, dram_info, &xe3lpd_3002_sa_info); + else if (DISPLAY_VER(display) >= 30) + tgl_get_bw_info(display, dram_info, &xe3lpd_sa_info); else if (DISPLAY_VERx100(display) >= 1401 && display->platform.dgfx && dram_info->type == INTEL_DRAM_GDDR_ECC) - xe2_hpd_get_bw_info(display, &xe2_hpd_ecc_sa_info); + xe2_hpd_get_bw_info(display, dram_info, &xe2_hpd_ecc_sa_info); else if (DISPLAY_VERx100(display) >= 1401 && display->platform.dgfx) - xe2_hpd_get_bw_info(display, &xe2_hpd_sa_info); + xe2_hpd_get_bw_info(display, dram_info, &xe2_hpd_sa_info); else if (DISPLAY_VER(display) >= 14) - tgl_get_bw_info(display, &mtl_sa_info); + tgl_get_bw_info(display, dram_info, &mtl_sa_info); else if (display->platform.dg2) dg2_get_bw_info(display); else if (display->platform.alderlake_p) - tgl_get_bw_info(display, &adlp_sa_info); + tgl_get_bw_info(display, dram_info, &adlp_sa_info); else if (display->platform.alderlake_s) - tgl_get_bw_info(display, &adls_sa_info); + tgl_get_bw_info(display, dram_info, &adls_sa_info); else if (display->platform.rocketlake) - tgl_get_bw_info(display, &rkl_sa_info); + tgl_get_bw_info(display, dram_info, &rkl_sa_info); else if (DISPLAY_VER(display) == 12) - tgl_get_bw_info(display, &tgl_sa_info); + tgl_get_bw_info(display, dram_info, &tgl_sa_info); else if (DISPLAY_VER(display) == 11) - icl_get_bw_info(display, &icl_sa_info); + icl_get_bw_info(display, dram_info, &icl_sa_info); } static unsigned int intel_bw_crtc_num_active_planes(const struct intel_crtc_state *crtc_state) @@ -862,6 +907,11 @@ static unsigned int intel_bw_data_rate(struct intel_display *display, return data_rate; } +struct intel_bw_state *to_intel_bw_state(struct intel_global_state *obj_state) +{ + return container_of(obj_state, struct intel_bw_state, base); +} + struct intel_bw_state * intel_atomic_get_old_bw_state(struct intel_atomic_state *state) { @@ -971,6 +1021,70 @@ static void icl_force_disable_sagv(struct intel_display *display, icl_pcode_restrict_qgv_points(display, bw_state->qgv_points_mask); } +void icl_sagv_pre_plane_update(struct intel_atomic_state *state) +{ + struct intel_display *display = to_intel_display(state); + const struct intel_bw_state *old_bw_state = + intel_atomic_get_old_bw_state(state); + const struct intel_bw_state *new_bw_state = + intel_atomic_get_new_bw_state(state); + u16 old_mask, new_mask; + + if (!new_bw_state) + return; + + old_mask = old_bw_state->qgv_points_mask; + new_mask = old_bw_state->qgv_points_mask | new_bw_state->qgv_points_mask; + + if (old_mask == new_mask) + return; + + WARN_ON(!new_bw_state->base.changed); + + drm_dbg_kms(display->drm, "Restricting QGV points: 0x%x -> 0x%x\n", + old_mask, new_mask); + + /* + * Restrict required qgv points before updating the configuration. + * According to BSpec we can't mask and unmask qgv points at the same + * time. Also masking should be done before updating the configuration + * and unmasking afterwards. + */ + icl_pcode_restrict_qgv_points(display, new_mask); +} + +void icl_sagv_post_plane_update(struct intel_atomic_state *state) +{ + struct intel_display *display = to_intel_display(state); + const struct intel_bw_state *old_bw_state = + intel_atomic_get_old_bw_state(state); + const struct intel_bw_state *new_bw_state = + intel_atomic_get_new_bw_state(state); + u16 old_mask, new_mask; + + if (!new_bw_state) + return; + + old_mask = old_bw_state->qgv_points_mask | new_bw_state->qgv_points_mask; + new_mask = new_bw_state->qgv_points_mask; + + if (old_mask == new_mask) + return; + + WARN_ON(!new_bw_state->base.changed); + + drm_dbg_kms(display->drm, "Relaxing QGV points: 0x%x -> 0x%x\n", + old_mask, new_mask); + + /* + * Allow required qgv points after updating the configuration. + * According to BSpec we can't mask and unmask qgv points at the same + * time. Also masking should be done before updating the configuration + * and unmasking afterwards. + */ + icl_pcode_restrict_qgv_points(display, new_mask); +} + static int mtl_find_qgv_points(struct intel_display *display, unsigned int data_rate, unsigned int num_active_planes, @@ -991,7 +1105,7 @@ static int mtl_find_qgv_points(struct intel_display *display, * for qgv peak bw in PM Demand request. So assign UINT_MAX if SAGV is * not enabled. PM Demand code will clamp the value for the register */ - if (!intel_can_enable_sagv(display, new_bw_state)) { + if (!intel_bw_can_enable_sagv(display, new_bw_state)) { new_bw_state->qgv_point_peakbw = U16_MAX; drm_dbg_kms(display->drm, "No SAGV, use UINT_MAX as peak bw."); return 0; @@ -1104,7 +1218,7 @@ static int icl_find_qgv_points(struct intel_display *display, * we can't enable SAGV due to the increased memory latency it may * cause. */ - if (!intel_can_enable_sagv(display, new_bw_state)) { + if (!intel_bw_can_enable_sagv(display, new_bw_state)) { qgv_points = icl_max_bw_qgv_point_mask(display, num_active_planes); drm_dbg_kms(display->drm, "No SAGV, using single QGV point mask 0x%x\n", qgv_points); @@ -1354,12 +1468,12 @@ int intel_bw_calc_min_cdclk(struct intel_atomic_state *state, * requirements. This can reduce back and forth * display blinking due to constant cdclk changes. */ - if (new_min_cdclk <= cdclk_state->bw_min_cdclk) + if (new_min_cdclk <= intel_cdclk_bw_min_cdclk(cdclk_state)) return 0; drm_dbg_kms(display->drm, "new bandwidth min cdclk (%d kHz) > old min cdclk (%d kHz)\n", - new_min_cdclk, cdclk_state->bw_min_cdclk); + new_min_cdclk, intel_cdclk_bw_min_cdclk(cdclk_state)); *need_cdclk_calc = true; return 0; @@ -1471,8 +1585,8 @@ static int intel_bw_check_sagv_mask(struct intel_atomic_state *state) if (!new_bw_state) return 0; - if (intel_can_enable_sagv(display, new_bw_state) != - intel_can_enable_sagv(display, old_bw_state)) { + if (intel_bw_can_enable_sagv(display, new_bw_state) != + intel_bw_can_enable_sagv(display, old_bw_state)) { ret = intel_atomic_serialize_global_state(&new_bw_state->base); if (ret) return ret; @@ -1518,8 +1632,8 @@ int intel_bw_atomic_check(struct intel_atomic_state *state, bool any_ms) new_bw_state = intel_atomic_get_new_bw_state(state); if (new_bw_state && - intel_can_enable_sagv(display, old_bw_state) != - intel_can_enable_sagv(display, new_bw_state)) + intel_bw_can_enable_sagv(display, old_bw_state) != + intel_bw_can_enable_sagv(display, new_bw_state)) changed = true; /* @@ -1641,3 +1755,32 @@ int intel_bw_init(struct intel_display *display) return 0; } + +bool intel_bw_pmdemand_needs_update(struct intel_atomic_state *state) +{ + const struct intel_bw_state *new_bw_state, *old_bw_state; + + new_bw_state = intel_atomic_get_new_bw_state(state); + old_bw_state = intel_atomic_get_old_bw_state(state); + + if (new_bw_state && + new_bw_state->qgv_point_peakbw != old_bw_state->qgv_point_peakbw) + return true; + + return false; +} + +bool intel_bw_can_enable_sagv(struct intel_display *display, + const struct intel_bw_state *bw_state) +{ + if (DISPLAY_VER(display) < 11 && + bw_state->active_pipes && !is_power_of_2(bw_state->active_pipes)) + return false; + + return bw_state->pipe_sagv_reject == 0; +} + +int intel_bw_qgv_point_peakbw(const struct intel_bw_state *bw_state) +{ + return bw_state->qgv_point_peakbw; +} diff --git a/drivers/gpu/drm/i915/display/intel_bw.h b/drivers/gpu/drm/i915/display/intel_bw.h index eb2cc883e9c1e3a1b34a641faf822e8559b7a1da..d51f50c9d30219248e5799851e3d32a41e177bd2 100644 --- a/drivers/gpu/drm/i915/display/intel_bw.h +++ b/drivers/gpu/drm/i915/display/intel_bw.h @@ -8,52 +8,14 @@ #include -#include "intel_display_limits.h" -#include "intel_display_power.h" -#include "intel_global_state.h" - struct intel_atomic_state; +struct intel_bw_state; struct intel_crtc; struct intel_crtc_state; struct intel_display; +struct intel_global_state; -struct intel_dbuf_bw { - unsigned int max_bw[I915_MAX_DBUF_SLICES]; - u8 active_planes[I915_MAX_DBUF_SLICES]; -}; - -struct intel_bw_state { - struct intel_global_state base; - struct intel_dbuf_bw dbuf_bw[I915_MAX_PIPES]; - - /* - * Contains a bit mask, used to determine, whether correspondent - * pipe allows SAGV or not. - */ - u8 pipe_sagv_reject; - - /* bitmask of active pipes */ - u8 active_pipes; - - /* - * From MTL onwards, to lock a QGV point, punit expects the peak BW of - * the selected QGV point as the parameter in multiples of 100MB/s - */ - u16 qgv_point_peakbw; - - /* - * Current QGV points mask, which restricts - * some particular SAGV states, not to confuse - * with pipe_sagv_mask. - */ - u16 qgv_points_mask; - - unsigned int data_rate[I915_MAX_PIPES]; - u8 num_active_planes[I915_MAX_PIPES]; -}; - -#define to_intel_bw_state(global_state) \ - container_of_const((global_state), struct intel_bw_state, base) +struct intel_bw_state *to_intel_bw_state(struct intel_global_state *obj_state); struct intel_bw_state * intel_atomic_get_old_bw_state(struct intel_atomic_state *state); @@ -67,8 +29,6 @@ intel_atomic_get_bw_state(struct intel_atomic_state *state); void intel_bw_init_hw(struct intel_display *display); int intel_bw_init(struct intel_display *display); int intel_bw_atomic_check(struct intel_atomic_state *state, bool any_ms); -int icl_pcode_restrict_qgv_points(struct intel_display *display, - u32 points_mask); int intel_bw_calc_min_cdclk(struct intel_atomic_state *state, bool *need_cdclk_calc); int intel_bw_min_cdclk(struct intel_display *display, @@ -76,4 +36,11 @@ int intel_bw_min_cdclk(struct intel_display *display, void intel_bw_update_hw_state(struct intel_display *display); void intel_bw_crtc_disable_noatomic(struct intel_crtc *crtc); +bool intel_bw_pmdemand_needs_update(struct intel_atomic_state *state); +bool intel_bw_can_enable_sagv(struct intel_display *display, + const struct intel_bw_state *bw_state); +void icl_sagv_pre_plane_update(struct intel_atomic_state *state); +void icl_sagv_post_plane_update(struct intel_atomic_state *state); +int intel_bw_qgv_point_peakbw(const struct intel_bw_state *bw_state); + #endif /* __INTEL_BW_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_cdclk.c b/drivers/gpu/drm/i915/display/intel_cdclk.c index b1718b491ffdab596f8313d518fa048243305afb..228aa64c1349be86af87578b7f93bcbe51efd46e 100644 --- a/drivers/gpu/drm/i915/display/intel_cdclk.c +++ b/drivers/gpu/drm/i915/display/intel_cdclk.c @@ -32,16 +32,17 @@ #include "i915_drv.h" #include "i915_reg.h" #include "intel_atomic.h" -#include "intel_atomic_plane.h" #include "intel_audio.h" #include "intel_bw.h" #include "intel_cdclk.h" #include "intel_crtc.h" #include "intel_de.h" +#include "intel_display_regs.h" #include "intel_display_types.h" #include "intel_mchbar_regs.h" #include "intel_pci_config.h" #include "intel_pcode.h" +#include "intel_plane.h" #include "intel_psr.h" #include "intel_vdsc.h" #include "skl_watermark.h" @@ -113,6 +114,42 @@ * dividers can be programmed correctly. */ +struct intel_cdclk_state { + struct intel_global_state base; + + /* + * Logical configuration of cdclk (used for all scaling, + * watermark, etc. calculations and checks). This is + * computed as if all enabled crtcs were active. + */ + struct intel_cdclk_config logical; + + /* + * Actual configuration of cdclk, can be different from the + * logical configuration only when all crtc's are DPMS off. + */ + struct intel_cdclk_config actual; + + /* minimum acceptable cdclk to satisfy bandwidth requirements */ + int bw_min_cdclk; + /* minimum acceptable cdclk for each pipe */ + int min_cdclk[I915_MAX_PIPES]; + /* minimum acceptable voltage level for each pipe */ + u8 min_voltage_level[I915_MAX_PIPES]; + + /* pipe to which cd2x update is synchronized */ + enum pipe pipe; + + /* forced minimum cdclk for glk+ audio w/a */ + int force_min_cdclk; + + /* bitmask of active pipes */ + u8 active_pipes; + + /* update cdclk with pipes disabled */ + bool disable_pipes; +}; + struct intel_cdclk_funcs { void (*get_cdclk)(struct intel_display *display, struct intel_cdclk_config *cdclk_config); @@ -567,20 +604,18 @@ static u8 vlv_calc_voltage_level(struct intel_display *display, int cdclk) static void vlv_get_cdclk(struct intel_display *display, struct intel_cdclk_config *cdclk_config) { - struct drm_i915_private *dev_priv = to_i915(display->drm); u32 val; - vlv_iosf_sb_get(dev_priv, - BIT(VLV_IOSF_SB_CCK) | BIT(VLV_IOSF_SB_PUNIT)); + vlv_iosf_sb_get(display->drm, BIT(VLV_IOSF_SB_CCK) | BIT(VLV_IOSF_SB_PUNIT)); - cdclk_config->vco = vlv_get_hpll_vco(dev_priv); - cdclk_config->cdclk = vlv_get_cck_clock(dev_priv, "cdclk", + cdclk_config->vco = vlv_get_hpll_vco(display->drm); + cdclk_config->cdclk = vlv_get_cck_clock(display->drm, "cdclk", CCK_DISPLAY_CLOCK_CONTROL, cdclk_config->vco); - val = vlv_punit_read(dev_priv, PUNIT_REG_DSPSSPM); + val = vlv_punit_read(display->drm, PUNIT_REG_DSPSSPM); - vlv_iosf_sb_put(dev_priv, + vlv_iosf_sb_put(display->drm, BIT(VLV_IOSF_SB_CCK) | BIT(VLV_IOSF_SB_PUNIT)); if (display->platform.valleyview) @@ -658,16 +693,16 @@ static void vlv_set_cdclk(struct intel_display *display, */ wakeref = intel_display_power_get(display, POWER_DOMAIN_DISPLAY_CORE); - vlv_iosf_sb_get(dev_priv, + vlv_iosf_sb_get(display->drm, BIT(VLV_IOSF_SB_CCK) | BIT(VLV_IOSF_SB_BUNIT) | BIT(VLV_IOSF_SB_PUNIT)); - val = vlv_punit_read(dev_priv, PUNIT_REG_DSPSSPM); + val = vlv_punit_read(display->drm, PUNIT_REG_DSPSSPM); val &= ~DSPFREQGUAR_MASK; val |= (cmd << DSPFREQGUAR_SHIFT); - vlv_punit_write(dev_priv, PUNIT_REG_DSPSSPM, val); - if (wait_for((vlv_punit_read(dev_priv, PUNIT_REG_DSPSSPM) & + vlv_punit_write(display->drm, PUNIT_REG_DSPSSPM, val); + if (wait_for((vlv_punit_read(display->drm, PUNIT_REG_DSPSSPM) & DSPFREQSTAT_MASK) == (cmd << DSPFREQSTAT_SHIFT), 50)) { drm_err(display->drm, @@ -681,12 +716,12 @@ static void vlv_set_cdclk(struct intel_display *display, cdclk) - 1; /* adjust cdclk divider */ - val = vlv_cck_read(dev_priv, CCK_DISPLAY_CLOCK_CONTROL); + val = vlv_cck_read(display->drm, CCK_DISPLAY_CLOCK_CONTROL); val &= ~CCK_FREQUENCY_VALUES; val |= divider; - vlv_cck_write(dev_priv, CCK_DISPLAY_CLOCK_CONTROL, val); + vlv_cck_write(display->drm, CCK_DISPLAY_CLOCK_CONTROL, val); - if (wait_for((vlv_cck_read(dev_priv, CCK_DISPLAY_CLOCK_CONTROL) & + if (wait_for((vlv_cck_read(display->drm, CCK_DISPLAY_CLOCK_CONTROL) & CCK_FREQUENCY_STATUS) == (divider << CCK_FREQUENCY_STATUS_SHIFT), 50)) drm_err(display->drm, @@ -694,7 +729,7 @@ static void vlv_set_cdclk(struct intel_display *display, } /* adjust self-refresh exit latency value */ - val = vlv_bunit_read(dev_priv, BUNIT_REG_BISOC); + val = vlv_bunit_read(display->drm, BUNIT_REG_BISOC); val &= ~0x7f; /* @@ -705,9 +740,9 @@ static void vlv_set_cdclk(struct intel_display *display, val |= 4500 / 250; /* 4.5 usec */ else val |= 3000 / 250; /* 3.0 usec */ - vlv_bunit_write(dev_priv, BUNIT_REG_BISOC, val); + vlv_bunit_write(display->drm, BUNIT_REG_BISOC, val); - vlv_iosf_sb_put(dev_priv, + vlv_iosf_sb_put(display->drm, BIT(VLV_IOSF_SB_CCK) | BIT(VLV_IOSF_SB_BUNIT) | BIT(VLV_IOSF_SB_PUNIT)); @@ -723,7 +758,6 @@ static void chv_set_cdclk(struct intel_display *display, const struct intel_cdclk_config *cdclk_config, enum pipe pipe) { - struct drm_i915_private *dev_priv = to_i915(display->drm); int cdclk = cdclk_config->cdclk; u32 val, cmd = cdclk_config->voltage_level; intel_wakeref_t wakeref; @@ -747,19 +781,19 @@ static void chv_set_cdclk(struct intel_display *display, */ wakeref = intel_display_power_get(display, POWER_DOMAIN_DISPLAY_CORE); - vlv_punit_get(dev_priv); - val = vlv_punit_read(dev_priv, PUNIT_REG_DSPSSPM); + vlv_punit_get(display->drm); + val = vlv_punit_read(display->drm, PUNIT_REG_DSPSSPM); val &= ~DSPFREQGUAR_MASK_CHV; val |= (cmd << DSPFREQGUAR_SHIFT_CHV); - vlv_punit_write(dev_priv, PUNIT_REG_DSPSSPM, val); - if (wait_for((vlv_punit_read(dev_priv, PUNIT_REG_DSPSSPM) & + vlv_punit_write(display->drm, PUNIT_REG_DSPSSPM, val); + if (wait_for((vlv_punit_read(display->drm, PUNIT_REG_DSPSSPM) & DSPFREQSTAT_MASK_CHV) == (cmd << DSPFREQSTAT_SHIFT_CHV), 50)) { drm_err(display->drm, "timed out waiting for CDclk change\n"); } - vlv_punit_put(dev_priv); + vlv_punit_put(display->drm); intel_update_cdclk(display); @@ -843,7 +877,6 @@ static void bdw_set_cdclk(struct intel_display *display, const struct intel_cdclk_config *cdclk_config, enum pipe pipe) { - struct drm_i915_private *dev_priv = to_i915(display->drm); int cdclk = cdclk_config->cdclk; int ret; @@ -856,7 +889,7 @@ static void bdw_set_cdclk(struct intel_display *display, "trying to change cdclk frequency with cdclk not enabled\n")) return; - ret = snb_pcode_write(&dev_priv->uncore, BDW_PCODE_DISPLAY_FREQ_CHANGE_REQ, 0x0); + ret = intel_pcode_write(display->drm, BDW_PCODE_DISPLAY_FREQ_CHANGE_REQ, 0x0); if (ret) { drm_err(display->drm, "failed to inform pcode about cdclk change\n"); @@ -884,8 +917,8 @@ static void bdw_set_cdclk(struct intel_display *display, LCPLL_CD_SOURCE_FCLK_DONE) == 0, 1)) drm_err(display->drm, "Switching back to LCPLL failed\n"); - snb_pcode_write(&dev_priv->uncore, HSW_PCODE_DE_WRITE_FREQ_REQ, - cdclk_config->voltage_level); + intel_pcode_write(display->drm, HSW_PCODE_DE_WRITE_FREQ_REQ, + cdclk_config->voltage_level); intel_de_write(display, CDCLK_FREQ, DIV_ROUND_CLOSEST(cdclk, 1000) - 1); @@ -1125,7 +1158,6 @@ static void skl_set_cdclk(struct intel_display *display, const struct intel_cdclk_config *cdclk_config, enum pipe pipe) { - struct drm_i915_private *dev_priv = to_i915(display->drm); int cdclk = cdclk_config->cdclk; int vco = cdclk_config->vco; u32 freq_select, cdclk_ctl; @@ -1142,10 +1174,10 @@ static void skl_set_cdclk(struct intel_display *display, drm_WARN_ON_ONCE(display->drm, display->platform.skylake && vco == 8640000); - ret = skl_pcode_request(&dev_priv->uncore, SKL_PCODE_CDCLK_CONTROL, - SKL_CDCLK_PREPARE_FOR_CHANGE, - SKL_CDCLK_READY_FOR_CHANGE, - SKL_CDCLK_READY_FOR_CHANGE, 3); + ret = intel_pcode_request(display->drm, SKL_PCODE_CDCLK_CONTROL, + SKL_CDCLK_PREPARE_FOR_CHANGE, + SKL_CDCLK_READY_FOR_CHANGE, + SKL_CDCLK_READY_FOR_CHANGE, 3); if (ret) { drm_err(display->drm, "Failed to inform PCU about cdclk change (%d)\n", ret); @@ -1188,8 +1220,8 @@ static void skl_set_cdclk(struct intel_display *display, intel_de_posting_read(display, CDCLK_CTL); /* inform PCU of the change */ - snb_pcode_write(&dev_priv->uncore, SKL_PCODE_CDCLK_CONTROL, - cdclk_config->voltage_level); + intel_pcode_write(display->drm, SKL_PCODE_CDCLK_CONTROL, + cdclk_config->voltage_level); intel_update_cdclk(display); } @@ -2125,7 +2157,6 @@ static void bxt_set_cdclk(struct intel_display *display, const struct intel_cdclk_config *cdclk_config, enum pipe pipe) { - struct drm_i915_private *dev_priv = to_i915(display->drm); struct intel_cdclk_config mid_cdclk_config; int cdclk = cdclk_config->cdclk; int ret = 0; @@ -2139,18 +2170,18 @@ static void bxt_set_cdclk(struct intel_display *display, if (DISPLAY_VER(display) >= 14 || display->platform.dg2) ; /* NOOP */ else if (DISPLAY_VER(display) >= 11) - ret = skl_pcode_request(&dev_priv->uncore, SKL_PCODE_CDCLK_CONTROL, - SKL_CDCLK_PREPARE_FOR_CHANGE, - SKL_CDCLK_READY_FOR_CHANGE, - SKL_CDCLK_READY_FOR_CHANGE, 3); + ret = intel_pcode_request(display->drm, SKL_PCODE_CDCLK_CONTROL, + SKL_CDCLK_PREPARE_FOR_CHANGE, + SKL_CDCLK_READY_FOR_CHANGE, + SKL_CDCLK_READY_FOR_CHANGE, 3); else /* * BSpec requires us to wait up to 150usec, but that leads to * timeouts; the 2ms used here is based on experiment. */ - ret = snb_pcode_write_timeout(&dev_priv->uncore, - HSW_PCODE_DE_WRITE_FREQ_REQ, - 0x80000000, 150, 2); + ret = intel_pcode_write_timeout(display->drm, + HSW_PCODE_DE_WRITE_FREQ_REQ, + 0x80000000, 2); if (ret) { drm_err(display->drm, @@ -2179,8 +2210,8 @@ static void bxt_set_cdclk(struct intel_display *display, * Display versions 14 and beyond */; else if (DISPLAY_VER(display) >= 11 && !display->platform.dg2) - ret = snb_pcode_write(&dev_priv->uncore, SKL_PCODE_CDCLK_CONTROL, - cdclk_config->voltage_level); + ret = intel_pcode_write(display->drm, SKL_PCODE_CDCLK_CONTROL, + cdclk_config->voltage_level); if (DISPLAY_VER(display) < 11) { /* * The timeout isn't specified, the 2ms used here is based on @@ -2188,10 +2219,9 @@ static void bxt_set_cdclk(struct intel_display *display, * FIXME: Waiting for the request completion could be delayed * until the next PCODE request based on BSpec. */ - ret = snb_pcode_write_timeout(&dev_priv->uncore, - HSW_PCODE_DE_WRITE_FREQ_REQ, - cdclk_config->voltage_level, - 150, 2); + ret = intel_pcode_write_timeout(display->drm, + HSW_PCODE_DE_WRITE_FREQ_REQ, + cdclk_config->voltage_level, 2); } if (ret) { drm_err(display->drm, @@ -2477,7 +2507,6 @@ static void intel_pcode_notify(struct intel_display *display, bool cdclk_update_valid, bool pipe_count_update_valid) { - struct drm_i915_private *i915 = to_i915(display->drm); int ret; u32 update_mask = 0; @@ -2492,11 +2521,11 @@ static void intel_pcode_notify(struct intel_display *display, if (pipe_count_update_valid) update_mask |= DISPLAY_TO_PCODE_PIPE_COUNT_VALID; - ret = skl_pcode_request(&i915->uncore, SKL_PCODE_CDCLK_CONTROL, - SKL_CDCLK_PREPARE_FOR_CHANGE | - update_mask, - SKL_CDCLK_READY_FOR_CHANGE, - SKL_CDCLK_READY_FOR_CHANGE, 3); + ret = intel_pcode_request(display->drm, SKL_PCODE_CDCLK_CONTROL, + SKL_CDCLK_PREPARE_FOR_CHANGE | + update_mask, + SKL_CDCLK_READY_FOR_CHANGE, + SKL_CDCLK_READY_FOR_CHANGE, 3); if (ret) drm_err(display->drm, "Failed to inform PCU about display config (err %d)\n", @@ -3388,7 +3417,9 @@ static int intel_compute_max_dotclk(struct intel_display *display) */ void intel_update_max_cdclk(struct intel_display *display) { - if (DISPLAY_VER(display) >= 30) { + if (DISPLAY_VERx100(display) >= 3002) { + display->cdclk.max_cdclk_freq = 480000; + } else if (DISPLAY_VER(display) >= 30) { display->cdclk.max_cdclk_freq = 691200; } else if (display->platform.jasperlake || display->platform.elkhartlake) { if (display->cdclk.hw.ref == 24000) @@ -3528,10 +3559,8 @@ static int pch_rawclk(struct intel_display *display) static int vlv_hrawclk(struct intel_display *display) { - struct drm_i915_private *dev_priv = to_i915(display->drm); - /* RAWCLK_FREQ_VLV register updated from power well code */ - return vlv_get_cck_clock_hpll(dev_priv, "hrawclk", + return vlv_get_cck_clock_hpll(display->drm, "hrawclk", CCK_DISPLAY_REF_CLOCK_CONTROL); } @@ -3841,3 +3870,60 @@ void intel_init_cdclk_hooks(struct intel_display *display) "Unknown platform. Assuming i830\n")) display->funcs.cdclk = &i830_cdclk_funcs; } + +int intel_cdclk_logical(const struct intel_cdclk_state *cdclk_state) +{ + return cdclk_state->logical.cdclk; +} + +int intel_cdclk_actual(const struct intel_cdclk_state *cdclk_state) +{ + return cdclk_state->actual.cdclk; +} + +int intel_cdclk_actual_voltage_level(const struct intel_cdclk_state *cdclk_state) +{ + return cdclk_state->actual.voltage_level; +} + +int intel_cdclk_min_cdclk(const struct intel_cdclk_state *cdclk_state, enum pipe pipe) +{ + return cdclk_state->min_cdclk[pipe]; +} + +int intel_cdclk_bw_min_cdclk(const struct intel_cdclk_state *cdclk_state) +{ + return cdclk_state->bw_min_cdclk; +} + +bool intel_cdclk_pmdemand_needs_update(struct intel_atomic_state *state) +{ + const struct intel_cdclk_state *new_cdclk_state, *old_cdclk_state; + + new_cdclk_state = intel_atomic_get_new_cdclk_state(state); + old_cdclk_state = intel_atomic_get_old_cdclk_state(state); + + if (new_cdclk_state && + (new_cdclk_state->actual.cdclk != old_cdclk_state->actual.cdclk || + new_cdclk_state->actual.voltage_level != old_cdclk_state->actual.voltage_level)) + return true; + + return false; +} + +void intel_cdclk_force_min_cdclk(struct intel_cdclk_state *cdclk_state, int force_min_cdclk) +{ + cdclk_state->force_min_cdclk = force_min_cdclk; +} + +void intel_cdclk_read_hw(struct intel_display *display) +{ + struct intel_cdclk_state *cdclk_state; + + cdclk_state = to_intel_cdclk_state(display->cdclk.obj.state); + + intel_update_cdclk(display); + intel_cdclk_dump_config(display, &display->cdclk.hw, "Current CDCLK"); + cdclk_state->actual = display->cdclk.hw; + cdclk_state->logical = display->cdclk.hw; +} diff --git a/drivers/gpu/drm/i915/display/intel_cdclk.h b/drivers/gpu/drm/i915/display/intel_cdclk.h index a1cefd455d92a8b4d3f57ceb2e2f986761319527..cacee598af0e2a8d0c379b011f65dcc610a7b9d7 100644 --- a/drivers/gpu/drm/i915/display/intel_cdclk.h +++ b/drivers/gpu/drm/i915/display/intel_cdclk.h @@ -8,10 +8,9 @@ #include -#include "intel_display_limits.h" -#include "intel_global_state.h" - +enum pipe; struct intel_atomic_state; +struct intel_cdclk_state; struct intel_crtc; struct intel_crtc_state; struct intel_display; @@ -23,42 +22,6 @@ struct intel_cdclk_config { bool joined_mbus; }; -struct intel_cdclk_state { - struct intel_global_state base; - - /* - * Logical configuration of cdclk (used for all scaling, - * watermark, etc. calculations and checks). This is - * computed as if all enabled crtcs were active. - */ - struct intel_cdclk_config logical; - - /* - * Actual configuration of cdclk, can be different from the - * logical configuration only when all crtc's are DPMS off. - */ - struct intel_cdclk_config actual; - - /* minimum acceptable cdclk to satisfy bandwidth requirements */ - int bw_min_cdclk; - /* minimum acceptable cdclk for each pipe */ - int min_cdclk[I915_MAX_PIPES]; - /* minimum acceptable voltage level for each pipe */ - u8 min_voltage_level[I915_MAX_PIPES]; - - /* pipe to which cd2x update is synchronized */ - enum pipe pipe; - - /* forced minimum cdclk for glk+ audio w/a */ - int force_min_cdclk; - - /* bitmask of active pipes */ - u8 active_pipes; - - /* update cdclk with pipes disabled */ - bool disable_pipes; -}; - void intel_cdclk_init_hw(struct intel_display *display); void intel_cdclk_uninit_hw(struct intel_display *display); void intel_init_cdclk_hooks(struct intel_display *display); @@ -97,4 +60,13 @@ void intel_cdclk_crtc_disable_noatomic(struct intel_crtc *crtc); int intel_cdclk_init(struct intel_display *display); void intel_cdclk_debugfs_register(struct intel_display *display); +int intel_cdclk_logical(const struct intel_cdclk_state *cdclk_state); +int intel_cdclk_actual(const struct intel_cdclk_state *cdclk_state); +int intel_cdclk_actual_voltage_level(const struct intel_cdclk_state *cdclk_state); +int intel_cdclk_min_cdclk(const struct intel_cdclk_state *cdclk_state, enum pipe pipe); +int intel_cdclk_bw_min_cdclk(const struct intel_cdclk_state *cdclk_state); +bool intel_cdclk_pmdemand_needs_update(struct intel_atomic_state *state); +void intel_cdclk_force_min_cdclk(struct intel_cdclk_state *cdclk_state, int force_min_cdclk); +void intel_cdclk_read_hw(struct intel_display *display); + #endif /* __INTEL_CDCLK_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_cmtg.c b/drivers/gpu/drm/i915/display/intel_cmtg.c index 82606ebae1de90851b9cd22616de413e9686bdd1..165138b95cb2fe774bff1e38e7c464b090cb0334 100644 --- a/drivers/gpu/drm/i915/display/intel_cmtg.c +++ b/drivers/gpu/drm/i915/display/intel_cmtg.c @@ -9,13 +9,13 @@ #include #include -#include "i915_reg.h" -#include "intel_crtc.h" #include "intel_cmtg.h" #include "intel_cmtg_regs.h" +#include "intel_crtc.h" #include "intel_de.h" #include "intel_display_device.h" #include "intel_display_power.h" +#include "intel_display_regs.h" /** * DOC: Common Primary Timing Generator (CMTG) diff --git a/drivers/gpu/drm/i915/display/intel_cmtg_regs.h b/drivers/gpu/drm/i915/display/intel_cmtg_regs.h index 668e41d65e862f1255795578281b9d6d14bb3542..945a355782841efd90018470c86f94eeba4043ef 100644 --- a/drivers/gpu/drm/i915/display/intel_cmtg_regs.h +++ b/drivers/gpu/drm/i915/display/intel_cmtg_regs.h @@ -6,7 +6,7 @@ #ifndef __INTEL_CMTG_REGS_H__ #define __INTEL_CMTG_REGS_H__ -#include "i915_reg_defs.h" +#include "intel_display_reg_defs.h" #define CMTG_CLK_SEL _MMIO(0x46160) #define CMTG_CLK_SEL_A_MASK REG_GENMASK(31, 29) diff --git a/drivers/gpu/drm/i915/display/intel_color.c b/drivers/gpu/drm/i915/display/intel_color.c index 98dddf72c0eb9749d5b3c4a90edb73216615bbfd..671db6926e4c89310eb2d64abea914b1d7f93eac 100644 --- a/drivers/gpu/drm/i915/display/intel_color.c +++ b/drivers/gpu/drm/i915/display/intel_color.c @@ -1339,8 +1339,8 @@ static void ilk_lut_write(const struct intel_crtc_state *crtc_state, { struct intel_display *display = to_intel_display(crtc_state); - if (crtc_state->dsb_color_vblank) - intel_dsb_reg_write(crtc_state->dsb_color_vblank, reg, val); + if (crtc_state->dsb_color) + intel_dsb_reg_write(crtc_state->dsb_color, reg, val); else intel_de_write_fw(display, reg, val); } @@ -1350,8 +1350,8 @@ static void ilk_lut_write_indexed(const struct intel_crtc_state *crtc_state, { struct intel_display *display = to_intel_display(crtc_state); - if (crtc_state->dsb_color_vblank) - intel_dsb_reg_write_indexed(crtc_state->dsb_color_vblank, reg, val); + if (crtc_state->dsb_color) + intel_dsb_reg_write_indexed(crtc_state->dsb_color, reg, val); else intel_de_write_fw(display, reg, val); } @@ -1389,7 +1389,7 @@ static void ilk_load_lut_8(const struct intel_crtc_state *crtc_state, for (i = 0; i < 256; i++) { ilk_lut_write(crtc_state, LGC_PALETTE(pipe, i), i9xx_lut_8(&lut[i])); - if (crtc_state->dsb_color_vblank) + if (crtc_state->dsb_color) ilk_lut_write(crtc_state, LGC_PALETTE(pipe, i), i9xx_lut_8(&lut[i])); } @@ -1917,7 +1917,7 @@ void intel_color_load_luts(const struct intel_crtc_state *crtc_state) { struct intel_display *display = to_intel_display(crtc_state); - if (crtc_state->dsb_color_vblank) + if (crtc_state->dsb_color) return; display->funcs.color->load_luts(crtc_state); @@ -1965,6 +1965,25 @@ void intel_color_modeset(const struct intel_crtc_state *crtc_state) } } +bool intel_color_uses_dsb(const struct intel_crtc_state *crtc_state) +{ + return crtc_state->dsb_color; +} + +bool intel_color_uses_chained_dsb(const struct intel_crtc_state *crtc_state) +{ + struct intel_display *display = to_intel_display(crtc_state); + + return crtc_state->dsb_color && !HAS_DOUBLE_BUFFERED_LUT(display); +} + +bool intel_color_uses_gosub_dsb(const struct intel_crtc_state *crtc_state) +{ + struct intel_display *display = to_intel_display(crtc_state); + + return crtc_state->dsb_color && HAS_DOUBLE_BUFFERED_LUT(display); +} + void intel_color_prepare_commit(struct intel_atomic_state *state, struct intel_crtc *crtc) { @@ -1982,47 +2001,53 @@ void intel_color_prepare_commit(struct intel_atomic_state *state, if (!crtc_state->pre_csc_lut && !crtc_state->post_csc_lut) return; - crtc_state->dsb_color_vblank = intel_dsb_prepare(state, crtc, INTEL_DSB_1, 1024); - if (!crtc_state->dsb_color_vblank) + if (HAS_DOUBLE_BUFFERED_LUT(display)) + crtc_state->dsb_color = intel_dsb_prepare(state, crtc, INTEL_DSB_0, 1024); + else + crtc_state->dsb_color = intel_dsb_prepare(state, crtc, INTEL_DSB_1, 1024); + + if (!intel_color_uses_dsb(crtc_state)) return; display->funcs.color->load_luts(crtc_state); - if (crtc_state->use_dsb) { - intel_vrr_send_push(crtc_state->dsb_color_vblank, crtc_state); - intel_dsb_wait_vblank_delay(state, crtc_state->dsb_color_vblank); - intel_vrr_check_push_sent(crtc_state->dsb_color_vblank, crtc_state); - intel_dsb_interrupt(crtc_state->dsb_color_vblank); + if (crtc_state->use_dsb && intel_color_uses_chained_dsb(crtc_state)) { + intel_vrr_send_push(crtc_state->dsb_color, crtc_state); + intel_dsb_wait_vblank_delay(state, crtc_state->dsb_color); + intel_vrr_check_push_sent(crtc_state->dsb_color, crtc_state); + intel_dsb_interrupt(crtc_state->dsb_color); } - intel_dsb_finish(crtc_state->dsb_color_vblank); + if (intel_color_uses_gosub_dsb(crtc_state)) + intel_dsb_gosub_finish(crtc_state->dsb_color); + else + intel_dsb_finish(crtc_state->dsb_color); } void intel_color_cleanup_commit(struct intel_crtc_state *crtc_state) { - if (crtc_state->dsb_color_vblank) { - intel_dsb_cleanup(crtc_state->dsb_color_vblank); - crtc_state->dsb_color_vblank = NULL; + if (crtc_state->dsb_color) { + intel_dsb_cleanup(crtc_state->dsb_color); + crtc_state->dsb_color = NULL; } } void intel_color_wait_commit(const struct intel_crtc_state *crtc_state) { - if (crtc_state->dsb_color_vblank) - intel_dsb_wait(crtc_state->dsb_color_vblank); -} - -bool intel_color_uses_dsb(const struct intel_crtc_state *crtc_state) -{ - return crtc_state->dsb_color_vblank; + if (crtc_state->dsb_color) + intel_dsb_wait(crtc_state->dsb_color); } static bool intel_can_preload_luts(struct intel_atomic_state *state, struct intel_crtc *crtc) { + struct intel_display *display = to_intel_display(state); const struct intel_crtc_state *old_crtc_state = intel_atomic_get_old_crtc_state(state, crtc); + if (HAS_DOUBLE_BUFFERED_LUT(display)) + return false; + return !old_crtc_state->post_csc_lut && !old_crtc_state->pre_csc_lut; } diff --git a/drivers/gpu/drm/i915/display/intel_color.h b/drivers/gpu/drm/i915/display/intel_color.h index 9d66457c1e893b572aaf8ee4e139105b37c23f6f..bf7a12ce9df0f809dbfa5f73f1dd8d510f90e4f9 100644 --- a/drivers/gpu/drm/i915/display/intel_color.h +++ b/drivers/gpu/drm/i915/display/intel_color.h @@ -24,6 +24,8 @@ void intel_color_prepare_commit(struct intel_atomic_state *state, struct intel_crtc *crtc); void intel_color_cleanup_commit(struct intel_crtc_state *crtc_state); bool intel_color_uses_dsb(const struct intel_crtc_state *crtc_state); +bool intel_color_uses_chained_dsb(const struct intel_crtc_state *crtc_state); +bool intel_color_uses_gosub_dsb(const struct intel_crtc_state *crtc_state); void intel_color_wait_commit(const struct intel_crtc_state *crtc_state); void intel_color_commit_noarm(struct intel_dsb *dsb, const struct intel_crtc_state *crtc_state); diff --git a/drivers/gpu/drm/i915/display/intel_combo_phy.c b/drivers/gpu/drm/i915/display/intel_combo_phy.c index f5cc38dbe5590730029708dd79d6c7ff8113e7cc..112749f97c262ac25ba5b8a2c83c11b16a96afca 100644 --- a/drivers/gpu/drm/i915/display/intel_combo_phy.c +++ b/drivers/gpu/drm/i915/display/intel_combo_phy.c @@ -5,11 +5,11 @@ #include -#include "i915_reg.h" #include "i915_utils.h" #include "intel_combo_phy.h" #include "intel_combo_phy_regs.h" #include "intel_de.h" +#include "intel_display_regs.h" #include "intel_display_types.h" #define for_each_combo_phy(__display, __phy) \ diff --git a/drivers/gpu/drm/i915/display/intel_combo_phy_regs.h b/drivers/gpu/drm/i915/display/intel_combo_phy_regs.h index ee41acdccf4e21c24ff8d89d20196f36e5776789..3694f95376c2efdb79ad5e78a39fddf913bc5197 100644 --- a/drivers/gpu/drm/i915/display/intel_combo_phy_regs.h +++ b/drivers/gpu/drm/i915/display/intel_combo_phy_regs.h @@ -6,7 +6,7 @@ #ifndef __INTEL_COMBO_PHY_REGS__ #define __INTEL_COMBO_PHY_REGS__ -#include "i915_reg_defs.h" +#include "intel_display_reg_defs.h" #define _ICL_COMBOPHY_A 0x162000 #define _ICL_COMBOPHY_B 0x6C000 diff --git a/drivers/gpu/drm/i915/display/intel_connector.c b/drivers/gpu/drm/i915/display/intel_connector.c index 18f3a1d2666259059c6500c57ca260ed2acb222f..42c923f416b326aba770a67f2a375b0a4f948949 100644 --- a/drivers/gpu/drm/i915/display/intel_connector.c +++ b/drivers/gpu/drm/i915/display/intel_connector.c @@ -64,10 +64,10 @@ static void intel_connector_modeset_retry_work_fn(struct work_struct *work) void intel_connector_queue_modeset_retry_work(struct intel_connector *connector) { - struct drm_i915_private *i915 = to_i915(connector->base.dev); + struct intel_display *display = to_intel_display(connector); drm_connector_get(&connector->base); - if (!queue_work(i915->unordered_wq, &connector->modeset_retry_work)) + if (!queue_work(display->wq.unordered, &connector->modeset_retry_work)) drm_connector_put(&connector->base); } @@ -208,8 +208,7 @@ enum pipe intel_connector_get_pipe(struct intel_connector *connector) { struct intel_display *display = to_intel_display(connector); - drm_WARN_ON(display->drm, - !drm_modeset_is_locked(&display->drm->mode_config.connection_mutex)); + drm_modeset_lock_assert_held(&display->drm->mode_config.connection_mutex); if (!connector->base.state->crtc) return INVALID_PIPE; diff --git a/drivers/gpu/drm/i915/display/intel_crt.c b/drivers/gpu/drm/i915/display/intel_crt.c index 38b50a779b6bb2e8d6c763a5b990221000a187b7..898c5d9e8f7adfe375d5ce6ac62ad918019df4d0 100644 --- a/drivers/gpu/drm/i915/display/intel_crt.c +++ b/drivers/gpu/drm/i915/display/intel_crt.c @@ -34,8 +34,6 @@ #include #include -#include "i915_irq.h" -#include "i915_reg.h" #include "intel_connector.h" #include "intel_crt.h" #include "intel_crt_regs.h" @@ -44,6 +42,7 @@ #include "intel_ddi_buf_trans.h" #include "intel_de.h" #include "intel_display_driver.h" +#include "intel_display_regs.h" #include "intel_display_types.h" #include "intel_fdi.h" #include "intel_fdi_regs.h" diff --git a/drivers/gpu/drm/i915/display/intel_crtc.c b/drivers/gpu/drm/i915/display/intel_crtc.c index 29cfc38f12e04645ec3a567fb5a41cd3daded524..a187db6df2d36ec7dbbba56fe7aa4e9f9b92751c 100644 --- a/drivers/gpu/drm/i915/display/intel_crtc.c +++ b/drivers/gpu/drm/i915/display/intel_crtc.c @@ -17,7 +17,6 @@ #include "i9xx_plane.h" #include "icl_dsi.h" #include "intel_atomic.h" -#include "intel_atomic_plane.h" #include "intel_color.h" #include "intel_crtc.h" #include "intel_cursor.h" @@ -29,6 +28,7 @@ #include "intel_dsi.h" #include "intel_fifo_underrun.h" #include "intel_pipe_crc.h" +#include "intel_plane.h" #include "intel_psr.h" #include "intel_sprite.h" #include "intel_vblank.h" @@ -417,10 +417,13 @@ int intel_crtc_get_pipe_from_crtc_id_ioctl(struct drm_device *dev, void *data, static bool intel_crtc_needs_vblank_work(const struct intel_crtc_state *crtc_state) { + struct intel_display *display = to_intel_display(crtc_state); + return crtc_state->hw.active && !crtc_state->preload_luts && !intel_crtc_needs_modeset(crtc_state) && - intel_crtc_needs_color_update(crtc_state) && + (intel_crtc_needs_color_update(crtc_state) && + !HAS_DOUBLE_BUFFERED_LUT(display)) && !intel_color_uses_dsb(crtc_state) && !crtc_state->use_dsb; } diff --git a/drivers/gpu/drm/i915/display/intel_cursor.c b/drivers/gpu/drm/i915/display/intel_cursor.c index 2fec5ba58373aea99afca76dd316fc43679be4a8..198e69efe9ac1ba13c6a3fece9016112985745a7 100644 --- a/drivers/gpu/drm/i915/display/intel_cursor.c +++ b/drivers/gpu/drm/i915/display/intel_cursor.c @@ -12,10 +12,8 @@ #include #include -#include "i915_reg.h" #include "i915_utils.h" #include "intel_atomic.h" -#include "intel_atomic_plane.h" #include "intel_cursor.h" #include "intel_cursor_regs.h" #include "intel_de.h" @@ -24,6 +22,7 @@ #include "intel_fb.h" #include "intel_fb_pin.h" #include "intel_frontbuffer.h" +#include "intel_plane.h" #include "intel_psr.h" #include "intel_psr_regs.h" #include "intel_vblank.h" @@ -159,10 +158,10 @@ static int intel_check_cursor(struct intel_crtc_state *crtc_state, return -EINVAL; } - ret = intel_atomic_plane_check_clipping(plane_state, crtc_state, - DRM_PLANE_NO_SCALING, - DRM_PLANE_NO_SCALING, - true); + ret = intel_plane_check_clipping(plane_state, crtc_state, + DRM_PLANE_NO_SCALING, + DRM_PLANE_NO_SCALING, + true); if (ret) return ret; diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c index a82b93cbc81d2c06e5519c130a91b4411b5ca3d1..ed8e640b96b0a1e3d55fbbddfc98c07ab1b67a02 100644 --- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c +++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c @@ -8,8 +8,8 @@ #include -#include "i915_reg.h" #include "i915_utils.h" +#include "intel_alpm.h" #include "intel_cx0_phy.h" #include "intel_cx0_phy_regs.h" #include "intel_ddi.h" @@ -39,7 +39,13 @@ bool intel_encoder_is_c10phy(struct intel_encoder *encoder) struct intel_display *display = to_intel_display(encoder); enum phy phy = intel_encoder_to_phy(encoder); - if (display->platform.pantherlake && phy == PHY_A) + /* PTL doesn't have a PHY connected to PORT B; as such, + * there will never be a case where PTL uses PHY B. + * WCL uses PORT A and B with the C10 PHY. + * Reusing the condition for WCL and extending it for PORT B + * should not cause any issues for PTL. + */ + if (display->platform.pantherlake && phy < PHY_C) return true; if ((display->platform.lunarlake || display->platform.meteorlake) && phy < PHY_C) @@ -3224,6 +3230,37 @@ void intel_mtl_pll_enable(struct intel_encoder *encoder, intel_cx0pll_enable(encoder, crtc_state); } +/* + * According to HAS we need to enable MAC Transmitting LFPS in the "PHY Common + * Control 0" PIPE register in case of AUX Less ALPM is going to be used. This + * function is doing that and is called by link retrain sequence. + */ +void intel_lnl_mac_transmit_lfps(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state) +{ + struct intel_display *display = to_intel_display(encoder); + u8 owned_lane_mask = intel_cx0_get_owned_lane_mask(encoder); + bool enable = intel_alpm_is_alpm_aux_less(enc_to_intel_dp(encoder), + crtc_state); + int i; + + if (DISPLAY_VER(display) < 20) + return; + + for (i = 0; i < 4; i++) { + int tx = i % 2 + 1; + u8 lane_mask = i < 2 ? INTEL_CX0_LANE0 : INTEL_CX0_LANE1; + + if (!(owned_lane_mask & lane_mask)) + continue; + + intel_cx0_rmw(encoder, lane_mask, PHY_CMN1_CONTROL(tx, 0), + CONTROL0_MAC_TRANSMIT_LFPS, + enable ? CONTROL0_MAC_TRANSMIT_LFPS : 0, + MB_WRITE_COMMITTED); + } +} + static u8 cx0_power_control_disable_val(struct intel_encoder *encoder) { struct intel_display *display = to_intel_display(encoder); diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.h b/drivers/gpu/drm/i915/display/intel_cx0_phy.h index a8f811ca5e7bcd42e0a0728be0a015b3dc7c64f4..c5a7b529955b0310233f924bdbf87ba96107dc0d 100644 --- a/drivers/gpu/drm/i915/display/intel_cx0_phy.h +++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.h @@ -43,5 +43,7 @@ void intel_cx0_phy_set_signal_levels(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state); int intel_mtl_tbt_calc_port_clock(struct intel_encoder *encoder); void intel_cx0_pll_power_save_wa(struct intel_display *display); +void intel_lnl_mac_transmit_lfps(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state); #endif /* __INTEL_CX0_PHY_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h index 59c22beaf1de50e121b5e79753b5c3bed27403db..77eae1d845f7916f30fbfe6f3f4d00691e3e1690 100644 --- a/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h +++ b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h @@ -6,8 +6,8 @@ #ifndef __INTEL_CX0_PHY_REGS_H__ #define __INTEL_CX0_PHY_REGS_H__ -#include "i915_reg_defs.h" #include "intel_display_limits.h" +#include "intel_display_reg_defs.h" /* DDI Buffer Control */ #define _DDI_CLK_VALFREQ_A 0x64030 @@ -285,6 +285,9 @@ #define PHY_CX0_TX_CONTROL(tx, control) (0x400 + ((tx) - 1) * 0x200 + (control)) #define CONTROL2_DISABLE_SINGLE_TX REG_BIT(6) +#define PHY_CMN1_CONTROL(tx, control) (0x800 + ((tx) - 1) * 0x200 + (control)) +#define CONTROL0_MAC_TRANSMIT_LFPS REG_BIT(1) + /* C20 Registers */ #define PHY_C20_WR_ADDRESS_L 0xC02 #define PHY_C20_WR_ADDRESS_H 0xC03 diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c index d58f8fc3732658d6b7cc2914dd8f57b3a046f4d6..0405396c7750eaf40dd35e25327b2d282c208c28 100644 --- a/drivers/gpu/drm/i915/display/intel_ddi.c +++ b/drivers/gpu/drm/i915/display/intel_ddi.c @@ -50,6 +50,7 @@ #include "intel_ddi_buf_trans.h" #include "intel_de.h" #include "intel_display_power.h" +#include "intel_display_regs.h" #include "intel_display_types.h" #include "intel_dkl_phy.h" #include "intel_dkl_phy_regs.h" @@ -72,11 +73,13 @@ #include "intel_lspcon.h" #include "intel_mg_phy_regs.h" #include "intel_modeset_lock.h" +#include "intel_panel.h" #include "intel_pfit.h" #include "intel_pps.h" #include "intel_psr.h" #include "intel_quirks.h" #include "intel_snps_phy.h" +#include "intel_step.h" #include "intel_tc.h" #include "intel_vdsc.h" #include "intel_vdsc_regs.h" @@ -236,7 +239,7 @@ static void intel_wait_ddi_buf_active(struct intel_encoder *encoder) port_name(port)); } -static u32 hsw_pll_to_ddi_pll_sel(const struct intel_shared_dpll *pll) +static u32 hsw_pll_to_ddi_pll_sel(const struct intel_dpll *pll) { switch (pll->info->id) { case DPLL_ID_WRPLL1: @@ -260,7 +263,7 @@ static u32 hsw_pll_to_ddi_pll_sel(const struct intel_shared_dpll *pll) static u32 icl_pll_to_ddi_clk_sel(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state) { - const struct intel_shared_dpll *pll = crtc_state->shared_dpll; + const struct intel_dpll *pll = crtc_state->intel_dpll; int clock = crtc_state->port_clock; const enum intel_dpll_id id = pll->info->id; @@ -1393,6 +1396,21 @@ static void tgl_dkl_phy_set_signal_levels(struct intel_encoder *encoder, for (ln = 0; ln < 2; ln++) { int level; + /* Wa_16011342517:adl-p */ + if (display->platform.alderlake_p && + IS_DISPLAY_STEP(display, STEP_A0, STEP_D0)) { + if ((intel_encoder_is_hdmi(encoder) && + crtc_state->port_clock == 594000) || + (intel_encoder_is_dp(encoder) && + crtc_state->port_clock == 162000)) { + intel_dkl_phy_rmw(display, DKL_TX_DPCNTL2(tc_port, ln), + LOADGEN_SHARING_PMD_DISABLE, 1); + } else { + intel_dkl_phy_rmw(display, DKL_TX_DPCNTL2(tc_port, ln), + LOADGEN_SHARING_PMD_DISABLE, 0); + } + } + intel_dkl_phy_write(display, DKL_TX_PMD_LANE_SUS(tc_port, ln), 0); level = intel_ddi_level(encoder, crtc_state, 2*ln+0); @@ -1561,7 +1579,7 @@ static bool _icl_ddi_is_clock_enabled(struct intel_display *display, i915_reg_t return !(intel_de_read(display, reg) & clk_off); } -static struct intel_shared_dpll * +static struct intel_dpll * _icl_ddi_get_pll(struct intel_display *display, i915_reg_t reg, u32 clk_sel_mask, u32 clk_sel_shift) { @@ -1569,14 +1587,14 @@ _icl_ddi_get_pll(struct intel_display *display, i915_reg_t reg, id = (intel_de_read(display, reg) & clk_sel_mask) >> clk_sel_shift; - return intel_get_shared_dpll_by_id(display, id); + return intel_get_dpll_by_id(display, id); } static void adls_ddi_enable_clock(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state) { struct intel_display *display = to_intel_display(encoder); - const struct intel_shared_dpll *pll = crtc_state->shared_dpll; + const struct intel_dpll *pll = crtc_state->intel_dpll; enum phy phy = intel_encoder_to_phy(encoder); if (drm_WARN_ON(display->drm, !pll)) @@ -1606,7 +1624,7 @@ static bool adls_ddi_is_clock_enabled(struct intel_encoder *encoder) ICL_DPCLKA_CFGCR0_DDI_CLK_OFF(phy)); } -static struct intel_shared_dpll *adls_ddi_get_pll(struct intel_encoder *encoder) +static struct intel_dpll *adls_ddi_get_pll(struct intel_encoder *encoder) { struct intel_display *display = to_intel_display(encoder); enum phy phy = intel_encoder_to_phy(encoder); @@ -1620,7 +1638,7 @@ static void rkl_ddi_enable_clock(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state) { struct intel_display *display = to_intel_display(encoder); - const struct intel_shared_dpll *pll = crtc_state->shared_dpll; + const struct intel_dpll *pll = crtc_state->intel_dpll; enum phy phy = intel_encoder_to_phy(encoder); if (drm_WARN_ON(display->drm, !pll)) @@ -1650,7 +1668,7 @@ static bool rkl_ddi_is_clock_enabled(struct intel_encoder *encoder) RKL_DPCLKA_CFGCR0_DDI_CLK_OFF(phy)); } -static struct intel_shared_dpll *rkl_ddi_get_pll(struct intel_encoder *encoder) +static struct intel_dpll *rkl_ddi_get_pll(struct intel_encoder *encoder) { struct intel_display *display = to_intel_display(encoder); enum phy phy = intel_encoder_to_phy(encoder); @@ -1664,7 +1682,7 @@ static void dg1_ddi_enable_clock(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state) { struct intel_display *display = to_intel_display(encoder); - const struct intel_shared_dpll *pll = crtc_state->shared_dpll; + const struct intel_dpll *pll = crtc_state->intel_dpll; enum phy phy = intel_encoder_to_phy(encoder); if (drm_WARN_ON(display->drm, !pll)) @@ -1703,7 +1721,7 @@ static bool dg1_ddi_is_clock_enabled(struct intel_encoder *encoder) DG1_DPCLKA_CFGCR0_DDI_CLK_OFF(phy)); } -static struct intel_shared_dpll *dg1_ddi_get_pll(struct intel_encoder *encoder) +static struct intel_dpll *dg1_ddi_get_pll(struct intel_encoder *encoder) { struct intel_display *display = to_intel_display(encoder); enum phy phy = intel_encoder_to_phy(encoder); @@ -1723,14 +1741,14 @@ static struct intel_shared_dpll *dg1_ddi_get_pll(struct intel_encoder *encoder) if (phy >= PHY_C) id += DPLL_ID_DG1_DPLL2; - return intel_get_shared_dpll_by_id(display, id); + return intel_get_dpll_by_id(display, id); } static void icl_ddi_combo_enable_clock(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state) { struct intel_display *display = to_intel_display(encoder); - const struct intel_shared_dpll *pll = crtc_state->shared_dpll; + const struct intel_dpll *pll = crtc_state->intel_dpll; enum phy phy = intel_encoder_to_phy(encoder); if (drm_WARN_ON(display->drm, !pll)) @@ -1760,7 +1778,7 @@ static bool icl_ddi_combo_is_clock_enabled(struct intel_encoder *encoder) ICL_DPCLKA_CFGCR0_DDI_CLK_OFF(phy)); } -struct intel_shared_dpll *icl_ddi_combo_get_pll(struct intel_encoder *encoder) +struct intel_dpll *icl_ddi_combo_get_pll(struct intel_encoder *encoder) { struct intel_display *display = to_intel_display(encoder); enum phy phy = intel_encoder_to_phy(encoder); @@ -1774,7 +1792,7 @@ static void jsl_ddi_tc_enable_clock(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state) { struct intel_display *display = to_intel_display(encoder); - const struct intel_shared_dpll *pll = crtc_state->shared_dpll; + const struct intel_dpll *pll = crtc_state->intel_dpll; enum port port = encoder->port; if (drm_WARN_ON(display->drm, !pll)) @@ -1817,7 +1835,7 @@ static void icl_ddi_tc_enable_clock(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state) { struct intel_display *display = to_intel_display(encoder); - const struct intel_shared_dpll *pll = crtc_state->shared_dpll; + const struct intel_dpll *pll = crtc_state->intel_dpll; enum tc_port tc_port = intel_encoder_to_tc(encoder); enum port port = encoder->port; @@ -1868,7 +1886,7 @@ static bool icl_ddi_tc_is_clock_enabled(struct intel_encoder *encoder) return !(tmp & ICL_DPCLKA_CFGCR0_TC_CLK_OFF(tc_port)); } -static struct intel_shared_dpll *icl_ddi_tc_get_pll(struct intel_encoder *encoder) +static struct intel_dpll *icl_ddi_tc_get_pll(struct intel_encoder *encoder) { struct intel_display *display = to_intel_display(encoder); enum tc_port tc_port = intel_encoder_to_tc(encoder); @@ -1895,10 +1913,10 @@ static struct intel_shared_dpll *icl_ddi_tc_get_pll(struct intel_encoder *encode return NULL; } - return intel_get_shared_dpll_by_id(display, id); + return intel_get_dpll_by_id(display, id); } -static struct intel_shared_dpll *bxt_ddi_get_pll(struct intel_encoder *encoder) +static struct intel_dpll *bxt_ddi_get_pll(struct intel_encoder *encoder) { struct intel_display *display = to_intel_display(encoder->base.dev); enum intel_dpll_id id; @@ -1918,14 +1936,14 @@ static struct intel_shared_dpll *bxt_ddi_get_pll(struct intel_encoder *encoder) return NULL; } - return intel_get_shared_dpll_by_id(display, id); + return intel_get_dpll_by_id(display, id); } static void skl_ddi_enable_clock(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state) { struct intel_display *display = to_intel_display(encoder); - const struct intel_shared_dpll *pll = crtc_state->shared_dpll; + const struct intel_dpll *pll = crtc_state->intel_dpll; enum port port = encoder->port; if (drm_WARN_ON(display->drm, !pll)) @@ -1967,7 +1985,7 @@ static bool skl_ddi_is_clock_enabled(struct intel_encoder *encoder) return !(intel_de_read(display, DPLL_CTRL2) & DPLL_CTRL2_DDI_CLK_OFF(port)); } -static struct intel_shared_dpll *skl_ddi_get_pll(struct intel_encoder *encoder) +static struct intel_dpll *skl_ddi_get_pll(struct intel_encoder *encoder) { struct intel_display *display = to_intel_display(encoder); enum port port = encoder->port; @@ -1986,14 +2004,14 @@ static struct intel_shared_dpll *skl_ddi_get_pll(struct intel_encoder *encoder) id = (tmp & DPLL_CTRL2_DDI_CLK_SEL_MASK(port)) >> DPLL_CTRL2_DDI_CLK_SEL_SHIFT(port); - return intel_get_shared_dpll_by_id(display, id); + return intel_get_dpll_by_id(display, id); } void hsw_ddi_enable_clock(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state) { struct intel_display *display = to_intel_display(encoder); - const struct intel_shared_dpll *pll = crtc_state->shared_dpll; + const struct intel_dpll *pll = crtc_state->intel_dpll; enum port port = encoder->port; if (drm_WARN_ON(display->drm, !pll)) @@ -2018,7 +2036,7 @@ bool hsw_ddi_is_clock_enabled(struct intel_encoder *encoder) return intel_de_read(display, PORT_CLK_SEL(port)) != PORT_CLK_SEL_NONE; } -static struct intel_shared_dpll *hsw_ddi_get_pll(struct intel_encoder *encoder) +static struct intel_dpll *hsw_ddi_get_pll(struct intel_encoder *encoder) { struct intel_display *display = to_intel_display(encoder); enum port port = encoder->port; @@ -2053,7 +2071,7 @@ static struct intel_shared_dpll *hsw_ddi_get_pll(struct intel_encoder *encoder) return NULL; } - return intel_get_shared_dpll_by_id(display, id); + return intel_get_dpll_by_id(display, id); } void intel_ddi_enable_clock(struct intel_encoder *encoder, @@ -2760,7 +2778,7 @@ static void tgl_ddi_pre_enable_dp(struct intel_atomic_state *state, * 4. Enable the port PLL. * * The PLL enabling itself was already done before this function by - * hsw_crtc_enable()->intel_enable_shared_dpll(). We need only + * hsw_crtc_enable()->intel_enable_dpll(). We need only * configure the PLL to port mapping here. */ intel_ddi_enable_clock(encoder, crtc_state); @@ -3354,6 +3372,8 @@ static void intel_ddi_enable_dp(struct intel_atomic_state *state, drm_connector_update_privacy_screen(conn_state); intel_edp_backlight_on(crtc_state, conn_state); + intel_panel_prepare(crtc_state, conn_state); + if (!intel_lspcon_active(dig_port) || intel_dp_has_hdmi_sink(&dig_port->dp)) intel_dp_set_infoframes(encoder, true, crtc_state, conn_state); @@ -3551,6 +3571,7 @@ static void intel_ddi_disable_dp(struct intel_atomic_state *state, intel_dp->link.active = false; + intel_panel_unprepare(old_conn_state); intel_psr_disable(intel_dp, old_crtc_state); intel_alpm_disable(intel_dp); intel_edp_backlight_off(old_conn_state); @@ -3647,7 +3668,7 @@ void intel_ddi_update_active_dpll(struct intel_atomic_state *state, for_each_intel_crtc_in_pipe_mask(display->drm, pipe_crtc, intel_crtc_joined_pipe_mask(crtc_state)) - intel_update_active_dpll(state, pipe_crtc, encoder); + intel_dpll_update_active(state, pipe_crtc, encoder); } /* @@ -3740,6 +3761,18 @@ static void mtl_ddi_prepare_link_retrain(struct intel_dp *intel_dp, intel_ddi_buf_enable(encoder, intel_dp->DP); intel_dp->DP |= DDI_BUF_CTL_ENABLE; + + /* + * 6.k If AUX-Less ALPM is going to be enabled: + * i. Configure PORT_ALPM_CTL and PORT_ALPM_LFPS_CTL here + */ + intel_alpm_port_configure(intel_dp, crtc_state); + + /* + * ii. Enable MAC Transmits LFPS in the "PHY Common Control 0" PIPE + * register + */ + intel_lnl_mac_transmit_lfps(encoder, crtc_state); } static void intel_ddi_prepare_link_retrain(struct intel_dp *intel_dp, @@ -4184,7 +4217,7 @@ static void intel_ddi_get_config(struct intel_encoder *encoder, void intel_ddi_get_clock(struct intel_encoder *encoder, struct intel_crtc_state *crtc_state, - struct intel_shared_dpll *pll) + struct intel_dpll *pll) { struct intel_display *display = to_intel_display(encoder); enum icl_port_dpll_id port_dpll_id = ICL_PORT_DPLL_DEFAULT; @@ -4200,7 +4233,7 @@ void intel_ddi_get_clock(struct intel_encoder *encoder, icl_set_active_port_dpll(crtc_state, port_dpll_id); - crtc_state->port_clock = intel_dpll_get_freq(display, crtc_state->shared_dpll, + crtc_state->port_clock = intel_dpll_get_freq(display, crtc_state->intel_dpll, &crtc_state->dpll_hw_state); } @@ -4254,7 +4287,7 @@ static void icl_ddi_combo_get_config(struct intel_encoder *encoder, intel_ddi_get_config(encoder, crtc_state); } -static bool icl_ddi_tc_pll_is_tbt(const struct intel_shared_dpll *pll) +static bool icl_ddi_tc_pll_is_tbt(const struct intel_dpll *pll) { return pll->info->id == DPLL_ID_ICL_TBTPLL; } @@ -4264,7 +4297,7 @@ icl_ddi_tc_port_pll_type(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state) { struct intel_display *display = to_intel_display(encoder); - const struct intel_shared_dpll *pll = crtc_state->shared_dpll; + const struct intel_dpll *pll = crtc_state->intel_dpll; if (drm_WARN_ON(display->drm, !pll)) return ICL_PORT_DPLL_DEFAULT; @@ -4287,7 +4320,7 @@ intel_ddi_port_pll_type(struct intel_encoder *encoder, static void icl_ddi_tc_get_clock(struct intel_encoder *encoder, struct intel_crtc_state *crtc_state, - struct intel_shared_dpll *pll) + struct intel_dpll *pll) { struct intel_display *display = to_intel_display(encoder); enum icl_port_dpll_id port_dpll_id; @@ -4310,10 +4343,10 @@ static void icl_ddi_tc_get_clock(struct intel_encoder *encoder, icl_set_active_port_dpll(crtc_state, port_dpll_id); - if (icl_ddi_tc_pll_is_tbt(crtc_state->shared_dpll)) + if (icl_ddi_tc_pll_is_tbt(crtc_state->intel_dpll)) crtc_state->port_clock = icl_calc_tbt_pll_link(display, encoder->port); else - crtc_state->port_clock = intel_dpll_get_freq(display, crtc_state->shared_dpll, + crtc_state->port_clock = intel_dpll_get_freq(display, crtc_state->intel_dpll, &crtc_state->dpll_hw_state); } diff --git a/drivers/gpu/drm/i915/display/intel_ddi.h b/drivers/gpu/drm/i915/display/intel_ddi.h index 353eb04079e9c5af88aa91110f3354223f06fc7f..f6f511bb04314ca122df5cf69491b2ce828865be 100644 --- a/drivers/gpu/drm/i915/display/intel_ddi.h +++ b/drivers/gpu/drm/i915/display/intel_ddi.h @@ -16,9 +16,9 @@ struct intel_crtc; struct intel_crtc_state; struct intel_display; struct intel_dp; +struct intel_dpll; struct intel_dpll_hw_state; struct intel_encoder; -struct intel_shared_dpll; enum pipe; enum port; enum transcoder; @@ -40,7 +40,7 @@ void intel_ddi_enable_clock(struct intel_encoder *encoder, void intel_ddi_disable_clock(struct intel_encoder *encoder); void intel_ddi_get_clock(struct intel_encoder *encoder, struct intel_crtc_state *crtc_state, - struct intel_shared_dpll *pll); + struct intel_dpll *pll); void hsw_ddi_enable_clock(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state); void hsw_ddi_disable_clock(struct intel_encoder *encoder); @@ -50,7 +50,7 @@ intel_ddi_port_pll_type(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state); void hsw_ddi_get_config(struct intel_encoder *encoder, struct intel_crtc_state *crtc_state); -struct intel_shared_dpll *icl_ddi_combo_get_pll(struct intel_encoder *encoder); +struct intel_dpll *icl_ddi_combo_get_pll(struct intel_encoder *encoder); void hsw_prepare_dp_ddi_buffers(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state); void intel_wait_ddi_buf_idle(struct intel_display *display, enum port port); diff --git a/drivers/gpu/drm/i915/display/intel_de.h b/drivers/gpu/drm/i915/display/intel_de.h index 54ce3e4f8fd909a144ecfe7b3d53ef5a92a94a8a..9ecdcf6b73e4d4abdea08c9cbfaf00053360de7c 100644 --- a/drivers/gpu/drm/i915/display/intel_de.h +++ b/drivers/gpu/drm/i915/display/intel_de.h @@ -107,10 +107,10 @@ intel_de_rmw(struct intel_display *display, i915_reg_t reg, u32 clear, u32 set) static inline int __intel_de_wait_for_register_nowl(struct intel_display *display, i915_reg_t reg, - u32 mask, u32 value, unsigned int timeout) + u32 mask, u32 value, unsigned int timeout_ms) { return intel_wait_for_register(__to_uncore(display), reg, mask, - value, timeout); + value, timeout_ms); } static inline int @@ -125,14 +125,14 @@ __intel_de_wait_for_register_atomic_nowl(struct intel_display *display, static inline int intel_de_wait(struct intel_display *display, i915_reg_t reg, - u32 mask, u32 value, unsigned int timeout) + u32 mask, u32 value, unsigned int timeout_ms) { int ret; intel_dmc_wl_get(display, reg); ret = __intel_de_wait_for_register_nowl(display, reg, mask, value, - timeout); + timeout_ms); intel_dmc_wl_put(display, reg); @@ -141,14 +141,14 @@ intel_de_wait(struct intel_display *display, i915_reg_t reg, static inline int intel_de_wait_fw(struct intel_display *display, i915_reg_t reg, - u32 mask, u32 value, unsigned int timeout) + u32 mask, u32 value, unsigned int timeout_ms, u32 *out_value) { int ret; intel_dmc_wl_get(display, reg); ret = intel_wait_for_register_fw(__to_uncore(display), reg, mask, - value, timeout); + value, timeout_ms, out_value); intel_dmc_wl_put(display, reg); @@ -176,16 +176,16 @@ intel_de_wait_custom(struct intel_display *display, i915_reg_t reg, static inline int intel_de_wait_for_set(struct intel_display *display, i915_reg_t reg, - u32 mask, unsigned int timeout) + u32 mask, unsigned int timeout_ms) { - return intel_de_wait(display, reg, mask, mask, timeout); + return intel_de_wait(display, reg, mask, mask, timeout_ms); } static inline int intel_de_wait_for_clear(struct intel_display *display, i915_reg_t reg, - u32 mask, unsigned int timeout) + u32 mask, unsigned int timeout_ms) { - return intel_de_wait(display, reg, mask, 0, timeout); + return intel_de_wait(display, reg, mask, 0, timeout_ms); } /* diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c index 6f0a0bc71b066aea8a2f501c885fff0c452c0f06..456fc4b04cda7fc69d671f35e1086817e3f612fc 100644 --- a/drivers/gpu/drm/i915/display/intel_display.c +++ b/drivers/gpu/drm/i915/display/intel_display.c @@ -57,7 +57,6 @@ #include "i9xx_wm.h" #include "intel_alpm.h" #include "intel_atomic.h" -#include "intel_atomic_plane.h" #include "intel_audio.h" #include "intel_bo.h" #include "intel_bw.h" @@ -67,13 +66,14 @@ #include "intel_crt.h" #include "intel_crtc.h" #include "intel_crtc_state_dump.h" +#include "intel_cursor.h" #include "intel_cursor_regs.h" #include "intel_cx0_phy.h" -#include "intel_cursor.h" #include "intel_ddi.h" #include "intel_de.h" #include "intel_display_driver.h" #include "intel_display_power.h" +#include "intel_display_regs.h" #include "intel_display_rpm.h" #include "intel_display_types.h" #include "intel_dmc.h" @@ -93,6 +93,7 @@ #include "intel_fbc.h" #include "intel_fdi.h" #include "intel_fifo_underrun.h" +#include "intel_flipq.h" #include "intel_frontbuffer.h" #include "intel_hdmi.h" #include "intel_hotplug.h" @@ -105,9 +106,9 @@ #include "intel_panel.h" #include "intel_pch_display.h" #include "intel_pch_refclk.h" -#include "intel_pcode.h" #include "intel_pfit.h" #include "intel_pipe_crc.h" +#include "intel_plane.h" #include "intel_plane_initial.h" #include "intel_pmdemand.h" #include "intel_pps.h" @@ -140,46 +141,47 @@ static void bdw_set_pipe_misc(struct intel_dsb *dsb, const struct intel_crtc_state *crtc_state); /* returns HPLL frequency in kHz */ -int vlv_get_hpll_vco(struct drm_i915_private *dev_priv) +int vlv_get_hpll_vco(struct drm_device *drm) { int hpll_freq, vco_freq[] = { 800, 1600, 2000, 2400 }; /* Obtain SKU information */ - hpll_freq = vlv_cck_read(dev_priv, CCK_FUSE_REG) & + hpll_freq = vlv_cck_read(drm, CCK_FUSE_REG) & CCK_FUSE_HPLL_FREQ_MASK; return vco_freq[hpll_freq] * 1000; } -int vlv_get_cck_clock(struct drm_i915_private *dev_priv, +int vlv_get_cck_clock(struct drm_device *drm, const char *name, u32 reg, int ref_freq) { u32 val; int divider; - val = vlv_cck_read(dev_priv, reg); + val = vlv_cck_read(drm, reg); divider = val & CCK_FREQUENCY_VALUES; - drm_WARN(&dev_priv->drm, (val & CCK_FREQUENCY_STATUS) != + drm_WARN(drm, (val & CCK_FREQUENCY_STATUS) != (divider << CCK_FREQUENCY_STATUS_SHIFT), "%s change in progress\n", name); return DIV_ROUND_CLOSEST(ref_freq << 1, divider + 1); } -int vlv_get_cck_clock_hpll(struct drm_i915_private *dev_priv, +int vlv_get_cck_clock_hpll(struct drm_device *drm, const char *name, u32 reg) { + struct drm_i915_private *dev_priv = to_i915(drm); int hpll; - vlv_cck_get(dev_priv); + vlv_cck_get(drm); if (dev_priv->hpll_freq == 0) - dev_priv->hpll_freq = vlv_get_hpll_vco(dev_priv); + dev_priv->hpll_freq = vlv_get_hpll_vco(drm); - hpll = vlv_get_cck_clock(dev_priv, name, reg, dev_priv->hpll_freq); + hpll = vlv_get_cck_clock(drm, name, reg, dev_priv->hpll_freq); - vlv_cck_put(dev_priv); + vlv_cck_put(drm); return hpll; } @@ -191,7 +193,7 @@ void intel_update_czclk(struct intel_display *display) if (!display->platform.valleyview && !display->platform.cherryview) return; - dev_priv->czclk_freq = vlv_get_cck_clock_hpll(dev_priv, "czclk", + dev_priv->czclk_freq = vlv_get_cck_clock_hpll(display->drm, "czclk", CCK_CZ_CLOCK_CONTROL); drm_dbg_kms(display->drm, "CZ clock rate: %d kHz\n", dev_priv->czclk_freq); @@ -1325,7 +1327,7 @@ static void intel_encoders_update_prepare(struct intel_atomic_state *state) if (intel_crtc_needs_modeset(new_crtc_state)) continue; - new_crtc_state->shared_dpll = old_crtc_state->shared_dpll; + new_crtc_state->intel_dpll = old_crtc_state->intel_dpll; new_crtc_state->dpll_hw_state = old_crtc_state->dpll_hw_state; } } @@ -1658,13 +1660,17 @@ static void hsw_crtc_enable(struct intel_atomic_state *state, if (drm_WARN_ON(display->drm, crtc->active)) return; - for_each_pipe_crtc_modeset_enable(display, pipe_crtc, new_crtc_state, i) - intel_dmc_enable_pipe(display, pipe_crtc->pipe); + for_each_pipe_crtc_modeset_enable(display, pipe_crtc, new_crtc_state, i) { + const struct intel_crtc_state *new_pipe_crtc_state = + intel_atomic_get_new_crtc_state(state, pipe_crtc); + + intel_dmc_enable_pipe(new_pipe_crtc_state); + } intel_encoders_pre_pll_enable(state, crtc); - if (new_crtc_state->shared_dpll) - intel_enable_shared_dpll(new_crtc_state); + if (new_crtc_state->intel_dpll) + intel_dpll_enable(new_crtc_state); intel_encoders_pre_enable(state, crtc); @@ -1793,12 +1799,16 @@ static void hsw_crtc_disable(struct intel_atomic_state *state, intel_encoders_disable(state, crtc); intel_encoders_post_disable(state, crtc); - intel_disable_shared_dpll(old_crtc_state); + intel_dpll_disable(old_crtc_state); intel_encoders_post_pll_disable(state, crtc); - for_each_pipe_crtc_modeset_disable(display, pipe_crtc, old_crtc_state, i) - intel_dmc_disable_pipe(display, pipe_crtc->pipe); + for_each_pipe_crtc_modeset_disable(display, pipe_crtc, old_crtc_state, i) { + const struct intel_crtc_state *old_pipe_crtc_state = + intel_atomic_get_old_crtc_state(state, pipe_crtc); + + intel_dmc_disable_pipe(old_pipe_crtc_state); + } } /* Prefer intel_encoder_is_combo() */ @@ -1959,7 +1969,7 @@ static void get_crtc_power_domains(struct intel_crtc_state *crtc_state, if (HAS_DDI(display) && crtc_state->has_audio) set_bit(POWER_DOMAIN_AUDIO_MMIO, mask->bits); - if (crtc_state->shared_dpll) + if (crtc_state->intel_dpll) set_bit(POWER_DOMAIN_DISPLAY_CORE, mask->bits); if (crtc_state->dsc.compression_enable) @@ -4159,7 +4169,7 @@ static u16 hsw_ips_linetime_wm(const struct intel_crtc_state *crtc_state, return 0; linetime_wm = DIV_ROUND_CLOSEST(pipe_mode->crtc_htotal * 1000 * 8, - cdclk_state->logical.cdclk); + intel_cdclk_logical(cdclk_state)); return min(linetime_wm, 0x1ff); } @@ -4225,7 +4235,7 @@ static int intel_crtc_atomic_check(struct intel_atomic_state *state, crtc_state->update_wm_post = true; if (intel_crtc_needs_modeset(crtc_state)) { - ret = intel_dpll_crtc_get_shared_dpll(state, crtc); + ret = intel_dpll_crtc_get_dpll(state, crtc); if (ret) return ret; } @@ -4318,6 +4328,22 @@ compute_sink_pipe_bpp(const struct drm_connector_state *conn_state, return 0; } +int intel_display_min_pipe_bpp(void) +{ + return 6 * 3; +} + +int intel_display_max_pipe_bpp(struct intel_display *display) +{ + if (display->platform.g4x || display->platform.valleyview || + display->platform.cherryview) + return 10*3; + else if (DISPLAY_VER(display) >= 5) + return 12*3; + else + return 8*3; +} + static int compute_baseline_pipe_bpp(struct intel_atomic_state *state, struct intel_crtc *crtc) @@ -4327,17 +4353,9 @@ compute_baseline_pipe_bpp(struct intel_atomic_state *state, intel_atomic_get_new_crtc_state(state, crtc); struct drm_connector *connector; struct drm_connector_state *connector_state; - int bpp, i; - - if (display->platform.g4x || display->platform.valleyview || - display->platform.cherryview) - bpp = 10*3; - else if (DISPLAY_VER(display) >= 5) - bpp = 12*3; - else - bpp = 8*3; + int i; - crtc_state->pipe_bpp = bpp; + crtc_state->pipe_bpp = intel_display_max_pipe_bpp(display); /* Clamp display bpp to connector max bpp */ for_each_new_connector_in_state(&state->base, connector, connector_state, i) { @@ -4501,7 +4519,7 @@ copy_joiner_crtc_state_modeset(struct intel_atomic_state *state, /* preserve some things from the slave's original crtc state */ saved_state->uapi = secondary_crtc_state->uapi; saved_state->scaler_state = secondary_crtc_state->scaler_state; - saved_state->shared_dpll = secondary_crtc_state->shared_dpll; + saved_state->intel_dpll = secondary_crtc_state->intel_dpll; saved_state->crc_enabled = secondary_crtc_state->crc_enabled; intel_crtc_free_hw_state(secondary_crtc_state); @@ -4564,7 +4582,7 @@ intel_crtc_prepare_cleared_state(struct intel_atomic_state *state, saved_state->uapi = crtc_state->uapi; saved_state->inherited = crtc_state->inherited; saved_state->scaler_state = crtc_state->scaler_state; - saved_state->shared_dpll = crtc_state->shared_dpll; + saved_state->intel_dpll = crtc_state->intel_dpll; saved_state->dpll_hw_state = crtc_state->dpll_hw_state; memcpy(saved_state->icl_port_dplls, crtc_state->icl_port_dplls, sizeof(saved_state->icl_port_dplls)); @@ -5318,7 +5336,7 @@ intel_pipe_config_compare(const struct intel_crtc_state *current_config, PIPE_CONF_CHECK_BOOL(double_wide); if (display->dpll.mgr) - PIPE_CONF_CHECK_P(shared_dpll); + PIPE_CONF_CHECK_P(intel_dpll); /* FIXME convert everything over the dpll_mgr */ if (display->dpll.mgr || HAS_GMCH(display)) @@ -5470,7 +5488,7 @@ static int intel_modeset_pipe(struct intel_atomic_state *state, if (ret) return ret; - ret = intel_atomic_add_affected_planes(state, crtc); + ret = intel_plane_add_affected(state, crtc); if (ret) return ret; @@ -6186,7 +6204,7 @@ static int intel_joiner_add_affected_crtcs(struct intel_atomic_state *state) if (ret) return ret; - ret = intel_atomic_add_affected_planes(state, crtc); + ret = intel_plane_add_affected(state, crtc); if (ret) return ret; } @@ -6428,7 +6446,7 @@ int intel_atomic_check(struct drm_device *dev, any_ms = true; - intel_release_shared_dplls(state, crtc); + intel_dpll_release(state, crtc); } if (any_ms && !check_digital_port_conflicts(state)) { @@ -6438,7 +6456,7 @@ int intel_atomic_check(struct drm_device *dev, goto fail; } - ret = intel_atomic_check_planes(state); + ret = intel_plane_atomic_check(state); if (ret) goto fail; @@ -6602,7 +6620,7 @@ static void commit_pipe_pre_planes(struct intel_atomic_state *state, intel_atomic_get_new_crtc_state(state, crtc); bool modeset = intel_crtc_needs_modeset(new_crtc_state); - drm_WARN_ON(display->drm, new_crtc_state->use_dsb); + drm_WARN_ON(display->drm, new_crtc_state->use_dsb || new_crtc_state->use_flipq); /* * During modesets pipe configuration was programmed as the @@ -6630,18 +6648,24 @@ static void commit_pipe_post_planes(struct intel_atomic_state *state, struct intel_display *display = to_intel_display(state); const struct intel_crtc_state *new_crtc_state = intel_atomic_get_new_crtc_state(state, crtc); + bool modeset = intel_crtc_needs_modeset(new_crtc_state); - drm_WARN_ON(display->drm, new_crtc_state->use_dsb); + drm_WARN_ON(display->drm, new_crtc_state->use_dsb || new_crtc_state->use_flipq); /* * Disable the scaler(s) after the plane(s) so that we don't * get a catastrophic underrun even if the two operations * end up happening in two different frames. */ - if (DISPLAY_VER(display) >= 9 && - !intel_crtc_needs_modeset(new_crtc_state)) + if (DISPLAY_VER(display) >= 9 && !modeset) skl_detach_scalers(NULL, new_crtc_state); + if (!modeset && + intel_crtc_needs_color_update(new_crtc_state) && + !intel_color_uses_dsb(new_crtc_state) && + HAS_DOUBLE_BUFFERED_LUT(display)) + intel_color_load_luts(new_crtc_state); + if (intel_crtc_vrr_enabling(state, crtc)) intel_vrr_enable(new_crtc_state); } @@ -6715,10 +6739,10 @@ static void intel_pre_update_crtc(struct intel_atomic_state *state, if (!modeset && intel_crtc_needs_color_update(new_crtc_state) && - !new_crtc_state->use_dsb) + !new_crtc_state->use_dsb && !new_crtc_state->use_flipq) intel_color_commit_noarm(NULL, new_crtc_state); - if (!new_crtc_state->use_dsb) + if (!new_crtc_state->use_dsb && !new_crtc_state->use_flipq) intel_crtc_planes_update_noarm(NULL, state, crtc); } @@ -6730,16 +6754,23 @@ static void intel_update_crtc(struct intel_atomic_state *state, struct intel_crtc_state *new_crtc_state = intel_atomic_get_new_crtc_state(state, crtc); - if (new_crtc_state->use_dsb) { + if (new_crtc_state->use_flipq) { + intel_flipq_enable(new_crtc_state); + + intel_crtc_prepare_vblank_event(new_crtc_state, &crtc->flipq_event); + + intel_flipq_add(crtc, INTEL_FLIPQ_PLANE_1, 0, INTEL_DSB_0, + new_crtc_state->dsb_commit); + } else if (new_crtc_state->use_dsb) { intel_crtc_prepare_vblank_event(new_crtc_state, &crtc->dsb_event); - intel_dsb_commit(new_crtc_state->dsb_commit, false); + intel_dsb_commit(new_crtc_state->dsb_commit); } else { /* Perform vblank evasion around commit operation */ intel_pipe_update_start(state, crtc); if (new_crtc_state->dsb_commit) - intel_dsb_commit(new_crtc_state->dsb_commit, false); + intel_dsb_commit(new_crtc_state->dsb_commit); commit_pipe_pre_planes(state, crtc); @@ -7168,7 +7199,17 @@ static void intel_atomic_dsb_prepare(struct intel_atomic_state *state, return; /* FIXME deal with everything */ + new_crtc_state->use_flipq = + intel_flipq_supported(display) && + !new_crtc_state->do_async_flip && + !new_crtc_state->vrr.enable && + !new_crtc_state->has_psr && + !intel_crtc_needs_modeset(new_crtc_state) && + !intel_crtc_needs_fastset(new_crtc_state) && + !intel_crtc_needs_color_update(new_crtc_state); + new_crtc_state->use_dsb = + !new_crtc_state->use_flipq && !new_crtc_state->do_async_flip && (DISPLAY_VER(display) >= 20 || !new_crtc_state->has_psr) && !intel_crtc_needs_modeset(new_crtc_state) && @@ -7184,7 +7225,9 @@ static void intel_atomic_dsb_finish(struct intel_atomic_state *state, struct intel_crtc_state *new_crtc_state = intel_atomic_get_new_crtc_state(state, crtc); - if (!new_crtc_state->use_dsb && !new_crtc_state->dsb_color_vblank) + if (!new_crtc_state->use_flipq && + !new_crtc_state->use_dsb && + !new_crtc_state->dsb_color) return; /* @@ -7193,14 +7236,20 @@ static void intel_atomic_dsb_finish(struct intel_atomic_state *state, * Double that for pipe stuff and other overhead. */ new_crtc_state->dsb_commit = intel_dsb_prepare(state, crtc, INTEL_DSB_0, - new_crtc_state->use_dsb ? 1024 : 16); + new_crtc_state->use_dsb || + new_crtc_state->use_flipq ? 1024 : 16); if (!new_crtc_state->dsb_commit) { + new_crtc_state->use_flipq = false; new_crtc_state->use_dsb = false; intel_color_cleanup_commit(new_crtc_state); return; } - if (new_crtc_state->use_dsb) { + if (new_crtc_state->use_flipq || new_crtc_state->use_dsb) { + /* Wa_18034343758 */ + if (new_crtc_state->use_flipq) + intel_flipq_wait_dmc_halt(new_crtc_state->dsb_commit, crtc); + if (intel_crtc_needs_color_update(new_crtc_state)) intel_color_commit_noarm(new_crtc_state->dsb_commit, new_crtc_state); @@ -7215,7 +7264,8 @@ static void intel_atomic_dsb_finish(struct intel_atomic_state *state, intel_psr_trigger_frame_change_event(new_crtc_state->dsb_commit, state, crtc); - intel_dsb_vblank_evade(state, new_crtc_state->dsb_commit); + if (new_crtc_state->use_dsb) + intel_dsb_vblank_evade(state, new_crtc_state->dsb_commit); if (intel_crtc_needs_color_update(new_crtc_state)) intel_color_commit_arm(new_crtc_state->dsb_commit, @@ -7231,19 +7281,27 @@ static void intel_atomic_dsb_finish(struct intel_atomic_state *state, skl_detach_scalers(new_crtc_state->dsb_commit, new_crtc_state); - if (!new_crtc_state->dsb_color_vblank) { - intel_dsb_wait_vblanks(new_crtc_state->dsb_commit, 1); - - intel_vrr_send_push(new_crtc_state->dsb_commit, new_crtc_state); - intel_dsb_wait_vblank_delay(state, new_crtc_state->dsb_commit); - intel_vrr_check_push_sent(new_crtc_state->dsb_commit, new_crtc_state); - intel_dsb_interrupt(new_crtc_state->dsb_commit); - } + /* Wa_18034343758 */ + if (new_crtc_state->use_flipq) + intel_flipq_unhalt_dmc(new_crtc_state->dsb_commit, crtc); } - if (new_crtc_state->dsb_color_vblank) + if (intel_color_uses_chained_dsb(new_crtc_state)) intel_dsb_chain(state, new_crtc_state->dsb_commit, - new_crtc_state->dsb_color_vblank, true); + new_crtc_state->dsb_color, true); + else if (intel_color_uses_gosub_dsb(new_crtc_state)) + intel_dsb_gosub(new_crtc_state->dsb_commit, + new_crtc_state->dsb_color); + + if (new_crtc_state->use_dsb && !intel_color_uses_chained_dsb(new_crtc_state)) { + intel_dsb_wait_vblanks(new_crtc_state->dsb_commit, 1); + + intel_vrr_send_push(new_crtc_state->dsb_commit, new_crtc_state); + intel_dsb_wait_vblank_delay(state, new_crtc_state->dsb_commit); + intel_vrr_check_push_sent(new_crtc_state->dsb_commit, + new_crtc_state); + intel_dsb_interrupt(new_crtc_state->dsb_commit); + } intel_dsb_finish(new_crtc_state->dsb_commit); } @@ -7366,6 +7424,7 @@ static void intel_atomic_commit_tail(struct intel_atomic_state *state) /* Now enable the clocks, plane, pipe, and connectors that we set up. */ display->funcs.display->commit_modeset_enables(state); + /* FIXME probably need to sequence this properly */ intel_program_dpkgc_latency(state); intel_wait_for_vblank_workers(state); @@ -7389,6 +7448,9 @@ static void intel_atomic_commit_tail(struct intel_atomic_state *state) if (!state->base.legacy_cursor_update && !new_crtc_state->use_dsb) intel_vrr_check_push_sent(NULL, new_crtc_state); + + if (new_crtc_state->use_flipq) + intel_flipq_disable(new_crtc_state); } /* @@ -7432,7 +7494,7 @@ static void intel_atomic_commit_tail(struct intel_atomic_state *state) * * FIXME get rid of this funny new->old swapping */ - old_crtc_state->dsb_color_vblank = fetch_and_zero(&new_crtc_state->dsb_color_vblank); + old_crtc_state->dsb_color = fetch_and_zero(&new_crtc_state->dsb_color); old_crtc_state->dsb_commit = fetch_and_zero(&new_crtc_state->dsb_commit); } @@ -7525,7 +7587,7 @@ static int intel_atomic_swap_state(struct intel_atomic_state *state) intel_atomic_swap_global_state(state); - intel_shared_dpll_swap_state(state); + intel_dpll_swap_state(state); intel_atomic_track_fbs(state); diff --git a/drivers/gpu/drm/i915/display/intel_display.h b/drivers/gpu/drm/i915/display/intel_display.h index 3b54a62c290af17f9867141f3613097fd62edab4..37e2ab301a80ef123bbdb4af9d4258b8fb6880ce 100644 --- a/drivers/gpu/drm/i915/display/intel_display.h +++ b/drivers/gpu/drm/i915/display/intel_display.h @@ -30,38 +30,21 @@ #include "i915_reg_defs.h" #include "intel_display_limits.h" -enum drm_scaling_filter; -struct dpll; struct drm_atomic_state; -struct drm_connector; struct drm_device; struct drm_display_mode; struct drm_encoder; -struct drm_file; -struct drm_format_info; -struct drm_framebuffer; -struct drm_i915_private; -struct drm_mode_fb_cmd2; struct drm_modeset_acquire_ctx; -struct drm_plane; -struct drm_plane_state; -struct i915_address_space; -struct i915_gtt_view; struct intel_atomic_state; struct intel_crtc; struct intel_crtc_state; struct intel_digital_port; struct intel_display; -struct intel_dp; struct intel_encoder; -struct intel_initial_plane_config; struct intel_link_m_n; struct intel_plane; struct intel_plane_state; struct intel_power_domain_mask; -struct pci_dev; -struct work_struct; - #define pipe_name(p) ((p) + 'A') @@ -452,10 +435,10 @@ void intel_enable_transcoder(const struct intel_crtc_state *new_crtc_state); void intel_disable_transcoder(const struct intel_crtc_state *old_crtc_state); void i830_enable_pipe(struct intel_display *display, enum pipe pipe); void i830_disable_pipe(struct intel_display *display, enum pipe pipe); -int vlv_get_hpll_vco(struct drm_i915_private *dev_priv); -int vlv_get_cck_clock(struct drm_i915_private *dev_priv, +int vlv_get_hpll_vco(struct drm_device *drm); +int vlv_get_cck_clock(struct drm_device *drm, const char *name, u32 reg, int ref_freq); -int vlv_get_cck_clock_hpll(struct drm_i915_private *dev_priv, +int vlv_get_cck_clock_hpll(struct drm_device *drm, const char *name, u32 reg); bool intel_has_pending_fb_unpin(struct intel_display *display); void intel_encoder_destroy(struct drm_encoder *encoder); @@ -524,6 +507,9 @@ void intel_plane_fixup_bitmasks(struct intel_crtc_state *crtc_state); bool intel_crtc_vrr_disabling(struct intel_atomic_state *state, struct intel_crtc *crtc); +int intel_display_min_pipe_bpp(void); +int intel_display_max_pipe_bpp(struct intel_display *display); + /* modesetting */ int intel_modeset_pipes_in_mask_early(struct intel_atomic_state *state, const char *reason, u8 pipe_mask); diff --git a/drivers/gpu/drm/i915/display/intel_display_conversion.c b/drivers/gpu/drm/i915/display/intel_display_conversion.c index 0578b68404da74f171df8980b13564c18fc35646..4d565935e2cc7ed481b422c1b9f1796dcb57dff6 100644 --- a/drivers/gpu/drm/i915/display/intel_display_conversion.c +++ b/drivers/gpu/drm/i915/display/intel_display_conversion.c @@ -2,10 +2,11 @@ /* Copyright © 2024 Intel Corporation */ #include "i915_drv.h" +#include "intel_display_conversion.h" struct intel_display *__i915_to_display(struct drm_i915_private *i915) { - return &i915->display; + return i915->display; } struct intel_display *__drm_to_display(struct drm_device *drm) diff --git a/drivers/gpu/drm/i915/display/intel_display_core.h b/drivers/gpu/drm/i915/display/intel_display_core.h index b4937e1023606a80b27a6c2d50a0b82eb66454f1..8c226406c5cd0a47bd5cd9ddf668e0825103c06f 100644 --- a/drivers/gpu/drm/i915/display/intel_display_core.h +++ b/drivers/gpu/drm/i915/display/intel_display_core.h @@ -42,7 +42,7 @@ struct intel_color_funcs; struct intel_crtc; struct intel_crtc_state; struct intel_dmc; -struct intel_dpll_funcs; +struct intel_dpll_global_funcs; struct intel_dpll_mgr; struct intel_fbdev; struct intel_fdi_funcs; @@ -122,11 +122,11 @@ struct intel_audio { * intel_{prepare,enable,disable}_shared_dpll. Must be global rather than per * dpll, because on some platforms plls share registers. */ -struct intel_dpll { +struct intel_dpll_global { struct mutex lock; - int num_shared_dpll; - struct intel_shared_dpll shared_dplls[I915_NUM_PLLS]; + int num_dpll; + struct intel_dpll dplls[I915_NUM_PLLS]; const struct intel_dpll_mgr *mgr; struct { @@ -300,7 +300,7 @@ struct intel_display { const struct intel_cdclk_funcs *cdclk; /* Display pll funcs */ - const struct intel_dpll_funcs *dpll; + const struct intel_dpll_global_funcs *dpll; /* irq display functions */ const struct intel_hotplug_funcs *hotplug; @@ -479,6 +479,12 @@ struct intel_display { u32 pipestat_irq_mask[I915_MAX_PIPES]; } irq; + struct { + /* protected by wm.wm_mutex */ + u16 linetime[I915_MAX_PIPES]; + bool disable[I915_MAX_PIPES]; + } pkgc; + struct { wait_queue_head_t waitqueue; @@ -538,6 +544,11 @@ struct intel_display { u32 block_time_us; } sagv; + struct { + /* LPT/WPT IOSF sideband protection */ + struct mutex lock; + } sbi; + struct { /* * DG2: Mask of PHYs that were not calibrated by the firmware @@ -565,12 +576,15 @@ struct intel_display { /* hipri wq for commit cleanups */ struct workqueue_struct *cleanup; + + /* unordered workqueue for all display unordered work */ + struct workqueue_struct *unordered; } wq; /* Grouping using named structs. Keep sorted. */ struct drm_dp_tunnel_mgr *dp_tunnel_mgr; struct intel_audio audio; - struct intel_dpll dpll; + struct intel_dpll_global dpll; struct intel_fbc *fbc[I915_MAX_FBCS]; struct intel_frontbuffer_tracking fb_tracking; struct intel_hotplug hotplug; diff --git a/drivers/gpu/drm/i915/display/intel_display_debugfs.c b/drivers/gpu/drm/i915/display/intel_display_debugfs.c index 8d0a1779dd193e199cf58f780196bcbc31f00d24..ce3f9810c42d29b00d3a6f9f189759dc1df460c1 100644 --- a/drivers/gpu/drm/i915/display/intel_display_debugfs.c +++ b/drivers/gpu/drm/i915/display/intel_display_debugfs.c @@ -4,6 +4,7 @@ */ #include +#include #include #include @@ -13,7 +14,6 @@ #include #include "hsw_ips.h" -#include "i915_irq.h" #include "i915_reg.h" #include "i9xx_wm_regs.h" #include "intel_alpm.h" @@ -25,6 +25,7 @@ #include "intel_display_debugfs_params.h" #include "intel_display_power.h" #include "intel_display_power_well.h" +#include "intel_display_regs.h" #include "intel_display_rpm.h" #include "intel_display_types.h" #include "intel_dmc.h" @@ -39,6 +40,7 @@ #include "intel_hdcp.h" #include "intel_hdmi.h" #include "intel_hotplug.h" +#include "intel_link_bw.h" #include "intel_panel.h" #include "intel_pps.h" #include "intel_psr.h" @@ -618,7 +620,7 @@ static int i915_shared_dplls_info(struct seq_file *m, void *unused) { struct intel_display *display = node_to_intel_display(m->private); struct drm_printer p = drm_seq_file_printer(m); - struct intel_shared_dpll *pll; + struct intel_dpll *pll; int i; drm_modeset_lock_all(display->drm); @@ -627,7 +629,7 @@ static int i915_shared_dplls_info(struct seq_file *m, void *unused) display->dpll.ref_clks.nssc, display->dpll.ref_clks.ssc); - for_each_shared_dpll(display, pll, i) { + for_each_dpll(display, pll, i) { drm_printf(&p, "DPLL%i: %s, id: %i\n", pll->index, pll->info->name, pll->info->id); drm_printf(&p, " pipe_mask: 0x%x, active: 0x%x, on: %s\n", @@ -972,7 +974,7 @@ static ssize_t i915_dsc_fec_support_write(struct file *file, return ret; drm_dbg(display->drm, "Got %s for DSC Enable\n", - (dsc_enable) ? "true" : "false"); + str_true_false(dsc_enable)); intel_dp->force_dsc_en = dsc_enable; *offp += len; @@ -1183,7 +1185,7 @@ static ssize_t i915_dsc_fractional_bpp_write(struct file *file, return ret; drm_dbg(display->drm, "Got %s for DSC Fractional BPP Enable\n", - (dsc_fractional_bpp_enable) ? "true" : "false"); + str_true_false(dsc_fractional_bpp_enable)); intel_dp->force_dsc_fractional_bpp_en = dsc_fractional_bpp_enable; *offp += len; @@ -1325,6 +1327,7 @@ void intel_connector_debugfs_add(struct intel_connector *connector) intel_psr_connector_debugfs_add(connector); intel_alpm_lobf_debugfs_add(connector); intel_dp_link_training_debugfs_add(connector); + intel_link_bw_connector_debugfs_add(connector); if (DISPLAY_VER(display) >= 11 && ((connector_type == DRM_MODE_CONNECTOR_DisplayPort && !connector->mst.dp) || diff --git a/drivers/gpu/drm/i915/display/intel_display_device.c b/drivers/gpu/drm/i915/display/intel_display_device.c index 90d7145986643c3011ecba9f2bb4daf59048564b..089cffabbad57b64f36170494e59aac4f2f4a686 100644 --- a/drivers/gpu/drm/i915/display/intel_display_device.c +++ b/drivers/gpu/drm/i915/display/intel_display_device.c @@ -18,6 +18,7 @@ #include "intel_display_params.h" #include "intel_display_power.h" #include "intel_display_reg_defs.h" +#include "intel_display_regs.h" #include "intel_display_types.h" #include "intel_fbc.h" #include "intel_step.h" @@ -1479,6 +1480,7 @@ static const struct { { 14, 1, &xe2_hpd_display }, { 20, 0, &xe2_lpd_display }, { 30, 0, &xe2_lpd_display }, + { 30, 2, &xe2_lpd_display }, }; static const struct intel_display_device_info * @@ -1621,13 +1623,17 @@ static void display_platforms_or(struct intel_display_platforms *dst, struct intel_display *intel_display_device_probe(struct pci_dev *pdev) { - struct intel_display *display = to_intel_display(pdev); + struct intel_display *display; const struct intel_display_device_info *info; struct intel_display_ip_ver ip_ver = {}; const struct platform_desc *desc; const struct subplatform_desc *subdesc; enum intel_step step; + display = kzalloc(sizeof(*display), GFP_KERNEL); + if (!display) + return ERR_PTR(-ENOMEM); + /* Add drm device backpointer as early as possible. */ display->drm = pci_get_drvdata(pdev); @@ -1708,7 +1714,11 @@ no_display: void intel_display_device_remove(struct intel_display *display) { + if (!display) + return; + intel_display_params_free(&display->params); + kfree(display); } static void __intel_display_device_info_runtime_init(struct intel_display *display) diff --git a/drivers/gpu/drm/i915/display/intel_display_device.h b/drivers/gpu/drm/i915/display/intel_display_device.h index 87c666792c0dacb06c3c5872490b1fa55c281aec..4308822f0415db14f8819533f66fd008cce46231 100644 --- a/drivers/gpu/drm/i915/display/intel_display_device.h +++ b/drivers/gpu/drm/i915/display/intel_display_device.h @@ -157,6 +157,7 @@ struct intel_display_platforms { #define HAS_DMC(__display) (DISPLAY_RUNTIME_INFO(__display)->has_dmc) #define HAS_DMC_WAKELOCK(__display) (DISPLAY_VER(__display) >= 20) #define HAS_DOUBLE_BUFFERED_M_N(__display) (DISPLAY_VER(__display) >= 9 || (__display)->platform.broadwell) +#define HAS_DOUBLE_BUFFERED_LUT(__display) (DISPLAY_VER(__display) >= 30) #define HAS_DOUBLE_WIDE(__display) (DISPLAY_VER(__display) < 4) #define HAS_DP20(__display) ((__display)->platform.dg2 || DISPLAY_VER(__display) >= 14) #define HAS_DPT(__display) (DISPLAY_VER(__display) >= 13) @@ -172,6 +173,7 @@ struct intel_display_platforms { #define HAS_GMBUS_BURST_READ(__display) (DISPLAY_VER(__display) >= 10 || (__display)->platform.kabylake) #define HAS_GMBUS_IRQ(__display) (DISPLAY_VER(__display) >= 4) #define HAS_GMCH(__display) (DISPLAY_INFO(__display)->has_gmch) +#define HAS_FDI(__display) (IS_DISPLAY_VER((__display), 5, 8) && !HAS_GMCH(__display)) #define HAS_HOTPLUG(__display) (DISPLAY_INFO(__display)->has_hotplug) #define HAS_HW_SAGV_WM(__display) (DISPLAY_VER(__display) >= 13 && !(__display)->platform.dgfx) #define HAS_IPC(__display) (DISPLAY_INFO(__display)->has_ipc) @@ -181,6 +183,7 @@ struct intel_display_platforms { #define HAS_MBUS_JOINING(__display) ((__display)->platform.alderlake_p || DISPLAY_VER(__display) >= 14) #define HAS_MSO(__display) (DISPLAY_VER(__display) >= 12) #define HAS_OVERLAY(__display) (DISPLAY_INFO(__display)->has_overlay) +#define HAS_PIPEDMC(__display) (DISPLAY_VER(__display) >= 12) #define HAS_PSR(__display) (DISPLAY_INFO(__display)->has_psr) #define HAS_PSR_HW_TRACKING(__display) (DISPLAY_INFO(__display)->has_psr_hw_tracking) #define HAS_PSR2_SEL_FETCH(__display) (DISPLAY_VER(__display) >= 12) @@ -189,9 +192,8 @@ struct intel_display_platforms { #define HAS_TRANSCODER(__display, trans) ((DISPLAY_RUNTIME_INFO(__display)->cpu_transcoder_mask & \ BIT(trans)) != 0) #define HAS_UNCOMPRESSED_JOINER(__display) (DISPLAY_VER(__display) >= 13) -#define HAS_ULTRAJOINER(__display) ((DISPLAY_VER(__display) >= 20 || \ - ((__display)->platform.dgfx && DISPLAY_VER(__display) == 14)) && \ - HAS_DSC(__display)) +#define HAS_ULTRAJOINER(__display) (((__display)->platform.dgfx && \ + DISPLAY_VER(__display) == 14) && HAS_DSC(__display)) #define HAS_VRR(__display) (DISPLAY_VER(__display) >= 11) #define INTEL_NUM_PIPES(__display) (hweight8(DISPLAY_RUNTIME_INFO(__display)->pipe_mask)) #define OVERLAY_NEEDS_PHYSICAL(__display) (DISPLAY_INFO(__display)->overlay_needs_physical) diff --git a/drivers/gpu/drm/i915/display/intel_display_driver.c b/drivers/gpu/drm/i915/display/intel_display_driver.c index 411fe7b918a739c0195b764568c0383d70d75741..8586ba102605afe8e564d757cb248252584650b3 100644 --- a/drivers/gpu/drm/i915/display/intel_display_driver.c +++ b/drivers/gpu/drm/i915/display/intel_display_driver.c @@ -27,6 +27,7 @@ #include "intel_cdclk.h" #include "intel_color.h" #include "intel_crtc.h" +#include "intel_display_core.h" #include "intel_display_debugfs.h" #include "intel_display_driver.h" #include "intel_display_irq.h" @@ -43,6 +44,7 @@ #include "intel_fbc.h" #include "intel_fbdev.h" #include "intel_fdi.h" +#include "intel_flipq.h" #include "intel_gmbus.h" #include "intel_hdcp.h" #include "intel_hotplug.h" @@ -83,16 +85,10 @@ bool intel_display_driver_probe_defer(struct pci_dev *pdev) void intel_display_driver_init_hw(struct intel_display *display) { - struct intel_cdclk_state *cdclk_state; - if (!HAS_DISPLAY(display)) return; - cdclk_state = to_intel_cdclk_state(display->cdclk.obj.state); - - intel_update_cdclk(display); - intel_cdclk_dump_config(display, &display->cdclk.hw, "Current CDCLK"); - cdclk_state->logical = cdclk_state->actual = display->cdclk.hw; + intel_cdclk_read_hw(display); intel_display_wa_apply(display); } @@ -241,12 +237,16 @@ int intel_display_driver_probe_noirq(struct intel_display *display) if (!HAS_DISPLAY(display)) return 0; - intel_dmc_init(display); + display->hotplug.dp_wq = alloc_ordered_workqueue("intel-dp", 0); + if (!display->hotplug.dp_wq) { + ret = -ENOMEM; + goto cleanup_vga_client_pw_domain_dmc; + } display->wq.modeset = alloc_ordered_workqueue("i915_modeset", 0); if (!display->wq.modeset) { ret = -ENOMEM; - goto cleanup_vga_client_pw_domain_dmc; + goto cleanup_wq_dp; } display->wq.flip = alloc_workqueue("i915_flip", WQ_HIGHPRI | @@ -262,27 +262,35 @@ int intel_display_driver_probe_noirq(struct intel_display *display) goto cleanup_wq_flip; } + display->wq.unordered = alloc_workqueue("display_unordered", 0, 0); + if (!display->wq.unordered) { + ret = -ENOMEM; + goto cleanup_wq_cleanup; + } + + intel_dmc_init(display); + intel_mode_config_init(display); ret = intel_cdclk_init(display); if (ret) - goto cleanup_wq_cleanup; + goto cleanup_wq_unordered; ret = intel_color_init(display); if (ret) - goto cleanup_wq_cleanup; + goto cleanup_wq_unordered; ret = intel_dbuf_init(display); if (ret) - goto cleanup_wq_cleanup; + goto cleanup_wq_unordered; ret = intel_bw_init(display); if (ret) - goto cleanup_wq_cleanup; + goto cleanup_wq_unordered; ret = intel_pmdemand_init(display); if (ret) - goto cleanup_wq_cleanup; + goto cleanup_wq_unordered; intel_init_quirks(display); @@ -290,12 +298,16 @@ int intel_display_driver_probe_noirq(struct intel_display *display) return 0; +cleanup_wq_unordered: + destroy_workqueue(display->wq.unordered); cleanup_wq_cleanup: destroy_workqueue(display->wq.cleanup); cleanup_wq_flip: destroy_workqueue(display->wq.flip); cleanup_wq_modeset: destroy_workqueue(display->wq.modeset); +cleanup_wq_dp: + destroy_workqueue(display->hotplug.dp_wq); cleanup_vga_client_pw_domain_dmc: intel_dmc_fini(display); intel_power_domains_driver_remove(display); @@ -466,7 +478,7 @@ int intel_display_driver_probe_nogem(struct intel_display *display) } intel_plane_possible_crtcs_init(display); - intel_shared_dpll_init(display); + intel_dpll_init(display); intel_fdi_pll_freq_update(display); intel_update_czclk(display); @@ -526,6 +538,8 @@ int intel_display_driver_probe(struct intel_display *display) */ intel_hdcp_component_init(display); + intel_flipq_init(display); + /* * Force all active planes to recompute their states. So that on * mode_setcrtc after probe, all the intel_plane_state variables @@ -591,6 +605,7 @@ void intel_display_driver_remove(struct intel_display *display) flush_workqueue(display->wq.flip); flush_workqueue(display->wq.modeset); flush_workqueue(display->wq.cleanup); + flush_workqueue(display->wq.unordered); /* * MST topology needs to be suspended so we don't have any calls to @@ -603,8 +618,6 @@ void intel_display_driver_remove(struct intel_display *display) /* part #2: call after irq uninstall */ void intel_display_driver_remove_noirq(struct intel_display *display) { - struct drm_i915_private *i915 = to_i915(display->drm); - if (!HAS_DISPLAY(display)) return; @@ -619,7 +632,7 @@ void intel_display_driver_remove_noirq(struct intel_display *display) intel_unregister_dsm_handler(); /* flush any delayed tasks or pending work */ - flush_workqueue(i915->unordered_wq); + flush_workqueue(display->wq.unordered); intel_hdcp_component_fini(display); @@ -631,9 +644,11 @@ void intel_display_driver_remove_noirq(struct intel_display *display) intel_gmbus_teardown(display); + destroy_workqueue(display->hotplug.dp_wq); destroy_workqueue(display->wq.flip); destroy_workqueue(display->wq.modeset); destroy_workqueue(display->wq.cleanup); + destroy_workqueue(display->wq.unordered); intel_fbc_cleanup(display); } diff --git a/drivers/gpu/drm/i915/display/intel_display_irq.c b/drivers/gpu/drm/i915/display/intel_display_irq.c index 3e73832e5e8132447fd5a5719dce4d489a547afe..fb25ec8adae3ff91e804eaf15713fe90d53a3487 100644 --- a/drivers/gpu/drm/i915/display/intel_display_irq.c +++ b/drivers/gpu/drm/i915/display/intel_display_irq.c @@ -9,14 +9,15 @@ #include "i915_irq.h" #include "i915_reg.h" #include "icl_dsi_regs.h" -#include "intel_atomic_plane.h" #include "intel_crtc.h" #include "intel_de.h" #include "intel_display_irq.h" +#include "intel_display_regs.h" #include "intel_display_rpm.h" #include "intel_display_rps.h" #include "intel_display_trace.h" #include "intel_display_types.h" +#include "intel_dmc.h" #include "intel_dmc_wl.h" #include "intel_dp_aux.h" #include "intel_dsb.h" @@ -25,6 +26,7 @@ #include "intel_gmbus.h" #include "intel_hotplug_irq.h" #include "intel_pipe_crc_regs.h" +#include "intel_plane.h" #include "intel_pmdemand.h" #include "intel_psr.h" #include "intel_psr_regs.h" @@ -1016,7 +1018,15 @@ static u32 gen8_de_port_aux_mask(struct intel_display *display) static u32 gen8_de_pipe_fault_mask(struct intel_display *display) { - if (DISPLAY_VER(display) >= 14) + if (DISPLAY_VER(display) >= 20) + return MTL_PLANE_ATS_FAULT | + GEN9_PIPE_CURSOR_FAULT | + GEN11_PIPE_PLANE5_FAULT | + GEN9_PIPE_PLANE4_FAULT | + GEN9_PIPE_PLANE3_FAULT | + GEN9_PIPE_PLANE2_FAULT | + GEN9_PIPE_PLANE1_FAULT; + else if (DISPLAY_VER(display) >= 14) return MTL_PIPEDMC_ATS_FAULT | MTL_PLANE_ATS_FAULT | GEN12_PIPEDMC_FAULT | @@ -1418,7 +1428,8 @@ void gen8_de_irq_handler(struct intel_display *display, u32 master_ctl) iir = intel_de_read(display, GEN8_DE_PIPE_IIR(pipe)); if (!iir) { drm_err_ratelimited(display->drm, - "The master control interrupt lied (DE PIPE)!\n"); + "The master control interrupt lied (DE PIPE %c)!\n", + pipe_name(pipe)); continue; } @@ -1441,6 +1452,9 @@ void gen8_de_irq_handler(struct intel_display *display, u32 master_ctl) intel_dsb_irq_handler(display, pipe, INTEL_DSB_2); } + if (HAS_PIPEDMC(display) && iir & GEN12_PIPEDMC_INTERRUPT) + intel_pipedmc_irq_handler(display, pipe); + if (iir & GEN8_PIPE_CDCLK_CRC_DONE) hsw_pipe_crc_irq_handler(display, pipe); @@ -2258,6 +2272,10 @@ void gen8_de_irq_postinstall(struct intel_display *display) GEN12_DSB_INT(INTEL_DSB_1) | GEN12_DSB_INT(INTEL_DSB_2); + /* TODO figure PIPEDMC interrupts for pre-LNL */ + if (DISPLAY_VER(display) >= 20) + de_pipe_masked |= GEN12_PIPEDMC_INTERRUPT; + de_pipe_enables = de_pipe_masked | GEN8_PIPE_VBLANK | GEN8_PIPE_FIFO_UNDERRUN | gen8_de_pipe_flip_done_mask(display); diff --git a/drivers/gpu/drm/i915/display/intel_display_params.c b/drivers/gpu/drm/i915/display/intel_display_params.c index c4f1ab43fc0c864a97226e216b9178749749049a..75316247ee8a858168732a2dadbbb9cf090d759f 100644 --- a/drivers/gpu/drm/i915/display/intel_display_params.c +++ b/drivers/gpu/drm/i915/display/intel_display_params.c @@ -62,6 +62,9 @@ intel_display_param_named_unsafe(enable_dpt, bool, 0400, intel_display_param_named_unsafe(enable_dsb, bool, 0400, "Enable display state buffer (DSB) (default: true)"); +intel_display_param_named_unsafe(enable_flipq, bool, 0400, + "Enable DMC flip queue (default: false)"); + intel_display_param_named_unsafe(enable_sagv, bool, 0400, "Enable system agent voltage/frequency scaling (SAGV) (default: true)"); diff --git a/drivers/gpu/drm/i915/display/intel_display_params.h b/drivers/gpu/drm/i915/display/intel_display_params.h index 5317138e6044bfc2e929bb7ac7603470a86c9798..784e6bae86154fd15d65efb08286f65a048f2fff 100644 --- a/drivers/gpu/drm/i915/display/intel_display_params.h +++ b/drivers/gpu/drm/i915/display/intel_display_params.h @@ -31,6 +31,7 @@ struct drm_printer; param(int, enable_dc, -1, 0400) \ param(bool, enable_dpt, true, 0400) \ param(bool, enable_dsb, true, 0600) \ + param(bool, enable_flipq, false, 0600) \ param(bool, enable_sagv, true, 0600) \ param(int, disable_power_well, -1, 0400) \ param(bool, enable_ips, true, 0600) \ diff --git a/drivers/gpu/drm/i915/display/intel_display_power.c b/drivers/gpu/drm/i915/display/intel_display_power.c index 16356523816fb8438d88851ab2a19ccc1b7662ef..273054c2232524be08f464aa8ba682e9d41e6976 100644 --- a/drivers/gpu/drm/i915/display/intel_display_power.c +++ b/drivers/gpu/drm/i915/display/intel_display_power.c @@ -5,6 +5,8 @@ #include +#include "soc/intel_dram.h" + #include "i915_drv.h" #include "i915_irq.h" #include "i915_reg.h" @@ -16,6 +18,7 @@ #include "intel_display_power.h" #include "intel_display_power_map.h" #include "intel_display_power_well.h" +#include "intel_display_regs.h" #include "intel_display_rpm.h" #include "intel_display_types.h" #include "intel_dmc.h" @@ -1254,10 +1257,8 @@ static u32 hsw_read_dcomp(struct intel_display *display) static void hsw_write_dcomp(struct intel_display *display, u32 val) { - struct drm_i915_private *dev_priv = to_i915(display->drm); - if (display->platform.haswell) { - if (snb_pcode_write(&dev_priv->uncore, GEN6_PCODE_WRITE_D_COMP, val)) + if (intel_pcode_write(display->drm, GEN6_PCODE_WRITE_D_COMP, val)) drm_dbg_kms(display->drm, "Failed to write to D_COMP\n"); } else { intel_de_write(display, D_COMP_BDW, val); @@ -1604,9 +1605,7 @@ static const struct buddy_page_mask wa_1409767108_buddy_page_masks[] = { static void tgl_bw_buddy_init(struct intel_display *display) { - struct drm_i915_private *dev_priv = to_i915(display->drm); - enum intel_dram_type type = dev_priv->dram_info.type; - u8 num_channels = dev_priv->dram_info.num_channels; + const struct dram_info *dram_info = intel_dram_info(display->drm); const struct buddy_page_mask *table; unsigned long abox_mask = DISPLAY_INFO(display)->abox_mask; int config, i; @@ -1623,8 +1622,8 @@ static void tgl_bw_buddy_init(struct intel_display *display) table = tgl_buddy_page_masks; for (config = 0; table[config].page_mask != 0; config++) - if (table[config].num_channels == num_channels && - table[config].type == type) + if (table[config].num_channels == dram_info->num_channels && + table[config].type == dram_info->type) break; if (table[config].page_mask == 0) { @@ -1883,12 +1882,11 @@ static void vlv_cmnlane_wa(struct intel_display *display) static bool vlv_punit_is_power_gated(struct intel_display *display, u32 reg0) { - struct drm_i915_private *dev_priv = to_i915(display->drm); bool ret; - vlv_punit_get(dev_priv); - ret = (vlv_punit_read(dev_priv, reg0) & SSPM0_SSC_MASK) == SSPM0_SSC_PWR_GATE; - vlv_punit_put(dev_priv); + vlv_punit_get(display->drm); + ret = (vlv_punit_read(display->drm, reg0) & SSPM0_SSC_MASK) == SSPM0_SSC_PWR_GATE; + vlv_punit_put(display->drm); return ret; } diff --git a/drivers/gpu/drm/i915/display/intel_display_power_map.c b/drivers/gpu/drm/i915/display/intel_display_power_map.c index ab1163744bc5953da900cbd935155b4050435f96..77268802b55eda64db1ad7997f8c7fdb15fdb9e2 100644 --- a/drivers/gpu/drm/i915/display/intel_display_power_map.c +++ b/drivers/gpu/drm/i915/display/intel_display_power_map.c @@ -5,12 +5,12 @@ #include -#include "i915_reg.h" #include "intel_display_core.h" #include "intel_display_power_map.h" #include "intel_display_power_well.h" +#include "intel_display_regs.h" #include "intel_display_types.h" -#include "vlv_sideband_reg.h" +#include "vlv_iosf_sb_reg.h" #define __LIST_INLINE_ELEMS(__elem_type, ...) \ ((__elem_type[]) { __VA_ARGS__ }) diff --git a/drivers/gpu/drm/i915/display/intel_display_power_well.c b/drivers/gpu/drm/i915/display/intel_display_power_well.c index b104bce0e14df676059e8376eb6955c557d7a2b9..48cac225a809f9a796a52967a03595c26b65d3d0 100644 --- a/drivers/gpu/drm/i915/display/intel_display_power_well.c +++ b/drivers/gpu/drm/i915/display/intel_display_power_well.c @@ -13,6 +13,7 @@ #include "intel_de.h" #include "intel_display_irq.h" #include "intel_display_power_well.h" +#include "intel_display_regs.h" #include "intel_display_rpm.h" #include "intel_display_types.h" #include "intel_dkl_phy.h" @@ -30,8 +31,20 @@ #include "intel_vga.h" #include "skl_watermark.h" #include "vlv_dpio_phy_regs.h" +#include "vlv_iosf_sb_reg.h" #include "vlv_sideband.h" -#include "vlv_sideband_reg.h" + +/* + * PG0 is HW controlled, so doesn't have a corresponding power well control knob + * + * {ICL,SKL}_DISP_PW1_IDX..{ICL,SKL}_DISP_PW4_IDX -> PG1..PG4 + */ +static enum skl_power_gate pw_idx_to_pg(struct intel_display *display, int pw_idx) +{ + int pw1_idx = DISPLAY_VER(display) >= 11 ? ICL_PW_CTL_IDX_PW_1 : SKL_PW_CTL_IDX_PW_1; + + return pw_idx - pw1_idx + SKL_PG1; +} struct i915_power_well_regs { i915_reg_t bios; @@ -307,8 +320,8 @@ static void hsw_wait_for_power_well_disable(struct intel_display *display, { const struct i915_power_well_regs *regs = power_well->desc->ops->regs; int pw_idx = i915_power_well_instance(power_well)->hsw.idx; - bool disabled; u32 reqs; + int ret; /* * Bspec doesn't require waiting for PWs to get disabled, but still do @@ -319,12 +332,18 @@ static void hsw_wait_for_power_well_disable(struct intel_display *display, * Skip the wait in case any of the request bits are set and print a * diagnostic message. */ - wait_for((disabled = !(intel_de_read(display, regs->driver) & - HSW_PWR_WELL_CTL_STATE(pw_idx))) || - (reqs = hsw_power_well_requesters(display, regs, pw_idx)), 1); - if (disabled) + reqs = hsw_power_well_requesters(display, regs, pw_idx); + + ret = intel_de_wait_for_clear(display, regs->driver, + HSW_PWR_WELL_CTL_STATE(pw_idx), + reqs ? 0 : 1); + if (!ret) return; + /* Refresh requesters in case they popped up during the wait. */ + if (!reqs) + reqs = hsw_power_well_requesters(display, regs, pw_idx); + drm_dbg_kms(display->drm, "%s forced on (bios:%d driver:%d kvmr:%d debug:%d)\n", intel_power_well_name(power_well), @@ -349,8 +368,7 @@ static void hsw_power_well_enable(struct intel_display *display, if (power_well->desc->has_fuses) { enum skl_power_gate pg; - pg = DISPLAY_VER(display) >= 11 ? ICL_PW_CTL_IDX_TO_PG(pw_idx) : - SKL_PW_CTL_IDX_TO_PG(pw_idx); + pg = pw_idx_to_pg(display, pw_idx); /* Wa_16013190616:adlp */ if (display->platform.alderlake_p && pg == SKL_PG1) @@ -374,8 +392,8 @@ static void hsw_power_well_enable(struct intel_display *display, if (power_well->desc->has_fuses) { enum skl_power_gate pg; - pg = DISPLAY_VER(display) >= 11 ? ICL_PW_CTL_IDX_TO_PG(pw_idx) : - SKL_PW_CTL_IDX_TO_PG(pw_idx); + pg = pw_idx_to_pg(display, pw_idx); + gen9_wait_for_power_well_fuses(display, pg); } @@ -485,8 +503,7 @@ static void icl_tc_cold_exit(struct intel_display *display) int ret, tries = 0; while (1) { - ret = snb_pcode_write_timeout(&i915->uncore, ICL_PCODE_EXIT_TCCOLD, 0, - 250, 1); + ret = intel_pcode_write(display->drm, ICL_PCODE_EXIT_TCCOLD, 0); if (ret != -EAGAIN || ++tries == 3) break; msleep(1); @@ -809,7 +826,6 @@ static void tgl_disable_dc3co(struct intel_display *display) static void assert_can_enable_dc5(struct intel_display *display) { - struct drm_i915_private __maybe_unused *dev_priv = to_i915(display->drm); enum i915_power_well_id high_pg; /* Power wells at this level and above must be disabled for DC5 entry */ @@ -829,7 +845,7 @@ static void assert_can_enable_dc5(struct intel_display *display) assert_display_rpm_held(display); - assert_dmc_loaded(display); + assert_main_dmc_loaded(display); } void gen9_enable_dc5(struct intel_display *display) @@ -860,7 +876,7 @@ static void assert_can_enable_dc6(struct intel_display *display) DC_STATE_EN_UPTO_DC6), "DC6 already programmed to be enabled.\n"); - assert_dmc_loaded(display); + assert_main_dmc_loaded(display); } void skl_enable_dc6(struct intel_display *display) @@ -1102,7 +1118,6 @@ static void i830_pipes_power_well_sync_hw(struct intel_display *display, static void vlv_set_power_well(struct intel_display *display, struct i915_power_well *power_well, bool enable) { - struct drm_i915_private *dev_priv = to_i915(display->drm); int pw_idx = i915_power_well_instance(power_well)->vlv.idx; u32 mask; u32 state; @@ -1112,29 +1127,29 @@ static void vlv_set_power_well(struct intel_display *display, state = enable ? PUNIT_PWRGT_PWR_ON(pw_idx) : PUNIT_PWRGT_PWR_GATE(pw_idx); - vlv_punit_get(dev_priv); + vlv_punit_get(display->drm); #define COND \ - ((vlv_punit_read(dev_priv, PUNIT_REG_PWRGT_STATUS) & mask) == state) + ((vlv_punit_read(display->drm, PUNIT_REG_PWRGT_STATUS) & mask) == state) if (COND) goto out; - ctrl = vlv_punit_read(dev_priv, PUNIT_REG_PWRGT_CTRL); + ctrl = vlv_punit_read(display->drm, PUNIT_REG_PWRGT_CTRL); ctrl &= ~mask; ctrl |= state; - vlv_punit_write(dev_priv, PUNIT_REG_PWRGT_CTRL, ctrl); + vlv_punit_write(display->drm, PUNIT_REG_PWRGT_CTRL, ctrl); if (wait_for(COND, 100)) drm_err(display->drm, "timeout setting power well state %08x (%08x)\n", state, - vlv_punit_read(dev_priv, PUNIT_REG_PWRGT_CTRL)); + vlv_punit_read(display->drm, PUNIT_REG_PWRGT_CTRL)); #undef COND out: - vlv_punit_put(dev_priv); + vlv_punit_put(display->drm); } static void vlv_power_well_enable(struct intel_display *display, @@ -1152,7 +1167,6 @@ static void vlv_power_well_disable(struct intel_display *display, static bool vlv_power_well_enabled(struct intel_display *display, struct i915_power_well *power_well) { - struct drm_i915_private *dev_priv = to_i915(display->drm); int pw_idx = i915_power_well_instance(power_well)->vlv.idx; bool enabled = false; u32 mask; @@ -1162,9 +1176,9 @@ static bool vlv_power_well_enabled(struct intel_display *display, mask = PUNIT_PWRGT_MASK(pw_idx); ctrl = PUNIT_PWRGT_PWR_ON(pw_idx); - vlv_punit_get(dev_priv); + vlv_punit_get(display->drm); - state = vlv_punit_read(dev_priv, PUNIT_REG_PWRGT_STATUS) & mask; + state = vlv_punit_read(display->drm, PUNIT_REG_PWRGT_STATUS) & mask; /* * We only ever set the power-on and power-gate states, anything * else is unexpected. @@ -1178,10 +1192,10 @@ static bool vlv_power_well_enabled(struct intel_display *display, * A transient state at this point would mean some unexpected party * is poking at the power controls too. */ - ctrl = vlv_punit_read(dev_priv, PUNIT_REG_PWRGT_CTRL) & mask; + ctrl = vlv_punit_read(display->drm, PUNIT_REG_PWRGT_CTRL) & mask; drm_WARN_ON(display->drm, ctrl != state); - vlv_punit_put(dev_priv); + vlv_punit_put(display->drm); return enabled; } @@ -1437,7 +1451,6 @@ static void assert_chv_phy_status(struct intel_display *display) static void chv_dpio_cmn_power_well_enable(struct intel_display *display, struct i915_power_well *power_well) { - struct drm_i915_private *dev_priv = to_i915(display->drm); enum i915_power_well_id id = i915_power_well_instance(power_well)->id; enum dpio_phy phy; u32 tmp; @@ -1461,30 +1474,30 @@ static void chv_dpio_cmn_power_well_enable(struct intel_display *display, drm_err(display->drm, "Display PHY %d is not power up\n", phy); - vlv_dpio_get(dev_priv); + vlv_dpio_get(display->drm); /* Enable dynamic power down */ - tmp = vlv_dpio_read(dev_priv, phy, CHV_CMN_DW28); + tmp = vlv_dpio_read(display->drm, phy, CHV_CMN_DW28); tmp |= DPIO_DYNPWRDOWNEN_CH0 | DPIO_CL1POWERDOWNEN | DPIO_SUS_CLK_CONFIG_GATE_CLKREQ; - vlv_dpio_write(dev_priv, phy, CHV_CMN_DW28, tmp); + vlv_dpio_write(display->drm, phy, CHV_CMN_DW28, tmp); if (id == VLV_DISP_PW_DPIO_CMN_BC) { - tmp = vlv_dpio_read(dev_priv, phy, CHV_CMN_DW6_CH1); + tmp = vlv_dpio_read(display->drm, phy, CHV_CMN_DW6_CH1); tmp |= DPIO_DYNPWRDOWNEN_CH1; - vlv_dpio_write(dev_priv, phy, CHV_CMN_DW6_CH1, tmp); + vlv_dpio_write(display->drm, phy, CHV_CMN_DW6_CH1, tmp); } else { /* * Force the non-existing CL2 off. BXT does this * too, so maybe it saves some power even though * CL2 doesn't exist? */ - tmp = vlv_dpio_read(dev_priv, phy, CHV_CMN_DW30); + tmp = vlv_dpio_read(display->drm, phy, CHV_CMN_DW30); tmp |= DPIO_CL2_LDOFUSE_PWRENB; - vlv_dpio_write(dev_priv, phy, CHV_CMN_DW30, tmp); + vlv_dpio_write(display->drm, phy, CHV_CMN_DW30, tmp); } - vlv_dpio_put(dev_priv); + vlv_dpio_put(display->drm); display->power.chv_phy_control |= PHY_COM_LANE_RESET_DEASSERT(phy); intel_de_write(display, DISPLAY_PHY_CONTROL, @@ -1535,7 +1548,6 @@ static void chv_dpio_cmn_power_well_disable(struct intel_display *display, static void assert_chv_phy_powergate(struct intel_display *display, enum dpio_phy phy, enum dpio_channel ch, bool override, unsigned int mask) { - struct drm_i915_private *dev_priv = to_i915(display->drm); u32 reg, val, expected, actual; /* @@ -1553,9 +1565,9 @@ static void assert_chv_phy_powergate(struct intel_display *display, enum dpio_ph else reg = CHV_CMN_DW6_CH1; - vlv_dpio_get(dev_priv); - val = vlv_dpio_read(dev_priv, phy, reg); - vlv_dpio_put(dev_priv); + vlv_dpio_get(display->drm); + val = vlv_dpio_read(display->drm, phy, reg); + vlv_dpio_put(display->drm); /* * This assumes !override is only used when the port is disabled. @@ -1665,14 +1677,13 @@ void chv_phy_powergate_lanes(struct intel_encoder *encoder, static bool chv_pipe_power_well_enabled(struct intel_display *display, struct i915_power_well *power_well) { - struct drm_i915_private *dev_priv = to_i915(display->drm); enum pipe pipe = PIPE_A; bool enabled; u32 state, ctrl; - vlv_punit_get(dev_priv); + vlv_punit_get(display->drm); - state = vlv_punit_read(dev_priv, PUNIT_REG_DSPSSPM) & DP_SSS_MASK(pipe); + state = vlv_punit_read(display->drm, PUNIT_REG_DSPSSPM) & DP_SSS_MASK(pipe); /* * We only ever set the power-on and power-gate states, anything * else is unexpected. @@ -1685,10 +1696,10 @@ static bool chv_pipe_power_well_enabled(struct intel_display *display, * A transient state at this point would mean some unexpected party * is poking at the power controls too. */ - ctrl = vlv_punit_read(dev_priv, PUNIT_REG_DSPSSPM) & DP_SSC_MASK(pipe); + ctrl = vlv_punit_read(display->drm, PUNIT_REG_DSPSSPM) & DP_SSC_MASK(pipe); drm_WARN_ON(display->drm, ctrl << 16 != state); - vlv_punit_put(dev_priv); + vlv_punit_put(display->drm); return enabled; } @@ -1697,36 +1708,35 @@ static void chv_set_pipe_power_well(struct intel_display *display, struct i915_power_well *power_well, bool enable) { - struct drm_i915_private *dev_priv = to_i915(display->drm); enum pipe pipe = PIPE_A; u32 state; u32 ctrl; state = enable ? DP_SSS_PWR_ON(pipe) : DP_SSS_PWR_GATE(pipe); - vlv_punit_get(dev_priv); + vlv_punit_get(display->drm); #define COND \ - ((vlv_punit_read(dev_priv, PUNIT_REG_DSPSSPM) & DP_SSS_MASK(pipe)) == state) + ((vlv_punit_read(display->drm, PUNIT_REG_DSPSSPM) & DP_SSS_MASK(pipe)) == state) if (COND) goto out; - ctrl = vlv_punit_read(dev_priv, PUNIT_REG_DSPSSPM); + ctrl = vlv_punit_read(display->drm, PUNIT_REG_DSPSSPM); ctrl &= ~DP_SSC_MASK(pipe); ctrl |= enable ? DP_SSC_PWR_ON(pipe) : DP_SSC_PWR_GATE(pipe); - vlv_punit_write(dev_priv, PUNIT_REG_DSPSSPM, ctrl); + vlv_punit_write(display->drm, PUNIT_REG_DSPSSPM, ctrl); if (wait_for(COND, 100)) drm_err(display->drm, "timeout setting power well state %08x (%08x)\n", state, - vlv_punit_read(dev_priv, PUNIT_REG_DSPSSPM)); + vlv_punit_read(display->drm, PUNIT_REG_DSPSSPM)); #undef COND out: - vlv_punit_put(dev_priv); + vlv_punit_put(display->drm); } static void chv_pipe_power_well_sync_hw(struct intel_display *display, @@ -1772,7 +1782,7 @@ tgl_tc_cold_request(struct intel_display *display, bool block) * Spec states that we should timeout the request after 200us * but the function below will timeout after 500us */ - ret = snb_pcode_read(&i915->uncore, TGL_PCODE_TCCOLD, &low_val, &high_val); + ret = intel_pcode_read(display->drm, TGL_PCODE_TCCOLD, &low_val, &high_val); if (ret == 0) { if (block && (low_val & TGL_PCODE_EXIT_TCCOLD_DATA_L_EXIT_FAILED)) diff --git a/drivers/gpu/drm/i915/display/intel_display_regs.h b/drivers/gpu/drm/i915/display/intel_display_regs.h new file mode 100644 index 0000000000000000000000000000000000000000..7bd09d981cd2d065b631043be26b11b8c478550c --- /dev/null +++ b/drivers/gpu/drm/i915/display/intel_display_regs.h @@ -0,0 +1,2932 @@ +/* SPDX-License-Identifier: MIT */ +/* Copyright © 2025 Intel Corporation */ + +#ifndef __INTEL_DISPLAY_REGS_H__ +#define __INTEL_DISPLAY_REGS_H__ + +#include "intel_display_reg_defs.h" + +#define _GEN7_PIPEA_DE_LOAD_SL 0x70068 +#define _GEN7_PIPEB_DE_LOAD_SL 0x71068 +#define GEN7_PIPE_DE_LOAD_SL(pipe) _MMIO_PIPE(pipe, _GEN7_PIPEA_DE_LOAD_SL, _GEN7_PIPEB_DE_LOAD_SL) + +#define DPIO_CTL _MMIO(VLV_DISPLAY_BASE + 0x2110) +#define DPIO_MODSEL1 (1 << 3) /* if ref clk b == 27 */ +#define DPIO_MODSEL0 (1 << 2) /* if ref clk a == 27 */ +#define DPIO_SFR_BYPASS (1 << 1) +#define DPIO_CMNRST (1 << 0) + +#define BXT_P_CR_GT_DISP_PWRON _MMIO(0x138090) +#define MIPIO_RST_CTRL (1 << 2) + +#define _BXT_PHY_CTL_DDI_A 0x64C00 +#define _BXT_PHY_CTL_DDI_B 0x64C10 +#define _BXT_PHY_CTL_DDI_C 0x64C20 +#define BXT_PHY_CMNLANE_POWERDOWN_ACK (1 << 10) +#define BXT_PHY_LANE_POWERDOWN_ACK (1 << 9) +#define BXT_PHY_LANE_ENABLED (1 << 8) +#define BXT_PHY_CTL(port) _MMIO_PORT(port, _BXT_PHY_CTL_DDI_A, \ + _BXT_PHY_CTL_DDI_B) + +#define _PHY_CTL_FAMILY_DDI 0x64C90 +#define _PHY_CTL_FAMILY_EDP 0x64C80 +#define _PHY_CTL_FAMILY_DDI_C 0x64CA0 +#define COMMON_RESET_DIS (1 << 31) +#define BXT_PHY_CTL_FAMILY(phy) \ + _MMIO(_PICK_EVEN_2RANGES(phy, 1, \ + _PHY_CTL_FAMILY_DDI, _PHY_CTL_FAMILY_DDI, \ + _PHY_CTL_FAMILY_EDP, _PHY_CTL_FAMILY_DDI_C)) + +/* UAIMI scratch pad register 1 */ +#define UAIMI_SPR1 _MMIO(0x4F074) +/* SKL VccIO mask */ +#define SKL_VCCIO_MASK 0x1 +/* SKL balance leg register */ +#define DISPIO_CR_TX_BMU_CR0 _MMIO(0x6C00C) +/* I_boost values */ +#define BALANCE_LEG_SHIFT(port) (8 + 3 * (port)) +#define BALANCE_LEG_MASK(port) (7 << (8 + 3 * (port))) +/* Balance leg disable bits */ +#define BALANCE_LEG_DISABLE_SHIFT 23 +#define BALANCE_LEG_DISABLE(port) (1 << (23 + (port))) + +#define ILK_GTT_FAULT _MMIO(0x44040) /* ilk/snb */ +#define GTT_FAULT_INVALID_GTT_PTE (1 << 7) +#define GTT_FAULT_INVALID_PTE_DATA (1 << 6) +#define GTT_FAULT_CURSOR_B_FAULT (1 << 5) +#define GTT_FAULT_CURSOR_A_FAULT (1 << 4) +#define GTT_FAULT_SPRITE_B_FAULT (1 << 3) +#define GTT_FAULT_SPRITE_A_FAULT (1 << 2) +#define GTT_FAULT_PRIMARY_B_FAULT (1 << 1) +#define GTT_FAULT_PRIMARY_A_FAULT (1 << 0) + +#define DERRMR _MMIO(0x44050) +/* Note that HBLANK events are reserved on bdw+ */ +#define DERRMR_PIPEA_SCANLINE (1 << 0) +#define DERRMR_PIPEA_PRI_FLIP_DONE (1 << 1) +#define DERRMR_PIPEA_SPR_FLIP_DONE (1 << 2) +#define DERRMR_PIPEA_VBLANK (1 << 3) +#define DERRMR_PIPEA_HBLANK (1 << 5) +#define DERRMR_PIPEB_SCANLINE (1 << 8) +#define DERRMR_PIPEB_PRI_FLIP_DONE (1 << 9) +#define DERRMR_PIPEB_SPR_FLIP_DONE (1 << 10) +#define DERRMR_PIPEB_VBLANK (1 << 11) +#define DERRMR_PIPEB_HBLANK (1 << 13) +/* Note that PIPEC is not a simple translation of PIPEA/PIPEB */ +#define DERRMR_PIPEC_SCANLINE (1 << 14) +#define DERRMR_PIPEC_PRI_FLIP_DONE (1 << 15) +#define DERRMR_PIPEC_SPR_FLIP_DONE (1 << 20) +#define DERRMR_PIPEC_VBLANK (1 << 21) +#define DERRMR_PIPEC_HBLANK (1 << 22) + +#define VLV_IRQ_REGS I915_IRQ_REGS(VLV_IMR, \ + VLV_IER, \ + VLV_IIR) + +#define VLV_EIR _MMIO(VLV_DISPLAY_BASE + 0x20b0) +#define VLV_EMR _MMIO(VLV_DISPLAY_BASE + 0x20b4) +#define VLV_ESR _MMIO(VLV_DISPLAY_BASE + 0x20b8) +#define VLV_ERROR_GUNIT_TLB_DATA (1 << 6) +#define VLV_ERROR_GUNIT_TLB_PTE (1 << 5) +#define VLV_ERROR_PAGE_TABLE (1 << 4) +#define VLV_ERROR_CLAIM (1 << 0) + +#define VLV_ERROR_REGS I915_ERROR_REGS(VLV_EMR, VLV_EIR) + +#define _MBUS_ABOX0_CTL 0x45038 +#define _MBUS_ABOX1_CTL 0x45048 +#define _MBUS_ABOX2_CTL 0x4504C +#define MBUS_ABOX_CTL(x) \ + _MMIO(_PICK_EVEN_2RANGES(x, 2, \ + _MBUS_ABOX0_CTL, _MBUS_ABOX1_CTL, \ + _MBUS_ABOX2_CTL, _MBUS_ABOX2_CTL)) + +#define MBUS_ABOX_BW_CREDIT_MASK (3 << 20) +#define MBUS_ABOX_BW_CREDIT(x) ((x) << 20) +#define MBUS_ABOX_B_CREDIT_MASK (0xF << 16) +#define MBUS_ABOX_B_CREDIT(x) ((x) << 16) +#define MBUS_ABOX_BT_CREDIT_POOL2_MASK (0x1F << 8) +#define MBUS_ABOX_BT_CREDIT_POOL2(x) ((x) << 8) +#define MBUS_ABOX_BT_CREDIT_POOL1_MASK (0x1F << 0) +#define MBUS_ABOX_BT_CREDIT_POOL1(x) ((x) << 0) + +#define IPS_CTL _MMIO(0x43408) +#define IPS_ENABLE REG_BIT(31) +#define IPS_FALSE_COLOR REG_BIT(4) + +/* + * Clock control & power management + */ +#define _DPLL_A 0x6014 +#define _DPLL_B 0x6018 +#define _CHV_DPLL_C 0x6030 +#define DPLL(dev_priv, pipe) _MMIO_BASE_PIPE3(DISPLAY_MMIO_BASE(dev_priv), \ + (pipe), _DPLL_A, _DPLL_B, _CHV_DPLL_C) + +#define VGA0 _MMIO(0x6000) +#define VGA1 _MMIO(0x6004) +#define VGA_PD _MMIO(0x6010) +#define VGA0_PD_P2_DIV_4 (1 << 7) +#define VGA0_PD_P1_DIV_2 (1 << 5) +#define VGA0_PD_P1_SHIFT 0 +#define VGA0_PD_P1_MASK (0x1f << 0) +#define VGA1_PD_P2_DIV_4 (1 << 15) +#define VGA1_PD_P1_DIV_2 (1 << 13) +#define VGA1_PD_P1_SHIFT 8 +#define VGA1_PD_P1_MASK (0x1f << 8) +#define DPLL_VCO_ENABLE (1 << 31) +#define DPLL_SDVO_HIGH_SPEED (1 << 30) +#define DPLL_DVO_2X_MODE (1 << 30) +#define DPLL_EXT_BUFFER_ENABLE_VLV (1 << 30) +#define DPLL_SYNCLOCK_ENABLE (1 << 29) +#define DPLL_REF_CLK_ENABLE_VLV (1 << 29) +#define DPLL_VGA_MODE_DIS (1 << 28) +#define DPLLB_MODE_DAC_SERIAL (1 << 26) /* i915 */ +#define DPLLB_MODE_LVDS (2 << 26) /* i915 */ +#define DPLL_MODE_MASK (3 << 26) +#define DPLL_DAC_SERIAL_P2_CLOCK_DIV_10 (0 << 24) /* i915 */ +#define DPLL_DAC_SERIAL_P2_CLOCK_DIV_5 (1 << 24) /* i915 */ +#define DPLLB_LVDS_P2_CLOCK_DIV_14 (0 << 24) /* i915 */ +#define DPLLB_LVDS_P2_CLOCK_DIV_7 (1 << 24) /* i915 */ +#define DPLL_P2_CLOCK_DIV_MASK 0x03000000 /* i915 */ +#define DPLL_FPA01_P1_POST_DIV_MASK 0x00ff0000 /* i915 */ +#define DPLL_FPA01_P1_POST_DIV_MASK_PINEVIEW 0x00ff8000 /* Pineview */ +#define DPLL_LOCK_VLV (1 << 15) +#define DPLL_INTEGRATED_CRI_CLK_VLV (1 << 14) +#define DPLL_INTEGRATED_REF_CLK_VLV (1 << 13) +#define DPLL_SSC_REF_CLK_CHV (1 << 13) +#define DPLL_PORTC_READY_MASK (0xf << 4) +#define DPLL_PORTB_READY_MASK (0xf) + +#define DPLL_FPA01_P1_POST_DIV_MASK_I830 0x001f0000 + +/* Additional CHV pll/phy registers */ +#define DPIO_PHY_STATUS _MMIO(VLV_DISPLAY_BASE + 0x6240) +#define DPLL_PORTD_READY_MASK (0xf) +#define DISPLAY_PHY_CONTROL _MMIO(VLV_DISPLAY_BASE + 0x60100) +#define PHY_CH_POWER_DOWN_OVRD_EN(phy, ch) (1 << (2 * (phy) + (ch) + 27)) +#define PHY_LDO_DELAY_0NS 0x0 +#define PHY_LDO_DELAY_200NS 0x1 +#define PHY_LDO_DELAY_600NS 0x2 +#define PHY_LDO_SEQ_DELAY(delay, phy) ((delay) << (2 * (phy) + 23)) +#define PHY_CH_POWER_DOWN_OVRD(mask, phy, ch) ((mask) << (8 * (phy) + 4 * (ch) + 11)) +#define PHY_CH_SU_PSR 0x1 +#define PHY_CH_DEEP_PSR 0x7 +#define PHY_CH_POWER_MODE(mode, phy, ch) ((mode) << (6 * (phy) + 3 * (ch) + 2)) +#define PHY_COM_LANE_RESET_DEASSERT(phy) (1 << (phy)) +#define DISPLAY_PHY_STATUS _MMIO(VLV_DISPLAY_BASE + 0x60104) +#define PHY_POWERGOOD(phy) (((phy) == DPIO_PHY0) ? (1 << 31) : (1 << 30)) +#define PHY_STATUS_CMN_LDO(phy, ch) (1 << (6 - (6 * (phy) + 3 * (ch)))) +#define PHY_STATUS_SPLINE_LDO(phy, ch, spline) (1 << (8 - (6 * (phy) + 3 * (ch) + (spline)))) + +/* + * The i830 generation, in LVDS mode, defines P1 as the bit number set within + * this field (only one bit may be set). + */ +#define DPLL_FPA01_P1_POST_DIV_MASK_I830_LVDS 0x003f0000 +#define DPLL_FPA01_P1_POST_DIV_SHIFT 16 +#define DPLL_FPA01_P1_POST_DIV_SHIFT_PINEVIEW 15 +/* i830, required in DVO non-gang */ +#define PLL_P2_DIVIDE_BY_4 (1 << 23) +#define PLL_P1_DIVIDE_BY_TWO (1 << 21) /* i830 */ +#define PLL_REF_INPUT_DREFCLK (0 << 13) +#define PLL_REF_INPUT_TVCLKINA (1 << 13) /* i830 */ +#define PLL_REF_INPUT_TVCLKINBC (2 << 13) /* SDVO TVCLKIN */ +#define PLLB_REF_INPUT_SPREADSPECTRUMIN (3 << 13) +#define PLL_REF_INPUT_MASK (3 << 13) +#define PLL_LOAD_PULSE_PHASE_SHIFT 9 +/* Ironlake */ +# define PLL_REF_SDVO_HDMI_MULTIPLIER_SHIFT 9 +# define PLL_REF_SDVO_HDMI_MULTIPLIER_MASK (7 << 9) +# define PLL_REF_SDVO_HDMI_MULTIPLIER(x) (((x) - 1) << 9) +# define DPLL_FPA1_P1_POST_DIV_SHIFT 0 +# define DPLL_FPA1_P1_POST_DIV_MASK 0xff + +/* + * Parallel to Serial Load Pulse phase selection. + * Selects the phase for the 10X DPLL clock for the PCIe + * digital display port. The range is 4 to 13; 10 or more + * is just a flip delay. The default is 6 + */ +#define PLL_LOAD_PULSE_PHASE_MASK (0xf << PLL_LOAD_PULSE_PHASE_SHIFT) +#define DISPLAY_RATE_SELECT_FPA1 (1 << 8) +/* + * SDVO multiplier for 945G/GM. Not used on 965. + */ +#define SDVO_MULTIPLIER_MASK 0x000000ff +#define SDVO_MULTIPLIER_SHIFT_HIRES 4 +#define SDVO_MULTIPLIER_SHIFT_VGA 0 + +#define _DPLL_A_MD 0x601c +#define _DPLL_B_MD 0x6020 +#define _CHV_DPLL_C_MD 0x603c +#define DPLL_MD(dev_priv, pipe) _MMIO_BASE_PIPE3(DISPLAY_MMIO_BASE(dev_priv), \ + (pipe), _DPLL_A_MD, _DPLL_B_MD, _CHV_DPLL_C_MD) + +/* + * UDI pixel divider, controlling how many pixels are stuffed into a packet. + * + * Value is pixels minus 1. Must be set to 1 pixel for SDVO. + */ +#define DPLL_MD_UDI_DIVIDER_MASK 0x3f000000 +#define DPLL_MD_UDI_DIVIDER_SHIFT 24 +/* UDI pixel divider for VGA, same as DPLL_MD_UDI_DIVIDER_MASK. */ +#define DPLL_MD_VGA_UDI_DIVIDER_MASK 0x003f0000 +#define DPLL_MD_VGA_UDI_DIVIDER_SHIFT 16 +/* + * SDVO/UDI pixel multiplier. + * + * SDVO requires that the bus clock rate be between 1 and 2 Ghz, and the bus + * clock rate is 10 times the DPLL clock. At low resolution/refresh rate + * modes, the bus rate would be below the limits, so SDVO allows for stuffing + * dummy bytes in the datastream at an increased clock rate, with both sides of + * the link knowing how many bytes are fill. + * + * So, for a mode with a dotclock of 65Mhz, we would want to double the clock + * rate to 130Mhz to get a bus rate of 1.30Ghz. The DPLL clock rate would be + * set to 130Mhz, and the SDVO multiplier set to 2x in this register and + * through an SDVO command. + * + * This register field has values of multiplication factor minus 1, with + * a maximum multiplier of 5 for SDVO. + */ +#define DPLL_MD_UDI_MULTIPLIER_MASK 0x00003f00 +#define DPLL_MD_UDI_MULTIPLIER_SHIFT 8 +/* + * SDVO/UDI pixel multiplier for VGA, same as DPLL_MD_UDI_MULTIPLIER_MASK. + * This best be set to the default value (3) or the CRT won't work. No, + * I don't entirely understand what this does... + */ +#define DPLL_MD_VGA_UDI_MULTIPLIER_MASK 0x0000003f +#define DPLL_MD_VGA_UDI_MULTIPLIER_SHIFT 0 + +#define RAWCLK_FREQ_VLV _MMIO(VLV_DISPLAY_BASE + 0x6024) + +#define _FPA0 0x6040 +#define _FPA1 0x6044 +#define _FPB0 0x6048 +#define _FPB1 0x604c +#define FP0(pipe) _MMIO_PIPE(pipe, _FPA0, _FPB0) +#define FP1(pipe) _MMIO_PIPE(pipe, _FPA1, _FPB1) +#define FP_N_DIV_MASK 0x003f0000 +#define FP_N_PINEVIEW_DIV_MASK 0x00ff0000 +#define FP_N_DIV_SHIFT 16 +#define FP_M1_DIV_MASK 0x00003f00 +#define FP_M1_DIV_SHIFT 8 +#define FP_M2_DIV_MASK 0x0000003f +#define FP_M2_PINEVIEW_DIV_MASK 0x000000ff +#define FP_M2_DIV_SHIFT 0 + +#define FW_BLC_SELF_VLV _MMIO(VLV_DISPLAY_BASE + 0x6500) +#define FW_CSPWRDWNEN (1 << 15) + +#define MI_ARB_VLV _MMIO(VLV_DISPLAY_BASE + 0x6504) + +#define CZCLK_CDCLK_FREQ_RATIO _MMIO(VLV_DISPLAY_BASE + 0x6508) +#define CDCLK_FREQ_SHIFT 4 +#define CDCLK_FREQ_MASK (0x1f << CDCLK_FREQ_SHIFT) +#define CZCLK_FREQ_MASK 0xf + +#define GCI_CONTROL _MMIO(VLV_DISPLAY_BASE + 0x650C) +#define PFI_CREDIT_63 (9 << 28) /* chv only */ +#define PFI_CREDIT_31 (8 << 28) /* chv only */ +#define PFI_CREDIT(x) (((x) - 8) << 28) /* 8-15 */ +#define PFI_CREDIT_RESEND (1 << 27) +#define VGA_FAST_MODE_DISABLE (1 << 14) + +#define GMBUSFREQ_VLV _MMIO(VLV_DISPLAY_BASE + 0x6510) + +#define PEG_BAND_GAP_DATA _MMIO(0x14d68) + +/* + * Overlay regs + */ +#define OVADD _MMIO(0x30000) +#define DOVSTA _MMIO(0x30008) +#define OC_BUF (0x3 << 20) +#define OGAMC5 _MMIO(0x30010) +#define OGAMC4 _MMIO(0x30014) +#define OGAMC3 _MMIO(0x30018) +#define OGAMC2 _MMIO(0x3001c) +#define OGAMC1 _MMIO(0x30020) +#define OGAMC0 _MMIO(0x30024) + +#define GEN9_CLKGATE_DIS_4 _MMIO(0x4653C) +#define BXT_GMBUS_GATING_DIS (1 << 14) +#define DG2_DPFC_GATING_DIS REG_BIT(31) + +#define GEN9_CLKGATE_DIS_5 _MMIO(0x46540) +#define DPCE_GATING_DIS REG_BIT(17) + +#define _CLKGATE_DIS_PSL_A 0x46520 +#define _CLKGATE_DIS_PSL_B 0x46524 +#define _CLKGATE_DIS_PSL_C 0x46528 +#define DUPS1_GATING_DIS (1 << 15) +#define DUPS2_GATING_DIS (1 << 19) +#define DUPS3_GATING_DIS (1 << 23) +#define CURSOR_GATING_DIS REG_BIT(28) +#define DPF_GATING_DIS (1 << 10) +#define DPF_RAM_GATING_DIS (1 << 9) +#define DPFR_GATING_DIS (1 << 8) + +#define CLKGATE_DIS_PSL(pipe) \ + _MMIO_PIPE(pipe, _CLKGATE_DIS_PSL_A, _CLKGATE_DIS_PSL_B) + +#define _CLKGATE_DIS_PSL_EXT_A 0x4654C +#define _CLKGATE_DIS_PSL_EXT_B 0x46550 +#define PIPEDMC_GATING_DIS REG_BIT(12) + +#define CLKGATE_DIS_PSL_EXT(pipe) \ + _MMIO_PIPE(pipe, _CLKGATE_DIS_PSL_EXT_A, _CLKGATE_DIS_PSL_EXT_B) + +/* + * Display engine regs + */ +/* Pipe/transcoder A timing regs */ +#define _TRANS_HTOTAL_A 0x60000 +#define _TRANS_HTOTAL_B 0x61000 +#define TRANS_HTOTAL(dev_priv, trans) _MMIO_TRANS2(dev_priv, (trans), _TRANS_HTOTAL_A) +#define HTOTAL_MASK REG_GENMASK(31, 16) +#define HTOTAL(htotal) REG_FIELD_PREP(HTOTAL_MASK, (htotal)) +#define HACTIVE_MASK REG_GENMASK(15, 0) +#define HACTIVE(hdisplay) REG_FIELD_PREP(HACTIVE_MASK, (hdisplay)) + +#define _TRANS_HBLANK_A 0x60004 +#define _TRANS_HBLANK_B 0x61004 +#define TRANS_HBLANK(dev_priv, trans) _MMIO_TRANS2(dev_priv, (trans), _TRANS_HBLANK_A) +#define HBLANK_END_MASK REG_GENMASK(31, 16) +#define HBLANK_END(hblank_end) REG_FIELD_PREP(HBLANK_END_MASK, (hblank_end)) +#define HBLANK_START_MASK REG_GENMASK(15, 0) +#define HBLANK_START(hblank_start) REG_FIELD_PREP(HBLANK_START_MASK, (hblank_start)) + +#define _TRANS_HSYNC_A 0x60008 +#define _TRANS_HSYNC_B 0x61008 +#define TRANS_HSYNC(dev_priv, trans) _MMIO_TRANS2(dev_priv, (trans), _TRANS_HSYNC_A) +#define HSYNC_END_MASK REG_GENMASK(31, 16) +#define HSYNC_END(hsync_end) REG_FIELD_PREP(HSYNC_END_MASK, (hsync_end)) +#define HSYNC_START_MASK REG_GENMASK(15, 0) +#define HSYNC_START(hsync_start) REG_FIELD_PREP(HSYNC_START_MASK, (hsync_start)) + +#define _TRANS_VTOTAL_A 0x6000c +#define _TRANS_VTOTAL_B 0x6100c +#define TRANS_VTOTAL(dev_priv, trans) _MMIO_TRANS2(dev_priv, (trans), _TRANS_VTOTAL_A) +#define VTOTAL_MASK REG_GENMASK(31, 16) +#define VTOTAL(vtotal) REG_FIELD_PREP(VTOTAL_MASK, (vtotal)) +#define VACTIVE_MASK REG_GENMASK(15, 0) +#define VACTIVE(vdisplay) REG_FIELD_PREP(VACTIVE_MASK, (vdisplay)) + +#define _TRANS_VBLANK_A 0x60010 +#define _TRANS_VBLANK_B 0x61010 +#define TRANS_VBLANK(dev_priv, trans) _MMIO_TRANS2(dev_priv, (trans), _TRANS_VBLANK_A) +#define VBLANK_END_MASK REG_GENMASK(31, 16) +#define VBLANK_END(vblank_end) REG_FIELD_PREP(VBLANK_END_MASK, (vblank_end)) +#define VBLANK_START_MASK REG_GENMASK(15, 0) +#define VBLANK_START(vblank_start) REG_FIELD_PREP(VBLANK_START_MASK, (vblank_start)) + +#define _TRANS_VSYNC_A 0x60014 +#define _TRANS_VSYNC_B 0x61014 +#define TRANS_VSYNC(dev_priv, trans) _MMIO_TRANS2(dev_priv, (trans), _TRANS_VSYNC_A) +#define VSYNC_END_MASK REG_GENMASK(31, 16) +#define VSYNC_END(vsync_end) REG_FIELD_PREP(VSYNC_END_MASK, (vsync_end)) +#define VSYNC_START_MASK REG_GENMASK(15, 0) +#define VSYNC_START(vsync_start) REG_FIELD_PREP(VSYNC_START_MASK, (vsync_start)) + +#define _PIPEASRC 0x6001c +#define _PIPEBSRC 0x6101c +#define PIPESRC(dev_priv, pipe) _MMIO_TRANS2(dev_priv, (pipe), _PIPEASRC) +#define PIPESRC_WIDTH_MASK REG_GENMASK(31, 16) +#define PIPESRC_WIDTH(w) REG_FIELD_PREP(PIPESRC_WIDTH_MASK, (w)) +#define PIPESRC_HEIGHT_MASK REG_GENMASK(15, 0) +#define PIPESRC_HEIGHT(h) REG_FIELD_PREP(PIPESRC_HEIGHT_MASK, (h)) + +#define _BCLRPAT_A 0x60020 +#define _BCLRPAT_B 0x61020 +#define BCLRPAT(dev_priv, trans) _MMIO_TRANS2(dev_priv, (trans), _BCLRPAT_A) + +#define _TRANS_VSYNCSHIFT_A 0x60028 +#define _TRANS_VSYNCSHIFT_B 0x61028 +#define TRANS_VSYNCSHIFT(dev_priv, trans) _MMIO_TRANS2(dev_priv, (trans), _TRANS_VSYNCSHIFT_A) + +#define _TRANS_MULT_A 0x6002c +#define _TRANS_MULT_B 0x6102c +#define TRANS_MULT(dev_priv, trans) _MMIO_TRANS2(dev_priv, (trans), _TRANS_MULT_A) + +/* Hotplug control (945+ only) */ +#define PORT_HOTPLUG_EN(dev_priv) _MMIO(DISPLAY_MMIO_BASE(dev_priv) + 0x61110) +#define PORTB_HOTPLUG_INT_EN (1 << 29) +#define PORTC_HOTPLUG_INT_EN (1 << 28) +#define PORTD_HOTPLUG_INT_EN (1 << 27) +#define SDVOB_HOTPLUG_INT_EN (1 << 26) +#define SDVOC_HOTPLUG_INT_EN (1 << 25) +#define TV_HOTPLUG_INT_EN (1 << 18) +#define CRT_HOTPLUG_INT_EN (1 << 9) +#define HOTPLUG_INT_EN_MASK (PORTB_HOTPLUG_INT_EN | \ + PORTC_HOTPLUG_INT_EN | \ + PORTD_HOTPLUG_INT_EN | \ + SDVOC_HOTPLUG_INT_EN | \ + SDVOB_HOTPLUG_INT_EN | \ + CRT_HOTPLUG_INT_EN) +#define CRT_HOTPLUG_FORCE_DETECT (1 << 3) +#define CRT_HOTPLUG_ACTIVATION_PERIOD_32 (0 << 8) +/* must use period 64 on GM45 according to docs */ +#define CRT_HOTPLUG_ACTIVATION_PERIOD_64 (1 << 8) +#define CRT_HOTPLUG_DAC_ON_TIME_2M (0 << 7) +#define CRT_HOTPLUG_DAC_ON_TIME_4M (1 << 7) +#define CRT_HOTPLUG_VOLTAGE_COMPARE_40 (0 << 5) +#define CRT_HOTPLUG_VOLTAGE_COMPARE_50 (1 << 5) +#define CRT_HOTPLUG_VOLTAGE_COMPARE_60 (2 << 5) +#define CRT_HOTPLUG_VOLTAGE_COMPARE_70 (3 << 5) +#define CRT_HOTPLUG_VOLTAGE_COMPARE_MASK (3 << 5) +#define CRT_HOTPLUG_DETECT_DELAY_1G (0 << 4) +#define CRT_HOTPLUG_DETECT_DELAY_2G (1 << 4) +#define CRT_HOTPLUG_DETECT_VOLTAGE_325MV (0 << 2) +#define CRT_HOTPLUG_DETECT_VOLTAGE_475MV (1 << 2) + +#define PORT_HOTPLUG_STAT(dev_priv) _MMIO(DISPLAY_MMIO_BASE(dev_priv) + 0x61114) +/* HDMI/DP bits are g4x+ */ +#define PORTD_HOTPLUG_LIVE_STATUS_G4X (1 << 27) +#define PORTC_HOTPLUG_LIVE_STATUS_G4X (1 << 28) +#define PORTB_HOTPLUG_LIVE_STATUS_G4X (1 << 29) +#define PORTD_HOTPLUG_INT_STATUS (3 << 21) +#define PORTD_HOTPLUG_INT_LONG_PULSE (2 << 21) +#define PORTD_HOTPLUG_INT_SHORT_PULSE (1 << 21) +#define PORTC_HOTPLUG_INT_STATUS (3 << 19) +#define PORTC_HOTPLUG_INT_LONG_PULSE (2 << 19) +#define PORTC_HOTPLUG_INT_SHORT_PULSE (1 << 19) +#define PORTB_HOTPLUG_INT_STATUS (3 << 17) +#define PORTB_HOTPLUG_INT_LONG_PULSE (2 << 17) +#define PORTB_HOTPLUG_INT_SHORT_PLUSE (1 << 17) +/* CRT/TV common between gen3+ */ +#define CRT_HOTPLUG_INT_STATUS (1 << 11) +#define TV_HOTPLUG_INT_STATUS (1 << 10) +#define CRT_HOTPLUG_MONITOR_MASK (3 << 8) +#define CRT_HOTPLUG_MONITOR_COLOR (3 << 8) +#define CRT_HOTPLUG_MONITOR_MONO (2 << 8) +#define CRT_HOTPLUG_MONITOR_NONE (0 << 8) +#define DP_AUX_CHANNEL_D_INT_STATUS_G4X (1 << 6) +#define DP_AUX_CHANNEL_C_INT_STATUS_G4X (1 << 5) +#define DP_AUX_CHANNEL_B_INT_STATUS_G4X (1 << 4) +#define DP_AUX_CHANNEL_MASK_INT_STATUS_G4X (7 << 4) + +/* SDVO is different across gen3/4 */ +#define SDVOC_HOTPLUG_INT_STATUS_G4X (1 << 3) +#define SDVOB_HOTPLUG_INT_STATUS_G4X (1 << 2) +/* + * Bspec seems to be seriously misleaded about the SDVO hpd bits on i965g/gm, + * since reality corrobates that they're the same as on gen3. But keep these + * bits here (and the comment!) to help any other lost wanderers back onto the + * right tracks. + */ +#define SDVOC_HOTPLUG_INT_STATUS_I965 (3 << 4) +#define SDVOB_HOTPLUG_INT_STATUS_I965 (3 << 2) +#define SDVOC_HOTPLUG_INT_STATUS_I915 (1 << 7) +#define SDVOB_HOTPLUG_INT_STATUS_I915 (1 << 6) +#define HOTPLUG_INT_STATUS_G4X (CRT_HOTPLUG_INT_STATUS | \ + SDVOB_HOTPLUG_INT_STATUS_G4X | \ + SDVOC_HOTPLUG_INT_STATUS_G4X | \ + PORTB_HOTPLUG_INT_STATUS | \ + PORTC_HOTPLUG_INT_STATUS | \ + PORTD_HOTPLUG_INT_STATUS) + +#define HOTPLUG_INT_STATUS_I915 (CRT_HOTPLUG_INT_STATUS | \ + SDVOB_HOTPLUG_INT_STATUS_I915 | \ + SDVOC_HOTPLUG_INT_STATUS_I915 | \ + PORTB_HOTPLUG_INT_STATUS | \ + PORTC_HOTPLUG_INT_STATUS | \ + PORTD_HOTPLUG_INT_STATUS) + +/* SDVO and HDMI port control. + * The same register may be used for SDVO or HDMI */ +#define _GEN3_SDVOB 0x61140 +#define _GEN3_SDVOC 0x61160 +#define GEN3_SDVOB _MMIO(_GEN3_SDVOB) +#define GEN3_SDVOC _MMIO(_GEN3_SDVOC) +#define GEN4_HDMIB GEN3_SDVOB +#define GEN4_HDMIC GEN3_SDVOC +#define VLV_HDMIB _MMIO(VLV_DISPLAY_BASE + 0x61140) +#define VLV_HDMIC _MMIO(VLV_DISPLAY_BASE + 0x61160) +#define CHV_HDMID _MMIO(VLV_DISPLAY_BASE + 0x6116C) +#define PCH_SDVOB _MMIO(0xe1140) +#define PCH_HDMIB PCH_SDVOB +#define PCH_HDMIC _MMIO(0xe1150) +#define PCH_HDMID _MMIO(0xe1160) + +#define PORT_DFT_I9XX _MMIO(0x61150) +#define DC_BALANCE_RESET (1 << 25) +#define PORT_DFT2_G4X(dev_priv) _MMIO(DISPLAY_MMIO_BASE(dev_priv) + 0x61154) +#define DC_BALANCE_RESET_VLV (1 << 31) +#define PIPE_SCRAMBLE_RESET_MASK ((1 << 14) | (0x3 << 0)) +#define PIPE_C_SCRAMBLE_RESET REG_BIT(14) /* chv */ +#define PIPE_B_SCRAMBLE_RESET REG_BIT(1) +#define PIPE_A_SCRAMBLE_RESET REG_BIT(0) + +/* Gen 3 SDVO bits: */ +#define SDVO_ENABLE (1 << 31) +#define SDVO_PIPE_SEL_SHIFT 30 +#define SDVO_PIPE_SEL_MASK (1 << 30) +#define SDVO_PIPE_SEL(pipe) ((pipe) << 30) +#define SDVO_STALL_SELECT (1 << 29) +#define SDVO_INTERRUPT_ENABLE (1 << 26) +/* + * 915G/GM SDVO pixel multiplier. + * Programmed value is multiplier - 1, up to 5x. + * \sa DPLL_MD_UDI_MULTIPLIER_MASK + */ +#define SDVO_PORT_MULTIPLY_MASK (7 << 23) +#define SDVO_PORT_MULTIPLY_SHIFT 23 +#define SDVO_PHASE_SELECT_MASK (15 << 19) +#define SDVO_PHASE_SELECT_DEFAULT (6 << 19) +#define SDVO_CLOCK_OUTPUT_INVERT (1 << 18) +#define SDVOC_GANG_MODE (1 << 16) /* Port C only */ +#define SDVO_BORDER_ENABLE (1 << 7) /* SDVO only */ +#define SDVOB_PCIE_CONCURRENCY (1 << 3) /* Port B only */ +#define SDVO_DETECTED (1 << 2) +/* Bits to be preserved when writing */ +#define SDVOB_PRESERVE_MASK ((1 << 17) | (1 << 16) | (1 << 14) | \ + SDVO_INTERRUPT_ENABLE) +#define SDVOC_PRESERVE_MASK ((1 << 17) | SDVO_INTERRUPT_ENABLE) + +/* Gen 4 SDVO/HDMI bits: */ +#define SDVO_COLOR_FORMAT_8bpc (0 << 26) +#define SDVO_COLOR_FORMAT_MASK (7 << 26) +#define SDVO_ENCODING_SDVO (0 << 10) +#define SDVO_ENCODING_HDMI (2 << 10) +#define HDMI_MODE_SELECT_HDMI (1 << 9) /* HDMI only */ +#define HDMI_MODE_SELECT_DVI (0 << 9) /* HDMI only */ +#define HDMI_COLOR_RANGE_16_235 (1 << 8) /* HDMI only */ +#define HDMI_AUDIO_ENABLE (1 << 6) /* HDMI only */ +/* VSYNC/HSYNC bits new with 965, default is to be set */ +#define SDVO_VSYNC_ACTIVE_HIGH (1 << 4) +#define SDVO_HSYNC_ACTIVE_HIGH (1 << 3) + +/* Gen 5 (IBX) SDVO/HDMI bits: */ +#define HDMI_COLOR_FORMAT_12bpc (3 << 26) /* HDMI only */ +#define SDVOB_HOTPLUG_ENABLE (1 << 23) /* SDVO only */ + +/* Gen 6 (CPT) SDVO/HDMI bits: */ +#define SDVO_PIPE_SEL_SHIFT_CPT 29 +#define SDVO_PIPE_SEL_MASK_CPT (3 << 29) +#define SDVO_PIPE_SEL_CPT(pipe) ((pipe) << 29) + +/* CHV SDVO/HDMI bits: */ +#define SDVO_PIPE_SEL_SHIFT_CHV 24 +#define SDVO_PIPE_SEL_MASK_CHV (3 << 24) +#define SDVO_PIPE_SEL_CHV(pipe) ((pipe) << 24) + +/* Video Data Island Packet control */ +#define VIDEO_DIP_DATA _MMIO(0x61178) +/* Read the description of VIDEO_DIP_DATA (before Haswell) or VIDEO_DIP_ECC + * (Haswell and newer) to see which VIDEO_DIP_DATA byte corresponds to each byte + * of the infoframe structure specified by CEA-861. */ +#define VIDEO_DIP_DATA_SIZE 32 +#define VIDEO_DIP_ASYNC_DATA_SIZE 36 +#define VIDEO_DIP_GMP_DATA_SIZE 36 +#define VIDEO_DIP_VSC_DATA_SIZE 36 +#define VIDEO_DIP_PPS_DATA_SIZE 132 +#define VIDEO_DIP_CTL _MMIO(0x61170) +/* Pre HSW: */ +#define VIDEO_DIP_ENABLE (1 << 31) +#define VIDEO_DIP_PORT(port) ((port) << 29) +#define VIDEO_DIP_PORT_MASK (3 << 29) +#define VIDEO_DIP_ENABLE_GCP (1 << 25) /* ilk+ */ +#define VIDEO_DIP_ENABLE_AVI (1 << 21) +#define VIDEO_DIP_ENABLE_VENDOR (2 << 21) +#define VIDEO_DIP_ENABLE_GAMUT (4 << 21) /* ilk+ */ +#define VIDEO_DIP_ENABLE_SPD (8 << 21) +#define VIDEO_DIP_SELECT_AVI (0 << 19) +#define VIDEO_DIP_SELECT_VENDOR (1 << 19) +#define VIDEO_DIP_SELECT_GAMUT (2 << 19) +#define VIDEO_DIP_SELECT_SPD (3 << 19) +#define VIDEO_DIP_SELECT_MASK (3 << 19) +#define VIDEO_DIP_FREQ_ONCE (0 << 16) +#define VIDEO_DIP_FREQ_VSYNC (1 << 16) +#define VIDEO_DIP_FREQ_2VSYNC (2 << 16) +#define VIDEO_DIP_FREQ_MASK (3 << 16) +/* HSW and later: */ +#define VIDEO_DIP_ENABLE_DRM_GLK (1 << 28) +#define PSR_VSC_BIT_7_SET (1 << 27) +#define VSC_SELECT_MASK (0x3 << 25) +#define VSC_SELECT_SHIFT 25 +#define VSC_DIP_HW_HEA_DATA (0 << 25) +#define VSC_DIP_HW_HEA_SW_DATA (1 << 25) +#define VSC_DIP_HW_DATA_SW_HEA (2 << 25) +#define VSC_DIP_SW_HEA_DATA (3 << 25) +#define VDIP_ENABLE_PPS (1 << 24) +#define VIDEO_DIP_ENABLE_VSC_HSW (1 << 20) +#define VIDEO_DIP_ENABLE_GCP_HSW (1 << 16) +#define VIDEO_DIP_ENABLE_AVI_HSW (1 << 12) +#define VIDEO_DIP_ENABLE_VS_HSW (1 << 8) +#define VIDEO_DIP_ENABLE_GMP_HSW (1 << 4) +#define VIDEO_DIP_ENABLE_SPD_HSW (1 << 0) +/* ADL and later: */ +#define VIDEO_DIP_ENABLE_AS_ADL REG_BIT(23) + +#define PCH_GTC_CTL _MMIO(0xe7000) +#define PCH_GTC_ENABLE (1 << 31) + +/* Display Port */ +#define DP_A _MMIO(0x64000) /* eDP */ +#define DP_B _MMIO(0x64100) +#define DP_C _MMIO(0x64200) +#define DP_D _MMIO(0x64300) +#define VLV_DP_B _MMIO(VLV_DISPLAY_BASE + 0x64100) +#define VLV_DP_C _MMIO(VLV_DISPLAY_BASE + 0x64200) +#define CHV_DP_D _MMIO(VLV_DISPLAY_BASE + 0x64300) +#define DP_PORT_EN REG_BIT(31) +#define DP_PIPE_SEL_MASK REG_GENMASK(30, 30) +#define DP_PIPE_SEL(pipe) REG_FIELD_PREP(DP_PIPE_SEL_MASK, (pipe)) +#define DP_PIPE_SEL_MASK_IVB REG_GENMASK(30, 29) +#define DP_PIPE_SEL_IVB(pipe) REG_FIELD_PREP(DP_PIPE_SEL_MASK_IVB, (pipe)) +#define DP_PIPE_SEL_SHIFT_CHV 16 +#define DP_PIPE_SEL_MASK_CHV REG_GENMASK(17, 16) +#define DP_PIPE_SEL_CHV(pipe) REG_FIELD_PREP(DP_PIPE_SEL_MASK_CHV, (pipe)) +#define DP_LINK_TRAIN_MASK REG_GENMASK(29, 28) +#define DP_LINK_TRAIN_PAT_1 REG_FIELD_PREP(DP_LINK_TRAIN_MASK, 0) +#define DP_LINK_TRAIN_PAT_2 REG_FIELD_PREP(DP_LINK_TRAIN_MASK, 1) +#define DP_LINK_TRAIN_PAT_IDLE REG_FIELD_PREP(DP_LINK_TRAIN_MASK, 2) +#define DP_LINK_TRAIN_OFF REG_FIELD_PREP(DP_LINK_TRAIN_MASK, 3) +#define DP_LINK_TRAIN_MASK_CPT REG_GENMASK(10, 8) +#define DP_LINK_TRAIN_PAT_1_CPT REG_FIELD_PREP(DP_LINK_TRAIN_MASK_CPT, 0) +#define DP_LINK_TRAIN_PAT_2_CPT REG_FIELD_PREP(DP_LINK_TRAIN_MASK_CPT, 1) +#define DP_LINK_TRAIN_PAT_IDLE_CPT REG_FIELD_PREP(DP_LINK_TRAIN_MASK_CPT, 2) +#define DP_LINK_TRAIN_OFF_CPT REG_FIELD_PREP(DP_LINK_TRAIN_MASK_CPT, 3) +#define DP_VOLTAGE_MASK REG_GENMASK(27, 25) +#define DP_VOLTAGE_0_4 REG_FIELD_PREP(DP_VOLTAGE_MASK, 0) +#define DP_VOLTAGE_0_6 REG_FIELD_PREP(DP_VOLTAGE_MASK, 1) +#define DP_VOLTAGE_0_8 REG_FIELD_PREP(DP_VOLTAGE_MASK, 2) +#define DP_VOLTAGE_1_2 REG_FIELD_PREP(DP_VOLTAGE_MASK, 3) +#define DP_PRE_EMPHASIS_MASK REG_GENMASK(24, 22) +#define DP_PRE_EMPHASIS_0 REG_FIELD_PREP(DP_PRE_EMPHASIS_MASK, 0) +#define DP_PRE_EMPHASIS_3_5 REG_FIELD_PREP(DP_PRE_EMPHASIS_MASK, 1) +#define DP_PRE_EMPHASIS_6 REG_FIELD_PREP(DP_PRE_EMPHASIS_MASK, 2) +#define DP_PRE_EMPHASIS_9_5 REG_FIELD_PREP(DP_PRE_EMPHASIS_MASK, 3) +#define DP_PORT_WIDTH_MASK REG_GENMASK(21, 19) +#define DP_PORT_WIDTH(width) REG_FIELD_PREP(DP_PORT_WIDTH_MASK, (width) - 1) +#define DP_ENHANCED_FRAMING REG_BIT(18) +#define EDP_PLL_FREQ_MASK REG_GENMASK(17, 16) +#define EDP_PLL_FREQ_270MHZ REG_FIELD_PREP(EDP_PLL_FREQ_MASK, 0) +#define EDP_PLL_FREQ_162MHZ REG_FIELD_PREP(EDP_PLL_FREQ_MASK, 1) +#define DP_PORT_REVERSAL REG_BIT(15) +#define EDP_PLL_ENABLE REG_BIT(14) +#define DP_CLOCK_OUTPUT_ENABLE REG_BIT(13) +#define DP_SCRAMBLING_DISABLE REG_BIT(12) +#define DP_SCRAMBLING_DISABLE_ILK REG_BIT(7) +#define DP_COLOR_RANGE_16_235 REG_BIT(8) +#define DP_AUDIO_OUTPUT_ENABLE REG_BIT(6) +#define DP_SYNC_VS_HIGH REG_BIT(4) +#define DP_SYNC_HS_HIGH REG_BIT(3) +#define DP_DETECTED REG_BIT(2) + +/* + * Computing GMCH M and N values for the Display Port link + * + * GMCH M/N = dot clock * bytes per pixel / ls_clk * # of lanes + * + * ls_clk (we assume) is the DP link clock (1.62 or 2.7 GHz) + * + * The GMCH value is used internally + * + * bytes_per_pixel is the number of bytes coming out of the plane, + * which is after the LUTs, so we want the bytes for our color format. + * For our current usage, this is always 3, one byte for R, G and B. + */ +#define _PIPEA_DATA_M_G4X 0x70050 +#define _PIPEB_DATA_M_G4X 0x71050 +#define PIPE_DATA_M_G4X(pipe) _MMIO_PIPE(pipe, _PIPEA_DATA_M_G4X, _PIPEB_DATA_M_G4X) +/* Transfer unit size for display port - 1, default is 0x3f (for TU size 64) */ +#define TU_SIZE_MASK REG_GENMASK(30, 25) +#define TU_SIZE(x) REG_FIELD_PREP(TU_SIZE_MASK, (x) - 1) /* default size 64 */ +#define DATA_LINK_M_N_MASK REG_GENMASK(23, 0) +#define DATA_LINK_N_MAX (0x800000) + +#define _PIPEA_DATA_N_G4X 0x70054 +#define _PIPEB_DATA_N_G4X 0x71054 +#define PIPE_DATA_N_G4X(pipe) _MMIO_PIPE(pipe, _PIPEA_DATA_N_G4X, _PIPEB_DATA_N_G4X) + +/* + * Computing Link M and N values for the Display Port link + * + * Link M / N = pixel_clock / ls_clk + * + * (the DP spec calls pixel_clock the 'strm_clk') + * + * The Link value is transmitted in the Main Stream + * Attributes and VB-ID. + */ +#define _PIPEA_LINK_M_G4X 0x70060 +#define _PIPEB_LINK_M_G4X 0x71060 +#define PIPE_LINK_M_G4X(pipe) _MMIO_PIPE(pipe, _PIPEA_LINK_M_G4X, _PIPEB_LINK_M_G4X) + +#define _PIPEA_LINK_N_G4X 0x70064 +#define _PIPEB_LINK_N_G4X 0x71064 +#define PIPE_LINK_N_G4X(pipe) _MMIO_PIPE(pipe, _PIPEA_LINK_N_G4X, _PIPEB_LINK_N_G4X) + +/* Pipe A */ +#define _PIPEADSL 0x70000 +#define PIPEDSL(dev_priv, pipe) _MMIO_PIPE2(dev_priv, pipe, _PIPEADSL) +#define PIPEDSL_CURR_FIELD REG_BIT(31) /* ctg+ */ +#define PIPEDSL_LINE_MASK REG_GENMASK(19, 0) + +#define _TRANSACONF 0x70008 +#define TRANSCONF(dev_priv, trans) _MMIO_PIPE2(dev_priv, (trans), _TRANSACONF) +#define TRANSCONF_ENABLE REG_BIT(31) +#define TRANSCONF_DOUBLE_WIDE REG_BIT(30) /* pre-i965 */ +#define TRANSCONF_STATE_ENABLE REG_BIT(30) /* i965+ */ +#define TRANSCONF_DSI_PLL_LOCKED REG_BIT(29) /* vlv & pipe A only */ +#define TRANSCONF_FRAME_START_DELAY_MASK REG_GENMASK(28, 27) /* pre-hsw */ +#define TRANSCONF_FRAME_START_DELAY(x) REG_FIELD_PREP(TRANSCONF_FRAME_START_DELAY_MASK, (x)) /* pre-hsw: 0-3 */ +#define TRANSCONF_PIPE_LOCKED REG_BIT(25) +#define TRANSCONF_FORCE_BORDER REG_BIT(25) +#define TRANSCONF_GAMMA_MODE_MASK_I9XX REG_BIT(24) /* gmch */ +#define TRANSCONF_GAMMA_MODE_MASK_ILK REG_GENMASK(25, 24) /* ilk-ivb */ +#define TRANSCONF_GAMMA_MODE_8BIT REG_FIELD_PREP(TRANSCONF_GAMMA_MODE_MASK, 0) +#define TRANSCONF_GAMMA_MODE_10BIT REG_FIELD_PREP(TRANSCONF_GAMMA_MODE_MASK, 1) +#define TRANSCONF_GAMMA_MODE_12BIT REG_FIELD_PREP(TRANSCONF_GAMMA_MODE_MASK_ILK, 2) /* ilk-ivb */ +#define TRANSCONF_GAMMA_MODE_SPLIT REG_FIELD_PREP(TRANSCONF_GAMMA_MODE_MASK_ILK, 3) /* ivb */ +#define TRANSCONF_GAMMA_MODE(x) REG_FIELD_PREP(TRANSCONF_GAMMA_MODE_MASK_ILK, (x)) /* pass in GAMMA_MODE_MODE_* */ +#define TRANSCONF_INTERLACE_MASK REG_GENMASK(23, 21) /* gen3+ */ +#define TRANSCONF_INTERLACE_PROGRESSIVE REG_FIELD_PREP(TRANSCONF_INTERLACE_MASK, 0) +#define TRANSCONF_INTERLACE_W_SYNC_SHIFT_PANEL REG_FIELD_PREP(TRANSCONF_INTERLACE_MASK, 4) /* gen4 only */ +#define TRANSCONF_INTERLACE_W_SYNC_SHIFT REG_FIELD_PREP(TRANSCONF_INTERLACE_MASK, 5) /* gen4 only */ +#define TRANSCONF_INTERLACE_W_FIELD_INDICATION REG_FIELD_PREP(TRANSCONF_INTERLACE_MASK, 6) +#define TRANSCONF_INTERLACE_FIELD_0_ONLY REG_FIELD_PREP(TRANSCONF_INTERLACE_MASK, 7) /* gen3 only */ +/* + * ilk+: PF/D=progressive fetch/display, IF/D=interlaced fetch/display, + * DBL=power saving pixel doubling, PF-ID* requires panel fitter + */ +#define TRANSCONF_INTERLACE_MASK_ILK REG_GENMASK(23, 21) /* ilk+ */ +#define TRANSCONF_INTERLACE_MASK_HSW REG_GENMASK(22, 21) /* hsw+ */ +#define TRANSCONF_INTERLACE_PF_PD_ILK REG_FIELD_PREP(TRANSCONF_INTERLACE_MASK_ILK, 0) +#define TRANSCONF_INTERLACE_PF_ID_ILK REG_FIELD_PREP(TRANSCONF_INTERLACE_MASK_ILK, 1) +#define TRANSCONF_INTERLACE_IF_ID_ILK REG_FIELD_PREP(TRANSCONF_INTERLACE_MASK_ILK, 3) +#define TRANSCONF_INTERLACE_IF_ID_DBL_ILK REG_FIELD_PREP(TRANSCONF_INTERLACE_MASK_ILK, 4) /* ilk/snb only */ +#define TRANSCONF_INTERLACE_PF_ID_DBL_ILK REG_FIELD_PREP(TRANSCONF_INTERLACE_MASK_ILK, 5) /* ilk/snb only */ +#define TRANSCONF_REFRESH_RATE_ALT_ILK REG_BIT(20) +#define TRANSCONF_MSA_TIMING_DELAY_MASK REG_GENMASK(19, 18) /* ilk/snb/ivb */ +#define TRANSCONF_MSA_TIMING_DELAY(x) REG_FIELD_PREP(TRANSCONF_MSA_TIMING_DELAY_MASK, (x)) +#define TRANSCONF_CXSR_DOWNCLOCK REG_BIT(16) +#define TRANSCONF_WGC_ENABLE REG_BIT(15) /* vlv/chv only */ +#define TRANSCONF_REFRESH_RATE_ALT_VLV REG_BIT(14) +#define TRANSCONF_COLOR_RANGE_SELECT REG_BIT(13) +#define TRANSCONF_OUTPUT_COLORSPACE_MASK REG_GENMASK(12, 11) /* ilk-ivb */ +#define TRANSCONF_OUTPUT_COLORSPACE_RGB REG_FIELD_PREP(TRANSCONF_OUTPUT_COLORSPACE_MASK, 0) /* ilk-ivb */ +#define TRANSCONF_OUTPUT_COLORSPACE_YUV601 REG_FIELD_PREP(TRANSCONF_OUTPUT_COLORSPACE_MASK, 1) /* ilk-ivb */ +#define TRANSCONF_OUTPUT_COLORSPACE_YUV709 REG_FIELD_PREP(TRANSCONF_OUTPUT_COLORSPACE_MASK, 2) /* ilk-ivb */ +#define TRANSCONF_OUTPUT_COLORSPACE_YUV_HSW REG_BIT(11) /* hsw only */ +#define TRANSCONF_BPC_MASK REG_GENMASK(7, 5) /* ctg-ivb */ +#define TRANSCONF_BPC_8 REG_FIELD_PREP(TRANSCONF_BPC_MASK, 0) +#define TRANSCONF_BPC_10 REG_FIELD_PREP(TRANSCONF_BPC_MASK, 1) +#define TRANSCONF_BPC_6 REG_FIELD_PREP(TRANSCONF_BPC_MASK, 2) +#define TRANSCONF_BPC_12 REG_FIELD_PREP(TRANSCONF_BPC_MASK, 3) +#define TRANSCONF_DITHER_EN REG_BIT(4) +#define TRANSCONF_DITHER_TYPE_MASK REG_GENMASK(3, 2) +#define TRANSCONF_DITHER_TYPE_SP REG_FIELD_PREP(TRANSCONF_DITHER_TYPE_MASK, 0) +#define TRANSCONF_DITHER_TYPE_ST1 REG_FIELD_PREP(TRANSCONF_DITHER_TYPE_MASK, 1) +#define TRANSCONF_DITHER_TYPE_ST2 REG_FIELD_PREP(TRANSCONF_DITHER_TYPE_MASK, 2) +#define TRANSCONF_DITHER_TYPE_TEMP REG_FIELD_PREP(TRANSCONF_DITHER_TYPE_MASK, 3) +#define TRANSCONF_PIXEL_COUNT_SCALING_MASK REG_GENMASK(1, 0) +#define TRANSCONF_PIXEL_COUNT_SCALING_X4 1 + +#define _PIPEASTAT 0x70024 +#define PIPESTAT(dev_priv, pipe) _MMIO_PIPE2(dev_priv, pipe, _PIPEASTAT) +#define PIPE_FIFO_UNDERRUN_STATUS (1UL << 31) +#define SPRITE1_FLIP_DONE_INT_EN_VLV (1UL << 30) +#define PIPE_CRC_ERROR_ENABLE (1UL << 29) +#define PIPE_CRC_DONE_ENABLE (1UL << 28) +#define PERF_COUNTER2_INTERRUPT_EN (1UL << 27) +#define PIPE_GMBUS_EVENT_ENABLE (1UL << 27) +#define PLANE_FLIP_DONE_INT_EN_VLV (1UL << 26) +#define PIPE_HOTPLUG_INTERRUPT_ENABLE (1UL << 26) +#define PIPE_VSYNC_INTERRUPT_ENABLE (1UL << 25) +#define PIPE_DISPLAY_LINE_COMPARE_ENABLE (1UL << 24) +#define PIPE_DPST_EVENT_ENABLE (1UL << 23) +#define SPRITE0_FLIP_DONE_INT_EN_VLV (1UL << 22) +#define PIPE_LEGACY_BLC_EVENT_ENABLE (1UL << 22) +#define PIPE_ODD_FIELD_INTERRUPT_ENABLE (1UL << 21) +#define PIPE_EVEN_FIELD_INTERRUPT_ENABLE (1UL << 20) +#define PIPE_B_PSR_INTERRUPT_ENABLE_VLV (1UL << 19) +#define PERF_COUNTER_INTERRUPT_EN (1UL << 19) +#define PIPE_HOTPLUG_TV_INTERRUPT_ENABLE (1UL << 18) /* pre-965 */ +#define PIPE_START_VBLANK_INTERRUPT_ENABLE (1UL << 18) /* 965 or later */ +#define PIPE_FRAMESTART_INTERRUPT_ENABLE (1UL << 17) +#define PIPE_VBLANK_INTERRUPT_ENABLE (1UL << 17) +#define PIPEA_HBLANK_INT_EN_VLV (1UL << 16) +#define PIPE_OVERLAY_UPDATED_ENABLE (1UL << 16) +#define SPRITE1_FLIP_DONE_INT_STATUS_VLV (1UL << 15) +#define SPRITE0_FLIP_DONE_INT_STATUS_VLV (1UL << 14) +#define PIPE_CRC_ERROR_INTERRUPT_STATUS (1UL << 13) +#define PIPE_CRC_DONE_INTERRUPT_STATUS (1UL << 12) +#define PERF_COUNTER2_INTERRUPT_STATUS (1UL << 11) +#define PIPE_GMBUS_INTERRUPT_STATUS (1UL << 11) +#define PLANE_FLIP_DONE_INT_STATUS_VLV (1UL << 10) +#define PIPE_HOTPLUG_INTERRUPT_STATUS (1UL << 10) +#define PIPE_VSYNC_INTERRUPT_STATUS (1UL << 9) +#define PIPE_DISPLAY_LINE_COMPARE_STATUS (1UL << 8) +#define PIPE_DPST_EVENT_STATUS (1UL << 7) +#define PIPE_A_PSR_STATUS_VLV (1UL << 6) +#define PIPE_LEGACY_BLC_EVENT_STATUS (1UL << 6) +#define PIPE_ODD_FIELD_INTERRUPT_STATUS (1UL << 5) +#define PIPE_EVEN_FIELD_INTERRUPT_STATUS (1UL << 4) +#define PIPE_B_PSR_STATUS_VLV (1UL << 3) +#define PERF_COUNTER_INTERRUPT_STATUS (1UL << 3) +#define PIPE_HOTPLUG_TV_INTERRUPT_STATUS (1UL << 2) /* pre-965 */ +#define PIPE_START_VBLANK_INTERRUPT_STATUS (1UL << 2) /* 965 or later */ +#define PIPE_FRAMESTART_INTERRUPT_STATUS (1UL << 1) +#define PIPE_VBLANK_INTERRUPT_STATUS (1UL << 1) +#define PIPE_HBLANK_INT_STATUS (1UL << 0) +#define PIPE_OVERLAY_UPDATED_STATUS (1UL << 0) +#define PIPESTAT_INT_ENABLE_MASK 0x7fff0000 +#define PIPESTAT_INT_STATUS_MASK 0x0000ffff + +#define _PIPE_ARB_CTL_A 0x70028 /* icl+ */ +#define PIPE_ARB_CTL(dev_priv, pipe) _MMIO_PIPE2(dev_priv, pipe, _PIPE_ARB_CTL_A) +#define PIPE_ARB_USE_PROG_SLOTS REG_BIT(13) + +#define _PIPE_MISC_A 0x70030 +#define _PIPE_MISC_B 0x71030 +#define PIPE_MISC(pipe) _MMIO_PIPE(pipe, _PIPE_MISC_A, _PIPE_MISC_B) +#define PIPE_MISC_YUV420_ENABLE REG_BIT(27) /* glk+ */ +#define PIPE_MISC_YUV420_MODE_FULL_BLEND REG_BIT(26) /* glk+ */ +#define PIPE_MISC_HDR_MODE_PRECISION REG_BIT(23) /* icl+ */ +#define PIPE_MISC_PSR_MASK_PRIMARY_FLIP REG_BIT(23) /* bdw */ +#define PIPE_MISC_PSR_MASK_SPRITE_ENABLE REG_BIT(22) /* bdw */ +#define PIPE_MISC_PSR_MASK_PIPE_REG_WRITE REG_BIT(21) /* skl+ */ +#define PIPE_MISC_PSR_MASK_CURSOR_MOVE REG_BIT(21) /* bdw */ +#define PIPE_MISC_PSR_MASK_VBLANK_VSYNC_INT REG_BIT(20) +#define PIPE_MISC_OUTPUT_COLORSPACE_YUV REG_BIT(11) +#define PIPE_MISC_PIXEL_ROUNDING_TRUNC REG_BIT(8) /* tgl+ */ +/* + * For Display < 13, Bits 5-7 of PIPE MISC represent DITHER BPC with + * valid values of: 6, 8, 10 BPC. + * ADLP+, the bits 5-7 represent PORT OUTPUT BPC with valid values of: + * 6, 8, 10, 12 BPC. + */ +#define PIPE_MISC_BPC_MASK REG_GENMASK(7, 5) +#define PIPE_MISC_BPC_8 REG_FIELD_PREP(PIPE_MISC_BPC_MASK, 0) +#define PIPE_MISC_BPC_10 REG_FIELD_PREP(PIPE_MISC_BPC_MASK, 1) +#define PIPE_MISC_BPC_6 REG_FIELD_PREP(PIPE_MISC_BPC_MASK, 2) +#define PIPE_MISC_BPC_12_ADLP REG_FIELD_PREP(PIPE_MISC_BPC_MASK, 4) /* adlp+ */ +#define PIPE_MISC_DITHER_ENABLE REG_BIT(4) +#define PIPE_MISC_DITHER_TYPE_MASK REG_GENMASK(3, 2) +#define PIPE_MISC_DITHER_TYPE_SP REG_FIELD_PREP(PIPE_MISC_DITHER_TYPE_MASK, 0) +#define PIPE_MISC_DITHER_TYPE_ST1 REG_FIELD_PREP(PIPE_MISC_DITHER_TYPE_MASK, 1) +#define PIPE_MISC_DITHER_TYPE_ST2 REG_FIELD_PREP(PIPE_MISC_DITHER_TYPE_MASK, 2) +#define PIPE_MISC_DITHER_TYPE_TEMP REG_FIELD_PREP(PIPE_MISC_DITHER_TYPE_MASK, 3) + +#define _PIPE_MISC2_A 0x7002C +#define _PIPE_MISC2_B 0x7102C +#define PIPE_MISC2(pipe) _MMIO_PIPE(pipe, _PIPE_MISC2_A, _PIPE_MISC2_B) +#define PIPE_MISC2_BUBBLE_COUNTER_MASK REG_GENMASK(31, 24) +#define PIPE_MISC2_BUBBLE_COUNTER_SCALER_EN REG_FIELD_PREP(PIPE_MISC2_BUBBLE_COUNTER_MASK, 80) +#define PIPE_MISC2_BUBBLE_COUNTER_SCALER_DIS REG_FIELD_PREP(PIPE_MISC2_BUBBLE_COUNTER_MASK, 20) +#define PIPE_MISC2_FLIP_INFO_PLANE_SEL_MASK REG_GENMASK(2, 0) /* tgl+ */ +#define PIPE_MISC2_FLIP_INFO_PLANE_SEL(plane_id) REG_FIELD_PREP(PIPE_MISC2_FLIP_INFO_PLANE_SEL_MASK, (plane_id)) + +#define DPINVGTT _MMIO(VLV_DISPLAY_BASE + 0x7002c) /* VLV/CHV only */ +#define DPINVGTT_EN_MASK_CHV REG_GENMASK(27, 16) +#define DPINVGTT_EN_MASK_VLV REG_GENMASK(23, 16) +#define SPRITEF_INVALID_GTT_INT_EN REG_BIT(27) +#define SPRITEE_INVALID_GTT_INT_EN REG_BIT(26) +#define PLANEC_INVALID_GTT_INT_EN REG_BIT(25) +#define CURSORC_INVALID_GTT_INT_EN REG_BIT(24) +#define CURSORB_INVALID_GTT_INT_EN REG_BIT(23) +#define CURSORA_INVALID_GTT_INT_EN REG_BIT(22) +#define SPRITED_INVALID_GTT_INT_EN REG_BIT(21) +#define SPRITEC_INVALID_GTT_INT_EN REG_BIT(20) +#define PLANEB_INVALID_GTT_INT_EN REG_BIT(19) +#define SPRITEB_INVALID_GTT_INT_EN REG_BIT(18) +#define SPRITEA_INVALID_GTT_INT_EN REG_BIT(17) +#define PLANEA_INVALID_GTT_INT_EN REG_BIT(16) +#define DPINVGTT_STATUS_MASK_CHV REG_GENMASK(11, 0) +#define DPINVGTT_STATUS_MASK_VLV REG_GENMASK(7, 0) +#define SPRITEF_INVALID_GTT_STATUS REG_BIT(11) +#define SPRITEE_INVALID_GTT_STATUS REG_BIT(10) +#define PLANEC_INVALID_GTT_STATUS REG_BIT(9) +#define CURSORC_INVALID_GTT_STATUS REG_BIT(8) +#define CURSORB_INVALID_GTT_STATUS REG_BIT(7) +#define CURSORA_INVALID_GTT_STATUS REG_BIT(6) +#define SPRITED_INVALID_GTT_STATUS REG_BIT(5) +#define SPRITEC_INVALID_GTT_STATUS REG_BIT(4) +#define PLANEB_INVALID_GTT_STATUS REG_BIT(3) +#define SPRITEB_INVALID_GTT_STATUS REG_BIT(2) +#define SPRITEA_INVALID_GTT_STATUS REG_BIT(1) +#define PLANEA_INVALID_GTT_STATUS REG_BIT(0) + +#define CBR1_VLV _MMIO(VLV_DISPLAY_BASE + 0x70400) +#define CBR_PND_DEADLINE_DISABLE (1 << 31) +#define CBR_PWM_CLOCK_MUX_SELECT (1 << 30) + +#define CBR4_VLV _MMIO(VLV_DISPLAY_BASE + 0x70450) +#define CBR_DPLLBMD_PIPE(pipe) (1 << (7 + (pipe) * 11)) /* pipes B and C */ + +/* + * The two pipe frame counter registers are not synchronized, so + * reading a stable value is somewhat tricky. The following code + * should work: + * + * do { + * high1 = ((INREG(PIPEAFRAMEHIGH) & PIPE_FRAME_HIGH_MASK) >> + * PIPE_FRAME_HIGH_SHIFT; + * low1 = ((INREG(PIPEAFRAMEPIXEL) & PIPE_FRAME_LOW_MASK) >> + * PIPE_FRAME_LOW_SHIFT); + * high2 = ((INREG(PIPEAFRAMEHIGH) & PIPE_FRAME_HIGH_MASK) >> + * PIPE_FRAME_HIGH_SHIFT); + * } while (high1 != high2); + * frame = (high1 << 8) | low1; + */ +#define _PIPEAFRAMEHIGH 0x70040 +#define PIPEFRAME(dev_priv, pipe) _MMIO_PIPE2(dev_priv, pipe, _PIPEAFRAMEHIGH) +#define PIPE_FRAME_HIGH_MASK 0x0000ffff +#define PIPE_FRAME_HIGH_SHIFT 0 + +#define _PIPEAFRAMEPIXEL 0x70044 +#define PIPEFRAMEPIXEL(dev_priv, pipe) _MMIO_PIPE2(dev_priv, pipe, _PIPEAFRAMEPIXEL) +#define PIPE_FRAME_LOW_MASK 0xff000000 +#define PIPE_FRAME_LOW_SHIFT 24 +#define PIPE_PIXEL_MASK 0x00ffffff +#define PIPE_PIXEL_SHIFT 0 + +/* GM45+ just has to be different */ +#define _PIPEA_FRMCOUNT_G4X 0x70040 +#define PIPE_FRMCOUNT_G4X(dev_priv, pipe) _MMIO_PIPE2(dev_priv, pipe, _PIPEA_FRMCOUNT_G4X) + +#define _PIPEA_FLIPCOUNT_G4X 0x70044 +#define PIPE_FLIPCOUNT_G4X(dev_priv, pipe) _MMIO_PIPE2(dev_priv, pipe, _PIPEA_FLIPCOUNT_G4X) + +/* CHV pipe B blender */ +#define _CHV_BLEND_A 0x60a00 +#define CHV_BLEND(dev_priv, pipe) _MMIO_TRANS2(dev_priv, pipe, _CHV_BLEND_A) +#define CHV_BLEND_MASK REG_GENMASK(31, 30) +#define CHV_BLEND_LEGACY REG_FIELD_PREP(CHV_BLEND_MASK, 0) +#define CHV_BLEND_ANDROID REG_FIELD_PREP(CHV_BLEND_MASK, 1) +#define CHV_BLEND_MPO REG_FIELD_PREP(CHV_BLEND_MASK, 2) + +#define _CHV_CANVAS_A 0x60a04 +#define CHV_CANVAS(dev_priv, pipe) _MMIO_TRANS2(dev_priv, pipe, _CHV_CANVAS_A) +#define CHV_CANVAS_RED_MASK REG_GENMASK(29, 20) +#define CHV_CANVAS_GREEN_MASK REG_GENMASK(19, 10) +#define CHV_CANVAS_BLUE_MASK REG_GENMASK(9, 0) + +/* Display/Sprite base address macros */ +#define DISP_BASEADDR_MASK (0xfffff000) +#define I915_LO_DISPBASE(val) ((val) & ~DISP_BASEADDR_MASK) +#define I915_HI_DISPBASE(val) ((val) & DISP_BASEADDR_MASK) + +/* + * VBIOS flags + * gen2: + * [00:06] alm,mgm + * [10:16] all + * [30:32] alm,mgm + * gen3+: + * [00:0f] all + * [10:1f] all + * [30:32] all + */ +#define SWF0(dev_priv, i) _MMIO(DISPLAY_MMIO_BASE(dev_priv) + 0x70410 + (i) * 4) +#define SWF1(dev_priv, i) _MMIO(DISPLAY_MMIO_BASE(dev_priv) + 0x71410 + (i) * 4) +#define SWF3(dev_priv, i) _MMIO(DISPLAY_MMIO_BASE(dev_priv) + 0x72414 + (i) * 4) +#define SWF_ILK(i) _MMIO(0x4F000 + (i) * 4) + +#define DIGITAL_PORT_HOTPLUG_CNTRL _MMIO(0x44030) +#define DIGITAL_PORTA_HOTPLUG_ENABLE (1 << 4) +#define DIGITAL_PORTA_PULSE_DURATION_2ms (0 << 2) /* pre-HSW */ +#define DIGITAL_PORTA_PULSE_DURATION_4_5ms (1 << 2) /* pre-HSW */ +#define DIGITAL_PORTA_PULSE_DURATION_6ms (2 << 2) /* pre-HSW */ +#define DIGITAL_PORTA_PULSE_DURATION_100ms (3 << 2) /* pre-HSW */ +#define DIGITAL_PORTA_PULSE_DURATION_MASK (3 << 2) /* pre-HSW */ +#define DIGITAL_PORTA_HOTPLUG_STATUS_MASK (3 << 0) +#define DIGITAL_PORTA_HOTPLUG_NO_DETECT (0 << 0) +#define DIGITAL_PORTA_HOTPLUG_SHORT_DETECT (1 << 0) +#define DIGITAL_PORTA_HOTPLUG_LONG_DETECT (2 << 0) + +/* refresh rate hardware control */ +#define RR_HW_CTL _MMIO(0x45300) +#define RR_HW_LOW_POWER_FRAMES_MASK 0xff +#define RR_HW_HIGH_POWER_FRAMES_MASK 0xff00 + +#define _PIPEA_DATA_M1 0x60030 +#define _PIPEB_DATA_M1 0x61030 +#define PIPE_DATA_M1(dev_priv, tran) _MMIO_TRANS2(dev_priv, tran, _PIPEA_DATA_M1) + +#define _PIPEA_DATA_N1 0x60034 +#define _PIPEB_DATA_N1 0x61034 +#define PIPE_DATA_N1(dev_priv, tran) _MMIO_TRANS2(dev_priv, tran, _PIPEA_DATA_N1) + +#define _PIPEA_DATA_M2 0x60038 +#define _PIPEB_DATA_M2 0x61038 +#define PIPE_DATA_M2(dev_priv, tran) _MMIO_TRANS2(dev_priv, tran, _PIPEA_DATA_M2) + +#define _PIPEA_DATA_N2 0x6003c +#define _PIPEB_DATA_N2 0x6103c +#define PIPE_DATA_N2(dev_priv, tran) _MMIO_TRANS2(dev_priv, tran, _PIPEA_DATA_N2) + +#define _PIPEA_LINK_M1 0x60040 +#define _PIPEB_LINK_M1 0x61040 +#define PIPE_LINK_M1(dev_priv, tran) _MMIO_TRANS2(dev_priv, tran, _PIPEA_LINK_M1) + +#define _PIPEA_LINK_N1 0x60044 +#define _PIPEB_LINK_N1 0x61044 +#define PIPE_LINK_N1(dev_priv, tran) _MMIO_TRANS2(dev_priv, tran, _PIPEA_LINK_N1) + +#define _PIPEA_LINK_M2 0x60048 +#define _PIPEB_LINK_M2 0x61048 +#define PIPE_LINK_M2(dev_priv, tran) _MMIO_TRANS2(dev_priv, tran, _PIPEA_LINK_M2) + +#define _PIPEA_LINK_N2 0x6004c +#define _PIPEB_LINK_N2 0x6104c +#define PIPE_LINK_N2(dev_priv, tran) _MMIO_TRANS2(dev_priv, tran, _PIPEA_LINK_N2) + +/* + * Skylake scalers + */ +#define _ID(id, a, b) _PICK_EVEN(id, a, b) +#define _PS_1A_CTRL 0x68180 +#define _PS_2A_CTRL 0x68280 +#define _PS_1B_CTRL 0x68980 +#define _PS_2B_CTRL 0x68A80 +#define _PS_1C_CTRL 0x69180 +#define SKL_PS_CTRL(pipe, id) _MMIO_PIPE(pipe, \ + _ID(id, _PS_1A_CTRL, _PS_2A_CTRL), \ + _ID(id, _PS_1B_CTRL, _PS_2B_CTRL)) +#define PS_SCALER_EN REG_BIT(31) +#define PS_SCALER_TYPE_MASK REG_BIT(30) /* icl+ */ +#define PS_SCALER_TYPE_NON_LINEAR REG_FIELD_PREP(PS_SCALER_TYPE_MASK, 0) +#define PS_SCALER_TYPE_LINEAR REG_FIELD_PREP(PS_SCALER_TYPE_MASK, 1) +#define SKL_PS_SCALER_MODE_MASK REG_GENMASK(29, 28) /* skl/bxt */ +#define SKL_PS_SCALER_MODE_DYN REG_FIELD_PREP(SKL_PS_SCALER_MODE_MASK, 0) +#define SKL_PS_SCALER_MODE_HQ REG_FIELD_PREP(SKL_PS_SCALER_MODE_MASK, 1) +#define SKL_PS_SCALER_MODE_NV12 REG_FIELD_PREP(SKL_PS_SCALER_MODE_MASK, 2) +#define PS_SCALER_MODE_MASK REG_BIT(29) /* glk-tgl */ +#define PS_SCALER_MODE_NORMAL REG_FIELD_PREP(PS_SCALER_MODE_MASK, 0) +#define PS_SCALER_MODE_PLANAR REG_FIELD_PREP(PS_SCALER_MODE_MASK, 1) +#define PS_ADAPTIVE_FILTERING_EN REG_BIT(28) /* icl+ */ +#define PS_BINDING_MASK REG_GENMASK(27, 25) +#define PS_BINDING_PIPE REG_FIELD_PREP(PS_BINDING_MASK, 0) +#define PS_BINDING_PLANE(plane_id) REG_FIELD_PREP(PS_BINDING_MASK, (plane_id) + 1) +#define PS_FILTER_MASK REG_GENMASK(24, 23) +#define PS_FILTER_MEDIUM REG_FIELD_PREP(PS_FILTER_MASK, 0) +#define PS_FILTER_PROGRAMMED REG_FIELD_PREP(PS_FILTER_MASK, 1) +#define PS_FILTER_EDGE_ENHANCE REG_FIELD_PREP(PS_FILTER_MASK, 2) +#define PS_FILTER_BILINEAR REG_FIELD_PREP(PS_FILTER_MASK, 3) +#define PS_ADAPTIVE_FILTER_MASK REG_BIT(22) /* icl+ */ +#define PS_ADAPTIVE_FILTER_MEDIUM REG_FIELD_PREP(PS_ADAPTIVE_FILTER_MASK, 0) +#define PS_ADAPTIVE_FILTER_EDGE_ENHANCE REG_FIELD_PREP(PS_ADAPTIVE_FILTER_MASK, 1) +#define PS_PIPE_SCALER_LOC_MASK REG_BIT(21) /* icl+ */ +#define PS_PIPE_SCALER_LOC_AFTER_OUTPUT_CSC REG_FIELD_PREP(PS_SCALER_LOCATION_MASK, 0) /* non-linear */ +#define PS_PIPE_SCALER_LOC_AFTER_CSC REG_FIELD_PREP(PS_SCALER_LOCATION_MASK, 1) /* linear */ +#define PS_VERT3TAP REG_BIT(21) /* skl/bxt */ +#define PS_VERT_INT_INVERT_FIELD REG_BIT(20) +#define PS_PROG_SCALE_FACTOR REG_BIT(19) /* tgl+ */ +#define PS_PWRUP_PROGRESS REG_BIT(17) +#define PS_V_FILTER_BYPASS REG_BIT(8) +#define PS_VADAPT_EN REG_BIT(7) /* skl/bxt */ +#define PS_VADAPT_MODE_MASK REG_GENMASK(6, 5) /* skl/bxt */ +#define PS_VADAPT_MODE_LEAST_ADAPT REG_FIELD_PREP(PS_VADAPT_MODE_MASK, 0) +#define PS_VADAPT_MODE_MOD_ADAPT REG_FIELD_PREP(PS_VADAPT_MODE_MASK, 1) +#define PS_VADAPT_MODE_MOST_ADAPT REG_FIELD_PREP(PS_VADAPT_MODE_MASK, 3) +#define PS_BINDING_Y_MASK REG_GENMASK(7, 5) /* icl-tgl */ +#define PS_BINDING_Y_PLANE(plane_id) REG_FIELD_PREP(PS_BINDING_Y_MASK, (plane_id) + 1) +#define PS_Y_VERT_FILTER_SELECT_MASK REG_BIT(4) /* glk+ */ +#define PS_Y_VERT_FILTER_SELECT(set) REG_FIELD_PREP(PS_Y_VERT_FILTER_SELECT_MASK, (set)) +#define PS_Y_HORZ_FILTER_SELECT_MASK REG_BIT(3) /* glk+ */ +#define PS_Y_HORZ_FILTER_SELECT(set) REG_FIELD_PREP(PS_Y_HORZ_FILTER_SELECT_MASK, (set)) +#define PS_UV_VERT_FILTER_SELECT_MASK REG_BIT(2) /* glk+ */ +#define PS_UV_VERT_FILTER_SELECT(set) REG_FIELD_PREP(PS_UV_VERT_FILTER_SELECT_MASK, (set)) +#define PS_UV_HORZ_FILTER_SELECT_MASK REG_BIT(1) /* glk+ */ +#define PS_UV_HORZ_FILTER_SELECT(set) REG_FIELD_PREP(PS_UV_HORZ_FILTER_SELECT_MASK, (set)) + +#define _PS_PWR_GATE_1A 0x68160 +#define _PS_PWR_GATE_2A 0x68260 +#define _PS_PWR_GATE_1B 0x68960 +#define _PS_PWR_GATE_2B 0x68A60 +#define _PS_PWR_GATE_1C 0x69160 +#define SKL_PS_PWR_GATE(pipe, id) _MMIO_PIPE(pipe, \ + _ID(id, _PS_PWR_GATE_1A, _PS_PWR_GATE_2A), \ + _ID(id, _PS_PWR_GATE_1B, _PS_PWR_GATE_2B)) +#define PS_PWR_GATE_DIS_OVERRIDE REG_BIT(31) +#define PS_PWR_GATE_SETTLING_TIME_MASK REG_GENMASK(4, 3) +#define PS_PWR_GATE_SETTLING_TIME_32 REG_FIELD_PREP(PS_PWR_GATE_SETTLING_TIME_MASK, 0) +#define PS_PWR_GATE_SETTLING_TIME_64 REG_FIELD_PREP(PS_PWR_GATE_SETTLING_TIME_MASK, 1) +#define PS_PWR_GATE_SETTLING_TIME_96 REG_FIELD_PREP(PS_PWR_GATE_SETTLING_TIME_MASK, 2) +#define PS_PWR_GATE_SETTLING_TIME_128 REG_FIELD_PREP(PS_PWR_GATE_SETTLING_TIME_MASK, 3) +#define PS_PWR_GATE_SLPEN_MASK REG_GENMASK(1, 0) +#define PS_PWR_GATE_SLPEN_8 REG_FIELD_PREP(PS_PWR_GATE_SLPEN_MASK, 0) +#define PS_PWR_GATE_SLPEN_16 REG_FIELD_PREP(PS_PWR_GATE_SLPEN_MASK, 1) +#define PS_PWR_GATE_SLPEN_24 REG_FIELD_PREP(PS_PWR_GATE_SLPEN_MASK, 2) +#define PS_PWR_GATE_SLPEN_32 REG_FIELD_PREP(PS_PWR_GATE_SLPEN_MASK, 3) + +#define _PS_WIN_POS_1A 0x68170 +#define _PS_WIN_POS_2A 0x68270 +#define _PS_WIN_POS_1B 0x68970 +#define _PS_WIN_POS_2B 0x68A70 +#define _PS_WIN_POS_1C 0x69170 +#define SKL_PS_WIN_POS(pipe, id) _MMIO_PIPE(pipe, \ + _ID(id, _PS_WIN_POS_1A, _PS_WIN_POS_2A), \ + _ID(id, _PS_WIN_POS_1B, _PS_WIN_POS_2B)) +#define PS_WIN_XPOS_MASK REG_GENMASK(31, 16) +#define PS_WIN_XPOS(x) REG_FIELD_PREP(PS_WIN_XPOS_MASK, (x)) +#define PS_WIN_YPOS_MASK REG_GENMASK(15, 0) +#define PS_WIN_YPOS(y) REG_FIELD_PREP(PS_WIN_YPOS_MASK, (y)) + +#define _PS_WIN_SZ_1A 0x68174 +#define _PS_WIN_SZ_2A 0x68274 +#define _PS_WIN_SZ_1B 0x68974 +#define _PS_WIN_SZ_2B 0x68A74 +#define _PS_WIN_SZ_1C 0x69174 +#define SKL_PS_WIN_SZ(pipe, id) _MMIO_PIPE(pipe, \ + _ID(id, _PS_WIN_SZ_1A, _PS_WIN_SZ_2A), \ + _ID(id, _PS_WIN_SZ_1B, _PS_WIN_SZ_2B)) +#define PS_WIN_XSIZE_MASK REG_GENMASK(31, 16) +#define PS_WIN_XSIZE(w) REG_FIELD_PREP(PS_WIN_XSIZE_MASK, (w)) +#define PS_WIN_YSIZE_MASK REG_GENMASK(15, 0) +#define PS_WIN_YSIZE(h) REG_FIELD_PREP(PS_WIN_YSIZE_MASK, (h)) + +#define _PS_VSCALE_1A 0x68184 +#define _PS_VSCALE_2A 0x68284 +#define _PS_VSCALE_1B 0x68984 +#define _PS_VSCALE_2B 0x68A84 +#define _PS_VSCALE_1C 0x69184 +#define SKL_PS_VSCALE(pipe, id) _MMIO_PIPE(pipe, \ + _ID(id, _PS_VSCALE_1A, _PS_VSCALE_2A), \ + _ID(id, _PS_VSCALE_1B, _PS_VSCALE_2B)) + +#define _PS_HSCALE_1A 0x68190 +#define _PS_HSCALE_2A 0x68290 +#define _PS_HSCALE_1B 0x68990 +#define _PS_HSCALE_2B 0x68A90 +#define _PS_HSCALE_1C 0x69190 +#define SKL_PS_HSCALE(pipe, id) _MMIO_PIPE(pipe, \ + _ID(id, _PS_HSCALE_1A, _PS_HSCALE_2A), \ + _ID(id, _PS_HSCALE_1B, _PS_HSCALE_2B)) + +#define _PS_VPHASE_1A 0x68188 +#define _PS_VPHASE_2A 0x68288 +#define _PS_VPHASE_1B 0x68988 +#define _PS_VPHASE_2B 0x68A88 +#define _PS_VPHASE_1C 0x69188 +#define SKL_PS_VPHASE(pipe, id) _MMIO_PIPE(pipe, \ + _ID(id, _PS_VPHASE_1A, _PS_VPHASE_2A), \ + _ID(id, _PS_VPHASE_1B, _PS_VPHASE_2B)) +#define PS_Y_PHASE_MASK REG_GENMASK(31, 16) +#define PS_Y_PHASE(x) REG_FIELD_PREP(PS_Y_PHASE_MASK, (x)) +#define PS_UV_RGB_PHASE_MASK REG_GENMASK(15, 0) +#define PS_UV_RGB_PHASE(x) REG_FIELD_PREP(PS_UV_RGB_PHASE_MASK, (x)) +#define PS_PHASE_MASK (0x7fff << 1) /* u2.13 */ +#define PS_PHASE_TRIP (1 << 0) + +#define _PS_HPHASE_1A 0x68194 +#define _PS_HPHASE_2A 0x68294 +#define _PS_HPHASE_1B 0x68994 +#define _PS_HPHASE_2B 0x68A94 +#define _PS_HPHASE_1C 0x69194 +#define SKL_PS_HPHASE(pipe, id) _MMIO_PIPE(pipe, \ + _ID(id, _PS_HPHASE_1A, _PS_HPHASE_2A), \ + _ID(id, _PS_HPHASE_1B, _PS_HPHASE_2B)) + +#define _PS_ECC_STAT_1A 0x681D0 +#define _PS_ECC_STAT_2A 0x682D0 +#define _PS_ECC_STAT_1B 0x689D0 +#define _PS_ECC_STAT_2B 0x68AD0 +#define _PS_ECC_STAT_1C 0x691D0 +#define SKL_PS_ECC_STAT(pipe, id) _MMIO_PIPE(pipe, \ + _ID(id, _PS_ECC_STAT_1A, _PS_ECC_STAT_2A), \ + _ID(id, _PS_ECC_STAT_1B, _PS_ECC_STAT_2B)) + +#define _PS_COEF_SET0_INDEX_1A 0x68198 +#define _PS_COEF_SET0_INDEX_2A 0x68298 +#define _PS_COEF_SET0_INDEX_1B 0x68998 +#define _PS_COEF_SET0_INDEX_2B 0x68A98 +#define GLK_PS_COEF_INDEX_SET(pipe, id, set) _MMIO_PIPE(pipe, \ + _ID(id, _PS_COEF_SET0_INDEX_1A, _PS_COEF_SET0_INDEX_2A) + (set) * 8, \ + _ID(id, _PS_COEF_SET0_INDEX_1B, _PS_COEF_SET0_INDEX_2B) + (set) * 8) +#define PS_COEF_INDEX_AUTO_INC REG_BIT(10) + +#define _PS_COEF_SET0_DATA_1A 0x6819C +#define _PS_COEF_SET0_DATA_2A 0x6829C +#define _PS_COEF_SET0_DATA_1B 0x6899C +#define _PS_COEF_SET0_DATA_2B 0x68A9C +#define GLK_PS_COEF_DATA_SET(pipe, id, set) _MMIO_PIPE(pipe, \ + _ID(id, _PS_COEF_SET0_DATA_1A, _PS_COEF_SET0_DATA_2A) + (set) * 8, \ + _ID(id, _PS_COEF_SET0_DATA_1B, _PS_COEF_SET0_DATA_2B) + (set) * 8) + +/* More Ivybridge lolz */ +#define DE_ERR_INT_IVB (1 << 30) +#define DE_GSE_IVB (1 << 29) +#define DE_PCH_EVENT_IVB (1 << 28) +#define DE_DP_A_HOTPLUG_IVB (1 << 27) +#define DE_AUX_CHANNEL_A_IVB (1 << 26) +#define DE_EDP_PSR_INT_HSW (1 << 19) +#define DE_SPRITEC_FLIP_DONE_IVB (1 << 14) +#define DE_PLANEC_FLIP_DONE_IVB (1 << 13) +#define DE_PIPEC_VBLANK_IVB (1 << 10) +#define DE_SPRITEB_FLIP_DONE_IVB (1 << 9) +#define DE_PLANEB_FLIP_DONE_IVB (1 << 8) +#define DE_PIPEB_VBLANK_IVB (1 << 5) +#define DE_SPRITEA_FLIP_DONE_IVB (1 << 4) +#define DE_PLANEA_FLIP_DONE_IVB (1 << 3) +#define DE_PLANE_FLIP_DONE_IVB(plane) (1 << (3 + 5 * (plane))) +#define DE_PIPEA_VBLANK_IVB (1 << 0) +#define DE_PIPE_VBLANK_IVB(pipe) (1 << ((pipe) * 5)) + +#define XELPD_DISPLAY_ERR_FATAL_MASK _MMIO(0x4421c) + +#define GEN8_DE_PIPE_ISR(pipe) _MMIO(0x44400 + (0x10 * (pipe))) +#define GEN8_DE_PIPE_IMR(pipe) _MMIO(0x44404 + (0x10 * (pipe))) +#define GEN8_DE_PIPE_IIR(pipe) _MMIO(0x44408 + (0x10 * (pipe))) +#define GEN8_DE_PIPE_IER(pipe) _MMIO(0x4440c + (0x10 * (pipe))) +#define GEN8_PIPE_FIFO_UNDERRUN REG_BIT(31) +#define GEN8_PIPE_CDCLK_CRC_ERROR REG_BIT(29) +#define GEN8_PIPE_CDCLK_CRC_DONE REG_BIT(28) +#define GEN12_PIPEDMC_INTERRUPT REG_BIT(26) /* tgl+ */ +#define GEN12_PIPEDMC_FAULT REG_BIT(25) /* tgl-mtl */ +#define MTL_PIPEDMC_ATS_FAULT REG_BIT(24) /* mtl */ +#define GEN12_PIPEDMC_FLIPQ_DONE REG_BIT(24) /* tgl-adl */ +#define GEN11_PIPE_PLANE7_FAULT REG_BIT(22) /* icl/tgl */ +#define GEN11_PIPE_PLANE6_FAULT REG_BIT(21) /* icl/tgl */ +#define GEN11_PIPE_PLANE5_FAULT REG_BIT(20) /* icl+ */ +#define GEN12_PIPE_VBLANK_UNMOD REG_BIT(19) /* tgl+ */ +#define MTL_PLANE_ATS_FAULT REG_BIT(18) /* mtl+ */ +#define GEN11_PIPE_PLANE7_FLIP_DONE REG_BIT(18) /* icl/tgl */ +#define MTL_PIPEDMC_FLIPQ_DONE REG_BIT(17) /* mtl */ +#define GEN11_PIPE_PLANE6_FLIP_DONE REG_BIT(17) /* icl/tgl */ +#define GEN11_PIPE_PLANE5_FLIP_DONE REG_BIT(16) /* icl+ */ +#define GEN12_DSB_2_INT REG_BIT(15) /* tgl+ */ +#define GEN12_DSB_1_INT REG_BIT(14) /* tgl+ */ +#define GEN12_DSB_0_INT REG_BIT(13) /* tgl+ */ +#define GEN12_DSB_INT(dsb_id) REG_BIT(13 + (dsb_id)) +#define GEN9_PIPE_CURSOR_FAULT REG_BIT(11) /* skl+ */ +#define GEN9_PIPE_PLANE4_FAULT REG_BIT(10) /* skl+ */ +#define GEN8_PIPE_CURSOR_FAULT REG_BIT(10) /* bdw */ +#define GEN9_PIPE_PLANE3_FAULT REG_BIT(9) /* skl+ */ +#define GEN8_PIPE_SPRITE_FAULT REG_BIT(9) /* bdw */ +#define GEN9_PIPE_PLANE2_FAULT REG_BIT(8) /* skl+ */ +#define GEN8_PIPE_PRIMARY_FAULT REG_BIT(8) /* bdw */ +#define GEN9_PIPE_PLANE1_FAULT REG_BIT(7) /* skl+ */ +#define GEN9_PIPE_PLANE4_FLIP_DONE REG_BIT(6) /* skl+ */ +#define GEN9_PIPE_PLANE3_FLIP_DONE REG_BIT(5) /* skl+ */ +#define GEN8_PIPE_SPRITE_FLIP_DONE REG_BIT(5) /* bdw */ +#define GEN9_PIPE_PLANE2_FLIP_DONE REG_BIT(4) /* skl+ */ +#define GEN8_PIPE_PRIMARY_FLIP_DONE REG_BIT(4) /* bdw */ +#define GEN9_PIPE_PLANE1_FLIP_DONE REG_BIT(3) /* skl+ */ +#define GEN9_PIPE_PLANE_FLIP_DONE(plane_id) \ + REG_BIT(((plane_id) >= PLANE_5 ? 16 - PLANE_5 : 3 - PLANE_1) + (plane_id)) /* skl+ */ +#define GEN8_PIPE_SCAN_LINE_EVENT REG_BIT(2) +#define GEN8_PIPE_VSYNC REG_BIT(1) +#define GEN8_PIPE_VBLANK REG_BIT(0) + +#define GEN8_DE_PIPE_IRQ_REGS(pipe) I915_IRQ_REGS(GEN8_DE_PIPE_IMR(pipe), \ + GEN8_DE_PIPE_IER(pipe), \ + GEN8_DE_PIPE_IIR(pipe)) + +#define _HPD_PIN_DDI(hpd_pin) ((hpd_pin) - HPD_PORT_A) +#define _HPD_PIN_TC(hpd_pin) ((hpd_pin) - HPD_PORT_TC1) + +#define GEN8_DE_PORT_ISR _MMIO(0x44440) +#define GEN8_DE_PORT_IMR _MMIO(0x44444) +#define GEN8_DE_PORT_IIR _MMIO(0x44448) +#define GEN8_DE_PORT_IER _MMIO(0x4444c) +#define DSI1_NON_TE (1 << 31) +#define DSI0_NON_TE (1 << 30) +#define ICL_AUX_CHANNEL_E (1 << 29) +#define ICL_AUX_CHANNEL_F (1 << 28) +#define GEN9_AUX_CHANNEL_D (1 << 27) +#define GEN9_AUX_CHANNEL_C (1 << 26) +#define GEN9_AUX_CHANNEL_B (1 << 25) +#define DSI1_TE (1 << 24) +#define DSI0_TE (1 << 23) +#define GEN8_DE_PORT_HOTPLUG(hpd_pin) REG_BIT(3 + _HPD_PIN_DDI(hpd_pin)) +#define BXT_DE_PORT_HOTPLUG_MASK (GEN8_DE_PORT_HOTPLUG(HPD_PORT_A) | \ + GEN8_DE_PORT_HOTPLUG(HPD_PORT_B) | \ + GEN8_DE_PORT_HOTPLUG(HPD_PORT_C)) +#define BDW_DE_PORT_HOTPLUG_MASK GEN8_DE_PORT_HOTPLUG(HPD_PORT_A) +#define BXT_DE_PORT_GMBUS (1 << 1) +#define GEN8_AUX_CHANNEL_A (1 << 0) +#define TGL_DE_PORT_AUX_USBC6 REG_BIT(13) +#define XELPD_DE_PORT_AUX_DDIE REG_BIT(13) +#define TGL_DE_PORT_AUX_USBC5 REG_BIT(12) +#define XELPD_DE_PORT_AUX_DDID REG_BIT(12) +#define TGL_DE_PORT_AUX_USBC4 REG_BIT(11) +#define TGL_DE_PORT_AUX_USBC3 REG_BIT(10) +#define TGL_DE_PORT_AUX_USBC2 REG_BIT(9) +#define TGL_DE_PORT_AUX_USBC1 REG_BIT(8) +#define TGL_DE_PORT_AUX_DDIC REG_BIT(2) +#define TGL_DE_PORT_AUX_DDIB REG_BIT(1) +#define TGL_DE_PORT_AUX_DDIA REG_BIT(0) + +#define GEN8_DE_PORT_IRQ_REGS I915_IRQ_REGS(GEN8_DE_PORT_IMR, \ + GEN8_DE_PORT_IER, \ + GEN8_DE_PORT_IIR) + +#define GEN8_DE_MISC_ISR _MMIO(0x44460) +#define GEN8_DE_MISC_IMR _MMIO(0x44464) +#define GEN8_DE_MISC_IIR _MMIO(0x44468) +#define GEN8_DE_MISC_IER _MMIO(0x4446c) +#define XELPDP_RM_TIMEOUT REG_BIT(29) +#define XELPDP_PMDEMAND_RSPTOUT_ERR REG_BIT(27) +#define GEN8_DE_MISC_GSE REG_BIT(27) +#define GEN8_DE_EDP_PSR REG_BIT(19) +#define XELPDP_PMDEMAND_RSP REG_BIT(3) +#define XE2LPD_DBUF_OVERLAP_DETECTED REG_BIT(1) + +#define GEN8_DE_MISC_IRQ_REGS I915_IRQ_REGS(GEN8_DE_MISC_IMR, \ + GEN8_DE_MISC_IER, \ + GEN8_DE_MISC_IIR) + +#define GEN11_DISPLAY_INT_CTL _MMIO(0x44200) +#define GEN11_DISPLAY_IRQ_ENABLE (1 << 31) +#define GEN11_AUDIO_CODEC_IRQ (1 << 24) +#define GEN11_DE_PCH_IRQ (1 << 23) +#define GEN11_DE_MISC_IRQ (1 << 22) +#define GEN11_DE_HPD_IRQ (1 << 21) +#define GEN11_DE_PORT_IRQ (1 << 20) +#define GEN11_DE_PIPE_C (1 << 18) +#define GEN11_DE_PIPE_B (1 << 17) +#define GEN11_DE_PIPE_A (1 << 16) + +#define GEN11_DE_HPD_ISR _MMIO(0x44470) +#define GEN11_DE_HPD_IMR _MMIO(0x44474) +#define GEN11_DE_HPD_IIR _MMIO(0x44478) +#define GEN11_DE_HPD_IER _MMIO(0x4447c) +#define GEN11_TC_HOTPLUG(hpd_pin) REG_BIT(16 + _HPD_PIN_TC(hpd_pin)) +#define GEN11_DE_TC_HOTPLUG_MASK (GEN11_TC_HOTPLUG(HPD_PORT_TC6) | \ + GEN11_TC_HOTPLUG(HPD_PORT_TC5) | \ + GEN11_TC_HOTPLUG(HPD_PORT_TC4) | \ + GEN11_TC_HOTPLUG(HPD_PORT_TC3) | \ + GEN11_TC_HOTPLUG(HPD_PORT_TC2) | \ + GEN11_TC_HOTPLUG(HPD_PORT_TC1)) +#define GEN11_TBT_HOTPLUG(hpd_pin) REG_BIT(_HPD_PIN_TC(hpd_pin)) +#define GEN11_DE_TBT_HOTPLUG_MASK (GEN11_TBT_HOTPLUG(HPD_PORT_TC6) | \ + GEN11_TBT_HOTPLUG(HPD_PORT_TC5) | \ + GEN11_TBT_HOTPLUG(HPD_PORT_TC4) | \ + GEN11_TBT_HOTPLUG(HPD_PORT_TC3) | \ + GEN11_TBT_HOTPLUG(HPD_PORT_TC2) | \ + GEN11_TBT_HOTPLUG(HPD_PORT_TC1)) + +#define GEN11_DE_HPD_IRQ_REGS I915_IRQ_REGS(GEN11_DE_HPD_IMR, \ + GEN11_DE_HPD_IER, \ + GEN11_DE_HPD_IIR) + +#define GEN11_TBT_HOTPLUG_CTL _MMIO(0x44030) +#define GEN11_TC_HOTPLUG_CTL _MMIO(0x44038) +#define GEN11_HOTPLUG_CTL_ENABLE(hpd_pin) (8 << (_HPD_PIN_TC(hpd_pin) * 4)) +#define GEN11_HOTPLUG_CTL_LONG_DETECT(hpd_pin) (2 << (_HPD_PIN_TC(hpd_pin) * 4)) +#define GEN11_HOTPLUG_CTL_SHORT_DETECT(hpd_pin) (1 << (_HPD_PIN_TC(hpd_pin) * 4)) +#define GEN11_HOTPLUG_CTL_NO_DETECT(hpd_pin) (0 << (_HPD_PIN_TC(hpd_pin) * 4)) + +#define PICAINTERRUPT_ISR _MMIO(0x16FE50) +#define PICAINTERRUPT_IMR _MMIO(0x16FE54) +#define PICAINTERRUPT_IIR _MMIO(0x16FE58) +#define PICAINTERRUPT_IER _MMIO(0x16FE5C) +#define XELPDP_DP_ALT_HOTPLUG(hpd_pin) REG_BIT(16 + _HPD_PIN_TC(hpd_pin)) +#define XELPDP_DP_ALT_HOTPLUG_MASK REG_GENMASK(19, 16) +#define XELPDP_AUX_TC(hpd_pin) REG_BIT(8 + _HPD_PIN_TC(hpd_pin)) +#define XELPDP_AUX_TC_MASK REG_GENMASK(11, 8) +#define XE2LPD_AUX_DDI(hpd_pin) REG_BIT(6 + _HPD_PIN_DDI(hpd_pin)) +#define XE2LPD_AUX_DDI_MASK REG_GENMASK(7, 6) +#define XELPDP_TBT_HOTPLUG(hpd_pin) REG_BIT(_HPD_PIN_TC(hpd_pin)) +#define XELPDP_TBT_HOTPLUG_MASK REG_GENMASK(3, 0) + +#define PICAINTERRUPT_IRQ_REGS I915_IRQ_REGS(PICAINTERRUPT_IMR, \ + PICAINTERRUPT_IER, \ + PICAINTERRUPT_IIR) + +#define XELPDP_PORT_HOTPLUG_CTL(hpd_pin) _MMIO(0x16F270 + (_HPD_PIN_TC(hpd_pin) * 0x200)) +#define XELPDP_TBT_HOTPLUG_ENABLE REG_BIT(6) +#define XELPDP_TBT_HPD_LONG_DETECT REG_BIT(5) +#define XELPDP_TBT_HPD_SHORT_DETECT REG_BIT(4) +#define XELPDP_DP_ALT_HOTPLUG_ENABLE REG_BIT(2) +#define XELPDP_DP_ALT_HPD_LONG_DETECT REG_BIT(1) +#define XELPDP_DP_ALT_HPD_SHORT_DETECT REG_BIT(0) + +#define XELPDP_INITIATE_PMDEMAND_REQUEST(dword) _MMIO(0x45230 + 4 * (dword)) +#define XELPDP_PMDEMAND_QCLK_GV_BW_MASK REG_GENMASK(31, 16) +#define XELPDP_PMDEMAND_VOLTAGE_INDEX_MASK REG_GENMASK(14, 12) +#define XELPDP_PMDEMAND_QCLK_GV_INDEX_MASK REG_GENMASK(11, 8) +#define XE3_PMDEMAND_PIPES_MASK REG_GENMASK(7, 4) +#define XELPDP_PMDEMAND_PIPES_MASK REG_GENMASK(7, 6) +#define XELPDP_PMDEMAND_DBUFS_MASK REG_GENMASK(5, 4) +#define XELPDP_PMDEMAND_PHYS_MASK REG_GENMASK(2, 0) + +#define XELPDP_PMDEMAND_REQ_ENABLE REG_BIT(31) +#define XELPDP_PMDEMAND_CDCLK_FREQ_MASK REG_GENMASK(30, 20) +#define XELPDP_PMDEMAND_DDICLK_FREQ_MASK REG_GENMASK(18, 8) +#define XELPDP_PMDEMAND_SCALERS_MASK REG_GENMASK(6, 4) +#define XELPDP_PMDEMAND_PLLS_MASK REG_GENMASK(2, 0) + +#define GEN12_DCPR_STATUS_1 _MMIO(0x46440) +#define XELPDP_PMDEMAND_INFLIGHT_STATUS REG_BIT(26) + +#define FUSE_STRAP _MMIO(0x42014) +#define ILK_INTERNAL_GRAPHICS_DISABLE REG_BIT(31) +#define ILK_INTERNAL_DISPLAY_DISABLE REG_BIT(30) +#define ILK_DISPLAY_DEBUG_DISABLE REG_BIT(29) +#define IVB_PIPE_C_DISABLE REG_BIT(28) +#define ILK_HDCP_DISABLE REG_BIT(25) +#define ILK_eDP_A_DISABLE REG_BIT(24) +#define HSW_CDCLK_LIMIT REG_BIT(24) +#define ILK_DESKTOP REG_BIT(23) +#define HSW_CPU_SSC_ENABLE REG_BIT(21) + +#define FUSE_STRAP3 _MMIO(0x42020) +#define HSW_REF_CLK_SELECT REG_BIT(1) + +#define CHICKEN_MISC_2 _MMIO(0x42084) +#define CHICKEN_MISC_DISABLE_DPT REG_BIT(30) /* adl,dg2 */ +#define BMG_DARB_HALF_BLK_END_BURST REG_BIT(27) +#define KBL_ARB_FILL_SPARE_14 REG_BIT(14) +#define KBL_ARB_FILL_SPARE_13 REG_BIT(13) +#define GLK_CL2_PWR_DOWN REG_BIT(12) +#define GLK_CL1_PWR_DOWN REG_BIT(11) +#define GLK_CL0_PWR_DOWN REG_BIT(10) + +#define CHICKEN_MISC_3 _MMIO(0x42088) +#define DP_MST_DPT_DPTP_ALIGN_WA(trans) REG_BIT(9 + (trans) - TRANSCODER_A) +#define DP_MST_SHORT_HBLANK_WA(trans) REG_BIT(5 + (trans) - TRANSCODER_A) +#define DP_MST_FEC_BS_JITTER_WA(trans) REG_BIT(0 + (trans) - TRANSCODER_A) + +#define CHICKEN_MISC_4 _MMIO(0x4208c) +#define CHICKEN_FBC_STRIDE_OVERRIDE REG_BIT(13) +#define CHICKEN_FBC_STRIDE_MASK REG_GENMASK(12, 0) +#define CHICKEN_FBC_STRIDE(x) REG_FIELD_PREP(CHICKEN_FBC_STRIDE_MASK, (x)) + +#define _CHICKEN_TRANS_A 0x420c0 +#define _CHICKEN_TRANS_B 0x420c4 +#define _CHICKEN_TRANS_C 0x420c8 +#define _CHICKEN_TRANS_EDP 0x420cc +#define _CHICKEN_TRANS_D 0x420d8 +#define _CHICKEN_TRANS(trans) _MMIO(_PICK((trans), \ + [TRANSCODER_EDP] = _CHICKEN_TRANS_EDP, \ + [TRANSCODER_A] = _CHICKEN_TRANS_A, \ + [TRANSCODER_B] = _CHICKEN_TRANS_B, \ + [TRANSCODER_C] = _CHICKEN_TRANS_C, \ + [TRANSCODER_D] = _CHICKEN_TRANS_D)) +#define _MTL_CHICKEN_TRANS_A 0x604e0 +#define _MTL_CHICKEN_TRANS_B 0x614e0 +#define _MTL_CHICKEN_TRANS(trans) _MMIO_TRANS((trans), \ + _MTL_CHICKEN_TRANS_A, \ + _MTL_CHICKEN_TRANS_B) +#define CHICKEN_TRANS(display, trans) (DISPLAY_VER(display) >= 14 ? _MTL_CHICKEN_TRANS(trans) : _CHICKEN_TRANS(trans)) +#define PIPE_VBLANK_WITH_DELAY REG_BIT(31) /* tgl+ */ +#define SKL_UNMASK_VBL_TO_PIPE_IN_SRD REG_BIT(30) /* skl+ */ +#define HSW_FRAME_START_DELAY_MASK REG_GENMASK(28, 27) +#define HSW_FRAME_START_DELAY(x) REG_FIELD_PREP(HSW_FRAME_START_DELAY_MASK, x) +#define VSC_DATA_SEL_SOFTWARE_CONTROL REG_BIT(25) /* GLK */ +#define FECSTALL_DIS_DPTSTREAM_DPTTG REG_BIT(23) +#define DDI_TRAINING_OVERRIDE_ENABLE REG_BIT(19) +#define ADLP_1_BASED_X_GRANULARITY REG_BIT(18) +#define DDI_TRAINING_OVERRIDE_VALUE REG_BIT(18) +#define DDIE_TRAINING_OVERRIDE_ENABLE REG_BIT(17) /* CHICKEN_TRANS_A only */ +#define DDIE_TRAINING_OVERRIDE_VALUE REG_BIT(16) /* CHICKEN_TRANS_A only */ +#define PSR2_ADD_VERTICAL_LINE_COUNT REG_BIT(15) +#define DP_FEC_BS_JITTER_WA REG_BIT(15) +#define PSR2_VSC_ENABLE_PROG_HEADER REG_BIT(12) +#define DP_DSC_INSERT_SF_AT_EOL_WA REG_BIT(4) +#define HDCP_LINE_REKEY_DISABLE REG_BIT(0) + +#define DISP_ARB_CTL2 _MMIO(0x45004) +#define DISP_DATA_PARTITION_5_6 REG_BIT(6) +#define DISP_IPC_ENABLE REG_BIT(3) + +#define GEN7_MSG_CTL _MMIO(0x45010) +#define WAIT_FOR_PCH_RESET_ACK (1 << 1) +#define WAIT_FOR_PCH_FLR_ACK (1 << 0) + +#define _BW_BUDDY0_CTL 0x45130 +#define _BW_BUDDY1_CTL 0x45140 +#define BW_BUDDY_CTL(x) _MMIO(_PICK_EVEN(x, \ + _BW_BUDDY0_CTL, \ + _BW_BUDDY1_CTL)) +#define BW_BUDDY_DISABLE REG_BIT(31) +#define BW_BUDDY_TLB_REQ_TIMER_MASK REG_GENMASK(21, 16) +#define BW_BUDDY_TLB_REQ_TIMER(x) REG_FIELD_PREP(BW_BUDDY_TLB_REQ_TIMER_MASK, x) + +#define _BW_BUDDY0_PAGE_MASK 0x45134 +#define _BW_BUDDY1_PAGE_MASK 0x45144 +#define BW_BUDDY_PAGE_MASK(x) _MMIO(_PICK_EVEN(x, \ + _BW_BUDDY0_PAGE_MASK, \ + _BW_BUDDY1_PAGE_MASK)) + +#define HSW_NDE_RSTWRN_OPT _MMIO(0x46408) +#define MTL_RESET_PICA_HANDSHAKE_EN REG_BIT(6) +#define RESET_PCH_HANDSHAKE_ENABLE REG_BIT(4) + +#define GEN11_CHICKEN_DCPR_2 _MMIO(0x46434) +#define DCPR_MASK_MAXLATENCY_MEMUP_CLR REG_BIT(27) +#define DCPR_MASK_LPMODE REG_BIT(26) +#define DCPR_SEND_RESP_IMM REG_BIT(25) +#define DCPR_CLEAR_MEMSTAT_DIS REG_BIT(24) + +#define XELPD_CHICKEN_DCPR_3 _MMIO(0x46438) +#define DMD_RSP_TIMEOUT_DISABLE REG_BIT(19) + +#define SKL_DFSM _MMIO(0x51000) +#define SKL_DFSM_DISPLAY_PM_DISABLE (1 << 27) +#define SKL_DFSM_DISPLAY_HDCP_DISABLE (1 << 25) +#define SKL_DFSM_CDCLK_LIMIT_MASK (3 << 23) +#define SKL_DFSM_CDCLK_LIMIT_675 (0 << 23) +#define SKL_DFSM_CDCLK_LIMIT_540 (1 << 23) +#define SKL_DFSM_CDCLK_LIMIT_450 (2 << 23) +#define SKL_DFSM_CDCLK_LIMIT_337_5 (3 << 23) +#define ICL_DFSM_DMC_DISABLE (1 << 23) +#define SKL_DFSM_PIPE_A_DISABLE (1 << 30) +#define SKL_DFSM_PIPE_B_DISABLE (1 << 21) +#define SKL_DFSM_PIPE_C_DISABLE (1 << 28) +#define TGL_DFSM_PIPE_D_DISABLE (1 << 22) +#define GLK_DFSM_DISPLAY_DSC_DISABLE (1 << 7) +#define XE2LPD_DFSM_DBUF_OVERLAP_DISABLE (1 << 3) + +#define XE2LPD_DE_CAP _MMIO(0x41100) +#define XE2LPD_DE_CAP_3DLUT_MASK REG_GENMASK(31, 30) +#define XE2LPD_DE_CAP_DSC_MASK REG_GENMASK(29, 28) +#define XE2LPD_DE_CAP_DSC_REMOVED 1 +#define XE2LPD_DE_CAP_SCALER_MASK REG_GENMASK(27, 26) +#define XE2LPD_DE_CAP_SCALER_SINGLE 1 + +#define SKL_DSSM _MMIO(0x51004) +#define ICL_DSSM_CDCLK_PLL_REFCLK_MASK (7 << 29) +#define ICL_DSSM_CDCLK_PLL_REFCLK_24MHz (0 << 29) +#define ICL_DSSM_CDCLK_PLL_REFCLK_19_2MHz (1 << 29) +#define ICL_DSSM_CDCLK_PLL_REFCLK_38_4MHz (2 << 29) + +/*GEN11 chicken */ +#define _PIPEA_CHICKEN 0x70038 +#define _PIPEB_CHICKEN 0x71038 +#define _PIPEC_CHICKEN 0x72038 +#define PIPE_CHICKEN(pipe) _MMIO_PIPE(pipe, _PIPEA_CHICKEN,\ + _PIPEB_CHICKEN) +#define UNDERRUN_RECOVERY_DISABLE_ADLP REG_BIT(30) +#define UNDERRUN_RECOVERY_ENABLE_DG2 REG_BIT(30) +#define PIXEL_ROUNDING_TRUNC_FB_PASSTHRU REG_BIT(15) +#define DG2_RENDER_CCSTAG_4_3_EN REG_BIT(12) +#define PER_PIXEL_ALPHA_BYPASS_EN REG_BIT(7) + +#define PCH_DISPLAY_BASE 0xc0000u + +/* south display engine interrupt: IBX */ +#define SDE_AUDIO_POWER_D (1 << 27) +#define SDE_AUDIO_POWER_C (1 << 26) +#define SDE_AUDIO_POWER_B (1 << 25) +#define SDE_AUDIO_POWER_SHIFT (25) +#define SDE_AUDIO_POWER_MASK (7 << SDE_AUDIO_POWER_SHIFT) +#define SDE_GMBUS (1 << 24) +#define SDE_AUDIO_HDCP_TRANSB (1 << 23) +#define SDE_AUDIO_HDCP_TRANSA (1 << 22) +#define SDE_AUDIO_HDCP_MASK (3 << 22) +#define SDE_AUDIO_TRANSB (1 << 21) +#define SDE_AUDIO_TRANSA (1 << 20) +#define SDE_AUDIO_TRANS_MASK (3 << 20) +#define SDE_POISON (1 << 19) +/* 18 reserved */ +#define SDE_FDI_RXB (1 << 17) +#define SDE_FDI_RXA (1 << 16) +#define SDE_FDI_MASK (3 << 16) +#define SDE_AUXD (1 << 15) +#define SDE_AUXC (1 << 14) +#define SDE_AUXB (1 << 13) +#define SDE_AUX_MASK (7 << 13) +/* 12 reserved */ +#define SDE_CRT_HOTPLUG (1 << 11) +#define SDE_PORTD_HOTPLUG (1 << 10) +#define SDE_PORTC_HOTPLUG (1 << 9) +#define SDE_PORTB_HOTPLUG (1 << 8) +#define SDE_SDVOB_HOTPLUG (1 << 6) +#define SDE_HOTPLUG_MASK (SDE_CRT_HOTPLUG | \ + SDE_SDVOB_HOTPLUG | \ + SDE_PORTB_HOTPLUG | \ + SDE_PORTC_HOTPLUG | \ + SDE_PORTD_HOTPLUG) +#define SDE_TRANSB_CRC_DONE (1 << 5) +#define SDE_TRANSB_CRC_ERR (1 << 4) +#define SDE_TRANSB_FIFO_UNDER (1 << 3) +#define SDE_TRANSA_CRC_DONE (1 << 2) +#define SDE_TRANSA_CRC_ERR (1 << 1) +#define SDE_TRANSA_FIFO_UNDER (1 << 0) +#define SDE_TRANS_MASK (0x3f) + +/* south display engine interrupt: CPT - CNP */ +#define SDE_AUDIO_POWER_D_CPT (1 << 31) +#define SDE_AUDIO_POWER_C_CPT (1 << 30) +#define SDE_AUDIO_POWER_B_CPT (1 << 29) +#define SDE_AUDIO_POWER_SHIFT_CPT 29 +#define SDE_AUDIO_POWER_MASK_CPT (7 << 29) +#define SDE_AUXD_CPT (1 << 27) +#define SDE_AUXC_CPT (1 << 26) +#define SDE_AUXB_CPT (1 << 25) +#define SDE_AUX_MASK_CPT (7 << 25) +#define SDE_PORTE_HOTPLUG_SPT (1 << 25) +#define SDE_PORTA_HOTPLUG_SPT (1 << 24) +#define SDE_PORTD_HOTPLUG_CPT (1 << 23) +#define SDE_PORTC_HOTPLUG_CPT (1 << 22) +#define SDE_PORTB_HOTPLUG_CPT (1 << 21) +#define SDE_CRT_HOTPLUG_CPT (1 << 19) +#define SDE_SDVOB_HOTPLUG_CPT (1 << 18) +#define SDE_HOTPLUG_MASK_CPT (SDE_CRT_HOTPLUG_CPT | \ + SDE_SDVOB_HOTPLUG_CPT | \ + SDE_PORTD_HOTPLUG_CPT | \ + SDE_PORTC_HOTPLUG_CPT | \ + SDE_PORTB_HOTPLUG_CPT) +#define SDE_HOTPLUG_MASK_SPT (SDE_PORTE_HOTPLUG_SPT | \ + SDE_PORTD_HOTPLUG_CPT | \ + SDE_PORTC_HOTPLUG_CPT | \ + SDE_PORTB_HOTPLUG_CPT | \ + SDE_PORTA_HOTPLUG_SPT) +#define SDE_GMBUS_CPT (1 << 17) +#define SDE_ERROR_CPT (1 << 16) +#define SDE_AUDIO_CP_REQ_C_CPT (1 << 10) +#define SDE_AUDIO_CP_CHG_C_CPT (1 << 9) +#define SDE_FDI_RXC_CPT (1 << 8) +#define SDE_AUDIO_CP_REQ_B_CPT (1 << 6) +#define SDE_AUDIO_CP_CHG_B_CPT (1 << 5) +#define SDE_FDI_RXB_CPT (1 << 4) +#define SDE_AUDIO_CP_REQ_A_CPT (1 << 2) +#define SDE_AUDIO_CP_CHG_A_CPT (1 << 1) +#define SDE_FDI_RXA_CPT (1 << 0) +#define SDE_AUDIO_CP_REQ_CPT (SDE_AUDIO_CP_REQ_C_CPT | \ + SDE_AUDIO_CP_REQ_B_CPT | \ + SDE_AUDIO_CP_REQ_A_CPT) +#define SDE_AUDIO_CP_CHG_CPT (SDE_AUDIO_CP_CHG_C_CPT | \ + SDE_AUDIO_CP_CHG_B_CPT | \ + SDE_AUDIO_CP_CHG_A_CPT) +#define SDE_FDI_MASK_CPT (SDE_FDI_RXC_CPT | \ + SDE_FDI_RXB_CPT | \ + SDE_FDI_RXA_CPT) + +/* south display engine interrupt: ICP/TGP/MTP */ +#define SDE_PICAINTERRUPT REG_BIT(31) +#define SDE_GMBUS_ICP (1 << 23) +#define SDE_TC_HOTPLUG_ICP(hpd_pin) REG_BIT(24 + _HPD_PIN_TC(hpd_pin)) +#define SDE_TC_HOTPLUG_DG2(hpd_pin) REG_BIT(25 + _HPD_PIN_TC(hpd_pin)) /* sigh */ +#define SDE_DDI_HOTPLUG_ICP(hpd_pin) REG_BIT(16 + _HPD_PIN_DDI(hpd_pin)) +#define SDE_DDI_HOTPLUG_MASK_ICP (SDE_DDI_HOTPLUG_ICP(HPD_PORT_D) | \ + SDE_DDI_HOTPLUG_ICP(HPD_PORT_C) | \ + SDE_DDI_HOTPLUG_ICP(HPD_PORT_B) | \ + SDE_DDI_HOTPLUG_ICP(HPD_PORT_A)) +#define SDE_TC_HOTPLUG_MASK_ICP (SDE_TC_HOTPLUG_ICP(HPD_PORT_TC6) | \ + SDE_TC_HOTPLUG_ICP(HPD_PORT_TC5) | \ + SDE_TC_HOTPLUG_ICP(HPD_PORT_TC4) | \ + SDE_TC_HOTPLUG_ICP(HPD_PORT_TC3) | \ + SDE_TC_HOTPLUG_ICP(HPD_PORT_TC2) | \ + SDE_TC_HOTPLUG_ICP(HPD_PORT_TC1)) + +#define SDE_IRQ_REGS I915_IRQ_REGS(SDEIMR, \ + SDEIER, \ + SDEIIR) + +#define SERR_INT _MMIO(0xc4040) +#define SERR_INT_POISON (1 << 31) +#define SERR_INT_TRANS_FIFO_UNDERRUN(pipe) (1 << ((pipe) * 3)) + +/* digital port hotplug */ +#define PCH_PORT_HOTPLUG _MMIO(0xc4030) /* SHOTPLUG_CTL */ +#define PORTA_HOTPLUG_ENABLE (1 << 28) /* LPT:LP+ & BXT */ +#define BXT_DDIA_HPD_INVERT (1 << 27) +#define PORTA_HOTPLUG_STATUS_MASK (3 << 24) /* SPT+ & BXT */ +#define PORTA_HOTPLUG_NO_DETECT (0 << 24) /* SPT+ & BXT */ +#define PORTA_HOTPLUG_SHORT_DETECT (1 << 24) /* SPT+ & BXT */ +#define PORTA_HOTPLUG_LONG_DETECT (2 << 24) /* SPT+ & BXT */ +#define PORTD_HOTPLUG_ENABLE (1 << 20) +#define PORTD_PULSE_DURATION_2ms (0 << 18) /* pre-LPT */ +#define PORTD_PULSE_DURATION_4_5ms (1 << 18) /* pre-LPT */ +#define PORTD_PULSE_DURATION_6ms (2 << 18) /* pre-LPT */ +#define PORTD_PULSE_DURATION_100ms (3 << 18) /* pre-LPT */ +#define PORTD_PULSE_DURATION_MASK (3 << 18) /* pre-LPT */ +#define PORTD_HOTPLUG_STATUS_MASK (3 << 16) +#define PORTD_HOTPLUG_NO_DETECT (0 << 16) +#define PORTD_HOTPLUG_SHORT_DETECT (1 << 16) +#define PORTD_HOTPLUG_LONG_DETECT (2 << 16) +#define PORTC_HOTPLUG_ENABLE (1 << 12) +#define BXT_DDIC_HPD_INVERT (1 << 11) +#define PORTC_PULSE_DURATION_2ms (0 << 10) /* pre-LPT */ +#define PORTC_PULSE_DURATION_4_5ms (1 << 10) /* pre-LPT */ +#define PORTC_PULSE_DURATION_6ms (2 << 10) /* pre-LPT */ +#define PORTC_PULSE_DURATION_100ms (3 << 10) /* pre-LPT */ +#define PORTC_PULSE_DURATION_MASK (3 << 10) /* pre-LPT */ +#define PORTC_HOTPLUG_STATUS_MASK (3 << 8) +#define PORTC_HOTPLUG_NO_DETECT (0 << 8) +#define PORTC_HOTPLUG_SHORT_DETECT (1 << 8) +#define PORTC_HOTPLUG_LONG_DETECT (2 << 8) +#define PORTB_HOTPLUG_ENABLE (1 << 4) +#define BXT_DDIB_HPD_INVERT (1 << 3) +#define PORTB_PULSE_DURATION_2ms (0 << 2) /* pre-LPT */ +#define PORTB_PULSE_DURATION_4_5ms (1 << 2) /* pre-LPT */ +#define PORTB_PULSE_DURATION_6ms (2 << 2) /* pre-LPT */ +#define PORTB_PULSE_DURATION_100ms (3 << 2) /* pre-LPT */ +#define PORTB_PULSE_DURATION_MASK (3 << 2) /* pre-LPT */ +#define PORTB_HOTPLUG_STATUS_MASK (3 << 0) +#define PORTB_HOTPLUG_NO_DETECT (0 << 0) +#define PORTB_HOTPLUG_SHORT_DETECT (1 << 0) +#define PORTB_HOTPLUG_LONG_DETECT (2 << 0) +#define BXT_DDI_HPD_INVERT_MASK (BXT_DDIA_HPD_INVERT | \ + BXT_DDIB_HPD_INVERT | \ + BXT_DDIC_HPD_INVERT) + +#define PCH_PORT_HOTPLUG2 _MMIO(0xc403C) /* SHOTPLUG_CTL2 SPT+ */ +#define PORTE_HOTPLUG_ENABLE (1 << 4) +#define PORTE_HOTPLUG_STATUS_MASK (3 << 0) +#define PORTE_HOTPLUG_NO_DETECT (0 << 0) +#define PORTE_HOTPLUG_SHORT_DETECT (1 << 0) +#define PORTE_HOTPLUG_LONG_DETECT (2 << 0) + +/* This register is a reuse of PCH_PORT_HOTPLUG register. The + * functionality covered in PCH_PORT_HOTPLUG is split into + * SHOTPLUG_CTL_DDI and SHOTPLUG_CTL_TC. + */ +#define SHOTPLUG_CTL_DDI _MMIO(0xc4030) +#define SHOTPLUG_CTL_DDI_HPD_ENABLE(hpd_pin) (0x8 << (_HPD_PIN_DDI(hpd_pin) * 4)) +#define SHOTPLUG_CTL_DDI_HPD_OUTPUT_DATA(hpd_pin) (0x4 << (_HPD_PIN_DDI(hpd_pin) * 4)) +#define SHOTPLUG_CTL_DDI_HPD_STATUS_MASK(hpd_pin) (0x3 << (_HPD_PIN_DDI(hpd_pin) * 4)) +#define SHOTPLUG_CTL_DDI_HPD_NO_DETECT(hpd_pin) (0x0 << (_HPD_PIN_DDI(hpd_pin) * 4)) +#define SHOTPLUG_CTL_DDI_HPD_SHORT_DETECT(hpd_pin) (0x1 << (_HPD_PIN_DDI(hpd_pin) * 4)) +#define SHOTPLUG_CTL_DDI_HPD_LONG_DETECT(hpd_pin) (0x2 << (_HPD_PIN_DDI(hpd_pin) * 4)) +#define SHOTPLUG_CTL_DDI_HPD_SHORT_LONG_DETECT(hpd_pin) (0x3 << (_HPD_PIN_DDI(hpd_pin) * 4)) + +#define SHOTPLUG_CTL_TC _MMIO(0xc4034) +#define ICP_TC_HPD_ENABLE(hpd_pin) (8 << (_HPD_PIN_TC(hpd_pin) * 4)) +#define ICP_TC_HPD_LONG_DETECT(hpd_pin) (2 << (_HPD_PIN_TC(hpd_pin) * 4)) +#define ICP_TC_HPD_SHORT_DETECT(hpd_pin) (1 << (_HPD_PIN_TC(hpd_pin) * 4)) + +#define SHPD_FILTER_CNT _MMIO(0xc4038) +#define SHPD_FILTER_CNT_500_ADJ 0x001D9 +#define SHPD_FILTER_CNT_250 0x000F8 + +#define _PCH_DPLL_A 0xc6014 +#define _PCH_DPLL_B 0xc6018 +#define PCH_DPLL(pll) _MMIO((pll) == 0 ? _PCH_DPLL_A : _PCH_DPLL_B) + +#define _PCH_FPA0 0xc6040 +#define _PCH_FPB0 0xc6048 +#define PCH_FP0(pll) _MMIO((pll) == 0 ? _PCH_FPA0 : _PCH_FPB0) +#define FP_CB_TUNE (0x3 << 22) + +#define _PCH_FPA1 0xc6044 +#define _PCH_FPB1 0xc604c +#define PCH_FP1(pll) _MMIO((pll) == 0 ? _PCH_FPA1 : _PCH_FPB1) + +#define PCH_DPLL_TEST _MMIO(0xc606c) + +#define PCH_DREF_CONTROL _MMIO(0xC6200) +#define DREF_CONTROL_MASK 0x7fc3 +#define DREF_CPU_SOURCE_OUTPUT_DISABLE (0 << 13) +#define DREF_CPU_SOURCE_OUTPUT_DOWNSPREAD (2 << 13) +#define DREF_CPU_SOURCE_OUTPUT_NONSPREAD (3 << 13) +#define DREF_CPU_SOURCE_OUTPUT_MASK (3 << 13) +#define DREF_SSC_SOURCE_DISABLE (0 << 11) +#define DREF_SSC_SOURCE_ENABLE (2 << 11) +#define DREF_SSC_SOURCE_MASK (3 << 11) +#define DREF_NONSPREAD_SOURCE_DISABLE (0 << 9) +#define DREF_NONSPREAD_CK505_ENABLE (1 << 9) +#define DREF_NONSPREAD_SOURCE_ENABLE (2 << 9) +#define DREF_NONSPREAD_SOURCE_MASK (3 << 9) +#define DREF_SUPERSPREAD_SOURCE_DISABLE (0 << 7) +#define DREF_SUPERSPREAD_SOURCE_ENABLE (2 << 7) +#define DREF_SUPERSPREAD_SOURCE_MASK (3 << 7) +#define DREF_SSC4_DOWNSPREAD (0 << 6) +#define DREF_SSC4_CENTERSPREAD (1 << 6) +#define DREF_SSC1_DISABLE (0 << 1) +#define DREF_SSC1_ENABLE (1 << 1) +#define DREF_SSC4_DISABLE (0) +#define DREF_SSC4_ENABLE (1) + +#define PCH_RAWCLK_FREQ _MMIO(0xc6204) +#define FDL_TP1_TIMER_SHIFT 12 +#define FDL_TP1_TIMER_MASK (3 << 12) +#define FDL_TP2_TIMER_SHIFT 10 +#define FDL_TP2_TIMER_MASK (3 << 10) +#define RAWCLK_FREQ_MASK 0x3ff +#define CNP_RAWCLK_DIV_MASK (0x3ff << 16) +#define CNP_RAWCLK_DIV(div) ((div) << 16) +#define CNP_RAWCLK_FRAC_MASK (0xf << 26) +#define CNP_RAWCLK_DEN(den) ((den) << 26) +#define ICP_RAWCLK_NUM(num) ((num) << 11) + +#define PCH_DPLL_TMR_CFG _MMIO(0xc6208) + +#define PCH_SSC4_PARMS _MMIO(0xc6210) +#define PCH_SSC4_AUX_PARMS _MMIO(0xc6214) + +#define PCH_DPLL_SEL _MMIO(0xc7000) +#define TRANS_DPLLB_SEL(pipe) (1 << ((pipe) * 4)) +#define TRANS_DPLLA_SEL(pipe) 0 +#define TRANS_DPLL_ENABLE(pipe) (1 << ((pipe) * 4 + 3)) + +/* transcoder */ +#define _PCH_TRANS_HTOTAL_A 0xe0000 +#define _PCH_TRANS_HTOTAL_B 0xe1000 +#define PCH_TRANS_HTOTAL(pipe) _MMIO_PIPE(pipe, _PCH_TRANS_HTOTAL_A, _PCH_TRANS_HTOTAL_B) +#define TRANS_HTOTAL_SHIFT 16 +#define TRANS_HACTIVE_SHIFT 0 + +#define _PCH_TRANS_HBLANK_A 0xe0004 +#define _PCH_TRANS_HBLANK_B 0xe1004 +#define PCH_TRANS_HBLANK(pipe) _MMIO_PIPE(pipe, _PCH_TRANS_HBLANK_A, _PCH_TRANS_HBLANK_B) +#define TRANS_HBLANK_END_SHIFT 16 +#define TRANS_HBLANK_START_SHIFT 0 + +#define _PCH_TRANS_HSYNC_A 0xe0008 +#define _PCH_TRANS_HSYNC_B 0xe1008 +#define PCH_TRANS_HSYNC(pipe) _MMIO_PIPE(pipe, _PCH_TRANS_HSYNC_A, _PCH_TRANS_HSYNC_B) +#define TRANS_HSYNC_END_SHIFT 16 +#define TRANS_HSYNC_START_SHIFT 0 + +#define _PCH_TRANS_VTOTAL_A 0xe000c +#define _PCH_TRANS_VTOTAL_B 0xe100c +#define PCH_TRANS_VTOTAL(pipe) _MMIO_PIPE(pipe, _PCH_TRANS_VTOTAL_A, _PCH_TRANS_VTOTAL_B) +#define TRANS_VTOTAL_SHIFT 16 +#define TRANS_VACTIVE_SHIFT 0 + +#define _PCH_TRANS_VBLANK_A 0xe0010 +#define _PCH_TRANS_VBLANK_B 0xe1010 +#define PCH_TRANS_VBLANK(pipe) _MMIO_PIPE(pipe, _PCH_TRANS_VBLANK_A, _PCH_TRANS_VBLANK_B) +#define TRANS_VBLANK_END_SHIFT 16 +#define TRANS_VBLANK_START_SHIFT 0 + +#define _PCH_TRANS_VSYNC_A 0xe0014 +#define _PCH_TRANS_VSYNC_B 0xe1014 +#define PCH_TRANS_VSYNC(pipe) _MMIO_PIPE(pipe, _PCH_TRANS_VSYNC_A, _PCH_TRANS_VSYNC_B) +#define TRANS_VSYNC_END_SHIFT 16 +#define TRANS_VSYNC_START_SHIFT 0 + +#define _PCH_TRANS_VSYNCSHIFT_A 0xe0028 +#define _PCH_TRANS_VSYNCSHIFT_B 0xe1028 +#define PCH_TRANS_VSYNCSHIFT(pipe) _MMIO_PIPE(pipe, _PCH_TRANS_VSYNCSHIFT_A, _PCH_TRANS_VSYNCSHIFT_B) + +#define _PCH_TRANSA_DATA_M1 0xe0030 +#define _PCH_TRANSB_DATA_M1 0xe1030 +#define PCH_TRANS_DATA_M1(pipe) _MMIO_PIPE(pipe, _PCH_TRANSA_DATA_M1, _PCH_TRANSB_DATA_M1) + +#define _PCH_TRANSA_DATA_N1 0xe0034 +#define _PCH_TRANSB_DATA_N1 0xe1034 +#define PCH_TRANS_DATA_N1(pipe) _MMIO_PIPE(pipe, _PCH_TRANSA_DATA_N1, _PCH_TRANSB_DATA_N1) + +#define _PCH_TRANSA_DATA_M2 0xe0038 +#define _PCH_TRANSB_DATA_M2 0xe1038 +#define PCH_TRANS_DATA_M2(pipe) _MMIO_PIPE(pipe, _PCH_TRANSA_DATA_M2, _PCH_TRANSB_DATA_M2) + +#define _PCH_TRANSA_DATA_N2 0xe003c +#define _PCH_TRANSB_DATA_N2 0xe103c +#define PCH_TRANS_DATA_N2(pipe) _MMIO_PIPE(pipe, _PCH_TRANSA_DATA_N2, _PCH_TRANSB_DATA_N2) + +#define _PCH_TRANSA_LINK_M1 0xe0040 +#define _PCH_TRANSB_LINK_M1 0xe1040 +#define PCH_TRANS_LINK_M1(pipe) _MMIO_PIPE(pipe, _PCH_TRANSA_LINK_M1, _PCH_TRANSB_LINK_M1) + +#define _PCH_TRANSA_LINK_N1 0xe0044 +#define _PCH_TRANSB_LINK_N1 0xe1044 +#define PCH_TRANS_LINK_N1(pipe) _MMIO_PIPE(pipe, _PCH_TRANSA_LINK_N1, _PCH_TRANSB_LINK_N1) + +#define _PCH_TRANSA_LINK_M2 0xe0048 +#define _PCH_TRANSB_LINK_M2 0xe1048 +#define PCH_TRANS_LINK_M2(pipe) _MMIO_PIPE(pipe, _PCH_TRANSA_LINK_M2, _PCH_TRANSB_LINK_M2) + +#define _PCH_TRANSA_LINK_N2 0xe004c +#define _PCH_TRANSB_LINK_N2 0xe104c +#define PCH_TRANS_LINK_N2(pipe) _MMIO_PIPE(pipe, _PCH_TRANSA_LINK_N2, _PCH_TRANSB_LINK_N2) + +/* Per-transcoder DIP controls (PCH) */ +#define _VIDEO_DIP_CTL_A 0xe0200 +#define _VIDEO_DIP_CTL_B 0xe1200 +#define TVIDEO_DIP_CTL(pipe) _MMIO_PIPE(pipe, _VIDEO_DIP_CTL_A, _VIDEO_DIP_CTL_B) + +#define _VIDEO_DIP_DATA_A 0xe0208 +#define _VIDEO_DIP_DATA_B 0xe1208 +#define TVIDEO_DIP_DATA(pipe) _MMIO_PIPE(pipe, _VIDEO_DIP_DATA_A, _VIDEO_DIP_DATA_B) + +#define _VIDEO_DIP_GCP_A 0xe0210 +#define _VIDEO_DIP_GCP_B 0xe1210 +#define TVIDEO_DIP_GCP(pipe) _MMIO_PIPE(pipe, _VIDEO_DIP_GCP_A, _VIDEO_DIP_GCP_B) +#define GCP_COLOR_INDICATION (1 << 2) +#define GCP_DEFAULT_PHASE_ENABLE (1 << 1) +#define GCP_AV_MUTE (1 << 0) + +/* Per-transcoder DIP controls (VLV) */ +#define _VLV_VIDEO_DIP_CTL_A 0x60200 +#define _VLV_VIDEO_DIP_CTL_B 0x61170 +#define _CHV_VIDEO_DIP_CTL_C 0x611f0 +#define VLV_TVIDEO_DIP_CTL(pipe) _MMIO_BASE_PIPE3(VLV_DISPLAY_BASE, (pipe), \ + _VLV_VIDEO_DIP_CTL_A, \ + _VLV_VIDEO_DIP_CTL_B, \ + _CHV_VIDEO_DIP_CTL_C) + +#define _VLV_VIDEO_DIP_DATA_A 0x60208 +#define _VLV_VIDEO_DIP_DATA_B 0x61174 +#define _CHV_VIDEO_DIP_DATA_C 0x611f4 +#define VLV_TVIDEO_DIP_DATA(pipe) _MMIO_BASE_PIPE3(VLV_DISPLAY_BASE, (pipe), \ + _VLV_VIDEO_DIP_DATA_A, \ + _VLV_VIDEO_DIP_DATA_B, \ + _CHV_VIDEO_DIP_DATA_C) + +#define _VLV_VIDEO_DIP_GDCP_PAYLOAD_A 0x60210 +#define _VLV_VIDEO_DIP_GDCP_PAYLOAD_B 0x61178 +#define _CHV_VIDEO_DIP_GDCP_PAYLOAD_C 0x611f8 +#define VLV_TVIDEO_DIP_GCP(pipe) _MMIO_BASE_PIPE3(VLV_DISPLAY_BASE, (pipe), \ + _VLV_VIDEO_DIP_GDCP_PAYLOAD_A, \ + _VLV_VIDEO_DIP_GDCP_PAYLOAD_B, \ + _CHV_VIDEO_DIP_GDCP_PAYLOAD_C) + +/* Haswell DIP controls */ +#define _HSW_VIDEO_DIP_CTL_A 0x60200 +#define _HSW_VIDEO_DIP_CTL_B 0x61200 +#define HSW_TVIDEO_DIP_CTL(dev_priv, trans) _MMIO_TRANS2(dev_priv, trans, _HSW_VIDEO_DIP_CTL_A) + +#define _HSW_VIDEO_DIP_AVI_DATA_A 0x60220 +#define _HSW_VIDEO_DIP_AVI_DATA_B 0x61220 +#define HSW_TVIDEO_DIP_AVI_DATA(dev_priv, trans, i) _MMIO_TRANS2(dev_priv, trans, _HSW_VIDEO_DIP_AVI_DATA_A + (i) * 4) + +#define _HSW_VIDEO_DIP_VS_DATA_A 0x60260 +#define _HSW_VIDEO_DIP_VS_DATA_B 0x61260 +#define HSW_TVIDEO_DIP_VS_DATA(dev_priv, trans, i) _MMIO_TRANS2(dev_priv, trans, _HSW_VIDEO_DIP_VS_DATA_A + (i) * 4) + +#define _HSW_VIDEO_DIP_SPD_DATA_A 0x602A0 +#define _HSW_VIDEO_DIP_SPD_DATA_B 0x612A0 +#define HSW_TVIDEO_DIP_SPD_DATA(dev_priv, trans, i) _MMIO_TRANS2(dev_priv, trans, _HSW_VIDEO_DIP_SPD_DATA_A + (i) * 4) + +#define _HSW_VIDEO_DIP_GMP_DATA_A 0x602E0 +#define _HSW_VIDEO_DIP_GMP_DATA_B 0x612E0 +#define HSW_TVIDEO_DIP_GMP_DATA(dev_priv, trans, i) _MMIO_TRANS2(dev_priv, trans, _HSW_VIDEO_DIP_GMP_DATA_A + (i) * 4) + +#define _HSW_VIDEO_DIP_VSC_DATA_A 0x60320 +#define _HSW_VIDEO_DIP_VSC_DATA_B 0x61320 +#define HSW_TVIDEO_DIP_VSC_DATA(dev_priv, trans, i) _MMIO_TRANS2(dev_priv, trans, _HSW_VIDEO_DIP_VSC_DATA_A + (i) * 4) + +/*ADLP and later: */ +#define _ADL_VIDEO_DIP_AS_DATA_A 0x60484 +#define _ADL_VIDEO_DIP_AS_DATA_B 0x61484 +#define ADL_TVIDEO_DIP_AS_SDP_DATA(dev_priv, trans, i) _MMIO_TRANS2(dev_priv, trans,\ + _ADL_VIDEO_DIP_AS_DATA_A + (i) * 4) + +#define _GLK_VIDEO_DIP_DRM_DATA_A 0x60440 +#define _GLK_VIDEO_DIP_DRM_DATA_B 0x61440 +#define GLK_TVIDEO_DIP_DRM_DATA(dev_priv, trans, i) _MMIO_TRANS2(dev_priv, trans, _GLK_VIDEO_DIP_DRM_DATA_A + (i) * 4) + +#define _HSW_VIDEO_DIP_AVI_ECC_A 0x60240 +#define _HSW_VIDEO_DIP_BVI_ECC_B 0x61240 +#define _HSW_VIDEO_DIP_VS_ECC_A 0x60280 +#define _HSW_VIDEO_DIP_VS_ECC_B 0x61280 +#define _HSW_VIDEO_DIP_SPD_ECC_A 0x602C0 +#define _HSW_VIDEO_DIP_SPD_ECC_B 0x612C0 +#define _HSW_VIDEO_DIP_GMP_ECC_A 0x60300 +#define _HSW_VIDEO_DIP_GMP_ECC_B 0x61300 +#define _HSW_VIDEO_DIP_VSC_ECC_A 0x60344 +#define _HSW_VIDEO_DIP_VSC_ECC_B 0x61344 + +#define _HSW_VIDEO_DIP_GCP_A 0x60210 +#define _HSW_VIDEO_DIP_GCP_B 0x61210 +#define HSW_TVIDEO_DIP_GCP(dev_priv, trans) _MMIO_TRANS2(dev_priv, trans, _HSW_VIDEO_DIP_GCP_A) + +#define _ICL_VIDEO_DIP_PPS_DATA_A 0x60350 +#define _ICL_VIDEO_DIP_PPS_DATA_B 0x61350 +#define ICL_VIDEO_DIP_PPS_DATA(dev_priv, trans, i) _MMIO_TRANS2(dev_priv, trans, _ICL_VIDEO_DIP_PPS_DATA_A + (i) * 4) + +#define _ICL_VIDEO_DIP_PPS_ECC_A 0x603D4 +#define _ICL_VIDEO_DIP_PPS_ECC_B 0x613D4 +#define ICL_VIDEO_DIP_PPS_ECC(dev_priv, trans, i) _MMIO_TRANS2(dev_priv, trans, _ICL_VIDEO_DIP_PPS_ECC_A + (i) * 4) + +#define _HSW_STEREO_3D_CTL_A 0x70020 +#define _HSW_STEREO_3D_CTL_B 0x71020 +#define HSW_STEREO_3D_CTL(dev_priv, trans) _MMIO_PIPE2(dev_priv, trans, _HSW_STEREO_3D_CTL_A) +#define S3D_ENABLE (1 << 31) + +#define _PCH_TRANSACONF 0xf0008 +#define _PCH_TRANSBCONF 0xf1008 +#define PCH_TRANSCONF(pipe) _MMIO_PIPE(pipe, _PCH_TRANSACONF, _PCH_TRANSBCONF) +#define LPT_TRANSCONF PCH_TRANSCONF(PIPE_A) /* lpt has only one transcoder */ +#define TRANS_ENABLE REG_BIT(31) +#define TRANS_STATE_ENABLE REG_BIT(30) +#define TRANS_FRAME_START_DELAY_MASK REG_GENMASK(28, 27) /* ibx */ +#define TRANS_FRAME_START_DELAY(x) REG_FIELD_PREP(TRANS_FRAME_START_DELAY_MASK, (x)) /* ibx: 0-3 */ +#define TRANS_INTERLACE_MASK REG_GENMASK(23, 21) +#define TRANS_INTERLACE_PROGRESSIVE REG_FIELD_PREP(TRANS_INTERLACE_MASK, 0) +#define TRANS_INTERLACE_LEGACY_VSYNC_IBX REG_FIELD_PREP(TRANS_INTERLACE_MASK, 2) /* ibx */ +#define TRANS_INTERLACE_INTERLACED REG_FIELD_PREP(TRANS_INTERLACE_MASK, 3) +#define TRANS_BPC_MASK REG_GENMASK(7, 5) /* ibx */ +#define TRANS_BPC_8 REG_FIELD_PREP(TRANS_BPC_MASK, 0) +#define TRANS_BPC_10 REG_FIELD_PREP(TRANS_BPC_MASK, 1) +#define TRANS_BPC_6 REG_FIELD_PREP(TRANS_BPC_MASK, 2) +#define TRANS_BPC_12 REG_FIELD_PREP(TRANS_BPC_MASK, 3) + +#define PCH_DP_B _MMIO(0xe4100) +#define PCH_DP_C _MMIO(0xe4200) +#define PCH_DP_D _MMIO(0xe4300) + +/* CPT */ +#define _TRANS_DP_CTL_A 0xe0300 +#define _TRANS_DP_CTL_B 0xe1300 +#define _TRANS_DP_CTL_C 0xe2300 +#define TRANS_DP_CTL(pipe) _MMIO_PIPE(pipe, _TRANS_DP_CTL_A, _TRANS_DP_CTL_B) +#define TRANS_DP_OUTPUT_ENABLE REG_BIT(31) +#define TRANS_DP_PORT_SEL_MASK REG_GENMASK(30, 29) +#define TRANS_DP_PORT_SEL_NONE REG_FIELD_PREP(TRANS_DP_PORT_SEL_MASK, 3) +#define TRANS_DP_PORT_SEL(port) REG_FIELD_PREP(TRANS_DP_PORT_SEL_MASK, (port) - PORT_B) +#define TRANS_DP_AUDIO_ONLY REG_BIT(26) +#define TRANS_DP_ENH_FRAMING REG_BIT(18) +#define TRANS_DP_BPC_MASK REG_GENMASK(10, 9) +#define TRANS_DP_BPC_8 REG_FIELD_PREP(TRANS_DP_BPC_MASK, 0) +#define TRANS_DP_BPC_10 REG_FIELD_PREP(TRANS_DP_BPC_MASK, 1) +#define TRANS_DP_BPC_6 REG_FIELD_PREP(TRANS_DP_BPC_MASK, 2) +#define TRANS_DP_BPC_12 REG_FIELD_PREP(TRANS_DP_BPC_MASK, 3) +#define TRANS_DP_VSYNC_ACTIVE_HIGH REG_BIT(4) +#define TRANS_DP_HSYNC_ACTIVE_HIGH REG_BIT(3) + +#define _TRANS_DP2_CTL_A 0x600a0 +#define _TRANS_DP2_CTL_B 0x610a0 +#define _TRANS_DP2_CTL_C 0x620a0 +#define _TRANS_DP2_CTL_D 0x630a0 +#define TRANS_DP2_CTL(trans) _MMIO_TRANS(trans, _TRANS_DP2_CTL_A, _TRANS_DP2_CTL_B) +#define TRANS_DP2_128B132B_CHANNEL_CODING REG_BIT(31) +#define TRANS_DP2_PANEL_REPLAY_ENABLE REG_BIT(30) +#define TRANS_DP2_DEBUG_ENABLE REG_BIT(23) + +#define _TRANS_DP2_VFREQHIGH_A 0x600a4 +#define _TRANS_DP2_VFREQHIGH_B 0x610a4 +#define _TRANS_DP2_VFREQHIGH_C 0x620a4 +#define _TRANS_DP2_VFREQHIGH_D 0x630a4 +#define TRANS_DP2_VFREQHIGH(trans) _MMIO_TRANS(trans, _TRANS_DP2_VFREQHIGH_A, _TRANS_DP2_VFREQHIGH_B) +#define TRANS_DP2_VFREQ_PIXEL_CLOCK_MASK REG_GENMASK(31, 8) +#define TRANS_DP2_VFREQ_PIXEL_CLOCK(clk_hz) REG_FIELD_PREP(TRANS_DP2_VFREQ_PIXEL_CLOCK_MASK, (clk_hz)) + +#define _TRANS_DP2_VFREQLOW_A 0x600a8 +#define _TRANS_DP2_VFREQLOW_B 0x610a8 +#define _TRANS_DP2_VFREQLOW_C 0x620a8 +#define _TRANS_DP2_VFREQLOW_D 0x630a8 +#define TRANS_DP2_VFREQLOW(trans) _MMIO_TRANS(trans, _TRANS_DP2_VFREQLOW_A, _TRANS_DP2_VFREQLOW_B) + +#define _DP_MIN_HBLANK_CTL_A 0x600ac +#define _DP_MIN_HBLANK_CTL_B 0x610ac +#define DP_MIN_HBLANK_CTL(trans) _MMIO_TRANS(trans, _DP_MIN_HBLANK_CTL_A, _DP_MIN_HBLANK_CTL_B) + +/* SNB eDP training params */ +/* SNB A-stepping */ +#define EDP_LINK_TRAIN_400MV_0DB_SNB_A (0x38 << 22) +#define EDP_LINK_TRAIN_400MV_6DB_SNB_A (0x02 << 22) +#define EDP_LINK_TRAIN_600MV_3_5DB_SNB_A (0x01 << 22) +#define EDP_LINK_TRAIN_800MV_0DB_SNB_A (0x0 << 22) +/* SNB B-stepping */ +#define EDP_LINK_TRAIN_400_600MV_0DB_SNB_B (0x0 << 22) +#define EDP_LINK_TRAIN_400MV_3_5DB_SNB_B (0x1 << 22) +#define EDP_LINK_TRAIN_400_600MV_6DB_SNB_B (0x3a << 22) +#define EDP_LINK_TRAIN_600_800MV_3_5DB_SNB_B (0x39 << 22) +#define EDP_LINK_TRAIN_800_1200MV_0DB_SNB_B (0x38 << 22) +#define EDP_LINK_TRAIN_VOL_EMP_MASK_SNB (0x3f << 22) + +/* IVB */ +#define EDP_LINK_TRAIN_400MV_0DB_IVB (0x24 << 22) +#define EDP_LINK_TRAIN_400MV_3_5DB_IVB (0x2a << 22) +#define EDP_LINK_TRAIN_400MV_6DB_IVB (0x2f << 22) +#define EDP_LINK_TRAIN_600MV_0DB_IVB (0x30 << 22) +#define EDP_LINK_TRAIN_600MV_3_5DB_IVB (0x36 << 22) +#define EDP_LINK_TRAIN_800MV_0DB_IVB (0x38 << 22) +#define EDP_LINK_TRAIN_800MV_3_5DB_IVB (0x3e << 22) + +/* legacy values */ +#define EDP_LINK_TRAIN_500MV_0DB_IVB (0x00 << 22) +#define EDP_LINK_TRAIN_1000MV_0DB_IVB (0x20 << 22) +#define EDP_LINK_TRAIN_500MV_3_5DB_IVB (0x02 << 22) +#define EDP_LINK_TRAIN_1000MV_3_5DB_IVB (0x22 << 22) +#define EDP_LINK_TRAIN_1000MV_6DB_IVB (0x23 << 22) + +#define EDP_LINK_TRAIN_VOL_EMP_MASK_IVB (0x3f << 22) + +#define VLV_CHICKEN_3 _MMIO(VLV_DISPLAY_BASE + 0x7040C) +#define PIXEL_OVERLAP_CNT_MASK (3 << 30) +#define PIXEL_OVERLAP_CNT_SHIFT 30 + +/* + * HSW - ICL power wells + * + * Platforms have up to 3 power well control register sets, each set + * controlling up to 16 power wells via a request/status HW flag tuple: + * - main (HSW_PWR_WELL_CTL[1-4]) + * - AUX (ICL_PWR_WELL_CTL_AUX[1-4]) + * - DDI (ICL_PWR_WELL_CTL_DDI[1-4]) + * Each control register set consists of up to 4 registers used by different + * sources that can request a power well to be enabled: + * - BIOS (HSW_PWR_WELL_CTL1/ICL_PWR_WELL_CTL_AUX1/ICL_PWR_WELL_CTL_DDI1) + * - DRIVER (HSW_PWR_WELL_CTL2/ICL_PWR_WELL_CTL_AUX2/ICL_PWR_WELL_CTL_DDI2) + * - KVMR (HSW_PWR_WELL_CTL3) (only in the main register set) + * - DEBUG (HSW_PWR_WELL_CTL4/ICL_PWR_WELL_CTL_AUX4/ICL_PWR_WELL_CTL_DDI4) + */ +#define HSW_PWR_WELL_CTL1 _MMIO(0x45400) +#define HSW_PWR_WELL_CTL2 _MMIO(0x45404) +#define HSW_PWR_WELL_CTL3 _MMIO(0x45408) +#define HSW_PWR_WELL_CTL4 _MMIO(0x4540C) +#define HSW_PWR_WELL_CTL_REQ(pw_idx) (0x2 << ((pw_idx) * 2)) +#define HSW_PWR_WELL_CTL_STATE(pw_idx) (0x1 << ((pw_idx) * 2)) + +/* HSW/BDW power well */ +#define HSW_PW_CTL_IDX_GLOBAL 15 + +/* SKL/BXT/GLK power wells */ +#define SKL_PW_CTL_IDX_PW_2 15 +#define SKL_PW_CTL_IDX_PW_1 14 +#define GLK_PW_CTL_IDX_AUX_C 10 +#define GLK_PW_CTL_IDX_AUX_B 9 +#define GLK_PW_CTL_IDX_AUX_A 8 +#define SKL_PW_CTL_IDX_DDI_D 4 +#define SKL_PW_CTL_IDX_DDI_C 3 +#define SKL_PW_CTL_IDX_DDI_B 2 +#define SKL_PW_CTL_IDX_DDI_A_E 1 +#define GLK_PW_CTL_IDX_DDI_A 1 +#define SKL_PW_CTL_IDX_MISC_IO 0 + +/* ICL/TGL - power wells */ +#define TGL_PW_CTL_IDX_PW_5 4 +#define ICL_PW_CTL_IDX_PW_4 3 +#define ICL_PW_CTL_IDX_PW_3 2 +#define ICL_PW_CTL_IDX_PW_2 1 +#define ICL_PW_CTL_IDX_PW_1 0 + +/* XE_LPD - power wells */ +#define XELPD_PW_CTL_IDX_PW_D 8 +#define XELPD_PW_CTL_IDX_PW_C 7 +#define XELPD_PW_CTL_IDX_PW_B 6 +#define XELPD_PW_CTL_IDX_PW_A 5 + +#define ICL_PWR_WELL_CTL_AUX1 _MMIO(0x45440) +#define ICL_PWR_WELL_CTL_AUX2 _MMIO(0x45444) +#define ICL_PWR_WELL_CTL_AUX4 _MMIO(0x4544C) +#define TGL_PW_CTL_IDX_AUX_TBT6 14 +#define TGL_PW_CTL_IDX_AUX_TBT5 13 +#define TGL_PW_CTL_IDX_AUX_TBT4 12 +#define ICL_PW_CTL_IDX_AUX_TBT4 11 +#define TGL_PW_CTL_IDX_AUX_TBT3 11 +#define ICL_PW_CTL_IDX_AUX_TBT3 10 +#define TGL_PW_CTL_IDX_AUX_TBT2 10 +#define ICL_PW_CTL_IDX_AUX_TBT2 9 +#define TGL_PW_CTL_IDX_AUX_TBT1 9 +#define ICL_PW_CTL_IDX_AUX_TBT1 8 +#define TGL_PW_CTL_IDX_AUX_TC6 8 +#define XELPD_PW_CTL_IDX_AUX_E 8 +#define TGL_PW_CTL_IDX_AUX_TC5 7 +#define XELPD_PW_CTL_IDX_AUX_D 7 +#define TGL_PW_CTL_IDX_AUX_TC4 6 +#define ICL_PW_CTL_IDX_AUX_F 5 +#define TGL_PW_CTL_IDX_AUX_TC3 5 +#define ICL_PW_CTL_IDX_AUX_E 4 +#define TGL_PW_CTL_IDX_AUX_TC2 4 +#define ICL_PW_CTL_IDX_AUX_D 3 +#define TGL_PW_CTL_IDX_AUX_TC1 3 +#define ICL_PW_CTL_IDX_AUX_C 2 +#define ICL_PW_CTL_IDX_AUX_B 1 +#define ICL_PW_CTL_IDX_AUX_A 0 + +#define ICL_PWR_WELL_CTL_DDI1 _MMIO(0x45450) +#define ICL_PWR_WELL_CTL_DDI2 _MMIO(0x45454) +#define ICL_PWR_WELL_CTL_DDI4 _MMIO(0x4545C) +#define XELPD_PW_CTL_IDX_DDI_E 8 +#define TGL_PW_CTL_IDX_DDI_TC6 8 +#define XELPD_PW_CTL_IDX_DDI_D 7 +#define TGL_PW_CTL_IDX_DDI_TC5 7 +#define TGL_PW_CTL_IDX_DDI_TC4 6 +#define ICL_PW_CTL_IDX_DDI_F 5 +#define TGL_PW_CTL_IDX_DDI_TC3 5 +#define ICL_PW_CTL_IDX_DDI_E 4 +#define TGL_PW_CTL_IDX_DDI_TC2 4 +#define ICL_PW_CTL_IDX_DDI_D 3 +#define TGL_PW_CTL_IDX_DDI_TC1 3 +#define ICL_PW_CTL_IDX_DDI_C 2 +#define ICL_PW_CTL_IDX_DDI_B 1 +#define ICL_PW_CTL_IDX_DDI_A 0 + +/* HSW - power well misc debug registers */ +#define HSW_PWR_WELL_CTL5 _MMIO(0x45410) +#define HSW_PWR_WELL_ENABLE_SINGLE_STEP (1 << 31) +#define HSW_PWR_WELL_PWR_GATE_OVERRIDE (1 << 20) +#define HSW_PWR_WELL_FORCE_ON (1 << 19) +#define HSW_PWR_WELL_CTL6 _MMIO(0x45414) + +/* SKL Fuse Status */ +enum skl_power_gate { + SKL_PG0, + SKL_PG1, + SKL_PG2, + ICL_PG3, + ICL_PG4, +}; + +#define SKL_FUSE_STATUS _MMIO(0x42000) +#define SKL_FUSE_DOWNLOAD_STATUS (1 << 31) +#define SKL_FUSE_PG_DIST_STATUS(pg) (1 << (27 - (pg))) + +/* Per-pipe DDI Function Control */ +#define _TRANS_DDI_FUNC_CTL_A 0x60400 +#define _TRANS_DDI_FUNC_CTL_B 0x61400 +#define _TRANS_DDI_FUNC_CTL_C 0x62400 +#define _TRANS_DDI_FUNC_CTL_D 0x63400 +#define _TRANS_DDI_FUNC_CTL_EDP 0x6F400 +#define _TRANS_DDI_FUNC_CTL_DSI0 0x6b400 +#define _TRANS_DDI_FUNC_CTL_DSI1 0x6bc00 +#define TRANS_DDI_FUNC_CTL(dev_priv, tran) _MMIO_TRANS2(dev_priv, tran, _TRANS_DDI_FUNC_CTL_A) + +#define TRANS_DDI_FUNC_ENABLE (1 << 31) +/* Those bits are ignored by pipe EDP since it can only connect to DDI A */ +#define TRANS_DDI_PORT_SHIFT 28 +#define TGL_TRANS_DDI_PORT_SHIFT 27 +#define TRANS_DDI_PORT_MASK (7 << TRANS_DDI_PORT_SHIFT) +#define TGL_TRANS_DDI_PORT_MASK (0xf << TGL_TRANS_DDI_PORT_SHIFT) +#define TRANS_DDI_SELECT_PORT(x) ((x) << TRANS_DDI_PORT_SHIFT) +#define TGL_TRANS_DDI_SELECT_PORT(x) (((x) + 1) << TGL_TRANS_DDI_PORT_SHIFT) +#define TRANS_DDI_MODE_SELECT_MASK (7 << 24) +#define TRANS_DDI_MODE_SELECT_HDMI (0 << 24) +#define TRANS_DDI_MODE_SELECT_DVI (1 << 24) +#define TRANS_DDI_MODE_SELECT_DP_SST (2 << 24) +#define TRANS_DDI_MODE_SELECT_DP_MST (3 << 24) +#define TRANS_DDI_MODE_SELECT_FDI_OR_128B132B (4 << 24) +#define TRANS_DDI_BPC_MASK (7 << 20) +#define TRANS_DDI_BPC_8 (0 << 20) +#define TRANS_DDI_BPC_10 (1 << 20) +#define TRANS_DDI_BPC_6 (2 << 20) +#define TRANS_DDI_BPC_12 (3 << 20) +#define TRANS_DDI_PORT_SYNC_MASTER_SELECT_MASK REG_GENMASK(19, 18) +#define TRANS_DDI_PORT_SYNC_MASTER_SELECT(x) REG_FIELD_PREP(TRANS_DDI_PORT_SYNC_MASTER_SELECT_MASK, (x)) +#define TRANS_DDI_PVSYNC (1 << 17) +#define TRANS_DDI_PHSYNC (1 << 16) +#define TRANS_DDI_PORT_SYNC_ENABLE REG_BIT(15) +#define XE3_TRANS_DDI_HDCP_LINE_REKEY_DISABLE REG_BIT(15) +#define TRANS_DDI_EDP_INPUT_MASK (7 << 12) +#define TRANS_DDI_EDP_INPUT_A_ON (0 << 12) +#define TRANS_DDI_EDP_INPUT_A_ONOFF (4 << 12) +#define TRANS_DDI_EDP_INPUT_B_ONOFF (5 << 12) +#define TRANS_DDI_EDP_INPUT_C_ONOFF (6 << 12) +#define TRANS_DDI_EDP_INPUT_D_ONOFF (7 << 12) +#define TRANS_DDI_HDCP_LINE_REKEY_DISABLE REG_BIT(12) +#define TRANS_DDI_MST_TRANSPORT_SELECT_MASK REG_GENMASK(11, 10) +#define TRANS_DDI_MST_TRANSPORT_SELECT(trans) \ + REG_FIELD_PREP(TRANS_DDI_MST_TRANSPORT_SELECT_MASK, trans) +#define TRANS_DDI_HDCP_SIGNALLING (1 << 9) +#define TRANS_DDI_DP_VC_PAYLOAD_ALLOC (1 << 8) +#define TRANS_DDI_HDMI_SCRAMBLER_CTS_ENABLE (1 << 7) +#define TRANS_DDI_HDMI_SCRAMBLER_RESET_FREQ (1 << 6) +#define TRANS_DDI_HDCP_SELECT REG_BIT(5) +#define TRANS_DDI_BFI_ENABLE (1 << 4) +#define TRANS_DDI_HIGH_TMDS_CHAR_RATE (1 << 4) +#define TRANS_DDI_PORT_WIDTH_MASK REG_GENMASK(3, 1) +#define TRANS_DDI_PORT_WIDTH(width) REG_FIELD_PREP(TRANS_DDI_PORT_WIDTH_MASK, (width) - 1) +#define TRANS_DDI_HDMI_SCRAMBLING (1 << 0) +#define TRANS_DDI_HDMI_SCRAMBLING_MASK (TRANS_DDI_HDMI_SCRAMBLER_CTS_ENABLE \ + | TRANS_DDI_HDMI_SCRAMBLER_RESET_FREQ \ + | TRANS_DDI_HDMI_SCRAMBLING) + +#define _TRANS_DDI_FUNC_CTL2_A 0x60404 +#define _TRANS_DDI_FUNC_CTL2_B 0x61404 +#define _TRANS_DDI_FUNC_CTL2_C 0x62404 +#define _TRANS_DDI_FUNC_CTL2_EDP 0x6f404 +#define _TRANS_DDI_FUNC_CTL2_DSI0 0x6b404 +#define _TRANS_DDI_FUNC_CTL2_DSI1 0x6bc04 +#define TRANS_DDI_FUNC_CTL2(dev_priv, tran) _MMIO_TRANS2(dev_priv, tran, _TRANS_DDI_FUNC_CTL2_A) +#define PORT_SYNC_MODE_ENABLE REG_BIT(4) +#define CMTG_SECONDARY_MODE REG_BIT(3) +#define PORT_SYNC_MODE_MASTER_SELECT_MASK REG_GENMASK(2, 0) +#define PORT_SYNC_MODE_MASTER_SELECT(x) REG_FIELD_PREP(PORT_SYNC_MODE_MASTER_SELECT_MASK, (x)) + +#define TRANS_CMTG_CHICKEN _MMIO(0x6fa90) +#define DISABLE_DPT_CLK_GATING REG_BIT(1) + +/* DisplayPort Transport Control */ +#define _DP_TP_CTL_A 0x64040 +#define _DP_TP_CTL_B 0x64140 +#define _TGL_DP_TP_CTL_A 0x60540 +#define DP_TP_CTL(port) _MMIO_PORT(port, _DP_TP_CTL_A, _DP_TP_CTL_B) +#define TGL_DP_TP_CTL(dev_priv, tran) _MMIO_TRANS2(dev_priv, (tran), _TGL_DP_TP_CTL_A) +#define DP_TP_CTL_ENABLE REG_BIT(31) +#define DP_TP_CTL_FEC_ENABLE REG_BIT(30) +#define DP_TP_CTL_MODE_MASK REG_BIT(27) +#define DP_TP_CTL_MODE_SST REG_FIELD_PREP(DP_TP_CTL_MODE_MASK, 0) +#define DP_TP_CTL_MODE_MST REG_FIELD_PREP(DP_TP_CTL_MODE_MASK, 1) +#define DP_TP_CTL_FORCE_ACT REG_BIT(25) +#define DP_TP_CTL_TRAIN_PAT4_SEL_MASK REG_GENMASK(20, 19) +#define DP_TP_CTL_TRAIN_PAT4_SEL_TP4A REG_FIELD_PREP(DP_TP_CTL_TRAIN_PAT4_SEL_MASK, 0) +#define DP_TP_CTL_TRAIN_PAT4_SEL_TP4B REG_FIELD_PREP(DP_TP_CTL_TRAIN_PAT4_SEL_MASK, 1) +#define DP_TP_CTL_TRAIN_PAT4_SEL_TP4C REG_FIELD_PREP(DP_TP_CTL_TRAIN_PAT4_SEL_MASK, 2) +#define DP_TP_CTL_ENHANCED_FRAME_ENABLE REG_BIT(18) +#define DP_TP_CTL_FDI_AUTOTRAIN REG_BIT(15) +#define DP_TP_CTL_LINK_TRAIN_MASK REG_GENMASK(10, 8) +#define DP_TP_CTL_LINK_TRAIN_PAT1 REG_FIELD_PREP(DP_TP_CTL_LINK_TRAIN_MASK, 0) +#define DP_TP_CTL_LINK_TRAIN_PAT2 REG_FIELD_PREP(DP_TP_CTL_LINK_TRAIN_MASK, 1) +#define DP_TP_CTL_LINK_TRAIN_PAT3 REG_FIELD_PREP(DP_TP_CTL_LINK_TRAIN_MASK, 4) +#define DP_TP_CTL_LINK_TRAIN_PAT4 REG_FIELD_PREP(DP_TP_CTL_LINK_TRAIN_MASK, 5) +#define DP_TP_CTL_LINK_TRAIN_IDLE REG_FIELD_PREP(DP_TP_CTL_LINK_TRAIN_MASK, 2) +#define DP_TP_CTL_LINK_TRAIN_NORMAL REG_FIELD_PREP(DP_TP_CTL_LINK_TRAIN_MASK, 3) +#define DP_TP_CTL_SCRAMBLE_DISABLE REG_BIT(7) + +/* DisplayPort Transport Status */ +#define _DP_TP_STATUS_A 0x64044 +#define _DP_TP_STATUS_B 0x64144 +#define _TGL_DP_TP_STATUS_A 0x60544 +#define DP_TP_STATUS(port) _MMIO_PORT(port, _DP_TP_STATUS_A, _DP_TP_STATUS_B) +#define TGL_DP_TP_STATUS(dev_priv, tran) _MMIO_TRANS2(dev_priv, (tran), _TGL_DP_TP_STATUS_A) +#define DP_TP_STATUS_FEC_ENABLE_LIVE REG_BIT(28) +#define DP_TP_STATUS_IDLE_DONE REG_BIT(25) +#define DP_TP_STATUS_ACT_SENT REG_BIT(24) +#define DP_TP_STATUS_MODE_STATUS_MST REG_BIT(23) +#define DP_TP_STATUS_STREAMS_ENABLED_MASK REG_GENMASK(18, 16) /* 17:16 on hsw but bit 18 mbz */ +#define DP_TP_STATUS_AUTOTRAIN_DONE REG_BIT(12) +#define DP_TP_STATUS_PAYLOAD_MAPPING_VC2_MASK REG_GENMASK(9, 8) +#define DP_TP_STATUS_PAYLOAD_MAPPING_VC1_MASK REG_GENMASK(5, 4) +#define DP_TP_STATUS_PAYLOAD_MAPPING_VC0_MASK REG_GENMASK(1, 0) + +/* DDI Buffer Control */ +#define _DDI_BUF_CTL_A 0x64000 +#define _DDI_BUF_CTL_B 0x64100 +/* Known as DDI_CTL_DE in MTL+ */ +#define DDI_BUF_CTL(port) _MMIO_PORT(port, _DDI_BUF_CTL_A, _DDI_BUF_CTL_B) +#define DDI_BUF_CTL_ENABLE REG_BIT(31) +#define XE2LPD_DDI_BUF_D2D_LINK_ENABLE REG_BIT(29) +#define XE2LPD_DDI_BUF_D2D_LINK_STATE REG_BIT(28) +#define DDI_BUF_EMP_MASK REG_GENMASK(27, 24) +#define DDI_BUF_TRANS_SELECT(n) REG_FIELD_PREP(DDI_BUF_EMP_MASK, (n)) +#define DDI_BUF_PHY_LINK_RATE_MASK REG_GENMASK(23, 20) +#define DDI_BUF_PHY_LINK_RATE(r) REG_FIELD_PREP(DDI_BUF_PHY_LINK_RATE_MASK, (r)) +#define DDI_BUF_PORT_DATA_MASK REG_GENMASK(19, 18) +#define DDI_BUF_PORT_DATA_10BIT REG_FIELD_PREP(DDI_BUF_PORT_DATA_MASK, 0) +#define DDI_BUF_PORT_DATA_20BIT REG_FIELD_PREP(DDI_BUF_PORT_DATA_MASK, 1) +#define DDI_BUF_PORT_DATA_40BIT REG_FIELD_PREP(DDI_BUF_PORT_DATA_MASK, 2) +#define DDI_BUF_PORT_REVERSAL REG_BIT(16) +#define DDI_BUF_LANE_STAGGER_DELAY_MASK REG_GENMASK(15, 8) +#define DDI_BUF_LANE_STAGGER_DELAY(symbols) REG_FIELD_PREP(DDI_BUF_LANE_STAGGER_DELAY_MASK, \ + (symbols)) +#define DDI_BUF_IS_IDLE REG_BIT(7) +#define DDI_BUF_CTL_TC_PHY_OWNERSHIP REG_BIT(6) +#define DDI_A_4_LANES REG_BIT(4) +#define DDI_PORT_WIDTH_MASK REG_GENMASK(3, 1) +#define DDI_PORT_WIDTH(width) REG_FIELD_PREP(DDI_PORT_WIDTH_MASK, \ + ((width) == 3 ? 4 : (width) - 1)) +#define DDI_PORT_WIDTH_SHIFT 1 +#define DDI_INIT_DISPLAY_DETECTED REG_BIT(0) + +/* DDI Buffer Translations */ +#define _DDI_BUF_TRANS_A 0x64E00 +#define _DDI_BUF_TRANS_B 0x64E60 +#define DDI_BUF_TRANS_LO(port, i) _MMIO(_PORT(port, _DDI_BUF_TRANS_A, _DDI_BUF_TRANS_B) + (i) * 8) +#define DDI_BUF_BALANCE_LEG_ENABLE (1 << 31) +#define DDI_BUF_TRANS_HI(port, i) _MMIO(_PORT(port, _DDI_BUF_TRANS_A, _DDI_BUF_TRANS_B) + (i) * 8 + 4) + +/* DDI DP Compliance Control */ +#define _DDI_DP_COMP_CTL_A 0x605F0 +#define _DDI_DP_COMP_CTL_B 0x615F0 +#define DDI_DP_COMP_CTL(pipe) _MMIO_PIPE(pipe, _DDI_DP_COMP_CTL_A, _DDI_DP_COMP_CTL_B) +#define DDI_DP_COMP_CTL_ENABLE (1 << 31) +#define DDI_DP_COMP_CTL_D10_2 (0 << 28) +#define DDI_DP_COMP_CTL_SCRAMBLED_0 (1 << 28) +#define DDI_DP_COMP_CTL_PRBS7 (2 << 28) +#define DDI_DP_COMP_CTL_CUSTOM80 (3 << 28) +#define DDI_DP_COMP_CTL_HBR2 (4 << 28) +#define DDI_DP_COMP_CTL_SCRAMBLED_1 (5 << 28) +#define DDI_DP_COMP_CTL_HBR2_RESET (0xFC << 0) + +/* DDI DP Compliance Pattern */ +#define _DDI_DP_COMP_PAT_A 0x605F4 +#define _DDI_DP_COMP_PAT_B 0x615F4 +#define DDI_DP_COMP_PAT(pipe, i) _MMIO(_PIPE(pipe, _DDI_DP_COMP_PAT_A, _DDI_DP_COMP_PAT_B) + (i) * 4) + +/* LPT PIXCLK_GATE */ +#define PIXCLK_GATE _MMIO(0xC6020) +#define PIXCLK_GATE_UNGATE (1 << 0) +#define PIXCLK_GATE_GATE (0 << 0) + +/* SPLL */ +#define SPLL_CTL _MMIO(0x46020) +#define SPLL_PLL_ENABLE (1 << 31) +#define SPLL_REF_BCLK (0 << 28) +#define SPLL_REF_MUXED_SSC (1 << 28) /* CPU SSC if fused enabled, PCH SSC otherwise */ +#define SPLL_REF_NON_SSC_HSW (2 << 28) +#define SPLL_REF_PCH_SSC_BDW (2 << 28) +#define SPLL_REF_LCPLL (3 << 28) +#define SPLL_REF_MASK (3 << 28) +#define SPLL_FREQ_810MHz (0 << 26) +#define SPLL_FREQ_1350MHz (1 << 26) +#define SPLL_FREQ_2700MHz (2 << 26) +#define SPLL_FREQ_MASK (3 << 26) + +/* WRPLL */ +#define _WRPLL_CTL1 0x46040 +#define _WRPLL_CTL2 0x46060 +#define WRPLL_CTL(pll) _MMIO_PIPE(pll, _WRPLL_CTL1, _WRPLL_CTL2) +#define WRPLL_PLL_ENABLE (1 << 31) +#define WRPLL_REF_BCLK (0 << 28) +#define WRPLL_REF_PCH_SSC (1 << 28) +#define WRPLL_REF_MUXED_SSC_BDW (2 << 28) /* CPU SSC if fused enabled, PCH SSC otherwise */ +#define WRPLL_REF_SPECIAL_HSW (2 << 28) /* muxed SSC (ULT), non-SSC (non-ULT) */ +#define WRPLL_REF_LCPLL (3 << 28) +#define WRPLL_REF_MASK (3 << 28) +/* WRPLL divider programming */ +#define WRPLL_DIVIDER_REFERENCE(x) ((x) << 0) +#define WRPLL_DIVIDER_REF_MASK (0xff) +#define WRPLL_DIVIDER_POST(x) ((x) << 8) +#define WRPLL_DIVIDER_POST_MASK (0x3f << 8) +#define WRPLL_DIVIDER_POST_SHIFT 8 +#define WRPLL_DIVIDER_FEEDBACK(x) ((x) << 16) +#define WRPLL_DIVIDER_FB_SHIFT 16 +#define WRPLL_DIVIDER_FB_MASK (0xff << 16) + +/* Port clock selection */ +#define _PORT_CLK_SEL_A 0x46100 +#define _PORT_CLK_SEL_B 0x46104 +#define PORT_CLK_SEL(port) _MMIO_PORT(port, _PORT_CLK_SEL_A, _PORT_CLK_SEL_B) +#define PORT_CLK_SEL_MASK REG_GENMASK(31, 29) +#define PORT_CLK_SEL_LCPLL_2700 REG_FIELD_PREP(PORT_CLK_SEL_MASK, 0) +#define PORT_CLK_SEL_LCPLL_1350 REG_FIELD_PREP(PORT_CLK_SEL_MASK, 1) +#define PORT_CLK_SEL_LCPLL_810 REG_FIELD_PREP(PORT_CLK_SEL_MASK, 2) +#define PORT_CLK_SEL_SPLL REG_FIELD_PREP(PORT_CLK_SEL_MASK, 3) +#define PORT_CLK_SEL_WRPLL(pll) REG_FIELD_PREP(PORT_CLK_SEL_MASK, 4 + (pll)) +#define PORT_CLK_SEL_WRPLL1 REG_FIELD_PREP(PORT_CLK_SEL_MASK, 4) +#define PORT_CLK_SEL_WRPLL2 REG_FIELD_PREP(PORT_CLK_SEL_MASK, 5) +#define PORT_CLK_SEL_NONE REG_FIELD_PREP(PORT_CLK_SEL_MASK, 7) + +/* On ICL+ this is the same as PORT_CLK_SEL, but all bits change. */ +#define DDI_CLK_SEL(port) PORT_CLK_SEL(port) +#define DDI_CLK_SEL_MASK REG_GENMASK(31, 28) +#define DDI_CLK_SEL_NONE REG_FIELD_PREP(DDI_CLK_SEL_MASK, 0x0) +#define DDI_CLK_SEL_MG REG_FIELD_PREP(DDI_CLK_SEL_MASK, 0x8) +#define DDI_CLK_SEL_TBT_162 REG_FIELD_PREP(DDI_CLK_SEL_MASK, 0xC) +#define DDI_CLK_SEL_TBT_270 REG_FIELD_PREP(DDI_CLK_SEL_MASK, 0xD) +#define DDI_CLK_SEL_TBT_540 REG_FIELD_PREP(DDI_CLK_SEL_MASK, 0xE) +#define DDI_CLK_SEL_TBT_810 REG_FIELD_PREP(DDI_CLK_SEL_MASK, 0xF) + +/* Transcoder clock selection */ +#define _TRANS_CLK_SEL_A 0x46140 +#define _TRANS_CLK_SEL_B 0x46144 +#define TRANS_CLK_SEL(tran) _MMIO_TRANS(tran, _TRANS_CLK_SEL_A, _TRANS_CLK_SEL_B) +/* For each transcoder, we need to select the corresponding port clock */ +#define TRANS_CLK_SEL_DISABLED (0x0 << 29) +#define TRANS_CLK_SEL_PORT(x) (((x) + 1) << 29) +#define TGL_TRANS_CLK_SEL_DISABLED (0x0 << 28) +#define TGL_TRANS_CLK_SEL_PORT(x) (((x) + 1) << 28) + +#define CDCLK_FREQ _MMIO(0x46200) + +#define _TRANSA_MSA_MISC 0x60410 +#define _TRANSB_MSA_MISC 0x61410 +#define _TRANSC_MSA_MISC 0x62410 +#define _TRANS_EDP_MSA_MISC 0x6f410 +#define TRANS_MSA_MISC(dev_priv, tran) _MMIO_TRANS2(dev_priv, tran, _TRANSA_MSA_MISC) +/* See DP_MSA_MISC_* for the bit definitions */ + +#define _TRANS_A_SET_CONTEXT_LATENCY 0x6007C +#define _TRANS_B_SET_CONTEXT_LATENCY 0x6107C +#define _TRANS_C_SET_CONTEXT_LATENCY 0x6207C +#define _TRANS_D_SET_CONTEXT_LATENCY 0x6307C +#define TRANS_SET_CONTEXT_LATENCY(dev_priv, tran) _MMIO_TRANS2(dev_priv, tran, _TRANS_A_SET_CONTEXT_LATENCY) +#define TRANS_SET_CONTEXT_LATENCY_MASK REG_GENMASK(15, 0) +#define TRANS_SET_CONTEXT_LATENCY_VALUE(x) REG_FIELD_PREP(TRANS_SET_CONTEXT_LATENCY_MASK, (x)) + +/* LCPLL Control */ +#define LCPLL_CTL _MMIO(0x130040) +#define LCPLL_PLL_DISABLE (1 << 31) +#define LCPLL_PLL_LOCK (1 << 30) +#define LCPLL_REF_NON_SSC (0 << 28) +#define LCPLL_REF_BCLK (2 << 28) +#define LCPLL_REF_PCH_SSC (3 << 28) +#define LCPLL_REF_MASK (3 << 28) +#define LCPLL_CLK_FREQ_MASK (3 << 26) +#define LCPLL_CLK_FREQ_450 (0 << 26) +#define LCPLL_CLK_FREQ_54O_BDW (1 << 26) +#define LCPLL_CLK_FREQ_337_5_BDW (2 << 26) +#define LCPLL_CLK_FREQ_675_BDW (3 << 26) +#define LCPLL_CD_CLOCK_DISABLE (1 << 25) +#define LCPLL_ROOT_CD_CLOCK_DISABLE (1 << 24) +#define LCPLL_CD2X_CLOCK_DISABLE (1 << 23) +#define LCPLL_POWER_DOWN_ALLOW (1 << 22) +#define LCPLL_CD_SOURCE_FCLK (1 << 21) +#define LCPLL_CD_SOURCE_FCLK_DONE (1 << 19) + +/* + * SKL Clocks + */ +/* CDCLK_CTL */ +#define CDCLK_CTL _MMIO(0x46000) +#define CDCLK_FREQ_SEL_MASK REG_GENMASK(27, 26) +#define CDCLK_FREQ_450_432 REG_FIELD_PREP(CDCLK_FREQ_SEL_MASK, 0) +#define CDCLK_FREQ_540 REG_FIELD_PREP(CDCLK_FREQ_SEL_MASK, 1) +#define CDCLK_FREQ_337_308 REG_FIELD_PREP(CDCLK_FREQ_SEL_MASK, 2) +#define CDCLK_FREQ_675_617 REG_FIELD_PREP(CDCLK_FREQ_SEL_MASK, 3) +#define MDCLK_SOURCE_SEL_MASK REG_GENMASK(25, 25) +#define MDCLK_SOURCE_SEL_CD2XCLK REG_FIELD_PREP(MDCLK_SOURCE_SEL_MASK, 0) +#define MDCLK_SOURCE_SEL_CDCLK_PLL REG_FIELD_PREP(MDCLK_SOURCE_SEL_MASK, 1) +#define BXT_CDCLK_CD2X_DIV_SEL_MASK REG_GENMASK(23, 22) +#define BXT_CDCLK_CD2X_DIV_SEL_1 REG_FIELD_PREP(BXT_CDCLK_CD2X_DIV_SEL_MASK, 0) +#define BXT_CDCLK_CD2X_DIV_SEL_1_5 REG_FIELD_PREP(BXT_CDCLK_CD2X_DIV_SEL_MASK, 1) +#define BXT_CDCLK_CD2X_DIV_SEL_2 REG_FIELD_PREP(BXT_CDCLK_CD2X_DIV_SEL_MASK, 2) +#define BXT_CDCLK_CD2X_DIV_SEL_4 REG_FIELD_PREP(BXT_CDCLK_CD2X_DIV_SEL_MASK, 3) +#define BXT_CDCLK_CD2X_PIPE(pipe) ((pipe) << 20) +#define CDCLK_DIVMUX_CD_OVERRIDE (1 << 19) +#define BXT_CDCLK_CD2X_PIPE_NONE BXT_CDCLK_CD2X_PIPE(3) +#define ICL_CDCLK_CD2X_PIPE(pipe) (_PICK(pipe, 0, 2, 6) << 19) +#define ICL_CDCLK_CD2X_PIPE_NONE (7 << 19) +#define TGL_CDCLK_CD2X_PIPE(pipe) BXT_CDCLK_CD2X_PIPE(pipe) +#define TGL_CDCLK_CD2X_PIPE_NONE ICL_CDCLK_CD2X_PIPE_NONE +#define BXT_CDCLK_SSA_PRECHARGE_ENABLE (1 << 16) +#define CDCLK_FREQ_DECIMAL_MASK (0x7ff) + +/* CDCLK_SQUASH_CTL */ +#define CDCLK_SQUASH_CTL _MMIO(0x46008) +#define CDCLK_SQUASH_ENABLE REG_BIT(31) +#define CDCLK_SQUASH_WINDOW_SIZE_MASK REG_GENMASK(27, 24) +#define CDCLK_SQUASH_WINDOW_SIZE(x) REG_FIELD_PREP(CDCLK_SQUASH_WINDOW_SIZE_MASK, (x)) +#define CDCLK_SQUASH_WAVEFORM_MASK REG_GENMASK(15, 0) +#define CDCLK_SQUASH_WAVEFORM(x) REG_FIELD_PREP(CDCLK_SQUASH_WAVEFORM_MASK, (x)) + +/* LCPLL_CTL */ +#define LCPLL1_CTL _MMIO(0x46010) +#define LCPLL2_CTL _MMIO(0x46014) +#define LCPLL_PLL_ENABLE (1 << 31) + +/* DPLL control1 */ +#define DPLL_CTRL1 _MMIO(0x6C058) +#define DPLL_CTRL1_HDMI_MODE(id) (1 << ((id) * 6 + 5)) +#define DPLL_CTRL1_SSC(id) (1 << ((id) * 6 + 4)) +#define DPLL_CTRL1_LINK_RATE_MASK(id) (7 << ((id) * 6 + 1)) +#define DPLL_CTRL1_LINK_RATE_SHIFT(id) ((id) * 6 + 1) +#define DPLL_CTRL1_LINK_RATE(linkrate, id) ((linkrate) << ((id) * 6 + 1)) +#define DPLL_CTRL1_OVERRIDE(id) (1 << ((id) * 6)) +#define DPLL_CTRL1_LINK_RATE_2700 0 +#define DPLL_CTRL1_LINK_RATE_1350 1 +#define DPLL_CTRL1_LINK_RATE_810 2 +#define DPLL_CTRL1_LINK_RATE_1620 3 +#define DPLL_CTRL1_LINK_RATE_1080 4 +#define DPLL_CTRL1_LINK_RATE_2160 5 + +/* DPLL control2 */ +#define DPLL_CTRL2 _MMIO(0x6C05C) +#define DPLL_CTRL2_DDI_CLK_OFF(port) (1 << ((port) + 15)) +#define DPLL_CTRL2_DDI_CLK_SEL_MASK(port) (3 << ((port) * 3 + 1)) +#define DPLL_CTRL2_DDI_CLK_SEL_SHIFT(port) ((port) * 3 + 1) +#define DPLL_CTRL2_DDI_CLK_SEL(clk, port) ((clk) << ((port) * 3 + 1)) +#define DPLL_CTRL2_DDI_SEL_OVERRIDE(port) (1 << ((port) * 3)) + +/* DPLL Status */ +#define DPLL_STATUS _MMIO(0x6C060) +#define DPLL_LOCK(id) (1 << ((id) * 8)) + +/* DPLL cfg */ +#define _DPLL1_CFGCR1 0x6C040 +#define _DPLL2_CFGCR1 0x6C048 +#define _DPLL3_CFGCR1 0x6C050 +#define DPLL_CFGCR1(id) _MMIO_PIPE((id) - SKL_DPLL1, _DPLL1_CFGCR1, _DPLL2_CFGCR1) +#define DPLL_CFGCR1_FREQ_ENABLE (1 << 31) +#define DPLL_CFGCR1_DCO_FRACTION_MASK (0x7fff << 9) +#define DPLL_CFGCR1_DCO_FRACTION(x) ((x) << 9) +#define DPLL_CFGCR1_DCO_INTEGER_MASK (0x1ff) + +#define _DPLL1_CFGCR2 0x6C044 +#define _DPLL2_CFGCR2 0x6C04C +#define _DPLL3_CFGCR2 0x6C054 +#define DPLL_CFGCR2(id) _MMIO_PIPE((id) - SKL_DPLL1, _DPLL1_CFGCR2, _DPLL2_CFGCR2) +#define DPLL_CFGCR2_QDIV_RATIO_MASK (0xff << 8) +#define DPLL_CFGCR2_QDIV_RATIO(x) ((x) << 8) +#define DPLL_CFGCR2_QDIV_MODE(x) ((x) << 7) +#define DPLL_CFGCR2_KDIV_MASK (3 << 5) +#define DPLL_CFGCR2_KDIV(x) ((x) << 5) +#define DPLL_CFGCR2_KDIV_5 (0 << 5) +#define DPLL_CFGCR2_KDIV_2 (1 << 5) +#define DPLL_CFGCR2_KDIV_3 (2 << 5) +#define DPLL_CFGCR2_KDIV_1 (3 << 5) +#define DPLL_CFGCR2_PDIV_MASK (7 << 2) +#define DPLL_CFGCR2_PDIV(x) ((x) << 2) +#define DPLL_CFGCR2_PDIV_1 (0 << 2) +#define DPLL_CFGCR2_PDIV_2 (1 << 2) +#define DPLL_CFGCR2_PDIV_3 (2 << 2) +#define DPLL_CFGCR2_PDIV_7 (4 << 2) +#define DPLL_CFGCR2_PDIV_7_INVALID (5 << 2) +#define DPLL_CFGCR2_CENTRAL_FREQ_MASK (3) + +/* ICL Clocks */ +#define ICL_DPCLKA_CFGCR0 _MMIO(0x164280) +#define ICL_DPCLKA_CFGCR0_DDI_CLK_OFF(phy) (1 << _PICK(phy, 10, 11, 24, 4, 5)) +#define RKL_DPCLKA_CFGCR0_DDI_CLK_OFF(phy) REG_BIT((phy) + 10) +#define ICL_DPCLKA_CFGCR0_TC_CLK_OFF(tc_port) (1 << ((tc_port) < TC_PORT_4 ? \ + (tc_port) + 12 : \ + (tc_port) - TC_PORT_4 + 21)) +#define ICL_DPCLKA_CFGCR0_DDI_CLK_SEL_SHIFT(phy) ((phy) * 2) +#define ICL_DPCLKA_CFGCR0_DDI_CLK_SEL_MASK(phy) (3 << ICL_DPCLKA_CFGCR0_DDI_CLK_SEL_SHIFT(phy)) +#define ICL_DPCLKA_CFGCR0_DDI_CLK_SEL(pll, phy) ((pll) << ICL_DPCLKA_CFGCR0_DDI_CLK_SEL_SHIFT(phy)) +#define RKL_DPCLKA_CFGCR0_DDI_CLK_SEL_SHIFT(phy) _PICK(phy, 0, 2, 4, 27) +#define RKL_DPCLKA_CFGCR0_DDI_CLK_SEL_MASK(phy) \ + (3 << RKL_DPCLKA_CFGCR0_DDI_CLK_SEL_SHIFT(phy)) +#define RKL_DPCLKA_CFGCR0_DDI_CLK_SEL(pll, phy) \ + ((pll) << RKL_DPCLKA_CFGCR0_DDI_CLK_SEL_SHIFT(phy)) + +/* + * DG1 Clocks + * First registers controls the first A and B, while the second register + * controls the phy C and D. The bits on these registers are the + * same, but refer to different phys + */ +#define _DG1_DPCLKA_CFGCR0 0x164280 +#define _DG1_DPCLKA1_CFGCR0 0x16C280 +#define _DG1_DPCLKA_PHY_IDX(phy) ((phy) % 2) +#define _DG1_DPCLKA_PLL_IDX(pll) ((pll) % 2) +#define DG1_DPCLKA_CFGCR0(phy) _MMIO_PHY((phy) / 2, \ + _DG1_DPCLKA_CFGCR0, \ + _DG1_DPCLKA1_CFGCR0) +#define DG1_DPCLKA_CFGCR0_DDI_CLK_OFF(phy) REG_BIT(_DG1_DPCLKA_PHY_IDX(phy) + 10) +#define DG1_DPCLKA_CFGCR0_DDI_CLK_SEL_SHIFT(phy) (_DG1_DPCLKA_PHY_IDX(phy) * 2) +#define DG1_DPCLKA_CFGCR0_DDI_CLK_SEL(pll, phy) (_DG1_DPCLKA_PLL_IDX(pll) << DG1_DPCLKA_CFGCR0_DDI_CLK_SEL_SHIFT(phy)) +#define DG1_DPCLKA_CFGCR0_DDI_CLK_SEL_MASK(phy) (0x3 << DG1_DPCLKA_CFGCR0_DDI_CLK_SEL_SHIFT(phy)) + +/* ADLS Clocks */ +#define _ADLS_DPCLKA_CFGCR0 0x164280 +#define _ADLS_DPCLKA_CFGCR1 0x1642BC +#define ADLS_DPCLKA_CFGCR(phy) _MMIO_PHY((phy) / 3, \ + _ADLS_DPCLKA_CFGCR0, \ + _ADLS_DPCLKA_CFGCR1) +#define ADLS_DPCLKA_CFGCR_DDI_SHIFT(phy) (((phy) % 3) * 2) +/* ADLS DPCLKA_CFGCR0 DDI mask */ +#define ADLS_DPCLKA_DDII_SEL_MASK REG_GENMASK(5, 4) +#define ADLS_DPCLKA_DDIB_SEL_MASK REG_GENMASK(3, 2) +#define ADLS_DPCLKA_DDIA_SEL_MASK REG_GENMASK(1, 0) +/* ADLS DPCLKA_CFGCR1 DDI mask */ +#define ADLS_DPCLKA_DDIK_SEL_MASK REG_GENMASK(3, 2) +#define ADLS_DPCLKA_DDIJ_SEL_MASK REG_GENMASK(1, 0) +#define ADLS_DPCLKA_CFGCR_DDI_CLK_SEL_MASK(phy) _PICK((phy), \ + ADLS_DPCLKA_DDIA_SEL_MASK, \ + ADLS_DPCLKA_DDIB_SEL_MASK, \ + ADLS_DPCLKA_DDII_SEL_MASK, \ + ADLS_DPCLKA_DDIJ_SEL_MASK, \ + ADLS_DPCLKA_DDIK_SEL_MASK) + +/* ICL PLL */ +#define _DPLL0_ENABLE 0x46010 +#define _DPLL1_ENABLE 0x46014 +#define _ADLS_DPLL2_ENABLE 0x46018 +#define _ADLS_DPLL3_ENABLE 0x46030 +#define PLL_ENABLE REG_BIT(31) +#define PLL_LOCK REG_BIT(30) +#define PLL_POWER_ENABLE REG_BIT(27) +#define PLL_POWER_STATE REG_BIT(26) +#define ICL_DPLL_ENABLE(pll) _MMIO(_PICK_EVEN_2RANGES(pll, 3, \ + _DPLL0_ENABLE, _DPLL1_ENABLE, \ + _ADLS_DPLL3_ENABLE, _ADLS_DPLL3_ENABLE)) + +#define _DG2_PLL3_ENABLE 0x4601C + +#define DG2_PLL_ENABLE(pll) _MMIO(_PICK_EVEN_2RANGES(pll, 3, \ + _DPLL0_ENABLE, _DPLL1_ENABLE, \ + _DG2_PLL3_ENABLE, _DG2_PLL3_ENABLE)) + +#define TBT_PLL_ENABLE _MMIO(0x46020) + +#define _MG_PLL1_ENABLE 0x46030 +#define _MG_PLL2_ENABLE 0x46034 +#define _MG_PLL3_ENABLE 0x46038 +#define _MG_PLL4_ENABLE 0x4603C +/* Bits are the same as _DPLL0_ENABLE */ +#define MG_PLL_ENABLE(tc_port) _MMIO_PORT((tc_port), _MG_PLL1_ENABLE, \ + _MG_PLL2_ENABLE) + +/* DG1 PLL */ +#define DG1_DPLL_ENABLE(pll) _MMIO(_PICK_EVEN_2RANGES(pll, 2, \ + _DPLL0_ENABLE, _DPLL1_ENABLE, \ + _MG_PLL1_ENABLE, _MG_PLL2_ENABLE)) + +/* ADL-P Type C PLL */ +#define PORTTC1_PLL_ENABLE 0x46038 +#define PORTTC2_PLL_ENABLE 0x46040 +#define ADLP_PORTTC_PLL_ENABLE(tc_port) _MMIO_PORT((tc_port), \ + PORTTC1_PLL_ENABLE, \ + PORTTC2_PLL_ENABLE) + +#define _ICL_DPLL0_CFGCR0 0x164000 +#define _ICL_DPLL1_CFGCR0 0x164080 +#define ICL_DPLL_CFGCR0(pll) _MMIO_PLL(pll, _ICL_DPLL0_CFGCR0, \ + _ICL_DPLL1_CFGCR0) +#define DPLL_CFGCR0_HDMI_MODE (1 << 30) +#define DPLL_CFGCR0_SSC_ENABLE (1 << 29) +#define DPLL_CFGCR0_SSC_ENABLE_ICL (1 << 25) +#define DPLL_CFGCR0_LINK_RATE_MASK (0xf << 25) +#define DPLL_CFGCR0_LINK_RATE_2700 (0 << 25) +#define DPLL_CFGCR0_LINK_RATE_1350 (1 << 25) +#define DPLL_CFGCR0_LINK_RATE_810 (2 << 25) +#define DPLL_CFGCR0_LINK_RATE_1620 (3 << 25) +#define DPLL_CFGCR0_LINK_RATE_1080 (4 << 25) +#define DPLL_CFGCR0_LINK_RATE_2160 (5 << 25) +#define DPLL_CFGCR0_LINK_RATE_3240 (6 << 25) +#define DPLL_CFGCR0_LINK_RATE_4050 (7 << 25) +#define DPLL_CFGCR0_DCO_FRACTION_MASK (0x7fff << 10) +#define DPLL_CFGCR0_DCO_FRACTION_SHIFT (10) +#define DPLL_CFGCR0_DCO_FRACTION(x) ((x) << 10) +#define DPLL_CFGCR0_DCO_INTEGER_MASK (0x3ff) + +#define _ICL_DPLL0_CFGCR1 0x164004 +#define _ICL_DPLL1_CFGCR1 0x164084 +#define ICL_DPLL_CFGCR1(pll) _MMIO_PLL(pll, _ICL_DPLL0_CFGCR1, \ + _ICL_DPLL1_CFGCR1) +#define DPLL_CFGCR1_QDIV_RATIO_MASK (0xff << 10) +#define DPLL_CFGCR1_QDIV_RATIO_SHIFT (10) +#define DPLL_CFGCR1_QDIV_RATIO(x) ((x) << 10) +#define DPLL_CFGCR1_QDIV_MODE_SHIFT (9) +#define DPLL_CFGCR1_QDIV_MODE(x) ((x) << 9) +#define DPLL_CFGCR1_KDIV_MASK (7 << 6) +#define DPLL_CFGCR1_KDIV_SHIFT (6) +#define DPLL_CFGCR1_KDIV(x) ((x) << 6) +#define DPLL_CFGCR1_KDIV_1 (1 << 6) +#define DPLL_CFGCR1_KDIV_2 (2 << 6) +#define DPLL_CFGCR1_KDIV_3 (4 << 6) +#define DPLL_CFGCR1_PDIV_MASK (0xf << 2) +#define DPLL_CFGCR1_PDIV_SHIFT (2) +#define DPLL_CFGCR1_PDIV(x) ((x) << 2) +#define DPLL_CFGCR1_PDIV_2 (1 << 2) +#define DPLL_CFGCR1_PDIV_3 (2 << 2) +#define DPLL_CFGCR1_PDIV_5 (4 << 2) +#define DPLL_CFGCR1_PDIV_7 (8 << 2) +#define DPLL_CFGCR1_CENTRAL_FREQ (3 << 0) +#define DPLL_CFGCR1_CENTRAL_FREQ_8400 (3 << 0) +#define TGL_DPLL_CFGCR1_CFSELOVRD_NORMAL_XTAL (0 << 0) + +#define _TGL_DPLL0_CFGCR0 0x164284 +#define _TGL_DPLL1_CFGCR0 0x16428C +#define _TGL_TBTPLL_CFGCR0 0x16429C +#define TGL_DPLL_CFGCR0(pll) _MMIO(_PICK_EVEN_2RANGES(pll, 2, \ + _TGL_DPLL0_CFGCR0, _TGL_DPLL1_CFGCR0, \ + _TGL_TBTPLL_CFGCR0, _TGL_TBTPLL_CFGCR0)) +#define RKL_DPLL_CFGCR0(pll) _MMIO_PLL(pll, _TGL_DPLL0_CFGCR0, \ + _TGL_DPLL1_CFGCR0) + +#define _TGL_DPLL0_DIV0 0x164B00 +#define _TGL_DPLL1_DIV0 0x164C00 +#define TGL_DPLL0_DIV0(pll) _MMIO_PLL(pll, _TGL_DPLL0_DIV0, _TGL_DPLL1_DIV0) +#define TGL_DPLL0_DIV0_AFC_STARTUP_MASK REG_GENMASK(27, 25) +#define TGL_DPLL0_DIV0_AFC_STARTUP(val) REG_FIELD_PREP(TGL_DPLL0_DIV0_AFC_STARTUP_MASK, (val)) + +#define _TGL_DPLL0_CFGCR1 0x164288 +#define _TGL_DPLL1_CFGCR1 0x164290 +#define _TGL_TBTPLL_CFGCR1 0x1642A0 +#define TGL_DPLL_CFGCR1(pll) _MMIO(_PICK_EVEN_2RANGES(pll, 2, \ + _TGL_DPLL0_CFGCR1, _TGL_DPLL1_CFGCR1, \ + _TGL_TBTPLL_CFGCR1, _TGL_TBTPLL_CFGCR1)) +#define RKL_DPLL_CFGCR1(pll) _MMIO_PLL(pll, _TGL_DPLL0_CFGCR1, \ + _TGL_DPLL1_CFGCR1) + +#define _DG1_DPLL2_CFGCR0 0x16C284 +#define _DG1_DPLL3_CFGCR0 0x16C28C +#define DG1_DPLL_CFGCR0(pll) _MMIO(_PICK_EVEN_2RANGES(pll, 2, \ + _TGL_DPLL0_CFGCR0, _TGL_DPLL1_CFGCR0, \ + _DG1_DPLL2_CFGCR0, _DG1_DPLL3_CFGCR0)) + +#define _DG1_DPLL2_CFGCR1 0x16C288 +#define _DG1_DPLL3_CFGCR1 0x16C290 +#define DG1_DPLL_CFGCR1(pll) _MMIO(_PICK_EVEN_2RANGES(pll, 2, \ + _TGL_DPLL0_CFGCR1, _TGL_DPLL1_CFGCR1, \ + _DG1_DPLL2_CFGCR1, _DG1_DPLL3_CFGCR1)) + +/* For ADL-S DPLL4_CFGCR0/1 are used to control DPLL2 */ +#define _ADLS_DPLL4_CFGCR0 0x164294 +#define _ADLS_DPLL3_CFGCR0 0x1642C0 +#define ADLS_DPLL_CFGCR0(pll) _MMIO(_PICK_EVEN_2RANGES(pll, 2, \ + _TGL_DPLL0_CFGCR0, _TGL_DPLL1_CFGCR0, \ + _ADLS_DPLL4_CFGCR0, _ADLS_DPLL3_CFGCR0)) + +#define _ADLS_DPLL4_CFGCR1 0x164298 +#define _ADLS_DPLL3_CFGCR1 0x1642C4 +#define ADLS_DPLL_CFGCR1(pll) _MMIO(_PICK_EVEN_2RANGES(pll, 2, \ + _TGL_DPLL0_CFGCR1, _TGL_DPLL1_CFGCR1, \ + _ADLS_DPLL4_CFGCR1, _ADLS_DPLL3_CFGCR1)) + +/* BXT display engine PLL */ +#define BXT_DE_PLL_CTL _MMIO(0x6d000) +#define BXT_DE_PLL_RATIO(x) (x) /* {60,65,100} * 19.2MHz */ +#define BXT_DE_PLL_RATIO_MASK 0xff + +#define BXT_DE_PLL_ENABLE _MMIO(0x46070) +#define BXT_DE_PLL_PLL_ENABLE (1 << 31) +#define BXT_DE_PLL_LOCK (1 << 30) +#define BXT_DE_PLL_FREQ_REQ (1 << 23) +#define BXT_DE_PLL_FREQ_REQ_ACK (1 << 22) +#define ICL_CDCLK_PLL_RATIO(x) (x) +#define ICL_CDCLK_PLL_RATIO_MASK 0xff + +/* GEN9 DC */ +#define DC_STATE_EN _MMIO(0x45504) +#define DC_STATE_DISABLE 0 +#define DC_STATE_EN_DC3CO REG_BIT(30) +#define DC_STATE_DC3CO_STATUS REG_BIT(29) +#define HOLD_PHY_CLKREQ_PG1_LATCH REG_BIT(21) +#define HOLD_PHY_PG1_LATCH REG_BIT(20) +#define DC_STATE_EN_UPTO_DC5 (1 << 0) +#define DC_STATE_EN_DC9 (1 << 3) +#define DC_STATE_EN_UPTO_DC6 (2 << 0) +#define DC_STATE_EN_UPTO_DC5_DC6_MASK 0x3 + +#define DC_STATE_DEBUG _MMIO(0x45520) +#define DC_STATE_DEBUG_MASK_CORES (1 << 0) +#define DC_STATE_DEBUG_MASK_MEMORY_UP (1 << 1) + +#define D_COMP_BDW _MMIO(0x138144) + +/* Pipe WM_LINETIME - watermark line time */ +#define _WM_LINETIME_A 0x45270 +#define _WM_LINETIME_B 0x45274 +#define WM_LINETIME(pipe) _MMIO_PIPE(pipe, _WM_LINETIME_A, _WM_LINETIME_B) +#define HSW_LINETIME_MASK REG_GENMASK(8, 0) +#define HSW_LINETIME(x) REG_FIELD_PREP(HSW_LINETIME_MASK, (x)) +#define HSW_IPS_LINETIME_MASK REG_GENMASK(24, 16) +#define HSW_IPS_LINETIME(x) REG_FIELD_PREP(HSW_IPS_LINETIME_MASK, (x)) + +/* SFUSE_STRAP */ +#define SFUSE_STRAP _MMIO(0xc2014) +#define SFUSE_STRAP_FUSE_LOCK (1 << 13) +#define SFUSE_STRAP_RAW_FREQUENCY (1 << 8) +#define SFUSE_STRAP_DISPLAY_DISABLED (1 << 7) +#define SFUSE_STRAP_CRT_DISABLED (1 << 6) +#define SFUSE_STRAP_DDIF_DETECTED (1 << 3) +#define SFUSE_STRAP_DDIB_DETECTED (1 << 2) +#define SFUSE_STRAP_DDIC_DETECTED (1 << 1) +#define SFUSE_STRAP_DDID_DETECTED (1 << 0) + +/* Gen4+ Timestamp and Pipe Frame time stamp registers */ +#define GEN4_TIMESTAMP _MMIO(0x2358) +#define ILK_TIMESTAMP_HI _MMIO(0x70070) +#define IVB_TIMESTAMP_CTR _MMIO(0x44070) + +/* g4x+, except vlv/chv! */ +#define _PIPE_FRMTMSTMP_A 0x70048 +#define _PIPE_FRMTMSTMP_B 0x71048 +#define PIPE_FRMTMSTMP(pipe) \ + _MMIO_PIPE(pipe, _PIPE_FRMTMSTMP_A, _PIPE_FRMTMSTMP_B) + +/* g4x+, except vlv/chv! */ +#define _PIPE_FLIPTMSTMP_A 0x7004C +#define _PIPE_FLIPTMSTMP_B 0x7104C +#define PIPE_FLIPTMSTMP(pipe) \ + _MMIO_PIPE(pipe, _PIPE_FLIPTMSTMP_A, _PIPE_FLIPTMSTMP_B) + +/* tgl+ */ +#define _PIPE_FLIPDONETMSTMP_A 0x70054 +#define _PIPE_FLIPDONETMSTMP_B 0x71054 +#define PIPE_FLIPDONETIMSTMP(pipe) \ + _MMIO_PIPE(pipe, _PIPE_FLIPDONETMSTMP_A, _PIPE_FLIPDONETMSTMP_B) + +#define _VLV_PIPE_MSA_MISC_A 0x70048 +#define VLV_PIPE_MSA_MISC(__display, pipe) \ + _MMIO_PIPE2(__display, pipe, _VLV_PIPE_MSA_MISC_A) +#define VLV_MSA_MISC1_HW_ENABLE REG_BIT(31) +#define VLV_MSA_MISC1_SW_S3D_MASK REG_GENMASK(2, 0) /* MSA MISC1 3:1 */ + +#define _ICL_PHY_MISC_A 0x64C00 +#define _ICL_PHY_MISC_B 0x64C04 +#define _DG2_PHY_MISC_TC1 0x64C14 /* TC1="PHY E" but offset as if "PHY F" */ +#define ICL_PHY_MISC(port) _MMIO_PORT(port, _ICL_PHY_MISC_A, _ICL_PHY_MISC_B) +#define DG2_PHY_MISC(port) ((port) == PHY_E ? _MMIO(_DG2_PHY_MISC_TC1) : \ + ICL_PHY_MISC(port)) +#define ICL_PHY_MISC_MUX_DDID (1 << 28) +#define ICL_PHY_MISC_DE_IO_COMP_PWR_DOWN (1 << 23) +#define DG2_PHY_DP_TX_ACK_MASK REG_GENMASK(23, 20) + +#define PORT_TX_DFLEXDPSP(fia) _MMIO_FIA((fia), 0x008A0) +#define MODULAR_FIA_MASK (1 << 4) +#define TC_LIVE_STATE_TBT(idx) (1 << ((idx) * 8 + 6)) +#define TC_LIVE_STATE_TC(idx) (1 << ((idx) * 8 + 5)) +#define DP_LANE_ASSIGNMENT_SHIFT(idx) ((idx) * 8) +#define DP_LANE_ASSIGNMENT_MASK(idx) (0xf << ((idx) * 8)) +#define DP_LANE_ASSIGNMENT(idx, x) ((x) << ((idx) * 8)) + +#define PORT_TX_DFLEXDPPMS(fia) _MMIO_FIA((fia), 0x00890) +#define DP_PHY_MODE_STATUS_COMPLETED(idx) (1 << (idx)) + +#define PORT_TX_DFLEXDPCSSS(fia) _MMIO_FIA((fia), 0x00894) +#define DP_PHY_MODE_STATUS_NOT_SAFE(idx) (1 << (idx)) + +#define PORT_TX_DFLEXPA1(fia) _MMIO_FIA((fia), 0x00880) +#define DP_PIN_ASSIGNMENT_SHIFT(idx) ((idx) * 4) +#define DP_PIN_ASSIGNMENT_MASK(idx) (0xf << ((idx) * 4)) +#define DP_PIN_ASSIGNMENT(idx, x) ((x) << ((idx) * 4)) + +#define _TCSS_DDI_STATUS_1 0x161500 +#define _TCSS_DDI_STATUS_2 0x161504 +#define TCSS_DDI_STATUS(tc) _MMIO(_PICK_EVEN(tc, \ + _TCSS_DDI_STATUS_1, \ + _TCSS_DDI_STATUS_2)) +#define TCSS_DDI_STATUS_PIN_ASSIGNMENT_MASK REG_GENMASK(28, 25) +#define TCSS_DDI_STATUS_READY REG_BIT(2) +#define TCSS_DDI_STATUS_HPD_LIVE_STATUS_TBT REG_BIT(1) +#define TCSS_DDI_STATUS_HPD_LIVE_STATUS_ALT REG_BIT(0) + +#define CLKREQ_POLICY _MMIO(0x101038) +#define CLKREQ_POLICY_MEM_UP_OVRD REG_BIT(1) + +#define CLKGATE_DIS_MISC _MMIO(0x46534) +#define CLKGATE_DIS_MISC_DMASC_GATING_DIS REG_BIT(21) + +#define _MTL_CLKGATE_DIS_TRANS_A 0x604E8 +#define _MTL_CLKGATE_DIS_TRANS_B 0x614E8 +#define MTL_CLKGATE_DIS_TRANS(dev_priv, trans) _MMIO_TRANS2(dev_priv, trans, _MTL_CLKGATE_DIS_TRANS_A) +#define MTL_CLKGATE_DIS_TRANS_DMASC_GATING_DIS REG_BIT(7) + +#define _MTL_PIPE_CLKGATE_DIS2_A 0x60114 +#define _MTL_PIPE_CLKGATE_DIS2_B 0x61114 +#define MTL_PIPE_CLKGATE_DIS2(pipe) _MMIO_PIPE(pipe, _MTL_PIPE_CLKGATE_DIS2_A, _MTL_PIPE_CLKGATE_DIS2_B) +#define MTL_DPFC_GATING_DIS REG_BIT(6) + +#define MTL_MEM_SS_INFO_QGV_POINT_OFFSET 0x45710 +#define MTL_MEM_SS_INFO_QGV_POINT_LOW(point) _MMIO(MTL_MEM_SS_INFO_QGV_POINT_OFFSET + (point) * 8) +#define MTL_TRCD_MASK REG_GENMASK(31, 24) +#define MTL_TRP_MASK REG_GENMASK(23, 16) +#define MTL_DCLK_MASK REG_GENMASK(15, 0) + +#define MTL_MEM_SS_INFO_QGV_POINT_HIGH(point) _MMIO(MTL_MEM_SS_INFO_QGV_POINT_OFFSET + (point) * 8 + 4) +#define MTL_TRAS_MASK REG_GENMASK(16, 8) +#define MTL_TRDPRE_MASK REG_GENMASK(7, 0) + + + +#endif /* __INTEL_DISPLAY_REGS_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_display_reset.c b/drivers/gpu/drm/i915/display/intel_display_reset.c index 1dbd3e841df3378f69b4014403c26b655d5e3f19..f5f38dca14d7f5cb56d665e72e845dfa06d988b3 100644 --- a/drivers/gpu/drm/i915/display/intel_display_reset.c +++ b/drivers/gpu/drm/i915/display/intel_display_reset.c @@ -8,6 +8,7 @@ #include "i915_drv.h" #include "intel_clock_gating.h" #include "intel_cx0_phy.h" +#include "intel_display_core.h" #include "intel_display_driver.h" #include "intel_display_reset.h" #include "intel_display_types.h" diff --git a/drivers/gpu/drm/i915/display/intel_display_rpm.c b/drivers/gpu/drm/i915/display/intel_display_rpm.c index 48da67dd0136188dbd38489eb101ae979c40e69a..56c4024201c16e8b3d9bab75ab0eafa15cbe98a4 100644 --- a/drivers/gpu/drm/i915/display/intel_display_rpm.c +++ b/drivers/gpu/drm/i915/display/intel_display_rpm.c @@ -2,6 +2,7 @@ /* Copyright © 2025 Intel Corporation */ #include "i915_drv.h" +#include "intel_display_core.h" #include "intel_display_rpm.h" #include "intel_runtime_pm.h" diff --git a/drivers/gpu/drm/i915/display/intel_display_rps.c b/drivers/gpu/drm/i915/display/intel_display_rps.c index 678b241159512b20fc4d6f60ca047bdbe9398fdc..82ea1ec482e417ac6e8ae4c451559f4c2fa470b4 100644 --- a/drivers/gpu/drm/i915/display/intel_display_rps.c +++ b/drivers/gpu/drm/i915/display/intel_display_rps.c @@ -9,6 +9,7 @@ #include "gt/intel_rps.h" #include "i915_drv.h" #include "i915_reg.h" +#include "intel_display_core.h" #include "intel_display_irq.h" #include "intel_display_rps.h" #include "intel_display_types.h" @@ -45,12 +46,13 @@ static int do_rps_boost(struct wait_queue_entry *_wait, void intel_display_rps_boost_after_vblank(struct drm_crtc *crtc, struct dma_fence *fence) { + struct intel_display *display = to_intel_display(crtc->dev); struct wait_rps_boost *wait; if (!dma_fence_is_i915(fence)) return; - if (DISPLAY_VER(to_i915(crtc->dev)) < 6) + if (DISPLAY_VER(display) < 6) return; if (drm_crtc_vblank_get(crtc)) diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h index 79c328d49ef07b26272bf81a794d6961021cc049..ce45261c4a8f4299293dd9dc485e63139d29063a 100644 --- a/drivers/gpu/drm/i915/display/intel_display_types.h +++ b/drivers/gpu/drm/i915/display/intel_display_types.h @@ -146,6 +146,8 @@ struct intel_framebuffer { unsigned int min_alignment; unsigned int vtd_guard; + + unsigned int (*panic_tiling)(unsigned int x, unsigned int y, unsigned int width); }; enum intel_hotplug_state { @@ -554,6 +556,10 @@ struct intel_connector { struct intel_dp *dp; } mst; + struct { + int force_bpp_x16; + } link; + /* Work struct to schedule a uevent on link train failure */ struct work_struct modeset_retry_work; @@ -595,7 +601,7 @@ struct intel_atomic_state { bool dpll_set, modeset; - struct intel_shared_dpll_state shared_dpll[I915_NUM_PLLS]; + struct intel_dpll_state dpll_state[I915_NUM_PLLS]; struct intel_dp_tunnel_inherited_state *inherited_dp_tunnels; @@ -1079,8 +1085,8 @@ struct intel_crtc_state { * haswell. */ struct dpll dpll; - /* Selected dpll when shared or NULL. */ - struct intel_shared_dpll *shared_dpll; + /* Selected dpll or NULL. */ + struct intel_dpll *intel_dpll; /* Actual register state of the dpll, for shared dpll cross-checking. */ struct intel_dpll_hw_state dpll_hw_state; @@ -1090,7 +1096,7 @@ struct intel_crtc_state { * setting shared_dpll and dpll_hw_state to one of these reserved ones. */ struct icl_port_dpll { - struct intel_shared_dpll *pll; + struct intel_dpll *pll; struct intel_dpll_hw_state hw_state; } icl_port_dplls[ICL_PORT_DPLL_COUNT]; @@ -1297,8 +1303,9 @@ struct intel_crtc_state { enum transcoder mst_master_transcoder; /* For DSB based pipe updates */ - struct intel_dsb *dsb_color_vblank, *dsb_commit; + struct intel_dsb *dsb_color, *dsb_commit; bool use_dsb; + bool use_flipq; u32 psr2_man_track_ctl; @@ -1365,6 +1372,21 @@ struct intel_pipe_crc { enum intel_pipe_crc_source source; }; +enum intel_flipq_id { + INTEL_FLIPQ_PLANE_1, + INTEL_FLIPQ_PLANE_2, + INTEL_FLIPQ_PLANE_3, + INTEL_FLIPQ_GENERAL, + INTEL_FLIPQ_FAST, + MAX_INTEL_FLIPQ, +}; + +struct intel_flipq { + u32 start_mmioaddr; + enum intel_flipq_id flipq_id; + u8 tail; +}; + struct intel_crtc { struct drm_crtc base; enum pipe pipe; @@ -1391,11 +1413,15 @@ struct intel_crtc { struct drm_pending_vblank_event *flip_done_event; /* armed event for DSB based updates */ struct drm_pending_vblank_event *dsb_event; + /* armed event for flip queue based updates */ + struct drm_pending_vblank_event *flipq_event; /* Access to these should be protected by display->irq.lock. */ bool cpu_fifo_underrun_disabled; bool pch_fifo_underrun_disabled; + struct intel_flipq flipq[MAX_INTEL_FLIPQ]; + /* per-pipe watermark state */ struct { /* watermarks currently being used */ @@ -1517,6 +1543,8 @@ struct intel_plane { bool async_flip); void (*enable_flip_done)(struct intel_plane *plane); void (*disable_flip_done)(struct intel_plane *plane); + /* For drm_panic */ + void (*disable_tiling)(struct intel_plane *plane); }; #define to_intel_atomic_state(x) container_of(x, struct intel_atomic_state, base) @@ -1669,7 +1697,9 @@ struct intel_dp { bool use_max_params; u8 dpcd[DP_RECEIVER_CAP_SIZE]; u8 psr_dpcd[EDP_PSR_RECEIVER_CAP_SIZE]; - u8 pr_dpcd; + u8 pr_dpcd[DP_PANEL_REPLAY_CAP_SIZE]; +#define INTEL_PR_DPCD_INDEX(pr_dpcd_register) ((pr_dpcd_register) - DP_PANEL_REPLAY_CAP_SUPPORT) + u8 downstream_ports[DP_MAX_DOWNSTREAM_PORTS]; u8 edp_dpcd[EDP_DISPLAY_CTL_CAP_SIZE]; u8 lttpr_common_caps[DP_LTTPR_COMMON_CAP_SIZE]; diff --git a/drivers/gpu/drm/i915/display/intel_display_wa.c b/drivers/gpu/drm/i915/display/intel_display_wa.c index da429c33291426622efa04de92ab3271c8c28bb7..f57280e9d041da536e9330f232fd4778ccbb3565 100644 --- a/drivers/gpu/drm/i915/display/intel_display_wa.c +++ b/drivers/gpu/drm/i915/display/intel_display_wa.c @@ -6,6 +6,7 @@ #include "i915_reg.h" #include "intel_de.h" #include "intel_display_core.h" +#include "intel_display_regs.h" #include "intel_display_wa.h" static void gen11_display_wa_apply(struct intel_display *display) diff --git a/drivers/gpu/drm/i915/display/intel_dkl_phy.c b/drivers/gpu/drm/i915/display/intel_dkl_phy.c index dad7192132ad9259f2600f9ea9ea3139b2e03d68..35e919eae3691e6264ed49cfb5ba420e6f5030bd 100644 --- a/drivers/gpu/drm/i915/display/intel_dkl_phy.c +++ b/drivers/gpu/drm/i915/display/intel_dkl_phy.c @@ -25,7 +25,9 @@ dkl_phy_set_hip_idx(struct intel_display *display, struct intel_dkl_phy_reg reg) { enum tc_port tc_port = DKL_REG_TC_PORT(reg); - drm_WARN_ON(display->drm, tc_port < TC_PORT_1 || tc_port >= I915_MAX_TC_PORTS); + if (drm_WARN_ON(display->drm, + tc_port < TC_PORT_1 || tc_port >= I915_MAX_TC_PORTS)) + return; intel_de_write(display, HIP_INDEX_REG(tc_port), diff --git a/drivers/gpu/drm/i915/display/intel_dkl_phy_regs.h b/drivers/gpu/drm/i915/display/intel_dkl_phy_regs.h index 56085b32956d877e40529147ac6089ba0b39eee1..f8ffeec29e933eb9a205ed0466f1ddf0abb191e4 100644 --- a/drivers/gpu/drm/i915/display/intel_dkl_phy_regs.h +++ b/drivers/gpu/drm/i915/display/intel_dkl_phy_regs.h @@ -8,6 +8,8 @@ #include +#include "intel_display_reg_defs.h" + struct intel_dkl_phy_reg { u32 reg:24; u32 bank_idx:4; @@ -151,6 +153,7 @@ struct intel_dkl_phy_reg { #define DKL_TX_DPCNTL2_CFG_LOADGENSELECT_TX1(val) REG_FIELD_PREP(DKL_TX_DPCNTL2_CFG_LOADGENSELECT_TX1_MASK, (val)) #define DKL_TX_DPCNTL2_CFG_LOADGENSELECT_TX2_MASK REG_GENMASK(6, 5) #define DKL_TX_DPCNTL2_CFG_LOADGENSELECT_TX2(val) REG_FIELD_PREP(DKL_TX_DPCNTL2_CFG_LOADGENSELECT_TX2_MASK, (val)) +#define LOADGEN_SHARING_PMD_DISABLE REG_BIT(12) #define _DKL_TX_FW_CALIB_LN0 0x02F8 #define _DKL_TX_FW_CALIB_LN1 0x12F8 diff --git a/drivers/gpu/drm/i915/display/intel_dmc.c b/drivers/gpu/drm/i915/display/intel_dmc.c index b58189d24e7e404442482a98c26897242d806665..744f51c0eab82cca405e99f65b135d9b235fba61 100644 --- a/drivers/gpu/drm/i915/display/intel_dmc.c +++ b/drivers/gpu/drm/i915/display/intel_dmc.c @@ -24,14 +24,22 @@ #include #include +#include + +#include +#include -#include "i915_drv.h" #include "i915_reg.h" +#include "i915_utils.h" +#include "intel_crtc.h" #include "intel_de.h" -#include "intel_display_rpm.h" #include "intel_display_power_well.h" +#include "intel_display_regs.h" +#include "intel_display_rpm.h" +#include "intel_display_types.h" #include "intel_dmc.h" #include "intel_dmc_regs.h" +#include "intel_flipq.h" #include "intel_step.h" /** @@ -176,7 +184,8 @@ static const char *dmc_firmware_default(struct intel_display *display, u32 *size const char *fw_path = NULL; u32 max_fw_size = 0; - if (DISPLAY_VERx100(display) == 3000) { + if (DISPLAY_VERx100(display) == 3002 || + DISPLAY_VERx100(display) == 3000) { fw_path = XE3LPD_DMC_PATH; max_fw_size = XE2LPD_DMC_MAX_FW_SIZE; } else if (DISPLAY_VERx100(display) == 2000) { @@ -425,29 +434,26 @@ static void disable_event_handler(struct intel_display *display, REG_FIELD_PREP(DMC_EVT_CTL_TYPE_MASK, DMC_EVT_CTL_TYPE_EDGE_0_1) | REG_FIELD_PREP(DMC_EVT_CTL_EVENT_ID_MASK, - DMC_EVT_CTL_EVENT_ID_FALSE)); + DMC_EVENT_FALSE)); intel_de_write(display, htp_reg, 0); } -static void disable_all_event_handlers(struct intel_display *display) +static void disable_all_event_handlers(struct intel_display *display, + enum intel_dmc_id dmc_id) { - enum intel_dmc_id dmc_id; + int handler; /* TODO: disable the event handlers on pre-GEN12 platforms as well */ if (DISPLAY_VER(display) < 12) return; - for_each_dmc_id(dmc_id) { - int handler; - - if (!has_dmc_id_fw(display, dmc_id)) - continue; + if (!has_dmc_id_fw(display, dmc_id)) + return; - for (handler = 0; handler < DMC_EVENT_HANDLER_COUNT_GEN12; handler++) - disable_event_handler(display, - DMC_EVT_CTL(display, dmc_id, handler), - DMC_EVT_HTP(display, dmc_id, handler)); - } + for (handler = 0; handler < DMC_EVENT_HANDLER_COUNT_GEN12; handler++) + disable_event_handler(display, + DMC_EVT_CTL(display, dmc_id, handler), + DMC_EVT_HTP(display, dmc_id, handler)); } static void adlp_pipedmc_clock_gating_wa(struct intel_display *display, bool enable) @@ -479,89 +485,36 @@ static void mtl_pipedmc_clock_gating_wa(struct intel_display *display) * for pipe A and B. */ intel_de_rmw(display, GEN9_CLKGATE_DIS_0, 0, - MTL_PIPEDMC_GATING_DIS_A | MTL_PIPEDMC_GATING_DIS_B); + MTL_PIPEDMC_GATING_DIS(PIPE_A) | + MTL_PIPEDMC_GATING_DIS(PIPE_B)); } static void pipedmc_clock_gating_wa(struct intel_display *display, bool enable) { - if (DISPLAY_VER(display) >= 14 && enable) + if (display->platform.meteorlake && enable) mtl_pipedmc_clock_gating_wa(display); else if (DISPLAY_VER(display) == 13) adlp_pipedmc_clock_gating_wa(display, enable); } -void intel_dmc_enable_pipe(struct intel_display *display, enum pipe pipe) +static u32 pipedmc_interrupt_mask(struct intel_display *display) { - enum intel_dmc_id dmc_id = PIPE_TO_DMC_ID(pipe); - - if (!is_valid_dmc_id(dmc_id) || !has_dmc_id_fw(display, dmc_id)) - return; - - if (DISPLAY_VER(display) >= 14) - intel_de_rmw(display, MTL_PIPEDMC_CONTROL, 0, PIPEDMC_ENABLE_MTL(pipe)); - else - intel_de_rmw(display, PIPEDMC_CONTROL(pipe), 0, PIPEDMC_ENABLE); -} - -void intel_dmc_disable_pipe(struct intel_display *display, enum pipe pipe) -{ - enum intel_dmc_id dmc_id = PIPE_TO_DMC_ID(pipe); - - if (!is_valid_dmc_id(dmc_id) || !has_dmc_id_fw(display, dmc_id)) - return; - - if (DISPLAY_VER(display) >= 14) - intel_de_rmw(display, MTL_PIPEDMC_CONTROL, PIPEDMC_ENABLE_MTL(pipe), 0); - else - intel_de_rmw(display, PIPEDMC_CONTROL(pipe), PIPEDMC_ENABLE, 0); -} - -/** - * intel_dmc_block_pkgc() - block PKG C-state - * @display: display instance - * @pipe: pipe which register use to block - * @block: block/unblock - * - * This interface is target for Wa_16025596647 usage. I.e. to set/clear - * PIPEDMC_BLOCK_PKGC_SW_BLOCK_PKGC_ALWAYS bit in PIPEDMC_BLOCK_PKGC_SW register. - */ -void intel_dmc_block_pkgc(struct intel_display *display, enum pipe pipe, - bool block) -{ - intel_de_rmw(display, PIPEDMC_BLOCK_PKGC_SW(pipe), - PIPEDMC_BLOCK_PKGC_SW_BLOCK_PKGC_ALWAYS, block ? - PIPEDMC_BLOCK_PKGC_SW_BLOCK_PKGC_ALWAYS : 0); + /* + * FIXME PIPEDMC_ERROR not enabled for now due to LNL pipe B + * triggering it during the first DC state transition. Figure + * out what is going on... + */ + return PIPEDMC_FLIPQ_PROG_DONE | + PIPEDMC_GTT_FAULT | + PIPEDMC_ATS_FAULT; } -/** - * intel_dmc_start_pkgc_exit_at_start_of_undelayed_vblank() - start of PKG - * C-state exit - * @display: display instance - * @pipe: pipe which register use to block - * @enable: enable/disable - * - * This interface is target for Wa_16025596647 usage. I.e. start the package C - * exit at the start of the undelayed vblank - */ -void intel_dmc_start_pkgc_exit_at_start_of_undelayed_vblank(struct intel_display *display, - enum pipe pipe, bool enable) +static u32 dmc_evt_ctl_disable(void) { - u32 val; - - if (enable) - val = DMC_EVT_CTL_ENABLE | DMC_EVT_CTL_RECURRING | - REG_FIELD_PREP(DMC_EVT_CTL_TYPE_MASK, - DMC_EVT_CTL_TYPE_EDGE_0_1) | - REG_FIELD_PREP(DMC_EVT_CTL_EVENT_ID_MASK, - DMC_EVT_CTL_EVENT_ID_VBLANK_A); - else - val = REG_FIELD_PREP(DMC_EVT_CTL_EVENT_ID_MASK, - DMC_EVT_CTL_EVENT_ID_FALSE) | - REG_FIELD_PREP(DMC_EVT_CTL_TYPE_MASK, - DMC_EVT_CTL_TYPE_EDGE_0_1); - - intel_de_write(display, MTL_PIPEDMC_EVT_CTL_4(pipe), - val); + return REG_FIELD_PREP(DMC_EVT_CTL_TYPE_MASK, + DMC_EVT_CTL_TYPE_EDGE_0_1) | + REG_FIELD_PREP(DMC_EVT_CTL_EVENT_ID_MASK, + DMC_EVENT_FALSE); } static bool is_dmc_evt_ctl_reg(struct intel_display *display, @@ -584,6 +537,15 @@ static bool is_dmc_evt_htp_reg(struct intel_display *display, return offset >= start && offset < end; } +static bool is_event_handler(struct intel_display *display, + enum intel_dmc_id dmc_id, + unsigned int event_id, + i915_reg_t reg, u32 data) +{ + return is_dmc_evt_ctl_reg(display, dmc_id, reg) && + REG_FIELD_GET(DMC_EVT_CTL_EVENT_ID_MASK, data) == event_id; +} + static bool disable_dmc_evt(struct intel_display *display, enum intel_dmc_id dmc_id, i915_reg_t reg, u32 data) @@ -597,12 +559,12 @@ static bool disable_dmc_evt(struct intel_display *display, /* also disable the flip queue event on the main DMC on TGL */ if (display->platform.tigerlake && - REG_FIELD_GET(DMC_EVT_CTL_EVENT_ID_MASK, data) == DMC_EVT_CTL_EVENT_ID_CLK_MSEC) + is_event_handler(display, dmc_id, MAINDMC_EVENT_CLK_MSEC, reg, data)) return true; /* also disable the HRR event on the main DMC on TGL/ADLS */ if ((display->platform.tigerlake || display->platform.alderlake_s) && - REG_FIELD_GET(DMC_EVT_CTL_EVENT_ID_MASK, data) == DMC_EVT_CTL_EVENT_ID_VBLANK_A) + is_event_handler(display, dmc_id, MAINDMC_EVENT_VBLANK_A, reg, data)) return true; return false; @@ -615,14 +577,267 @@ static u32 dmc_mmiodata(struct intel_display *display, if (disable_dmc_evt(display, dmc_id, dmc->dmc_info[dmc_id].mmioaddr[i], dmc->dmc_info[dmc_id].mmiodata[i])) - return REG_FIELD_PREP(DMC_EVT_CTL_TYPE_MASK, - DMC_EVT_CTL_TYPE_EDGE_0_1) | - REG_FIELD_PREP(DMC_EVT_CTL_EVENT_ID_MASK, - DMC_EVT_CTL_EVENT_ID_FALSE); + return dmc_evt_ctl_disable(); else return dmc->dmc_info[dmc_id].mmiodata[i]; } +static void dmc_load_mmio(struct intel_display *display, enum intel_dmc_id dmc_id) +{ + struct intel_dmc *dmc = display_to_dmc(display); + int i; + + for (i = 0; i < dmc->dmc_info[dmc_id].mmio_count; i++) { + intel_de_write(display, dmc->dmc_info[dmc_id].mmioaddr[i], + dmc_mmiodata(display, dmc, dmc_id, i)); + } +} + +static void dmc_load_program(struct intel_display *display, enum intel_dmc_id dmc_id) +{ + struct intel_dmc *dmc = display_to_dmc(display); + int i; + + disable_all_event_handlers(display, dmc_id); + + preempt_disable(); + + for (i = 0; i < dmc->dmc_info[dmc_id].dmc_fw_size; i++) { + intel_de_write_fw(display, + DMC_PROGRAM(dmc->dmc_info[dmc_id].start_mmioaddr, i), + dmc->dmc_info[dmc_id].payload[i]); + } + + preempt_enable(); + + dmc_load_mmio(display, dmc_id); +} + +static void assert_dmc_loaded(struct intel_display *display, + enum intel_dmc_id dmc_id) +{ + struct intel_dmc *dmc = display_to_dmc(display); + u32 expected, found; + int i; + + if (!is_valid_dmc_id(dmc_id) || !has_dmc_id_fw(display, dmc_id)) + return; + + found = intel_de_read(display, DMC_PROGRAM(dmc->dmc_info[dmc_id].start_mmioaddr, 0)); + expected = dmc->dmc_info[dmc_id].payload[0]; + + drm_WARN(display->drm, found != expected, + "DMC %d program storage start incorrect (expected 0x%x, current 0x%x)\n", + dmc_id, expected, found); + + for (i = 0; i < dmc->dmc_info[dmc_id].mmio_count; i++) { + i915_reg_t reg = dmc->dmc_info[dmc_id].mmioaddr[i]; + + found = intel_de_read(display, reg); + expected = dmc_mmiodata(display, dmc, dmc_id, i); + + /* once set DMC_EVT_CTL_ENABLE can't be cleared :/ */ + if (is_dmc_evt_ctl_reg(display, dmc_id, reg)) { + found &= ~DMC_EVT_CTL_ENABLE; + expected &= ~DMC_EVT_CTL_ENABLE; + } + + drm_WARN(display->drm, found != expected, + "DMC %d mmio[%d]/0x%x incorrect (expected 0x%x, current 0x%x)\n", + dmc_id, i, i915_mmio_reg_offset(reg), expected, found); + } +} + +void assert_main_dmc_loaded(struct intel_display *display) +{ + assert_dmc_loaded(display, DMC_FW_MAIN); +} + +static bool need_pipedmc_load_program(struct intel_display *display) +{ + /* On TGL/derivatives pipe DMC state is lost when PG1 is disabled */ + return DISPLAY_VER(display) == 12; +} + +static bool need_pipedmc_load_mmio(struct intel_display *display, enum pipe pipe) +{ + /* + * PTL: + * - pipe A/B DMC doesn't need save/restore + * - pipe C/D DMC is in PG0, needs manual save/restore + */ + if (DISPLAY_VER(display) == 30) + return pipe >= PIPE_C; + + /* + * FIXME LNL unclear, main DMC firmware has the pipe DMC A/B PG0 + * save/restore, but so far unable to see the loss of pipe DMC state + * in action. Are we just failing to turn off PG0 due to some other + * SoC level stuff? + */ + if (DISPLAY_VER(display) == 20) + return false; + + /* + * FIXME BMG untested, main DMC firmware has the + * pipe DMC A/B PG0 save/restore... + */ + if (display->platform.battlemage) + return false; + + /* + * DG2: + * - Pipe DMCs presumably in PG0? + * - No DC6, and even DC9 doesn't seem to result + * in loss of DMC state for whatever reason + */ + if (display->platform.dg2) + return false; + + /* + * ADL/MTL: + * - pipe A/B DMC is in PG0, saved/restored by the main DMC + * - pipe C/D DMC is in PG0, needs manual save/restore + */ + if (IS_DISPLAY_VER(display, 13, 14)) + return pipe >= PIPE_C; + + return false; +} + +static bool can_enable_pipedmc(const struct intel_crtc_state *crtc_state) +{ + struct intel_display *display = to_intel_display(crtc_state); + + /* + * On TGL/derivatives pipe DMC state is lost when PG1 is disabled. + * Do not even enable the pipe DMC when that can happen outside + * of driver control (PSR+DC5/6). + */ + if (DISPLAY_VER(display) == 12 && crtc_state->has_psr) + return false; + + return true; +} + +void intel_dmc_enable_pipe(const struct intel_crtc_state *crtc_state) +{ + struct intel_display *display = to_intel_display(crtc_state); + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); + enum pipe pipe = crtc->pipe; + enum intel_dmc_id dmc_id = PIPE_TO_DMC_ID(pipe); + + if (!is_valid_dmc_id(dmc_id) || !has_dmc_id_fw(display, dmc_id)) + return; + + if (!can_enable_pipedmc(crtc_state)) { + intel_dmc_disable_pipe(crtc_state); + return; + } + + if (need_pipedmc_load_program(display)) + dmc_load_program(display, dmc_id); + else if (need_pipedmc_load_mmio(display, pipe)) + dmc_load_mmio(display, dmc_id); + + assert_dmc_loaded(display, dmc_id); + + if (DISPLAY_VER(display) >= 20) { + intel_flipq_reset(display, pipe); + + intel_de_write(display, PIPEDMC_INTERRUPT(pipe), pipedmc_interrupt_mask(display)); + intel_de_write(display, PIPEDMC_INTERRUPT_MASK(pipe), ~pipedmc_interrupt_mask(display)); + } + + if (DISPLAY_VER(display) >= 14) + intel_de_rmw(display, MTL_PIPEDMC_CONTROL, 0, PIPEDMC_ENABLE_MTL(pipe)); + else + intel_de_rmw(display, PIPEDMC_CONTROL(pipe), 0, PIPEDMC_ENABLE); +} + +void intel_dmc_disable_pipe(const struct intel_crtc_state *crtc_state) +{ + struct intel_display *display = to_intel_display(crtc_state); + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); + enum pipe pipe = crtc->pipe; + enum intel_dmc_id dmc_id = PIPE_TO_DMC_ID(pipe); + + if (!is_valid_dmc_id(dmc_id) || !has_dmc_id_fw(display, dmc_id)) + return; + + if (DISPLAY_VER(display) >= 14) + intel_de_rmw(display, MTL_PIPEDMC_CONTROL, PIPEDMC_ENABLE_MTL(pipe), 0); + else + intel_de_rmw(display, PIPEDMC_CONTROL(pipe), PIPEDMC_ENABLE, 0); + + if (DISPLAY_VER(display) >= 20) { + intel_de_write(display, PIPEDMC_INTERRUPT_MASK(pipe), ~0); + intel_de_write(display, PIPEDMC_INTERRUPT(pipe), pipedmc_interrupt_mask(display)); + + intel_flipq_reset(display, pipe); + } +} + +static void dmc_configure_event(struct intel_display *display, + enum intel_dmc_id dmc_id, + unsigned int event_id, + bool enable) +{ + struct intel_dmc *dmc = display_to_dmc(display); + int num_handlers = 0; + int i; + + for (i = 0; i < dmc->dmc_info[dmc_id].mmio_count; i++) { + i915_reg_t reg = dmc->dmc_info[dmc_id].mmioaddr[i]; + u32 data = dmc->dmc_info[dmc_id].mmiodata[i]; + + if (!is_event_handler(display, dmc_id, event_id, reg, data)) + continue; + + intel_de_write(display, reg, enable ? data : dmc_evt_ctl_disable()); + num_handlers++; + } + + drm_WARN_ONCE(display->drm, num_handlers != 1, + "DMC %d has %d handlers for event 0x%x\n", + dmc_id, num_handlers, event_id); +} + +/** + * intel_dmc_block_pkgc() - block PKG C-state + * @display: display instance + * @pipe: pipe which register use to block + * @block: block/unblock + * + * This interface is target for Wa_16025596647 usage. I.e. to set/clear + * PIPEDMC_BLOCK_PKGC_SW_BLOCK_PKGC_ALWAYS bit in PIPEDMC_BLOCK_PKGC_SW register. + */ +void intel_dmc_block_pkgc(struct intel_display *display, enum pipe pipe, + bool block) +{ + intel_de_rmw(display, PIPEDMC_BLOCK_PKGC_SW(pipe), + PIPEDMC_BLOCK_PKGC_SW_BLOCK_PKGC_ALWAYS, block ? + PIPEDMC_BLOCK_PKGC_SW_BLOCK_PKGC_ALWAYS : 0); +} + +/** + * intel_dmc_start_pkgc_exit_at_start_of_undelayed_vblank() - start of PKG + * C-state exit + * @display: display instance + * @pipe: pipe which register use to block + * @enable: enable/disable + * + * This interface is target for Wa_16025596647 usage. I.e. start the package C + * exit at the start of the undelayed vblank + */ +void intel_dmc_start_pkgc_exit_at_start_of_undelayed_vblank(struct intel_display *display, + enum pipe pipe, bool enable) +{ + enum intel_dmc_id dmc_id = PIPE_TO_DMC_ID(pipe); + + dmc_configure_event(display, dmc_id, PIPEDMC_EVENT_VBLANK, enable); +} + /** * intel_dmc_load_program() - write the firmware from memory to register. * @display: display instance @@ -634,37 +849,26 @@ static u32 dmc_mmiodata(struct intel_display *display, void intel_dmc_load_program(struct intel_display *display) { struct i915_power_domains *power_domains = &display->power.domains; - struct intel_dmc *dmc = display_to_dmc(display); enum intel_dmc_id dmc_id; - u32 i; if (!intel_dmc_has_payload(display)) return; - pipedmc_clock_gating_wa(display, true); - - disable_all_event_handlers(display); - assert_display_rpm_held(display); - preempt_disable(); + pipedmc_clock_gating_wa(display, true); for_each_dmc_id(dmc_id) { - for (i = 0; i < dmc->dmc_info[dmc_id].dmc_fw_size; i++) { - intel_de_write_fw(display, - DMC_PROGRAM(dmc->dmc_info[dmc_id].start_mmioaddr, i), - dmc->dmc_info[dmc_id].payload[i]); - } + dmc_load_program(display, dmc_id); + assert_dmc_loaded(display, dmc_id); } - preempt_enable(); - - for_each_dmc_id(dmc_id) { - for (i = 0; i < dmc->dmc_info[dmc_id].mmio_count; i++) { - intel_de_write(display, dmc->dmc_info[dmc_id].mmioaddr[i], - dmc_mmiodata(display, dmc, dmc_id, i)); - } - } + if (DISPLAY_VER(display) >= 20) + intel_de_write(display, DMC_FQ_W2_PTS_CFG_SEL, + PIPE_D_DMC_W2_PTS_CONFIG_SELECT(PIPE_D) | + PIPE_C_DMC_W2_PTS_CONFIG_SELECT(PIPE_C) | + PIPE_B_DMC_W2_PTS_CONFIG_SELECT(PIPE_B) | + PIPE_A_DMC_W2_PTS_CONFIG_SELECT(PIPE_A)); power_domains->dc_state = 0; @@ -682,26 +886,17 @@ void intel_dmc_load_program(struct intel_display *display) */ void intel_dmc_disable_program(struct intel_display *display) { + enum intel_dmc_id dmc_id; + if (!intel_dmc_has_payload(display)) return; pipedmc_clock_gating_wa(display, true); - disable_all_event_handlers(display); - pipedmc_clock_gating_wa(display, false); -} -void assert_dmc_loaded(struct intel_display *display) -{ - struct intel_dmc *dmc = display_to_dmc(display); + for_each_dmc_id(dmc_id) + disable_all_event_handlers(display, dmc_id); - drm_WARN_ONCE(display->drm, !dmc, "DMC not initialized\n"); - drm_WARN_ONCE(display->drm, dmc && - !intel_de_read(display, DMC_PROGRAM(dmc->dmc_info[DMC_FW_MAIN].start_mmioaddr, 0)), - "DMC program storage start is NULL\n"); - drm_WARN_ONCE(display->drm, !intel_de_read(display, DMC_SSP_BASE), - "DMC SSP Base Not fine\n"); - drm_WARN_ONCE(display->drm, !intel_de_read(display, DMC_HTP_SKL), - "DMC HTP Not fine\n"); + pipedmc_clock_gating_wa(display, false); } static bool fw_info_matches_stepping(const struct intel_fw_info *fw_info, @@ -1120,7 +1315,6 @@ out: */ void intel_dmc_init(struct intel_display *display) { - struct drm_i915_private *i915 = to_i915(display->drm); struct intel_dmc *dmc; if (!HAS_DMC(display)) @@ -1163,7 +1357,7 @@ void intel_dmc_init(struct intel_display *display) display->dmc.dmc = dmc; drm_dbg_kms(display->drm, "Loading %s\n", dmc->fw_path); - queue_work(i915->unordered_wq, &dmc->work); + queue_work(display->wq.unordered, &dmc->work); return; @@ -1194,6 +1388,17 @@ void intel_dmc_suspend(struct intel_display *display) intel_dmc_runtime_pm_put(display); } +void intel_dmc_wait_fw_load(struct intel_display *display) +{ + struct intel_dmc *dmc = display_to_dmc(display); + + if (!HAS_DMC(display)) + return; + + if (dmc) + flush_work(&dmc->work); +} + /** * intel_dmc_resume() - init DMC firmware during system resume * @display: display instance @@ -1403,3 +1608,73 @@ void intel_dmc_debugfs_register(struct intel_display *display) debugfs_create_file("i915_dmc_info", 0444, minor->debugfs_root, display, &intel_dmc_debugfs_status_fops); } + +void intel_pipedmc_irq_handler(struct intel_display *display, enum pipe pipe) +{ + struct intel_crtc *crtc = intel_crtc_for_pipe(display, pipe); + u32 tmp = 0, int_vector; + + if (DISPLAY_VER(display) >= 20) { + tmp = intel_de_read(display, PIPEDMC_INTERRUPT(pipe)); + intel_de_write(display, PIPEDMC_INTERRUPT(pipe), tmp); + + if (tmp & PIPEDMC_FLIPQ_PROG_DONE) { + spin_lock(&display->drm->event_lock); + + if (crtc->flipq_event) { + /* + * Update vblank counter/timestamp in case it + * hasn't been done yet for this frame. + */ + drm_crtc_accurate_vblank_count(&crtc->base); + + drm_crtc_send_vblank_event(&crtc->base, crtc->flipq_event); + crtc->flipq_event = NULL; + } + + spin_unlock(&display->drm->event_lock); + } + + if (tmp & PIPEDMC_ATS_FAULT) + drm_err_ratelimited(display->drm, "[CRTC:%d:%s] PIPEDMC ATS fault\n", + crtc->base.base.id, crtc->base.name); + if (tmp & PIPEDMC_GTT_FAULT) + drm_err_ratelimited(display->drm, "[CRTC:%d:%s] PIPEDMC GTT fault\n", + crtc->base.base.id, crtc->base.name); + if (tmp & PIPEDMC_ERROR) + drm_err(display->drm, "[CRTC:%d:%s]] PIPEDMC error\n", + crtc->base.base.id, crtc->base.name); + } + + int_vector = intel_de_read(display, PIPEDMC_STATUS(pipe)) & PIPEDMC_INT_VECTOR_MASK; + if (tmp == 0 && int_vector != 0) + drm_err(display->drm, "[CRTC:%d:%s]] PIPEDMC interrupt vector 0x%x\n", + crtc->base.base.id, crtc->base.name, tmp); +} + +void intel_pipedmc_enable_event(struct intel_crtc *crtc, + enum pipedmc_event_id event) +{ + struct intel_display *display = to_intel_display(crtc); + enum intel_dmc_id dmc_id = PIPE_TO_DMC_ID(crtc->pipe); + + dmc_configure_event(display, dmc_id, event, true); +} + +void intel_pipedmc_disable_event(struct intel_crtc *crtc, + enum pipedmc_event_id event) +{ + struct intel_display *display = to_intel_display(crtc); + enum intel_dmc_id dmc_id = PIPE_TO_DMC_ID(crtc->pipe); + + dmc_configure_event(display, dmc_id, event, false); +} + +u32 intel_pipedmc_start_mmioaddr(struct intel_crtc *crtc) +{ + struct intel_display *display = to_intel_display(crtc); + struct intel_dmc *dmc = display_to_dmc(display); + enum intel_dmc_id dmc_id = PIPE_TO_DMC_ID(crtc->pipe); + + return dmc ? dmc->dmc_info[dmc_id].start_mmioaddr : 0; +} diff --git a/drivers/gpu/drm/i915/display/intel_dmc.h b/drivers/gpu/drm/i915/display/intel_dmc.h index bd1c459b00757b450af5df9ebaccbf7140da0d55..40e9dcb033cc6af1736c75a220de0a872f7673cb 100644 --- a/drivers/gpu/drm/i915/display/intel_dmc.h +++ b/drivers/gpu/drm/i915/display/intel_dmc.h @@ -9,15 +9,19 @@ #include enum pipe; +enum pipedmc_event_id; struct drm_printer; +struct intel_crtc; +struct intel_crtc_state; struct intel_display; struct intel_dmc_snapshot; void intel_dmc_init(struct intel_display *display); void intel_dmc_load_program(struct intel_display *display); +void intel_dmc_wait_fw_load(struct intel_display *display); void intel_dmc_disable_program(struct intel_display *display); -void intel_dmc_enable_pipe(struct intel_display *display, enum pipe pipe); -void intel_dmc_disable_pipe(struct intel_display *display, enum pipe pipe); +void intel_dmc_enable_pipe(const struct intel_crtc_state *crtc_state); +void intel_dmc_disable_pipe(const struct intel_crtc_state *crtc_state); void intel_dmc_block_pkgc(struct intel_display *display, enum pipe pipe, bool block); void intel_dmc_start_pkgc_exit_at_start_of_undelayed_vblank(struct intel_display *display, @@ -32,6 +36,16 @@ struct intel_dmc_snapshot *intel_dmc_snapshot_capture(struct intel_display *disp void intel_dmc_snapshot_print(const struct intel_dmc_snapshot *snapshot, struct drm_printer *p); void intel_dmc_update_dc6_allowed_count(struct intel_display *display, bool start_tracking); -void assert_dmc_loaded(struct intel_display *display); +void assert_main_dmc_loaded(struct intel_display *display); + +void intel_pipedmc_irq_handler(struct intel_display *display, enum pipe pipe); + +u32 intel_pipedmc_start_mmioaddr(struct intel_crtc *crtc); +void intel_pipedmc_enable_event(struct intel_crtc *crtc, + enum pipedmc_event_id event); +void intel_pipedmc_disable_event(struct intel_crtc *crtc, + enum pipedmc_event_id event); + +void intel_pipedmc_irq_handler(struct intel_display *display, enum pipe pipe); #endif /* __INTEL_DMC_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_dmc_regs.h b/drivers/gpu/drm/i915/display/intel_dmc_regs.h index e16ea3f16ed88915f34961b2888088efc85830a0..c5aa49921cb9ff1a642bb141481c91ba5e648c58 100644 --- a/drivers/gpu/drm/i915/display/intel_dmc_regs.h +++ b/drivers/gpu/drm/i915/display/intel_dmc_regs.h @@ -6,7 +6,273 @@ #ifndef __INTEL_DMC_REGS_H__ #define __INTEL_DMC_REGS_H__ -#include "i915_reg_defs.h" +#include "intel_display_reg_defs.h" + +enum dmc_event_id { + DMC_EVENT_TRUE = 0x0, + DMC_EVENT_FALSE = 0x1, +}; + +enum maindmc_event_id { + MAINDMC_EVENT_CMP_ZERO = 0x8, + MAINDMC_EVENT_CMP_ODD = 0x9, + MAINDMC_EVENT_CMP_NEG = 0xa, + MAINDMC_EVENT_CMP_CARRY = 0xb, + + MAINDMC_EVENT_TMR0_DONE = 0x14, + MAINDMC_EVENT_TMR1_DONE = 0x15, + MAINDMC_EVENT_TMR2_DONE = 0x16, + MAINDMC_EVENT_COUNT0_DONE = 0x17, + MAINDMC_EVENT_COUNT1_DONE = 0x18, + MAINDMC_EVENT_PERF_CNTR_DARBF = 0x19, + + MAINDMC_EVENT_SCANLINE_INRANGE_FQ_A_TRIGGER = 0x22, + MAINDMC_EVENT_SCANLINE_INRANGE_FQ_B_TRIGGER = 0x23, + MAINDMC_EVENT_SCANLINE_INRANGE_FQ_C_TRIGGER = 0x24, + MAINDMC_EVENT_SCANLINE_INRANGE_FQ_D_TRIGGER = 0x25, + MAINDMC_EVENT_1KHZ_FQ_A_TRIGGER = 0x26, + MAINDMC_EVENT_1KHZ_FQ_B_TRIGGER = 0x27, + MAINDMC_EVENT_1KHZ_FQ_C_TRIGGER = 0x28, + MAINDMC_EVENT_1KHZ_FQ_D_TRIGGER = 0x29, + MAINDMC_EVENT_SCANLINE_COMP_A = 0x2a, + MAINDMC_EVENT_SCANLINE_COMP_B = 0x2b, + MAINDMC_EVENT_SCANLINE_COMP_C = 0x2c, + MAINDMC_EVENT_SCANLINE_COMP_D = 0x2d, + MAINDMC_EVENT_VBLANK_DELAYED_A = 0x2e, + MAINDMC_EVENT_VBLANK_DELAYED_B = 0x2f, + MAINDMC_EVENT_VBLANK_DELAYED_C = 0x30, + MAINDMC_EVENT_VBLANK_DELAYED_D = 0x31, + MAINDMC_EVENT_VBLANK_A = 0x32, + MAINDMC_EVENT_VBLANK_B = 0x33, + MAINDMC_EVENT_VBLANK_C = 0x34, + MAINDMC_EVENT_VBLANK_D = 0x35, + MAINDMC_EVENT_HBLANK_A = 0x36, + MAINDMC_EVENT_HBLANK_B = 0x37, + MAINDMC_EVENT_HBLANK_C = 0x38, + MAINDMC_EVENT_HBLANK_D = 0x39, + MAINDMC_EVENT_VSYNC_A = 0x3a, + MAINDMC_EVENT_VSYNC_B = 0x3b, + MAINDMC_EVENT_VSYNC_C = 0x3c, + MAINDMC_EVENT_VSYNC_D = 0x3d, + MAINDMC_EVENT_SCANLINE_A = 0x3e, + MAINDMC_EVENT_SCANLINE_B = 0x3f, + MAINDMC_EVENT_SCANLINE_C = 0x40, + MAINDMC_EVENT_SCANLINE_D = 0x41, + + MAINDMC_EVENT_PLANE1_FLIP_A = 0x42, + MAINDMC_EVENT_PLANE2_FLIP_A = 0x43, + MAINDMC_EVENT_PLANE3_FLIP_A = 0x44, + MAINDMC_EVENT_PLANE4_FLIP_A = 0x45, + MAINDMC_EVENT_PLANE5_FLIP_A = 0x46, + MAINDMC_EVENT_PLANE6_FLIP_A = 0x47, + MAINDMC_EVENT_PLANE7_FLIP_A = 0x48, + MAINDMC_EVENT_PLANE1_FLIP_B = 0x49, + MAINDMC_EVENT_PLANE2_FLIP_B = 0x4a, + MAINDMC_EVENT_PLANE3_FLIP_B = 0x4b, + MAINDMC_EVENT_PLANE4_FLIP_B = 0x4c, + MAINDMC_EVENT_PLANE5_FLIP_B = 0x4d, + MAINDMC_EVENT_PLANE6_FLIP_B = 0x4e, + MAINDMC_EVENT_PLANE7_FLIP_B = 0x4f, + MAINDMC_EVENT_PLANE1_FLIP_C = 0x50, + MAINDMC_EVENT_PLANE2_FLIP_C = 0x51, + MAINDMC_EVENT_PLANE3_FLIP_C = 0x52, + MAINDMC_EVENT_PLANE4_FLIP_C = 0x53, + MAINDMC_EVENT_PLANE5_FLIP_C = 0x54, + MAINDMC_EVENT_PLANE6_FLIP_C = 0x55, + MAINDMC_EVENT_PLANE7_FLIP_C = 0x56, + MAINDMC_EVENT_PLANE1_FLIP_D = 0x57, + MAINDMC_EVENT_PLANE2_FLIP_D = 0x58, + MAINDMC_EVENT_PLANE3_FLIP_D = 0x59, + MAINDMC_EVENT_PLANE4_FLIP_D = 0x5a, + MAINDMC_EVENT_PLANE5_FLIP_D = 0x5b, + MAINDMC_EVENT_PLANE6_FLIP_D = 0x5c, + MAINDMC_EVENT_PLANE7_FLIP_D = 0x5d, + MAINDMC_EVENT_PLANE1_FLIP_DONE_A = 0x5e, + MAINDMC_EVENT_PLANE2_FLIP_DONE_A = 0x5f, + MAINDMC_EVENT_PLANE3_FLIP_DONE_A = 0x60, + MAINDMC_EVENT_PLANE4_FLIP_DONE_A = 0x61, + MAINDMC_EVENT_PLANE5_FLIP_DONE_A = 0x62, + MAINDMC_EVENT_PLANE6_FLIP_DONE_A = 0x63, + MAINDMC_EVENT_PLANE7_FLIP_DONE_A = 0x64, + MAINDMC_EVENT_PLANE1_FLIP_DONE_B = 0x65, + MAINDMC_EVENT_PLANE2_FLIP_DONE_B = 0x66, + MAINDMC_EVENT_PLANE3_FLIP_DONE_B = 0x67, + MAINDMC_EVENT_PLANE4_FLIP_DONE_B = 0x68, + MAINDMC_EVENT_PLANE5_FLIP_DONE_B = 0x69, + MAINDMC_EVENT_PLANE6_FLIP_DONE_B = 0x6a, + MAINDMC_EVENT_PLANE7_FLIP_DONE_B = 0x6b, + MAINDMC_EVENT_PLANE1_FLIP_DONE_C = 0x6c, + MAINDMC_EVENT_PLANE2_FLIP_DONE_C = 0x6d, + MAINDMC_EVENT_PLANE3_FLIP_DONE_C = 0x6e, + MAINDMC_EVENT_PLANE4_FLIP_DONE_C = 0x6f, + MAINDMC_EVENT_PLANE5_FLIP_DONE_C = 0x70, + MAINDMC_EVENT_PLANE6_FLIP_DONE_C = 0x71, + MAINDMC_EVENT_PLANE7_FLIP_DONE_C = 0x72, + MAINDMC_EVENT_PLANE1_FLIP_DONE_D = 0x73, + MAINDMC_EVENT_PLANE2_FLIP_DONE_D = 0x74, + MAINDMC_EVENT_PLANE3_FLIP_DONE_D = 0x75, + MAINDMC_EVENT_PLANE4_FLIP_DONE_D = 0x76, + MAINDMC_EVENT_PLANE5_FLIP_DONE_D = 0x77, + MAINDMC_EVENT_PLANE6_FLIP_DONE_D = 0x78, + MAINDMC_EVENT_PLANE7_FLIP_DONE_D = 0x79, + + MAINDMC_EVENT_WIDI_GTT_FAULT_SL1 = 0x7d, + MAINDMC_EVENT_WIDI_GTT_FAULT_SL2 = 0x7e, + MAINDMC_EVENT_WIDI_CAP_ACTIVE_SL1 = 0x7f, + MAINDMC_EVENT_WIDI_CAP_ACTIVE_SL2 = 0x80, + + MAINDMC_EVENT_RENUKE_A = 0x85, + MAINDMC_EVENT_RENUKE_B = 0x86, + MAINDMC_EVENT_RENUKE_C = 0x87, + MAINDMC_EVENT_RENUKE_D = 0x88, + MAINDMC_EVENT_DPFC_FIFO_FULL_A = 0x89, + MAINDMC_EVENT_DPFC_FIFO_FULL_B = 0x8a, + MAINDMC_EVENT_DPFC_FIFO_FULL_C = 0x8b, + MAINDMC_EVENT_DPFC_FIFO_FULL_D = 0x8c, + MAINDMC_EVENT_DPFC_PIXEL_CNT_MISMATCH_A = 0x8d, + MAINDMC_EVENT_DPFC_PIXEL_CNT_MISMATCH_B = 0x8e, + MAINDMC_EVENT_DPFC_PIXEL_CNT_MISMATCH_C = 0x8f, + MAINDMC_EVENT_DPFC_PIXEL_CNT_MISMATCH_D = 0x90, + MAINDMC_EVENT_DPFC_COMPTAG_UNDERRUN_A = 0x91, + MAINDMC_EVENT_DPFC_COMPTAG_UNDERRUN_B = 0x92, + MAINDMC_EVENT_DPFC_COMPTAG_UNDERRUN_C = 0x93, + MAINDMC_EVENT_DPFC_COMPTAG_UNDERRUN_D = 0x94, + MAINDMC_EVENT_DPFC_FIFO_NOT_EMPTY_A = 0x95, + MAINDMC_EVENT_DPFC_FIFO_NOT_EMPTY_B = 0x96, + MAINDMC_EVENT_DPFC_FIFO_NOT_EMPTY_C = 0x97, + MAINDMC_EVENT_DPFC_FIFO_NOT_EMPTY_D = 0x98, + MAINDMC_EVENT_DPFC_COMPTAG_MISMATCH_A = 0x99, + MAINDMC_EVENT_DPFC_COMPTAG_MISMATCH_B = 0x9a, + MAINDMC_EVENT_DPFC_COMPTAG_MISMATCH_C = 0x9b, + MAINDMC_EVENT_DPFC_COMPTAG_MISMATCH_D = 0x9c, + MAINDMC_EVENT_DISP_PCH_INT = 0x9d, + MAINDMC_EVENT_GTT_ERR = 0x9e, + MAINDMC_EVENT_VTD_ERR = 0x9f, + MAINDMC_EVENT_FULL_FQ_WAKE_TRIGGER_A = 0xa0, + MAINDMC_EVENT_FULL_FQ_WAKE_TRIGGER_B = 0xa1, + MAINDMC_EVENT_FULL_FQ_WAKE_TRIGGER_C = 0xa2, + MAINDMC_EVENT_FULL_FQ_WAKE_TRIGGER_D = 0xa3, + MAINDMC_EVENT_PIPEDMC_CHICKEN_FW_EVENT_A = 0xa4, + MAINDMC_EVENT_PIPEDMC_CHICKEN_FW_EVENT_B = 0xa5, + MAINDMC_EVENT_PIPEDMC_CHICKEN_FW_EVENT_C = 0xa6, + MAINDMC_EVENT_PIPEDMC_CHICKEN_FW_EVENT_D = 0xa7, + + MAINDMC_EVENT_DC_CLOCK_OFF_START_EDP = 0xb2, + MAINDMC_EVENT_DC_CLOCK_OFF_START_DSI = 0xb3, + MAINDMC_EVENT_DCPR_DMC_CSR_START = 0xb4, + MAINDMC_EVENT_IN_PSR = 0xb5, + + MAINDMC_EVENT_IN_MEMUP = 0xb7, + MAINDMC_EVENT_IN_VGA = 0xb8, + + MAINDMC_EVENT_IN_KVM_SESSION = 0xba, + MAINDMC_EVENT_DEWAKE = 0xbb, + + MAINDMC_EVENT_TRAP_HIT = 0xbd, + MAINDMC_EVENT_CLK_USEC = 0xbe, + MAINDMC_EVENT_CLK_MSEC = 0xbf, + + MAINDMC_EVENT_CHICKEN1 = 0xc8, + MAINDMC_EVENT_CHICKEN2 = 0xc9, + MAINDMC_EVENT_CHICKEN3 = 0xca, + MAINDMC_EVENT_DDT_UBP = 0xcb, + + MAINDMC_EVENT_HP_LATENCY = 0xcd, + MAINDMC_EVENT_LP_LATENCY = 0xce, + MAINDMC_EVENT_WIDI_LP_REQ_SL1 = 0xcf, + MAINDMC_EVENT_WIDI_LP_REQ_SL2 = 0xd0, + + MAINDMC_EVENT_DG_DMC_EVT_0 = 0xd3, + MAINDMC_EVENT_DG_DMC_EVT_1 = 0xd4, + MAINDMC_EVENT_DG_DMC_EVT_2 = 0xd5, + MAINDMC_EVENT_DG_DMC_EVT_3 = 0xd6, + MAINDMC_EVENT_DG_DMC_EVT_4 = 0xd7, + MAINDMC_EVENT_DACFE_CLK_STOP = 0xd8, + MAINDMC_EVENT_DACFE_AZILIA_SDI_WAKE = 0xd9, + MAINDMC_EVENT_AUDIO_DOUBLE_FUNC_GRP_RST = 0xda, + MAINDMC_EVENT_AUDIO_CMD_VALID = 0xdb, + MAINDMC_EVENT_AUDIO_FRM_SYNC_BCLK = 0xdc, + MAINDMC_EVENT_AUDIO_FRM_SYNC_CDCLK = 0xdd, + MAINDMC_EVENT_AUDIO_PRESENCE_DETECT_A = 0xde, + MAINDMC_EVENT_AUDIO_PRESENCE_DETECT_B = 0xdf, + MAINDMC_EVENT_AUDIO_PRESENCE_DETECT_C = 0xe0, + MAINDMC_EVENT_AUDIO_PRESENCE_DETECT_E = 0xe1, + MAINDMC_EVENT_CMTG_SCANLINE_IN_GB_DC6v = 0xe2, + MAINDMC_EVENT_DCPR_CMTG_SCANLINE_OUTSIDE_GB = 0xe3, + MAINDMC_EVENT_DC6v_BACKWARD_COMPAT = 0xe4, + MAINDMC_EVENT_DPMA_PM_ABORT = 0xe5, + + MAINDMC_EVENT_STACK_OVF = 0xfc, + MAINDMC_EVENT_NO_CLAIM = 0xfd, + MAINDMC_EVENT_UNK_CMD = 0xfe, + MAINDMC_EVENT_HTP_MOD = 0xff, +}; + +enum pipedmc_event_id { + PIPEDMC_EVENT_TMR0_DONE = 0x14, + PIPEDMC_EVENT_TMR1_DONE = 0x15, + PIPEDMC_EVENT_TMR2_DONE = 0x16, + PIPEDMC_EVENT_COUNT0_DONE = 0x17, + PIPEDMC_EVENT_COUNT1_DONE = 0x18, + PIPEDMC_EVENT_PGA_PGB_RESTORE_DONE = 0x19, + PIPEDMC_EVENT_PG1_PG2_RESTORE_DONE = 0x1a, + PIPEDMC_EVENT_PGA_PGB_SAVE_DONE = 0x1b, + PIPEDMC_EVENT_PG1_PG2_SAVE_DONE = 0x1c, + + PIPEDMC_EVENT_FULL_FQ_WAKE_TRIGGER = 0x2b, + PIPEDMC_EVENT_1KHZ_FQ_TRIGGER = 0x2c, + PIPEDMC_EVENT_SCANLINE_INRANGE_FQ_TRIGGER = 0x2d, + PIPEDMC_EVENT_SCANLINE_INRANGE = 0x2e, + PIPEDMC_EVENT_SCANLINE_OUTRANGE = 0x2f, + PIPEDMC_EVENT_SCANLINE_EQUAL = 0x30, + PIPEDMC_EVENT_DELAYED_VBLANK = 0x31, + PIPEDMC_EVENT_VBLANK = 0x32, + PIPEDMC_EVENT_HBLANK = 0x33, + PIPEDMC_EVENT_VSYNC = 0x34, + PIPEDMC_EVENT_SCANLINE_FROM_DMUX = 0x35, + PIPEDMC_EVENT_PLANE1_FLIP = 0x36, + PIPEDMC_EVENT_PLANE2_FLIP = 0x37, + PIPEDMC_EVENT_PLANE3_FLIP = 0x38, + PIPEDMC_EVENT_PLANE4_FLIP = 0x39, + PIPEDMC_EVENT_PLANE5_FLIP = 0x3a, + PIPEDMC_EVENT_PLANE6_FLIP = 0x3b, + PIPEDMC_EVENT_PLANE7_FLIP = 0x3c, + PIPEDMC_EVENT_ADAPTIVE_DCB_TRIGGER = 0x3d, + + PIPEDMC_EVENT_PLANE1_FLIP_DONE = 0x56, + PIPEDMC_EVENT_PLANE2_FLIP_DONE = 0x57, + PIPEDMC_EVENT_PLANE3_FLIP_DONE = 0x58, + PIPEDMC_EVENT_PLANE4_FLIP_DONE = 0x59, + PIPEDMC_EVENT_PLANE5_FLIP_DONE = 0x5a, + PIPEDMC_EVENT_PLANE6_FLIP_DONE = 0x5b, + PIPEDMC_EVENT_PLANE7_FLIP_DONE = 0x5c, + + PIPEDMC_EVENT_GTT_ERR = 0x9b, + + PIPEDMC_EVENT_IN_PSR = 0xb5, + PIPEDMC_EVENT_DSI_DMC_IDLE = 0xb6, + PIPEDMC_EVENT_PSR2_DMC_IDLE = 0xb7, + PIPEDMC_EVENT_IN_VGA = 0xb8, + + PIPEDMC_EVENT_TRAP_HIT = 0xbd, + PIPEDMC_EVENT_CLK_USEC = 0xbe, + PIPEDMC_EVENT_CLK_MSEC = 0xbf, + + PIPEDMC_EVENT_CHICKEN1 = 0xc8, + PIPEDMC_EVENT_CHICKEN2 = 0xc9, + PIPEDMC_EVENT_CHICKEN3 = 0xca, + PIPEDMC_EVENT_DDT_UBP = 0xcb, + + PIPEDMC_EVENT_LP_LATENCY = 0xce, + + PIPEDMC_EVENT_LACE_PART_A_HIST_TRIGGER = 0xdf, + PIPEDMC_EVENT_LACE_PART_B_HIST_TRIGGER = 0xe0, + + PIPEDMC_EVENT_STACK_OVF = 0xfc, + PIPEDMC_EVENT_NO_CLAIM = 0xfd, + PIPEDMC_EVENT_UNK_CMD = 0xfe, + PIPEDMC_EVENT_HTP_MOD = 0xff, +}; #define DMC_PROGRAM(addr, i) _MMIO((addr) + (i) * 4) #define DMC_SSP_BASE_ADDR_GEN9 0x00002FC0 @@ -21,11 +287,170 @@ #define MTL_PIPEDMC_CONTROL _MMIO(0x45250) #define PIPEDMC_ENABLE_MTL(pipe) REG_BIT(((pipe) - PIPE_A) * 4) -#define _MTL_PIPEDMC_EVT_CTL_4_A 0x5f044 -#define _MTL_PIPEDMC_EVT_CTL_4_B 0x5f444 -#define MTL_PIPEDMC_EVT_CTL_4(pipe) _MMIO_PIPE(pipe, \ - _MTL_PIPEDMC_EVT_CTL_4_A, \ - _MTL_PIPEDMC_EVT_CTL_4_B) +#define _PIPEDMC_LOAD_HTP_A 0x5f000 +#define _PIPEDMC_LOAD_HTP_B 0x5f400 +#define PIPEDMC_LOAD_HTP(pipe) _MMIO_PIPE((pipe), _PIPEDMC_LOAD_HTP_A, _PIPEDMC_LOAD_HTP_B) + +#define _PIPEDMC_CTL_A 0x5f064 +#define _PIPEDMC_CTL_B 0x5f464 +#define PIPEDMC_CTL(pipe) _MMIO_PIPE((pipe), _PIPEDMC_CTL_A, _PIPEDMC_CTL_B) +#define PIPEDMC_HALT REG_BIT(31) +#define PIPEDMC_STEP REG_BIT(27) +#define PIPEDMC_CLOCKGATE REG_BIT(23) + +#define _PIPEDMC_STATUS_A 0x5f06c +#define _PIPEDMC_STATUS_B 0x5f46c +#define PIPEDMC_STATUS(pipe) _MMIO_PIPE((pipe), _PIPEDMC_STATUS_A, _PIPEDMC_STATUS_B) +#define PIPEDMC_SSP REG_GENMASK(31, 16) +#define PIPEDMC_INT_VECTOR_MASK REG_GENMASK(15, 8) +/* PIPEDMC_INT_VECTOR values defined by firmware */ +#define PIPEDMC_INT_VECTOR_SCANLINE_COMP_ERROR REG_FIELD_PREP(PIPEDMC_INT_VECTOR_MASK, 0x1) +#define PIPEDMC_INT_VECTOR_DC6V_FLIPQ_OVERLAP_ERROR REG_FIELD_PREP(PIPEDMC_INT_VECTOR_MASK, 0x2) +#define PIPEDMC_INT_VECTOR_FLIPQ_PROG_DONE REG_FIELD_PREP(PIPEDMC_INT_VECTOR_MASK, 0xff) /* Wa_16018781658:lnl[a0] */ +#define PIPEDMC_EVT_PENDING REG_GENMASK(7, 0) + +#define _PIPEDMC_FQ_CTRL_A 0x5f078 +#define _PIPEDMC_FQ_CTRL_B 0x5f478 +#define PIPEDMC_FQ_CTRL(pipe) _MMIO_PIPE((pipe), _PIPEDMC_FQ_CTRL_A, _PIPEDMC_FQ_CTRL_B) +#define PIPEDMC_FQ_CTRL_ENABLE REG_BIT(31) +#define PIPEDMC_FQ_CTRL_ASYNC REG_BIT(29) +#define PIPEDMC_FQ_CTRL_PREEMPT REG_BIT(0) + +#define _PIPEDMC_FQ_STATUS_A 0x5f098 +#define _PIPEDMC_FQ_STATUS_B 0x5f498 +#define PIPEDMC_FQ_STATUS(pipe) _MMIO_PIPE((pipe), _PIPEDMC_FQ_STATUS_A, _PIPEDMC_FQ_STATUS_B) +#define PIPEDMC_FQ_STATUS_BUSY REG_BIT(31) +#define PIPEDMC_FQ_STATUS_W2_LIVE_STATUS REG_BIT(1) +#define PIPEDMC_FQ_STATUS_W1_LIVE_STATUS REG_BIT(0) + +#define _PIPEDMC_FPQ_ATOMIC_TP_A 0x5f0a0 +#define _PIPEDMC_FPQ_ATOMIC_TP_B 0x5f4a0 +#define PIPEDMC_FPQ_ATOMIC_TP(pipe) _MMIO_PIPE((pipe), _PIPEDMC_FPQ_ATOMIC_TP_A, _PIPEDMC_FPQ_ATOMIC_TP_B) +#define PIPEDMC_FPQ_PLANEQ_3_TP_MASK REG_GENMASK(31, 26) +#define PIPEDMC_FPQ_PLANEQ_3_TP(tail) REG_FIELD_PREP(PIPEDMC_FPQ_PLANEQ_3_TP_MASK, (tail)) +#define PIPEDMC_FPQ_PLANEQ_2_TP_MASK REG_GENMASK(24, 19) +#define PIPEDMC_FPQ_PLANEQ_2_TP(tail) REG_FIELD_PREP(PIPEDMC_FPQ_PLANEQ_2_TP_MASK, (tail)) +#define PIPEDMC_FPQ_PLANEQ_1_TP_MASK REG_GENMASK(17, 12) +#define PIPEDMC_FPQ_PLANEQ_1_TP(tail) REG_FIELD_PREP(PIPEDMC_FPQ_PLANEQ_1_TP_MASK, (tail)) +#define PIPEDMC_FPQ_FASTQ_TP_MASK REG_GENMASK(10, 6) +#define PIPEDMC_FPQ_FASTQ_TP(tail) REG_FIELD_PREP(PIPEDMC_FPQ_FASTQ_TP_MASK, (tail)) +#define PIPEDMC_FPQ_GENERALQ_TP_MASK REG_GENMASK(4, 0) +#define PIPEDMC_FPQ_GENERALQ_TP(tail) REG_FIELD_PREP(PIPEDMC_FPQ_GENERALQ_TP_MASK, (tail)) + +#define _PIPEDMC_FPQ_LINES_TO_W1_A 0x5f0a4 +#define _PIPEDMC_FPQ_LINES_TO_W1_B 0x5f4a4 +#define PIPEDMC_FPQ_LINES_TO_W1 _MMIO_PIPE((pipe), _PIPEDMC_FPQ_LINES_TO_W1_A, _PIPEDMC_FPQ_LINES_TO_W1_B) + +#define _PIPEDMC_FPQ_LINES_TO_W2_A 0x5f0a8 +#define _PIPEDMC_FPQ_LINES_TO_W2_B 0x5f4a8 +#define PIPEDMC_FPQ_LINES_TO_W2 _MMIO_PIPE((pipe), _PIPEDMC_FPQ_LINES_TO_W2_A, _PIPEDMC_FPQ_LINES_TO_W2_B) + +#define _PIPEDMC_SCANLINECMP_A 0x5f11c +#define _PIPEDMC_SCANLINECMP_B 0x5f51c +#define PIPEDMC_SCANLINECMP(pipe) _MMIO_PIPE((pipe), _PIPEDMC_SCANLINECMP_A, _PIPEDMC_SCANLINECMP_B) +#define PIPEDMC_SCANLINECMP_EN REG_BIT(31) +#define PIPEDMC_SCANLINE_NUMBER REG_GENMASK(20, 0) + +#define _PIPEDMC_SCANLINECMPLOWER_A 0x5f120 +#define _PIPEDMC_SCANLINECMPLOWER_B 0x5f520 +#define PIPEDMC_SCANLINECMPLOWER(pipe) _MMIO_PIPE((pipe), _PIPEDMC_SCANLINECMPLOWER_A, _PIPEDMC_SCANLINECMPLOWER_B) +#define PIPEDMC_SCANLINEINRANGECMP_EN REG_BIT(31) +#define PIPEDMC_SCANLINEOUTRANGECMP_EN REG_BIT(30) +#define PIPEDMC_SCANLINE_LOWER_MASK REG_GENMASK(20, 0) +#define PIPEDMC_SCANLINE_LOWER(scanline) REG_FIELD_PREP(PIPEDMC_SCANLINE_LOWER_MASK, (scanline)) + +#define _PIPEDMC_SCANLINECMPUPPER_A 0x5f124 +#define _PIPEDMC_SCANLINECMPUPPER_B 0x5f524 +#define PIPEDMC_SCANLINECMPUPPER(pipe) _MMIO_PIPE((pipe), _PIPEDMC_SCANLINECMPUPPER_A, _PIPEDMC_SCANLINECMPUPPER_B) +#define PIPEDMC_SCANLINE_UPPER_MASK REG_GENMASK(20, 0) +#define PIPEDMC_SCANLINE_UPPER(scanline) REG_FIELD_PREP(PIPEDMC_SCANLINE_UPPER_MASK, (scanline)) + +#define _MMIO_PIPEDMC_FPQ(pipe, fq_id, \ + reg_fpq1_a, reg_fpq2_a, reg_fpq3_a, reg_fpq4_a, \ + reg_fpq1_b, reg_fpq2_b, reg_fpq3_b, reg_fpq4_b) \ + _MMIO(_PICK_EVEN_2RANGES((fq_id), INTEL_FLIPQ_PLANE_3, \ + _PIPE((pipe), (reg_fpq1_a), (reg_fpq1_b)), \ + _PIPE((pipe), (reg_fpq2_a), (reg_fpq2_b)), \ + _PIPE((pipe), (reg_fpq3_a), (reg_fpq3_b)), \ + _PIPE((pipe), (reg_fpq4_a), (reg_fpq4_b)))) + +#define _PIPEDMC_FPQ1_HP_A 0x5f128 +#define _PIPEDMC_FPQ2_HP_A 0x5f138 +#define _PIPEDMC_FPQ3_HP_A 0x5f168 +#define _PIPEDMC_FPQ4_HP_A 0x5f174 +#define _PIPEDMC_FPQ5_HP_A 0x5f180 +#define _PIPEDMC_FPQ1_HP_B 0x5f528 +#define _PIPEDMC_FPQ2_HP_B 0x5f538 +#define _PIPEDMC_FPQ3_HP_B 0x5f568 +#define _PIPEDMC_FPQ4_HP_B 0x5f574 +#define _PIPEDMC_FPQ5_HP_B 0x5f580 +#define PIPEDMC_FPQ_HP(pipe, fq_id) _MMIO_PIPEDMC_FPQ((pipe), (fq_id), \ + _PIPEDMC_FPQ1_HP_A, _PIPEDMC_FPQ2_HP_A, \ + _PIPEDMC_FPQ3_HP_A, _PIPEDMC_FPQ4_HP_A, \ + _PIPEDMC_FPQ1_HP_B, _PIPEDMC_FPQ2_HP_B, \ + _PIPEDMC_FPQ3_HP_B, _PIPEDMC_FPQ4_HP_B) + +#define _PIPEDMC_FPQ1_TP_A 0x5f12c +#define _PIPEDMC_FPQ2_TP_A 0x5f13c +#define _PIPEDMC_FPQ3_TP_A 0x5f16c +#define _PIPEDMC_FPQ4_TP_A 0x5f178 +#define _PIPEDMC_FPQ5_TP_A 0x5f184 +#define _PIPEDMC_FPQ1_TP_B 0x5f52c +#define _PIPEDMC_FPQ2_TP_B 0x5f53c +#define _PIPEDMC_FPQ3_TP_B 0x5f56c +#define _PIPEDMC_FPQ4_TP_B 0x5f578 +#define _PIPEDMC_FPQ5_TP_B 0x5f584 +#define PIPEDMC_FPQ_TP(pipe, fq_id) _MMIO_PIPEDMC_FPQ((pipe), (fq_id), \ + _PIPEDMC_FPQ1_TP_A, _PIPEDMC_FPQ2_TP_A, \ + _PIPEDMC_FPQ3_TP_A, _PIPEDMC_FPQ4_TP_A, \ + _PIPEDMC_FPQ1_TP_B, _PIPEDMC_FPQ2_TP_B, \ + _PIPEDMC_FPQ3_TP_B, _PIPEDMC_FPQ4_TP_B) + +#define _PIPEDMC_FPQ1_CHP_A 0x5f130 +#define _PIPEDMC_FPQ2_CHP_A 0x5f140 +#define _PIPEDMC_FPQ3_CHP_A 0x5f170 +#define _PIPEDMC_FPQ4_CHP_A 0x5f17c +#define _PIPEDMC_FPQ5_CHP_A 0x5f188 +#define _PIPEDMC_FPQ1_CHP_B 0x5f530 +#define _PIPEDMC_FPQ2_CHP_B 0x5f540 +#define _PIPEDMC_FPQ3_CHP_B 0x5f570 +#define _PIPEDMC_FPQ4_CHP_B 0x5f57c +#define _PIPEDMC_FPQ5_CHP_B 0x5f588 +#define PIPEDMC_FPQ_CHP(pipe, fq_id) _MMIO_PIPEDMC_FPQ((pipe), (fq_id), \ + _PIPEDMC_FPQ1_CHP_A, _PIPEDMC_FPQ2_CHP_A, \ + _PIPEDMC_FPQ3_CHP_A, _PIPEDMC_FPQ4_CHP_A, \ + _PIPEDMC_FPQ1_CHP_B, _PIPEDMC_FPQ2_CHP_B, \ + _PIPEDMC_FPQ3_CHP_B, _PIPEDMC_FPQ4_CHP_B) + +#define _PIPEDMC_FPQ_TS_A 0x5f134 +#define _PIPEDMC_FPQ_TS_B 0x5f534 +#define PIPEDMC_FPQ_TS(pipe) _MMIO_PIPE((pipe), _PIPEDMC_FPQ_TS_A, _PIPEDMC_FPQ_TS_B) + +#define _PIPEDMC_SCANLINE_RO_A 0x5f144 +#define _PIPEDMC_SCANLINE_RO_B 0x5f544 +#define PIPEDMC_SCANLINE_RO(pipe) _MMIO_PIPE((pipe), _PIPEDMC_SCANLINE_RO_A, _PIPEDMC_SCANLINE_RO_B) + +#define _PIPEDMC_FPQ_CTL1_A 0x5f160 +#define _PIPEDMC_FPQ_CTL1_B 0x5f560 +#define PIPEDMC_FPQ_CTL1(pipe) _MMIO_PIPE((pipe), _PIPEDMC_FPQ_CTL1_A, _PIPEDMC_FPQ_CTL1_B) +#define PIPEDMC_SW_DMC_WAKE REG_BIT(0) + +#define _PIPEDMC_FPQ_CTL2_A 0x5f164 +#define _PIPEDMC_FPQ_CTL2_B 0x5f564 +#define PIPEDMC_FPQ_CTL2(pipe) _MMIO_PIPE((pipe), _PIPEDMC_FPQ_CTL2_A, _PIPEDMC_FPQ_CTL2_B) +#define PIPEDMC_DMC_INT_AT_DELAYED_VBLANK REG_BIT(1) +#define PIPEDMC_W1_DMC_WAKE REG_BIT(0) + +#define _PIPEDMC_INTERRUPT_A 0x5f190 /* lnl+ */ +#define _PIPEDMC_INTERRUPT_B 0x5f590 /* lnl+ */ +#define PIPEDMC_INTERRUPT(pipe) _MMIO_PIPE((pipe), _PIPEDMC_INTERRUPT_A, _PIPEDMC_INTERRUPT_B) +#define _PIPEDMC_INTERRUPT_MASK_A 0x5f194 /* lnl+ */ +#define _PIPEDMC_INTERRUPT_MASK_B 0x5f594 /* lnl+ */ +#define PIPEDMC_INTERRUPT_MASK(pipe) _MMIO_PIPE((pipe), _PIPEDMC_INTERRUPT_MASK_A, _PIPEDMC_INTERRUPT_MASK_B) +#define PIPEDMC_FLIPQ_PROG_DONE REG_BIT(3) +#define PIPEDMC_ERROR REG_BIT(2) +#define PIPEDMC_GTT_FAULT REG_BIT(1) +#define PIPEDMC_ATS_FAULT REG_BIT(0) #define PIPEDMC_BLOCK_PKGC_SW_A 0x5f1d0 #define PIPEDMC_BLOCK_PKGC_SW_B 0x5F5d0 @@ -71,12 +496,7 @@ #define DMC_EVT_CTL_TYPE_LEVEL_1 1 #define DMC_EVT_CTL_TYPE_EDGE_1_0 2 #define DMC_EVT_CTL_TYPE_EDGE_0_1 3 - #define DMC_EVT_CTL_EVENT_ID_MASK REG_GENMASK(15, 8) -#define DMC_EVT_CTL_EVENT_ID_FALSE 0x01 -#define DMC_EVT_CTL_EVENT_ID_VBLANK_A 0x32 /* main DMC */ -/* An event handler scheduled to run at a 1 kHz frequency. */ -#define DMC_EVT_CTL_EVENT_ID_CLK_MSEC 0xbf #define DMC_HTP_ADDR_SKL 0x00500034 #define DMC_SSP_BASE _MMIO(0x8F074) @@ -117,4 +537,51 @@ #define DMC_WAKELOCK_CTL_REQ REG_BIT(31) #define DMC_WAKELOCK_CTL_ACK REG_BIT(15) +#define DMC_FQ_W2_PTS_CFG_SEL _MMIO(0x8f240) +#define PIPE_D_DMC_W2_PTS_CONFIG_SELECT_MASK REG_GENMASK(26, 24) +#define PIPE_D_DMC_W2_PTS_CONFIG_SELECT(pipe) REG_FIELD_PREP(PIPE_D_DMC_W2_PTS_CONFIG_SELECT_MASK, (pipe)) +#define PIPE_C_DMC_W2_PTS_CONFIG_SELECT_MASK REG_GENMASK(18, 16) +#define PIPE_C_DMC_W2_PTS_CONFIG_SELECT(pipe) REG_FIELD_PREP(PIPE_C_DMC_W2_PTS_CONFIG_SELECT_MASK, (pipe)) +#define PIPE_B_DMC_W2_PTS_CONFIG_SELECT_MASK REG_GENMASK(10, 8) +#define PIPE_B_DMC_W2_PTS_CONFIG_SELECT(pipe) REG_FIELD_PREP(PIPE_B_DMC_W2_PTS_CONFIG_SELECT_MASK, (pipe)) +#define PIPE_A_DMC_W2_PTS_CONFIG_SELECT_MASK REG_GENMASK(2, 0) +#define PIPE_A_DMC_W2_PTS_CONFIG_SELECT(pipe) REG_FIELD_PREP(PIPE_A_DMC_W2_PTS_CONFIG_SELECT_MASK, (pipe)) + +/* plane/general flip queue entries */ +#define PIPEDMC_FQ_RAM(start_mmioaddr, i) _MMIO((start_mmioaddr) + (i) * 4) +/* LNL */ +/* DW0 pts */ +/* DW1 head */ +/* DW2 size/etc. */ +#define LNL_FQ_INTERRUPT REG_BIT(31) +#define LNL_FQ_DSB_ID_MASK REG_GENMASK(30, 29) +#define LNL_FQ_DSB_ID(dsb_id) REG_FIELD_PREP(LNL_FQ_DSB_ID_MASK, (dsb_id)) +#define LNL_FQ_EXECUTED REG_BIT(28) +#define LNL_FQ_DSB_SIZE_MASK REG_GENMASK(15, 0) +#define LNL_FQ_DSB_SIZE(size) REG_FIELD_PREP(LNL_FQ_DSB_SIZE_MASK, (size)) +/* DW3 reserved (plane queues) */ +/* DW3 second DSB head (general queue) */ +/* DW4 second DSB size/etc. (general queue) */ +/* DW5 reserved (general queue) */ + +/* PTL+ */ +/* DW0 pts */ +/* DW1 reserved */ +/* DW2 size/etc. */ +#define PTL_FQ_INTERRUPT REG_BIT(31) +#define PTL_FQ_NEED_PUSH REG_BIT(30) +#define PTL_FQ_BLOCK_PUSH REG_BIT(29) +#define PTL_FQ_EXECUTED REG_BIT(28) +#define PTL_FQ_DSB_ID_MASK REG_GENMASK(25, 24) +#define PTL_FQ_DSB_ID(dsb_id) REG_FIELD_PREP(PTL_FQ_DSB_ID_MASK, (dsb_id)) +#define PTL_FQ_DSB_SIZE_MASK REG_GENMASK(15, 0) +#define PTL_FQ_DSB_SIZE(size) REG_FIELD_PREP(PTL_FQ_DSB_SIZE_MASK, (size)) +/* DW3 head */ +/* DW4 second DSB size/etc. (general queue) */ +/* DW5 second DSB head (general queue) */ + +/* undocumented magic DMC variables */ +#define PTL_PIPEDMC_EXEC_TIME_LINES(start_mmioaddr) _MMIO((start_mmioaddr) + 0x6b8) +#define PTL_PIPEDMC_END_OF_EXEC_GB(start_mmioaddr) _MMIO((start_mmioaddr) + 0x6c0) + #endif /* __INTEL_DMC_REGS_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_dmc_wl.c b/drivers/gpu/drm/i915/display/intel_dmc_wl.c index 7e2ce0c2f6c391c063237b28c092254769a09533..b3bb89ba34f990ca916b1908071ba8647c4bf569 100644 --- a/drivers/gpu/drm/i915/display/intel_dmc_wl.c +++ b/drivers/gpu/drm/i915/display/intel_dmc_wl.c @@ -7,9 +7,8 @@ #include -#include "i915_drv.h" -#include "i915_reg.h" #include "intel_de.h" +#include "intel_display_regs.h" #include "intel_dmc_regs.h" #include "intel_dmc_wl.h" @@ -155,12 +154,11 @@ static const struct intel_dmc_wl_range xe3lpd_dc3co_dmc_ranges[] = { static void __intel_dmc_wl_release(struct intel_display *display) { - struct drm_i915_private *i915 = to_i915(display->drm); struct intel_dmc_wl *wl = &display->wl; WARN_ON(refcount_read(&wl->refcount)); - queue_delayed_work(i915->unordered_wq, &wl->work, + queue_delayed_work(display->wq.unordered, &wl->work, msecs_to_jiffies(DMC_WAKELOCK_HOLD_TIME)); } diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c index 640c43bf62d4c5d51b1b76a1ca848d4564c90d85..f48912f308df7fd26c9d089e8f1bd096bfc8df95 100644 --- a/drivers/gpu/drm/i915/display/intel_dp.c +++ b/drivers/gpu/drm/i915/display/intel_dp.c @@ -27,6 +27,8 @@ #include #include +#include +#include #include #include #include @@ -34,7 +36,6 @@ #include #include #include - #include #include @@ -49,8 +50,6 @@ #include #include "g4x_dp.h" -#include "i915_irq.h" -#include "i915_reg.h" #include "i915_utils.h" #include "intel_alpm.h" #include "intel_atomic.h" @@ -64,6 +63,7 @@ #include "intel_ddi.h" #include "intel_de.h" #include "intel_display_driver.h" +#include "intel_display_regs.h" #include "intel_display_rpm.h" #include "intel_display_types.h" #include "intel_dp.h" @@ -847,7 +847,7 @@ small_joiner_ram_size_bits(struct intel_display *display) return 6144 * 8; } -u32 intel_dp_dsc_nearest_valid_bpp(struct intel_display *display, u32 bpp, u32 pipe_bpp) +static u32 intel_dp_dsc_nearest_valid_bpp(struct intel_display *display, u32 bpp, u32 pipe_bpp) { u32 bits_per_pixel = bpp; int i; @@ -939,6 +939,7 @@ static u32 ultrajoiner_ram_max_bpp(u32 mode_hdisplay) return ultrajoiner_ram_bits() / mode_hdisplay; } +/* TODO: return a bpp_x16 value */ static u32 get_max_compressed_bpp_with_joiner(struct intel_display *display, u32 mode_clock, u32 mode_hdisplay, @@ -955,6 +956,7 @@ u32 get_max_compressed_bpp_with_joiner(struct intel_display *display, return max_bpp; } +/* TODO: return a bpp_x16 value */ u16 intel_dp_dsc_get_max_compressed_bpp(struct intel_display *display, u32 link_clock, u32 lane_count, u32 mode_clock, u32 mode_hdisplay, @@ -1195,7 +1197,7 @@ intel_dp_output_format(struct intel_connector *connector, int intel_dp_min_bpp(enum intel_output_format output_format) { if (output_format == INTEL_OUTPUT_FORMAT_RGB) - return 6 * 3; + return intel_display_min_pipe_bpp(); else return 8 * 3; } @@ -2067,7 +2069,7 @@ int intel_dp_dsc_sink_max_compressed_bpp(const struct intel_connector *connector pipe_config, bpc) >> 4; } -static int dsc_src_min_compressed_bpp(void) +int intel_dp_dsc_min_src_compressed_bpp(void) { /* Min Compressed bpp supported by source is 8 */ return 8; @@ -2099,7 +2101,7 @@ static int dsc_src_max_compressed_bpp(struct intel_dp *intel_dp) /* * Note: for pre-13 display you still need to check the validity of each step. */ -static int intel_dp_dsc_bpp_step_x16(const struct intel_connector *connector) +int intel_dp_dsc_bpp_step_x16(const struct intel_connector *connector) { struct intel_display *display = to_intel_display(connector); u8 incr = drm_dp_dsc_sink_bpp_incr(connector->dp.dsc_dpcd); @@ -2107,12 +2109,19 @@ static int intel_dp_dsc_bpp_step_x16(const struct intel_connector *connector) if (DISPLAY_VER(display) < 14 || !incr) return fxp_q4_from_int(1); + if (connector->mst.dp && + !connector->link.force_bpp_x16 && !connector->mst.dp->force_dsc_fractional_bpp_en) + return fxp_q4_from_int(1); + /* fxp q4 */ return fxp_q4_from_int(1) / incr; } -/* Note: This is not universally usable! */ -static bool intel_dp_dsc_valid_bpp(struct intel_dp *intel_dp, int bpp_x16) +/* + * Note: for bpp_x16 to be valid it must be also within the source/sink's + * min..max bpp capability range. + */ +bool intel_dp_dsc_valid_compressed_bpp(struct intel_dp *intel_dp, int bpp_x16) { struct intel_display *display = to_intel_display(intel_dp); int i; @@ -2150,24 +2159,16 @@ static int dsc_compute_compressed_bpp(struct intel_dp *intel_dp, const struct intel_connector *connector = to_intel_connector(conn_state->connector); const struct drm_display_mode *adjusted_mode = &pipe_config->hw.adjusted_mode; int output_bpp; - int dsc_min_bpp; - int dsc_max_bpp; int min_bpp_x16, max_bpp_x16, bpp_step_x16; int dsc_joiner_max_bpp; int num_joined_pipes = intel_crtc_num_joined_pipes(pipe_config); int bpp_x16; int ret; - dsc_min_bpp = fxp_q4_to_int_roundup(limits->link.min_bpp_x16); - dsc_joiner_max_bpp = get_max_compressed_bpp_with_joiner(display, adjusted_mode->clock, adjusted_mode->hdisplay, num_joined_pipes); - dsc_max_bpp = min(dsc_joiner_max_bpp, fxp_q4_to_int(limits->link.max_bpp_x16)); - - /* FIXME: remove the round trip via integers */ - min_bpp_x16 = fxp_q4_from_int(dsc_min_bpp); - max_bpp_x16 = fxp_q4_from_int(dsc_max_bpp); + max_bpp_x16 = min(fxp_q4_from_int(dsc_joiner_max_bpp), limits->link.max_bpp_x16); bpp_step_x16 = intel_dp_dsc_bpp_step_x16(connector); @@ -2175,8 +2176,12 @@ static int dsc_compute_compressed_bpp(struct intel_dp *intel_dp, output_bpp = intel_dp_output_bpp(pipe_config->output_format, pipe_bpp); max_bpp_x16 = min(max_bpp_x16, fxp_q4_from_int(output_bpp) - bpp_step_x16); + drm_WARN_ON(display->drm, !is_power_of_2(bpp_step_x16)); + min_bpp_x16 = round_up(limits->link.min_bpp_x16, bpp_step_x16); + max_bpp_x16 = round_down(max_bpp_x16, bpp_step_x16); + for (bpp_x16 = max_bpp_x16; bpp_x16 >= min_bpp_x16; bpp_x16 -= bpp_step_x16) { - if (!intel_dp_dsc_valid_bpp(intel_dp, bpp_x16)) + if (!intel_dp_dsc_valid_compressed_bpp(intel_dp, bpp_x16)) continue; ret = dsc_compute_link_config(intel_dp, @@ -2479,7 +2484,7 @@ intel_dp_compute_config_link_bpp_limits(struct intel_dp *intel_dp, int dsc_src_min_bpp, dsc_sink_min_bpp, dsc_min_bpp; int dsc_src_max_bpp, dsc_sink_max_bpp, dsc_max_bpp; - dsc_src_min_bpp = dsc_src_min_compressed_bpp(); + dsc_src_min_bpp = intel_dp_dsc_min_src_compressed_bpp(); dsc_sink_min_bpp = intel_dp_dsc_sink_min_compressed_bpp(crtc_state); dsc_min_bpp = max(dsc_src_min_bpp, dsc_sink_min_bpp); limits->link.min_bpp_x16 = fxp_q4_from_int(dsc_min_bpp); @@ -3721,6 +3726,9 @@ static void intel_dp_get_pcon_dsc_cap(struct intel_dp *intel_dp) memset(intel_dp->pcon_dsc_dpcd, 0, sizeof(intel_dp->pcon_dsc_dpcd)); + if (!drm_dp_is_branch(intel_dp->dpcd)) + return; + if (drm_dp_dpcd_read(&intel_dp->aux, DP_PCON_DSC_ENCODER, intel_dp->pcon_dsc_dpcd, sizeof(intel_dp->pcon_dsc_dpcd)) < 0) @@ -5785,6 +5793,28 @@ intel_dp_detect_sdp_caps(struct intel_dp *intel_dp) drm_dp_as_sdp_supported(&intel_dp->aux, intel_dp->dpcd); } +static bool intel_dp_needs_dpcd_probe(struct intel_dp *intel_dp, bool force_on_external) +{ + struct intel_connector *connector = intel_dp->attached_connector; + + if (intel_dp_is_edp(intel_dp)) + return false; + + if (force_on_external) + return true; + + if (intel_dp->is_mst) + return false; + + return drm_edid_has_quirk(&connector->base, DRM_EDID_QUIRK_DP_DPCD_PROBE); +} + +void intel_dp_dpcd_set_probe(struct intel_dp *intel_dp, bool force_on_external) +{ + drm_dp_dpcd_set_probe(&intel_dp->aux, + intel_dp_needs_dpcd_probe(intel_dp, force_on_external)); +} + static int intel_dp_detect(struct drm_connector *_connector, struct drm_modeset_acquire_ctx *ctx, @@ -5913,6 +5943,8 @@ out_unset_edid: if (status != connector_status_connected && !intel_dp->is_mst) intel_dp_unset_edid(intel_dp); + intel_dp_dpcd_set_probe(intel_dp, false); + if (!intel_dp_is_edp(intel_dp)) drm_dp_set_subconnector_property(&connector->base, status, @@ -5943,6 +5975,8 @@ intel_dp_force(struct drm_connector *_connector) return; intel_dp_set_edid(intel_dp); + + intel_dp_dpcd_set_probe(intel_dp, false); } static int intel_dp_get_modes(struct drm_connector *_connector) @@ -6315,10 +6349,11 @@ intel_dp_hpd_pulse(struct intel_digital_port *dig_port, bool long_hpd) * complete the DP tunnel BW request for the latter connector/encoder * waiting for this encoder's DPRX read, perform a dummy read here. */ - if (long_hpd) + if (long_hpd) { + intel_dp_dpcd_set_probe(intel_dp, true); + intel_dp_read_dprx_caps(intel_dp, dpcd); - if (long_hpd) { intel_dp->reset_link_params = true; intel_dp_invalidate_source_oui(intel_dp); diff --git a/drivers/gpu/drm/i915/display/intel_dp.h b/drivers/gpu/drm/i915/display/intel_dp.h index 742ae26ac4a9a9adef98856f55f5087e2b41c197..0657f568119662828344cfddbe876f2acf2596d9 100644 --- a/drivers/gpu/drm/i915/display/intel_dp.h +++ b/drivers/gpu/drm/i915/display/intel_dp.h @@ -147,6 +147,7 @@ int intel_dp_dsc_sink_min_compressed_bpp(const struct intel_crtc_state *pipe_con int intel_dp_dsc_sink_max_compressed_bpp(const struct intel_connector *connector, const struct intel_crtc_state *pipe_config, int bpc); +bool intel_dp_dsc_valid_compressed_bpp(struct intel_dp *intel_dp, int bpp_x16); u8 intel_dp_dsc_get_slice_count(const struct intel_connector *connector, int mode_clock, int mode_hdisplay, int num_joined_pipes); @@ -173,8 +174,6 @@ bool intel_dp_supports_dsc(struct intel_dp *intel_dp, const struct intel_connector *connector, const struct intel_crtc_state *crtc_state); -u32 intel_dp_dsc_nearest_valid_bpp(struct intel_display *display, u32 bpp, u32 pipe_bpp); - void intel_ddi_update_pipe(struct intel_atomic_state *state, struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state, @@ -209,7 +208,11 @@ bool intel_dp_has_connector(struct intel_dp *intel_dp, const struct drm_connector_state *conn_state); int intel_dp_dsc_max_src_input_bpc(struct intel_display *display); int intel_dp_dsc_min_src_input_bpc(void); +int intel_dp_dsc_min_src_compressed_bpp(void); int intel_dp_compute_min_hblank(struct intel_crtc_state *crtc_state, const struct drm_connector_state *conn_state); +int intel_dp_dsc_bpp_step_x16(const struct intel_connector *connector); +void intel_dp_dpcd_set_probe(struct intel_dp *intel_dp, bool force_on_external); + #endif /* __INTEL_DP_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_dp_aux.c b/drivers/gpu/drm/i915/display/intel_dp_aux.c index bf8e8e0cc19c9acf972adb21b6a017d124f429b7..829a7c0fbe4febaafe3ba8e026c7d161f3544c0f 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_aux.c +++ b/drivers/gpu/drm/i915/display/intel_dp_aux.c @@ -5,7 +5,6 @@ #include -#include "i915_reg.h" #include "i915_utils.h" #include "intel_de.h" #include "intel_display_types.h" @@ -835,6 +834,8 @@ void intel_dp_aux_init(struct intel_dp *intel_dp) intel_dp->aux.transfer = intel_dp_aux_transfer; cpu_latency_qos_add_request(&intel_dp->pm_qos, PM_QOS_DEFAULT_VALUE); + + intel_dp_dpcd_set_probe(intel_dp, true); } static enum aux_ch default_aux_ch(struct intel_encoder *encoder) diff --git a/drivers/gpu/drm/i915/display/intel_dp_hdcp.c b/drivers/gpu/drm/i915/display/intel_dp_hdcp.c index cc312596fb77b07eccfb209e724483b81c25a900..bd757db8592742d63d0c27654adb1cd6b690b5b7 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_hdcp.c +++ b/drivers/gpu/drm/i915/display/intel_dp_hdcp.c @@ -11,9 +11,9 @@ #include #include -#include "i915_reg.h" #include "intel_ddi.h" #include "intel_de.h" +#include "intel_display_regs.h" #include "intel_display_types.h" #include "intel_dp.h" #include "intel_dp_hdcp.h" @@ -805,10 +805,16 @@ intel_dp_mst_hdcp2_stream_encryption(struct intel_connector *connector, enum pipe pipe = (enum pipe)cpu_transcoder; enum port port = dig_port->base.port; int ret; - - drm_WARN_ON(display->drm, enable && - !!(intel_de_read(display, HDCP2_AUTH_STREAM(display, cpu_transcoder, port)) - & AUTH_STREAM_TYPE) != data->streams[0].stream_type); + u32 val; + u8 stream_type; + + if (DISPLAY_VER(display) < 30) { + val = intel_de_read(display, + HDCP2_AUTH_STREAM(display, cpu_transcoder, port)); + stream_type = REG_FIELD_GET(AUTH_STREAM_TYPE_MASK, val); + drm_WARN_ON(display->drm, enable && + stream_type != data->streams[0].stream_type); + } ret = intel_dp_mst_toggle_hdcp_stream_select(connector, enable); if (ret) @@ -824,6 +830,14 @@ intel_dp_mst_hdcp2_stream_encryption(struct intel_connector *connector, return -ETIMEDOUT; } + if (DISPLAY_VER(display) >= 30) { + val = intel_de_read(display, + HDCP2_STREAM_STATUS(display, cpu_transcoder, port)); + stream_type = REG_FIELD_GET(STREAM_TYPE_STATUS_MASK, val); + drm_WARN_ON(display->drm, enable && + stream_type != data->streams[0].stream_type); + } + return 0; } diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.c b/drivers/gpu/drm/i915/display/intel_dp_mst.c index e1501b13f08fb6946886d8bd47aeef341bac8358..74497c9a0554d1a173508fd1d413d6f0f8c22245 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_mst.c +++ b/drivers/gpu/drm/i915/display/intel_dp_mst.c @@ -23,6 +23,9 @@ * */ +#include +#include + #include #include #include @@ -30,7 +33,6 @@ #include #include -#include "i915_reg.h" #include "i915_utils.h" #include "intel_atomic.h" #include "intel_audio.h" @@ -39,6 +41,7 @@ #include "intel_ddi.h" #include "intel_de.h" #include "intel_display_driver.h" +#include "intel_display_regs.h" #include "intel_display_types.h" #include "intel_dp.h" #include "intel_dp_hdcp.h" @@ -135,6 +138,7 @@ static bool intel_dp_mst_inc_active_streams(struct intel_dp *intel_dp) return intel_dp->mst.active_streams++ == 0; } +/* TODO: return a bpp_x16 value */ static int intel_dp_mst_max_dpt_bpp(const struct intel_crtc_state *crtc_state, bool dsc) { @@ -241,6 +245,15 @@ static int intel_dp_mst_dsc_get_slice_count(const struct intel_connector *connec num_joined_pipes); } +static void mst_stream_update_slots(const struct intel_crtc_state *crtc_state, + struct drm_dp_mst_topology_state *topology_state) +{ + u8 link_coding_cap = intel_dp_is_uhbr(crtc_state) ? + DP_CAP_ANSI_128B132B : DP_CAP_ANSI_8B10B; + + drm_dp_mst_update_slots(topology_state, link_coding_cap); +} + int intel_dp_mtp_tu_compute_config(struct intel_dp *intel_dp, struct intel_crtc_state *crtc_state, struct drm_connector_state *conn_state, @@ -263,6 +276,12 @@ int intel_dp_mtp_tu_compute_config(struct intel_dp *intel_dp, fxp_q4_to_frac(max_bpp_x16) || fxp_q4_to_frac(bpp_step_x16))); + if (!bpp_step_x16) { + /* Allow using zero step only to indicate single try for a given bpp. */ + drm_WARN_ON(display->drm, min_bpp_x16 != max_bpp_x16); + bpp_step_x16 = 1; + } + if (is_mst) { mst_state = drm_atomic_get_mst_topology_state(state, &intel_dp->mst.mgr); if (IS_ERR(mst_state)) @@ -270,6 +289,8 @@ int intel_dp_mtp_tu_compute_config(struct intel_dp *intel_dp, mst_state->pbn_div = drm_dp_get_vc_payload_bw(crtc_state->port_clock, crtc_state->lane_count); + + mst_stream_update_slots(crtc_state, mst_state); } if (dsc) { @@ -298,12 +319,20 @@ int intel_dp_mtp_tu_compute_config(struct intel_dp *intel_dp, } } + drm_WARN_ON(display->drm, min_bpp_x16 % bpp_step_x16 || max_bpp_x16 % bpp_step_x16); + for (bpp_x16 = max_bpp_x16; bpp_x16 >= min_bpp_x16; bpp_x16 -= bpp_step_x16) { int local_bw_overhead; int link_bpp_x16; drm_dbg_kms(display->drm, "Trying bpp " FXP_Q4_FMT "\n", FXP_Q4_ARGS(bpp_x16)); + if (dsc && !intel_dp_dsc_valid_compressed_bpp(intel_dp, bpp_x16)) { + /* SST must have validated the single bpp tried here already earlier. */ + drm_WARN_ON(display->drm, !is_mst); + continue; + } + link_bpp_x16 = dsc ? bpp_x16 : fxp_q4_from_int(intel_dp_output_bpp(crtc_state->output_format, fxp_q4_to_int(bpp_x16))); @@ -367,6 +396,10 @@ int intel_dp_mtp_tu_compute_config(struct intel_dp *intel_dp, slots = drm_dp_atomic_find_time_slots(state, &intel_dp->mst.mgr, connector->mst.port, dfixed_trunc(pbn)); + + /* TODO: Check this already in drm_dp_atomic_find_time_slots(). */ + if (slots > mst_state->total_avail_slots) + slots = -EINVAL; } else { /* Same as above for remote_tu */ crtc_state->dp_m_n.tu = ALIGN(crtc_state->dp_m_n.tu, @@ -386,10 +419,6 @@ int intel_dp_mtp_tu_compute_config(struct intel_dp *intel_dp, break; } - - /* Allow using zero step to indicate one try */ - if (!bpp_step_x16) - break; } if (slots < 0) { @@ -437,7 +466,8 @@ static int mst_stream_dsc_compute_link_config(struct intel_dp *intel_dp, int num_bpc; u8 dsc_bpc[3] = {}; int min_bpp, max_bpp, sink_min_bpp, sink_max_bpp; - int min_compressed_bpp, max_compressed_bpp; + int min_compressed_bpp_x16, max_compressed_bpp_x16; + int bpp_step_x16; max_bpp = limits->pipe.max_bpp; min_bpp = limits->pipe.min_bpp; @@ -462,46 +492,28 @@ static int mst_stream_dsc_compute_link_config(struct intel_dp *intel_dp, crtc_state->pipe_bpp = max_bpp; - max_compressed_bpp = fxp_q4_to_int(limits->link.max_bpp_x16); - min_compressed_bpp = fxp_q4_to_int_roundup(limits->link.min_bpp_x16); + min_compressed_bpp_x16 = limits->link.min_bpp_x16; + max_compressed_bpp_x16 = limits->link.max_bpp_x16; - drm_dbg_kms(display->drm, "DSC Sink supported compressed min bpp %d compressed max bpp %d\n", - min_compressed_bpp, max_compressed_bpp); + drm_dbg_kms(display->drm, + "DSC Sink supported compressed min bpp " FXP_Q4_FMT " compressed max bpp " FXP_Q4_FMT "\n", + FXP_Q4_ARGS(min_compressed_bpp_x16), FXP_Q4_ARGS(max_compressed_bpp_x16)); - /* Align compressed bpps according to our own constraints */ - max_compressed_bpp = intel_dp_dsc_nearest_valid_bpp(display, max_compressed_bpp, - crtc_state->pipe_bpp); - min_compressed_bpp = intel_dp_dsc_nearest_valid_bpp(display, min_compressed_bpp, - crtc_state->pipe_bpp); + bpp_step_x16 = intel_dp_dsc_bpp_step_x16(connector); + + max_compressed_bpp_x16 = min(max_compressed_bpp_x16, fxp_q4_from_int(crtc_state->pipe_bpp) - bpp_step_x16); + + drm_WARN_ON(display->drm, !is_power_of_2(bpp_step_x16)); + min_compressed_bpp_x16 = round_up(min_compressed_bpp_x16, bpp_step_x16); + max_compressed_bpp_x16 = round_down(max_compressed_bpp_x16, bpp_step_x16); crtc_state->lane_count = limits->max_lane_count; crtc_state->port_clock = limits->max_rate; return intel_dp_mtp_tu_compute_config(intel_dp, crtc_state, conn_state, - fxp_q4_from_int(min_compressed_bpp), - fxp_q4_from_int(max_compressed_bpp), - fxp_q4_from_int(1), true); -} - -static int mst_stream_update_slots(struct intel_dp *intel_dp, - struct intel_crtc_state *crtc_state, - struct drm_connector_state *conn_state) -{ - struct intel_display *display = to_intel_display(intel_dp); - struct drm_dp_mst_topology_mgr *mgr = &intel_dp->mst.mgr; - struct drm_dp_mst_topology_state *topology_state; - u8 link_coding_cap = intel_dp_is_uhbr(crtc_state) ? - DP_CAP_ANSI_128B132B : DP_CAP_ANSI_8B10B; - - topology_state = drm_atomic_get_mst_topology_state(conn_state->state, mgr); - if (IS_ERR(topology_state)) { - drm_dbg_kms(display->drm, "slot update failed\n"); - return PTR_ERR(topology_state); - } - - drm_dp_mst_update_slots(topology_state, link_coding_cap); - - return 0; + min_compressed_bpp_x16, + max_compressed_bpp_x16, + bpp_step_x16, true); } static int mode_hblank_period_ns(const struct drm_display_mode *mode) @@ -706,10 +718,6 @@ static int mst_stream_compute_config(struct intel_encoder *encoder, pipe_config->dp_m_n.tu); } - if (ret) - return ret; - - ret = mst_stream_update_slots(intel_dp, pipe_config, conn_state); if (ret) return ret; diff --git a/drivers/gpu/drm/i915/display/intel_dp_test.c b/drivers/gpu/drm/i915/display/intel_dp_test.c index bd61f3c3ec91175387a31712d90460390c22ce44..6ed5012c5fac06a3146310fa4dae82f0b1c0e1eb 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_test.c +++ b/drivers/gpu/drm/i915/display/intel_dp_test.c @@ -10,9 +10,9 @@ #include #include -#include "i915_reg.h" #include "intel_ddi.h" #include "intel_de.h" +#include "intel_display_regs.h" #include "intel_display_types.h" #include "intel_dp.h" #include "intel_dp_link_training.h" diff --git a/drivers/gpu/drm/i915/display/intel_dpio_phy.c b/drivers/gpu/drm/i915/display/intel_dpio_phy.c index 69f2421394205f0186a674d8e141ce3febaa802c..3f77ad92c156359992ed874a8c0b7e48eed18771 100644 --- a/drivers/gpu/drm/i915/display/intel_dpio_phy.c +++ b/drivers/gpu/drm/i915/display/intel_dpio_phy.c @@ -21,13 +21,15 @@ * DEALINGS IN THE SOFTWARE. */ +#include + #include "bxt_dpio_phy_regs.h" -#include "i915_drv.h" -#include "i915_reg.h" +#include "i915_utils.h" #include "intel_ddi.h" #include "intel_ddi_buf_trans.h" #include "intel_de.h" #include "intel_display_power_well.h" +#include "intel_display_regs.h" #include "intel_display_types.h" #include "intel_dp.h" #include "intel_dpio_phy.h" @@ -426,7 +428,7 @@ static void _bxt_dpio_phy_init(struct intel_display *display, enum dpio_phy phy) * use 1ms due to occasional timeouts observed with that. */ if (intel_de_wait_fw(display, BXT_PORT_CL1CM_DW0(phy), - PHY_RESERVED | PHY_POWER_GOOD, PHY_POWER_GOOD, 1)) + PHY_RESERVED | PHY_POWER_GOOD, PHY_POWER_GOOD, 1, NULL)) drm_err(display->drm, "timeout during PHY%d power on\n", phy); @@ -715,53 +717,53 @@ void chv_set_phy_signal_level(struct intel_encoder *encoder, u32 deemph_reg_value, u32 margin_reg_value, bool uniq_trans_scale) { - struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + struct intel_display *display = to_intel_display(encoder); struct intel_digital_port *dig_port = enc_to_dig_port(encoder); enum dpio_channel ch = vlv_dig_port_to_channel(dig_port); enum dpio_phy phy = vlv_dig_port_to_phy(dig_port); u32 val; int i; - vlv_dpio_get(dev_priv); + vlv_dpio_get(display->drm); /* Clear calc init */ - val = vlv_dpio_read(dev_priv, phy, VLV_PCS01_DW10(ch)); + val = vlv_dpio_read(display->drm, phy, VLV_PCS01_DW10(ch)); val &= ~(DPIO_PCS_SWING_CALC_TX0_TX2 | DPIO_PCS_SWING_CALC_TX1_TX3); val &= ~(DPIO_PCS_TX1DEEMP_MASK | DPIO_PCS_TX2DEEMP_MASK); val |= DPIO_PCS_TX1DEEMP_9P5 | DPIO_PCS_TX2DEEMP_9P5; - vlv_dpio_write(dev_priv, phy, VLV_PCS01_DW10(ch), val); + vlv_dpio_write(display->drm, phy, VLV_PCS01_DW10(ch), val); if (crtc_state->lane_count > 2) { - val = vlv_dpio_read(dev_priv, phy, VLV_PCS23_DW10(ch)); + val = vlv_dpio_read(display->drm, phy, VLV_PCS23_DW10(ch)); val &= ~(DPIO_PCS_SWING_CALC_TX0_TX2 | DPIO_PCS_SWING_CALC_TX1_TX3); val &= ~(DPIO_PCS_TX1DEEMP_MASK | DPIO_PCS_TX2DEEMP_MASK); val |= DPIO_PCS_TX1DEEMP_9P5 | DPIO_PCS_TX2DEEMP_9P5; - vlv_dpio_write(dev_priv, phy, VLV_PCS23_DW10(ch), val); + vlv_dpio_write(display->drm, phy, VLV_PCS23_DW10(ch), val); } - val = vlv_dpio_read(dev_priv, phy, VLV_PCS01_DW9(ch)); + val = vlv_dpio_read(display->drm, phy, VLV_PCS01_DW9(ch)); val &= ~(DPIO_PCS_TX1MARGIN_MASK | DPIO_PCS_TX2MARGIN_MASK); val |= DPIO_PCS_TX1MARGIN_000 | DPIO_PCS_TX2MARGIN_000; - vlv_dpio_write(dev_priv, phy, VLV_PCS01_DW9(ch), val); + vlv_dpio_write(display->drm, phy, VLV_PCS01_DW9(ch), val); if (crtc_state->lane_count > 2) { - val = vlv_dpio_read(dev_priv, phy, VLV_PCS23_DW9(ch)); + val = vlv_dpio_read(display->drm, phy, VLV_PCS23_DW9(ch)); val &= ~(DPIO_PCS_TX1MARGIN_MASK | DPIO_PCS_TX2MARGIN_MASK); val |= DPIO_PCS_TX1MARGIN_000 | DPIO_PCS_TX2MARGIN_000; - vlv_dpio_write(dev_priv, phy, VLV_PCS23_DW9(ch), val); + vlv_dpio_write(display->drm, phy, VLV_PCS23_DW9(ch), val); } /* Program swing deemph */ for (i = 0; i < crtc_state->lane_count; i++) { - val = vlv_dpio_read(dev_priv, phy, CHV_TX_DW4(ch, i)); + val = vlv_dpio_read(display->drm, phy, CHV_TX_DW4(ch, i)); val &= ~DPIO_SWING_DEEMPH9P5_MASK; val |= DPIO_SWING_DEEMPH9P5(deemph_reg_value); - vlv_dpio_write(dev_priv, phy, CHV_TX_DW4(ch, i), val); + vlv_dpio_write(display->drm, phy, CHV_TX_DW4(ch, i), val); } /* Program swing margin */ for (i = 0; i < crtc_state->lane_count; i++) { - val = vlv_dpio_read(dev_priv, phy, CHV_TX_DW2(ch, i)); + val = vlv_dpio_read(display->drm, phy, CHV_TX_DW2(ch, i)); val &= ~DPIO_SWING_MARGIN000_MASK; val |= DPIO_SWING_MARGIN000(margin_reg_value); @@ -774,7 +776,7 @@ void chv_set_phy_signal_level(struct intel_encoder *encoder, val &= ~DPIO_UNIQ_TRANS_SCALE_MASK; val |= DPIO_UNIQ_TRANS_SCALE(0x9a); - vlv_dpio_write(dev_priv, phy, CHV_TX_DW2(ch, i), val); + vlv_dpio_write(display->drm, phy, CHV_TX_DW2(ch, i), val); } /* @@ -784,70 +786,70 @@ void chv_set_phy_signal_level(struct intel_encoder *encoder, * 27 for ch0 and ch1. */ for (i = 0; i < crtc_state->lane_count; i++) { - val = vlv_dpio_read(dev_priv, phy, CHV_TX_DW3(ch, i)); + val = vlv_dpio_read(display->drm, phy, CHV_TX_DW3(ch, i)); if (uniq_trans_scale) val |= DPIO_TX_UNIQ_TRANS_SCALE_EN; else val &= ~DPIO_TX_UNIQ_TRANS_SCALE_EN; - vlv_dpio_write(dev_priv, phy, CHV_TX_DW3(ch, i), val); + vlv_dpio_write(display->drm, phy, CHV_TX_DW3(ch, i), val); } /* Start swing calculation */ - val = vlv_dpio_read(dev_priv, phy, VLV_PCS01_DW10(ch)); + val = vlv_dpio_read(display->drm, phy, VLV_PCS01_DW10(ch)); val |= DPIO_PCS_SWING_CALC_TX0_TX2 | DPIO_PCS_SWING_CALC_TX1_TX3; - vlv_dpio_write(dev_priv, phy, VLV_PCS01_DW10(ch), val); + vlv_dpio_write(display->drm, phy, VLV_PCS01_DW10(ch), val); if (crtc_state->lane_count > 2) { - val = vlv_dpio_read(dev_priv, phy, VLV_PCS23_DW10(ch)); + val = vlv_dpio_read(display->drm, phy, VLV_PCS23_DW10(ch)); val |= DPIO_PCS_SWING_CALC_TX0_TX2 | DPIO_PCS_SWING_CALC_TX1_TX3; - vlv_dpio_write(dev_priv, phy, VLV_PCS23_DW10(ch), val); + vlv_dpio_write(display->drm, phy, VLV_PCS23_DW10(ch), val); } - vlv_dpio_put(dev_priv); + vlv_dpio_put(display->drm); } static void __chv_data_lane_soft_reset(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state, bool reset) { - struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + struct intel_display *display = to_intel_display(encoder); struct intel_digital_port *dig_port = enc_to_dig_port(encoder); enum dpio_channel ch = vlv_dig_port_to_channel(dig_port); enum dpio_phy phy = vlv_dig_port_to_phy(dig_port); u32 val; - val = vlv_dpio_read(dev_priv, phy, VLV_PCS01_DW0(ch)); + val = vlv_dpio_read(display->drm, phy, VLV_PCS01_DW0(ch)); if (reset) val &= ~(DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET); else val |= DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET; - vlv_dpio_write(dev_priv, phy, VLV_PCS01_DW0(ch), val); + vlv_dpio_write(display->drm, phy, VLV_PCS01_DW0(ch), val); if (crtc_state->lane_count > 2) { - val = vlv_dpio_read(dev_priv, phy, VLV_PCS23_DW0(ch)); + val = vlv_dpio_read(display->drm, phy, VLV_PCS23_DW0(ch)); if (reset) val &= ~(DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET); else val |= DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET; - vlv_dpio_write(dev_priv, phy, VLV_PCS23_DW0(ch), val); + vlv_dpio_write(display->drm, phy, VLV_PCS23_DW0(ch), val); } - val = vlv_dpio_read(dev_priv, phy, VLV_PCS01_DW1(ch)); + val = vlv_dpio_read(display->drm, phy, VLV_PCS01_DW1(ch)); val |= CHV_PCS_REQ_SOFTRESET_EN; if (reset) val &= ~DPIO_PCS_CLK_SOFT_RESET; else val |= DPIO_PCS_CLK_SOFT_RESET; - vlv_dpio_write(dev_priv, phy, VLV_PCS01_DW1(ch), val); + vlv_dpio_write(display->drm, phy, VLV_PCS01_DW1(ch), val); if (crtc_state->lane_count > 2) { - val = vlv_dpio_read(dev_priv, phy, VLV_PCS23_DW1(ch)); + val = vlv_dpio_read(display->drm, phy, VLV_PCS23_DW1(ch)); val |= CHV_PCS_REQ_SOFTRESET_EN; if (reset) val &= ~DPIO_PCS_CLK_SOFT_RESET; else val |= DPIO_PCS_CLK_SOFT_RESET; - vlv_dpio_write(dev_priv, phy, VLV_PCS23_DW1(ch), val); + vlv_dpio_write(display->drm, phy, VLV_PCS23_DW1(ch), val); } } @@ -855,11 +857,11 @@ void chv_data_lane_soft_reset(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state, bool reset) { - struct drm_i915_private *i915 = to_i915(encoder->base.dev); + struct intel_display *display = to_intel_display(encoder); - vlv_dpio_get(i915); + vlv_dpio_get(display->drm); __chv_data_lane_soft_reset(encoder, crtc_state, reset); - vlv_dpio_put(i915); + vlv_dpio_put(display->drm); } void chv_phy_pre_pll_enable(struct intel_encoder *encoder, @@ -867,7 +869,6 @@ void chv_phy_pre_pll_enable(struct intel_encoder *encoder, { struct intel_display *display = to_intel_display(encoder); struct intel_digital_port *dig_port = enc_to_dig_port(encoder); - struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); enum dpio_channel ch = vlv_dig_port_to_channel(dig_port); enum dpio_phy phy = vlv_dig_port_to_phy(dig_port); @@ -886,47 +887,47 @@ void chv_phy_pre_pll_enable(struct intel_encoder *encoder, chv_phy_powergate_lanes(encoder, true, lane_mask); - vlv_dpio_get(dev_priv); + vlv_dpio_get(display->drm); /* Assert data lane reset */ __chv_data_lane_soft_reset(encoder, crtc_state, true); /* program left/right clock distribution */ if (pipe != PIPE_B) { - val = vlv_dpio_read(dev_priv, phy, CHV_CMN_DW5_CH0); + val = vlv_dpio_read(display->drm, phy, CHV_CMN_DW5_CH0); val &= ~(CHV_BUFLEFTENA1_MASK | CHV_BUFRIGHTENA1_MASK); if (ch == DPIO_CH0) val |= CHV_BUFLEFTENA1_FORCE; if (ch == DPIO_CH1) val |= CHV_BUFRIGHTENA1_FORCE; - vlv_dpio_write(dev_priv, phy, CHV_CMN_DW5_CH0, val); + vlv_dpio_write(display->drm, phy, CHV_CMN_DW5_CH0, val); } else { - val = vlv_dpio_read(dev_priv, phy, CHV_CMN_DW1_CH1); + val = vlv_dpio_read(display->drm, phy, CHV_CMN_DW1_CH1); val &= ~(CHV_BUFLEFTENA2_MASK | CHV_BUFRIGHTENA2_MASK); if (ch == DPIO_CH0) val |= CHV_BUFLEFTENA2_FORCE; if (ch == DPIO_CH1) val |= CHV_BUFRIGHTENA2_FORCE; - vlv_dpio_write(dev_priv, phy, CHV_CMN_DW1_CH1, val); + vlv_dpio_write(display->drm, phy, CHV_CMN_DW1_CH1, val); } /* program clock channel usage */ - val = vlv_dpio_read(dev_priv, phy, VLV_PCS01_DW8(ch)); + val = vlv_dpio_read(display->drm, phy, VLV_PCS01_DW8(ch)); val |= DPIO_PCS_USEDCLKCHANNEL_OVRRIDE; if (pipe == PIPE_B) val |= DPIO_PCS_USEDCLKCHANNEL; else val &= ~DPIO_PCS_USEDCLKCHANNEL; - vlv_dpio_write(dev_priv, phy, VLV_PCS01_DW8(ch), val); + vlv_dpio_write(display->drm, phy, VLV_PCS01_DW8(ch), val); if (crtc_state->lane_count > 2) { - val = vlv_dpio_read(dev_priv, phy, VLV_PCS23_DW8(ch)); + val = vlv_dpio_read(display->drm, phy, VLV_PCS23_DW8(ch)); val |= DPIO_PCS_USEDCLKCHANNEL_OVRRIDE; if (pipe == PIPE_B) val |= DPIO_PCS_USEDCLKCHANNEL; else val &= ~DPIO_PCS_USEDCLKCHANNEL; - vlv_dpio_write(dev_priv, phy, VLV_PCS23_DW8(ch), val); + vlv_dpio_write(display->drm, phy, VLV_PCS23_DW8(ch), val); } /* @@ -934,38 +935,38 @@ void chv_phy_pre_pll_enable(struct intel_encoder *encoder, * matches the pipe, but here we need to * pick the CL based on the port. */ - val = vlv_dpio_read(dev_priv, phy, CHV_CMN_DW19(ch)); + val = vlv_dpio_read(display->drm, phy, CHV_CMN_DW19(ch)); if (pipe == PIPE_B) val |= CHV_CMN_USEDCLKCHANNEL; else val &= ~CHV_CMN_USEDCLKCHANNEL; - vlv_dpio_write(dev_priv, phy, CHV_CMN_DW19(ch), val); + vlv_dpio_write(display->drm, phy, CHV_CMN_DW19(ch), val); - vlv_dpio_put(dev_priv); + vlv_dpio_put(display->drm); } void chv_phy_pre_encoder_enable(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state) { + struct intel_display *display = to_intel_display(encoder); struct intel_dp *intel_dp = enc_to_intel_dp(encoder); struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp); - struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); enum dpio_channel ch = vlv_dig_port_to_channel(dig_port); enum dpio_phy phy = vlv_dig_port_to_phy(dig_port); int data, i, stagger; u32 val; - vlv_dpio_get(dev_priv); + vlv_dpio_get(display->drm); /* allow hardware to manage TX FIFO reset source */ - val = vlv_dpio_read(dev_priv, phy, VLV_PCS01_DW11(ch)); + val = vlv_dpio_read(display->drm, phy, VLV_PCS01_DW11(ch)); val &= ~DPIO_LANEDESKEW_STRAP_OVRD; - vlv_dpio_write(dev_priv, phy, VLV_PCS01_DW11(ch), val); + vlv_dpio_write(display->drm, phy, VLV_PCS01_DW11(ch), val); if (crtc_state->lane_count > 2) { - val = vlv_dpio_read(dev_priv, phy, VLV_PCS23_DW11(ch)); + val = vlv_dpio_read(display->drm, phy, VLV_PCS23_DW11(ch)); val &= ~DPIO_LANEDESKEW_STRAP_OVRD; - vlv_dpio_write(dev_priv, phy, VLV_PCS23_DW11(ch), val); + vlv_dpio_write(display->drm, phy, VLV_PCS23_DW11(ch), val); } /* Program Tx lane latency optimal setting*/ @@ -975,7 +976,7 @@ void chv_phy_pre_encoder_enable(struct intel_encoder *encoder, data = 0; else data = (i == 1) ? 0 : DPIO_UPAR; - vlv_dpio_write(dev_priv, phy, CHV_TX_DW14(ch, i), data); + vlv_dpio_write(display->drm, phy, CHV_TX_DW14(ch, i), data); } /* Data lane stagger programming */ @@ -990,17 +991,17 @@ void chv_phy_pre_encoder_enable(struct intel_encoder *encoder, else stagger = 0x2; - val = vlv_dpio_read(dev_priv, phy, VLV_PCS01_DW11(ch)); + val = vlv_dpio_read(display->drm, phy, VLV_PCS01_DW11(ch)); val |= DPIO_TX2_STAGGER_MASK(0x1f); - vlv_dpio_write(dev_priv, phy, VLV_PCS01_DW11(ch), val); + vlv_dpio_write(display->drm, phy, VLV_PCS01_DW11(ch), val); if (crtc_state->lane_count > 2) { - val = vlv_dpio_read(dev_priv, phy, VLV_PCS23_DW11(ch)); + val = vlv_dpio_read(display->drm, phy, VLV_PCS23_DW11(ch)); val |= DPIO_TX2_STAGGER_MASK(0x1f); - vlv_dpio_write(dev_priv, phy, VLV_PCS23_DW11(ch), val); + vlv_dpio_write(display->drm, phy, VLV_PCS23_DW11(ch), val); } - vlv_dpio_write(dev_priv, phy, VLV_PCS01_DW12(ch), + vlv_dpio_write(display->drm, phy, VLV_PCS01_DW12(ch), DPIO_LANESTAGGER_STRAP(stagger) | DPIO_LANESTAGGER_STRAP_OVRD | DPIO_TX1_STAGGER_MASK(0x1f) | @@ -1008,7 +1009,7 @@ void chv_phy_pre_encoder_enable(struct intel_encoder *encoder, DPIO_TX2_STAGGER_MULT(0)); if (crtc_state->lane_count > 2) { - vlv_dpio_write(dev_priv, phy, VLV_PCS23_DW12(ch), + vlv_dpio_write(display->drm, phy, VLV_PCS23_DW12(ch), DPIO_LANESTAGGER_STRAP(stagger) | DPIO_LANESTAGGER_STRAP_OVRD | DPIO_TX1_STAGGER_MASK(0x1f) | @@ -1019,7 +1020,7 @@ void chv_phy_pre_encoder_enable(struct intel_encoder *encoder, /* Deassert data lane reset */ __chv_data_lane_soft_reset(encoder, crtc_state, false); - vlv_dpio_put(dev_priv); + vlv_dpio_put(display->drm); } void chv_phy_release_cl2_override(struct intel_encoder *encoder) @@ -1036,25 +1037,25 @@ void chv_phy_release_cl2_override(struct intel_encoder *encoder) void chv_phy_post_pll_disable(struct intel_encoder *encoder, const struct intel_crtc_state *old_crtc_state) { - struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + struct intel_display *display = to_intel_display(encoder); enum dpio_phy phy = vlv_dig_port_to_phy(enc_to_dig_port(encoder)); enum pipe pipe = to_intel_crtc(old_crtc_state->uapi.crtc)->pipe; u32 val; - vlv_dpio_get(dev_priv); + vlv_dpio_get(display->drm); /* disable left/right clock distribution */ if (pipe != PIPE_B) { - val = vlv_dpio_read(dev_priv, phy, CHV_CMN_DW5_CH0); + val = vlv_dpio_read(display->drm, phy, CHV_CMN_DW5_CH0); val &= ~(CHV_BUFLEFTENA1_MASK | CHV_BUFRIGHTENA1_MASK); - vlv_dpio_write(dev_priv, phy, CHV_CMN_DW5_CH0, val); + vlv_dpio_write(display->drm, phy, CHV_CMN_DW5_CH0, val); } else { - val = vlv_dpio_read(dev_priv, phy, CHV_CMN_DW1_CH1); + val = vlv_dpio_read(display->drm, phy, CHV_CMN_DW1_CH1); val &= ~(CHV_BUFLEFTENA2_MASK | CHV_BUFRIGHTENA2_MASK); - vlv_dpio_write(dev_priv, phy, CHV_CMN_DW1_CH1, val); + vlv_dpio_write(display->drm, phy, CHV_CMN_DW1_CH1, val); } - vlv_dpio_put(dev_priv); + vlv_dpio_put(display->drm); /* * Leave the power down bit cleared for at least one @@ -1073,97 +1074,97 @@ void vlv_set_phy_signal_level(struct intel_encoder *encoder, u32 demph_reg_value, u32 preemph_reg_value, u32 uniqtranscale_reg_value, u32 tx3_demph) { - struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + struct intel_display *display = to_intel_display(encoder); struct intel_digital_port *dig_port = enc_to_dig_port(encoder); enum dpio_channel ch = vlv_dig_port_to_channel(dig_port); enum dpio_phy phy = vlv_dig_port_to_phy(dig_port); - vlv_dpio_get(dev_priv); + vlv_dpio_get(display->drm); - vlv_dpio_write(dev_priv, phy, VLV_TX_DW5_GRP(ch), 0x00000000); - vlv_dpio_write(dev_priv, phy, VLV_TX_DW4_GRP(ch), demph_reg_value); - vlv_dpio_write(dev_priv, phy, VLV_TX_DW2_GRP(ch), - uniqtranscale_reg_value); - vlv_dpio_write(dev_priv, phy, VLV_TX_DW3_GRP(ch), 0x0C782040); + vlv_dpio_write(display->drm, phy, VLV_TX_DW5_GRP(ch), 0x00000000); + vlv_dpio_write(display->drm, phy, VLV_TX_DW4_GRP(ch), demph_reg_value); + vlv_dpio_write(display->drm, phy, VLV_TX_DW2_GRP(ch), + uniqtranscale_reg_value); + vlv_dpio_write(display->drm, phy, VLV_TX_DW3_GRP(ch), 0x0C782040); if (tx3_demph) - vlv_dpio_write(dev_priv, phy, VLV_TX_DW4(ch, 3), tx3_demph); + vlv_dpio_write(display->drm, phy, VLV_TX_DW4(ch, 3), tx3_demph); - vlv_dpio_write(dev_priv, phy, VLV_PCS_DW11_GRP(ch), 0x00030000); - vlv_dpio_write(dev_priv, phy, VLV_PCS_DW9_GRP(ch), preemph_reg_value); - vlv_dpio_write(dev_priv, phy, VLV_TX_DW5_GRP(ch), DPIO_TX_OCALINIT_EN); + vlv_dpio_write(display->drm, phy, VLV_PCS_DW11_GRP(ch), 0x00030000); + vlv_dpio_write(display->drm, phy, VLV_PCS_DW9_GRP(ch), preemph_reg_value); + vlv_dpio_write(display->drm, phy, VLV_TX_DW5_GRP(ch), DPIO_TX_OCALINIT_EN); - vlv_dpio_put(dev_priv); + vlv_dpio_put(display->drm); } void vlv_phy_pre_pll_enable(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state) { + struct intel_display *display = to_intel_display(encoder); struct intel_digital_port *dig_port = enc_to_dig_port(encoder); - struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); enum dpio_channel ch = vlv_dig_port_to_channel(dig_port); enum dpio_phy phy = vlv_dig_port_to_phy(dig_port); /* Program Tx lane resets to default */ - vlv_dpio_get(dev_priv); + vlv_dpio_get(display->drm); - vlv_dpio_write(dev_priv, phy, VLV_PCS_DW0_GRP(ch), + vlv_dpio_write(display->drm, phy, VLV_PCS_DW0_GRP(ch), DPIO_PCS_TX_LANE2_RESET | DPIO_PCS_TX_LANE1_RESET); - vlv_dpio_write(dev_priv, phy, VLV_PCS_DW1_GRP(ch), + vlv_dpio_write(display->drm, phy, VLV_PCS_DW1_GRP(ch), DPIO_PCS_CLK_CRI_RXEB_EIOS_EN | DPIO_PCS_CLK_CRI_RXDIGFILTSG_EN | DPIO_PCS_CLK_DATAWIDTH_8_10 | DPIO_PCS_CLK_SOFT_RESET); /* Fix up inter-pair skew failure */ - vlv_dpio_write(dev_priv, phy, VLV_PCS_DW12_GRP(ch), 0x00750f00); - vlv_dpio_write(dev_priv, phy, VLV_TX_DW11_GRP(ch), 0x00001500); - vlv_dpio_write(dev_priv, phy, VLV_TX_DW14_GRP(ch), 0x40400000); + vlv_dpio_write(display->drm, phy, VLV_PCS_DW12_GRP(ch), 0x00750f00); + vlv_dpio_write(display->drm, phy, VLV_TX_DW11_GRP(ch), 0x00001500); + vlv_dpio_write(display->drm, phy, VLV_TX_DW14_GRP(ch), 0x40400000); - vlv_dpio_put(dev_priv); + vlv_dpio_put(display->drm); } void vlv_phy_pre_encoder_enable(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state) { + struct intel_display *display = to_intel_display(encoder); struct intel_dp *intel_dp = enc_to_intel_dp(encoder); struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp); - struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); enum dpio_channel ch = vlv_dig_port_to_channel(dig_port); enum dpio_phy phy = vlv_dig_port_to_phy(dig_port); enum pipe pipe = crtc->pipe; u32 val; - vlv_dpio_get(dev_priv); + vlv_dpio_get(display->drm); /* Enable clock channels for this port */ val = DPIO_PCS_USEDCLKCHANNEL_OVRRIDE; if (pipe == PIPE_B) val |= DPIO_PCS_USEDCLKCHANNEL; val |= 0xc4; - vlv_dpio_write(dev_priv, phy, VLV_PCS_DW8_GRP(ch), val); + vlv_dpio_write(display->drm, phy, VLV_PCS_DW8_GRP(ch), val); /* Program lane clock */ - vlv_dpio_write(dev_priv, phy, VLV_PCS_DW14_GRP(ch), 0x00760018); - vlv_dpio_write(dev_priv, phy, VLV_PCS_DW23_GRP(ch), 0x00400888); + vlv_dpio_write(display->drm, phy, VLV_PCS_DW14_GRP(ch), 0x00760018); + vlv_dpio_write(display->drm, phy, VLV_PCS_DW23_GRP(ch), 0x00400888); - vlv_dpio_put(dev_priv); + vlv_dpio_put(display->drm); } void vlv_phy_reset_lanes(struct intel_encoder *encoder, const struct intel_crtc_state *old_crtc_state) { + struct intel_display *display = to_intel_display(encoder); struct intel_digital_port *dig_port = enc_to_dig_port(encoder); - struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); enum dpio_channel ch = vlv_dig_port_to_channel(dig_port); enum dpio_phy phy = vlv_dig_port_to_phy(dig_port); - vlv_dpio_get(dev_priv); - vlv_dpio_write(dev_priv, phy, VLV_PCS_DW0_GRP(ch), 0x00000000); - vlv_dpio_write(dev_priv, phy, VLV_PCS_DW1_GRP(ch), 0x00e00060); - vlv_dpio_put(dev_priv); + vlv_dpio_get(display->drm); + vlv_dpio_write(display->drm, phy, VLV_PCS_DW0_GRP(ch), 0x00000000); + vlv_dpio_write(display->drm, phy, VLV_PCS_DW1_GRP(ch), 0x00e00060); + vlv_dpio_put(display->drm); } void vlv_wait_port_ready(struct intel_encoder *encoder, diff --git a/drivers/gpu/drm/i915/display/intel_dpll.c b/drivers/gpu/drm/i915/display/intel_dpll.c index a9e9b98d0bf9c9c6ee33b48ff9306003d64962da..f969c5399a511d093eb8d589c5ea96f758cd423b 100644 --- a/drivers/gpu/drm/i915/display/intel_dpll.c +++ b/drivers/gpu/drm/i915/display/intel_dpll.c @@ -6,13 +6,14 @@ #include #include -#include "i915_drv.h" -#include "i915_reg.h" +#include + #include "intel_atomic.h" #include "intel_crtc.h" #include "intel_cx0_phy.h" #include "intel_de.h" #include "intel_display.h" +#include "intel_display_regs.h" #include "intel_display_types.h" #include "intel_dpio_phy.h" #include "intel_dpll.h" @@ -24,11 +25,11 @@ #include "vlv_dpio_phy_regs.h" #include "vlv_sideband.h" -struct intel_dpll_funcs { +struct intel_dpll_global_funcs { int (*crtc_compute_clock)(struct intel_atomic_state *state, struct intel_crtc *crtc); - int (*crtc_get_shared_dpll)(struct intel_atomic_state *state, - struct intel_crtc *crtc); + int (*crtc_get_dpll)(struct intel_atomic_state *state, + struct intel_crtc *crtc); }; struct intel_limit { @@ -513,8 +514,8 @@ void i9xx_crtc_clock_get(struct intel_crtc_state *crtc_state) void vlv_crtc_clock_get(struct intel_crtc_state *crtc_state) { + struct intel_display *display = to_intel_display(crtc_state); struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); - struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); enum dpio_channel ch = vlv_pipe_to_channel(crtc->pipe); enum dpio_phy phy = vlv_pipe_to_phy(crtc->pipe); const struct i9xx_dpll_hw_state *hw_state = &crtc_state->dpll_hw_state.i9xx; @@ -526,9 +527,9 @@ void vlv_crtc_clock_get(struct intel_crtc_state *crtc_state) if ((hw_state->dpll & DPLL_VCO_ENABLE) == 0) return; - vlv_dpio_get(dev_priv); - tmp = vlv_dpio_read(dev_priv, phy, VLV_PLL_DW3(ch)); - vlv_dpio_put(dev_priv); + vlv_dpio_get(display->drm); + tmp = vlv_dpio_read(display->drm, phy, VLV_PLL_DW3(ch)); + vlv_dpio_put(display->drm); clock.m1 = REG_FIELD_GET(DPIO_M1_DIV_MASK, tmp); clock.m2 = REG_FIELD_GET(DPIO_M2_DIV_MASK, tmp); @@ -541,8 +542,8 @@ void vlv_crtc_clock_get(struct intel_crtc_state *crtc_state) void chv_crtc_clock_get(struct intel_crtc_state *crtc_state) { + struct intel_display *display = to_intel_display(crtc_state); struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); - struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); enum dpio_channel ch = vlv_pipe_to_channel(crtc->pipe); enum dpio_phy phy = vlv_pipe_to_phy(crtc->pipe); const struct i9xx_dpll_hw_state *hw_state = &crtc_state->dpll_hw_state.i9xx; @@ -554,13 +555,13 @@ void chv_crtc_clock_get(struct intel_crtc_state *crtc_state) if ((hw_state->dpll & DPLL_VCO_ENABLE) == 0) return; - vlv_dpio_get(dev_priv); - cmn_dw13 = vlv_dpio_read(dev_priv, phy, CHV_CMN_DW13(ch)); - pll_dw0 = vlv_dpio_read(dev_priv, phy, CHV_PLL_DW0(ch)); - pll_dw1 = vlv_dpio_read(dev_priv, phy, CHV_PLL_DW1(ch)); - pll_dw2 = vlv_dpio_read(dev_priv, phy, CHV_PLL_DW2(ch)); - pll_dw3 = vlv_dpio_read(dev_priv, phy, CHV_PLL_DW3(ch)); - vlv_dpio_put(dev_priv); + vlv_dpio_get(display->drm); + cmn_dw13 = vlv_dpio_read(display->drm, phy, CHV_CMN_DW13(ch)); + pll_dw0 = vlv_dpio_read(display->drm, phy, CHV_PLL_DW0(ch)); + pll_dw1 = vlv_dpio_read(display->drm, phy, CHV_PLL_DW1(ch)); + pll_dw2 = vlv_dpio_read(display->drm, phy, CHV_PLL_DW2(ch)); + pll_dw3 = vlv_dpio_read(display->drm, phy, CHV_PLL_DW3(ch)); + vlv_dpio_put(display->drm); clock.m1 = REG_FIELD_GET(DPIO_CHV_M1_DIV_MASK, pll_dw1) == DPIO_CHV_M1_DIV_BY_2 ? 2 : 0; clock.m2 = REG_FIELD_GET(DPIO_CHV_M2_DIV_MASK, pll_dw0) << 22; @@ -1161,7 +1162,7 @@ static int hsw_crtc_compute_clock(struct intel_atomic_state *state, intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DSI)) return 0; - ret = intel_compute_shared_dplls(state, crtc, encoder); + ret = intel_dpll_compute(state, crtc, encoder); if (ret) return ret; @@ -1176,8 +1177,8 @@ static int hsw_crtc_compute_clock(struct intel_atomic_state *state, return 0; } -static int hsw_crtc_get_shared_dpll(struct intel_atomic_state *state, - struct intel_crtc *crtc) +static int hsw_crtc_get_dpll(struct intel_atomic_state *state, + struct intel_crtc *crtc) { struct intel_display *display = to_intel_display(state); struct intel_crtc_state *crtc_state = @@ -1189,7 +1190,7 @@ static int hsw_crtc_get_shared_dpll(struct intel_atomic_state *state, intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DSI)) return 0; - return intel_reserve_shared_dplls(state, crtc, encoder); + return intel_dpll_reserve(state, crtc, encoder); } static int dg2_crtc_compute_clock(struct intel_atomic_state *state, @@ -1223,7 +1224,7 @@ static int mtl_crtc_compute_clock(struct intel_atomic_state *state, if (ret) return ret; - /* TODO: Do the readback via intel_compute_shared_dplls() */ + /* TODO: Do the readback via intel_dpll_compute() */ crtc_state->port_clock = intel_cx0pll_calc_port_clock(encoder, &crtc_state->dpll_hw_state.cx0pll); crtc_state->hw.adjusted_mode.crtc_clock = intel_crtc_dotclock(crtc_state); @@ -1394,7 +1395,7 @@ static int ilk_crtc_compute_clock(struct intel_atomic_state *state, ilk_compute_dpll(crtc_state, &crtc_state->dpll, &crtc_state->dpll); - ret = intel_compute_shared_dplls(state, crtc, NULL); + ret = intel_dpll_compute(state, crtc, NULL); if (ret) return ret; @@ -1404,8 +1405,8 @@ static int ilk_crtc_compute_clock(struct intel_atomic_state *state, return ret; } -static int ilk_crtc_get_shared_dpll(struct intel_atomic_state *state, - struct intel_crtc *crtc) +static int ilk_crtc_get_dpll(struct intel_atomic_state *state, + struct intel_crtc *crtc) { struct intel_crtc_state *crtc_state = intel_atomic_get_new_crtc_state(state, crtc); @@ -1414,7 +1415,7 @@ static int ilk_crtc_get_shared_dpll(struct intel_atomic_state *state, if (!crtc_state->has_pch_encoder) return 0; - return intel_reserve_shared_dplls(state, crtc, NULL); + return intel_dpll_reserve(state, crtc, NULL); } static u32 vlv_dpll(const struct intel_crtc_state *crtc_state) @@ -1690,45 +1691,45 @@ static int i8xx_crtc_compute_clock(struct intel_atomic_state *state, return 0; } -static const struct intel_dpll_funcs mtl_dpll_funcs = { +static const struct intel_dpll_global_funcs mtl_dpll_funcs = { .crtc_compute_clock = mtl_crtc_compute_clock, }; -static const struct intel_dpll_funcs dg2_dpll_funcs = { +static const struct intel_dpll_global_funcs dg2_dpll_funcs = { .crtc_compute_clock = dg2_crtc_compute_clock, }; -static const struct intel_dpll_funcs hsw_dpll_funcs = { +static const struct intel_dpll_global_funcs hsw_dpll_funcs = { .crtc_compute_clock = hsw_crtc_compute_clock, - .crtc_get_shared_dpll = hsw_crtc_get_shared_dpll, + .crtc_get_dpll = hsw_crtc_get_dpll, }; -static const struct intel_dpll_funcs ilk_dpll_funcs = { +static const struct intel_dpll_global_funcs ilk_dpll_funcs = { .crtc_compute_clock = ilk_crtc_compute_clock, - .crtc_get_shared_dpll = ilk_crtc_get_shared_dpll, + .crtc_get_dpll = ilk_crtc_get_dpll, }; -static const struct intel_dpll_funcs chv_dpll_funcs = { +static const struct intel_dpll_global_funcs chv_dpll_funcs = { .crtc_compute_clock = chv_crtc_compute_clock, }; -static const struct intel_dpll_funcs vlv_dpll_funcs = { +static const struct intel_dpll_global_funcs vlv_dpll_funcs = { .crtc_compute_clock = vlv_crtc_compute_clock, }; -static const struct intel_dpll_funcs g4x_dpll_funcs = { +static const struct intel_dpll_global_funcs g4x_dpll_funcs = { .crtc_compute_clock = g4x_crtc_compute_clock, }; -static const struct intel_dpll_funcs pnv_dpll_funcs = { +static const struct intel_dpll_global_funcs pnv_dpll_funcs = { .crtc_compute_clock = pnv_crtc_compute_clock, }; -static const struct intel_dpll_funcs i9xx_dpll_funcs = { +static const struct intel_dpll_global_funcs i9xx_dpll_funcs = { .crtc_compute_clock = i9xx_crtc_compute_clock, }; -static const struct intel_dpll_funcs i8xx_dpll_funcs = { +static const struct intel_dpll_global_funcs i8xx_dpll_funcs = { .crtc_compute_clock = i8xx_crtc_compute_clock, }; @@ -1758,8 +1759,8 @@ int intel_dpll_crtc_compute_clock(struct intel_atomic_state *state, return 0; } -int intel_dpll_crtc_get_shared_dpll(struct intel_atomic_state *state, - struct intel_crtc *crtc) +int intel_dpll_crtc_get_dpll(struct intel_atomic_state *state, + struct intel_crtc *crtc) { struct intel_display *display = to_intel_display(state); struct intel_crtc_state *crtc_state = @@ -1767,15 +1768,15 @@ int intel_dpll_crtc_get_shared_dpll(struct intel_atomic_state *state, int ret; drm_WARN_ON(display->drm, !intel_crtc_needs_modeset(crtc_state)); - drm_WARN_ON(display->drm, !crtc_state->hw.enable && crtc_state->shared_dpll); + drm_WARN_ON(display->drm, !crtc_state->hw.enable && crtc_state->intel_dpll); - if (!crtc_state->hw.enable || crtc_state->shared_dpll) + if (!crtc_state->hw.enable || crtc_state->intel_dpll) return 0; - if (!display->funcs.dpll->crtc_get_shared_dpll) + if (!display->funcs.dpll->crtc_get_dpll) return 0; - ret = display->funcs.dpll->crtc_get_shared_dpll(state, crtc); + ret = display->funcs.dpll->crtc_get_dpll(state, crtc); if (ret) { drm_dbg_kms(display->drm, "[CRTC:%d:%s] Couldn't get a shared DPLL\n", crtc->base.base.id, crtc->base.name); @@ -1871,45 +1872,43 @@ void i9xx_enable_pll(const struct intel_crtc_state *crtc_state) static void vlv_pllb_recal_opamp(struct intel_display *display, enum dpio_phy phy, enum dpio_channel ch) { - struct drm_i915_private *dev_priv = to_i915(display->drm); u32 tmp; /* * PLLB opamp always calibrates to max value of 0x3f, force enable it * and set it to a reasonable value instead. */ - tmp = vlv_dpio_read(dev_priv, phy, VLV_PLL_DW17(ch)); + tmp = vlv_dpio_read(display->drm, phy, VLV_PLL_DW17(ch)); tmp &= 0xffffff00; tmp |= 0x00000030; - vlv_dpio_write(dev_priv, phy, VLV_PLL_DW17(ch), tmp); + vlv_dpio_write(display->drm, phy, VLV_PLL_DW17(ch), tmp); - tmp = vlv_dpio_read(dev_priv, phy, VLV_REF_DW11); + tmp = vlv_dpio_read(display->drm, phy, VLV_REF_DW11); tmp &= 0x00ffffff; tmp |= 0x8c000000; - vlv_dpio_write(dev_priv, phy, VLV_REF_DW11, tmp); + vlv_dpio_write(display->drm, phy, VLV_REF_DW11, tmp); - tmp = vlv_dpio_read(dev_priv, phy, VLV_PLL_DW17(ch)); + tmp = vlv_dpio_read(display->drm, phy, VLV_PLL_DW17(ch)); tmp &= 0xffffff00; - vlv_dpio_write(dev_priv, phy, VLV_PLL_DW17(ch), tmp); + vlv_dpio_write(display->drm, phy, VLV_PLL_DW17(ch), tmp); - tmp = vlv_dpio_read(dev_priv, phy, VLV_REF_DW11); + tmp = vlv_dpio_read(display->drm, phy, VLV_REF_DW11); tmp &= 0x00ffffff; tmp |= 0xb0000000; - vlv_dpio_write(dev_priv, phy, VLV_REF_DW11, tmp); + vlv_dpio_write(display->drm, phy, VLV_REF_DW11, tmp); } static void vlv_prepare_pll(const struct intel_crtc_state *crtc_state) { struct intel_display *display = to_intel_display(crtc_state); struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); - struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); const struct dpll *clock = &crtc_state->dpll; enum dpio_channel ch = vlv_pipe_to_channel(crtc->pipe); enum dpio_phy phy = vlv_pipe_to_phy(crtc->pipe); enum pipe pipe = crtc->pipe; u32 tmp, coreclk; - vlv_dpio_get(dev_priv); + vlv_dpio_get(display->drm); /* See eDP HDMI DPIO driver vbios notes doc */ @@ -1918,15 +1917,15 @@ static void vlv_prepare_pll(const struct intel_crtc_state *crtc_state) vlv_pllb_recal_opamp(display, phy, ch); /* Set up Tx target for periodic Rcomp update */ - vlv_dpio_write(dev_priv, phy, VLV_PCS_DW17_BCAST, 0x0100000f); + vlv_dpio_write(display->drm, phy, VLV_PCS_DW17_BCAST, 0x0100000f); /* Disable target IRef on PLL */ - tmp = vlv_dpio_read(dev_priv, phy, VLV_PLL_DW16(ch)); + tmp = vlv_dpio_read(display->drm, phy, VLV_PLL_DW16(ch)); tmp &= 0x00ffffff; - vlv_dpio_write(dev_priv, phy, VLV_PLL_DW16(ch), tmp); + vlv_dpio_write(display->drm, phy, VLV_PLL_DW16(ch), tmp); /* Disable fast lock */ - vlv_dpio_write(dev_priv, phy, VLV_CMN_DW0, 0x610); + vlv_dpio_write(display->drm, phy, VLV_CMN_DW0, 0x610); /* Set idtafcrecal before PLL is enabled */ tmp = DPIO_M1_DIV(clock->m1) | @@ -1942,48 +1941,42 @@ static void vlv_prepare_pll(const struct intel_crtc_state *crtc_state) * Note: don't use the DAC post divider as it seems unstable. */ tmp |= DPIO_S1_DIV(DPIO_S1_DIV_HDMIDP); - vlv_dpio_write(dev_priv, phy, VLV_PLL_DW3(ch), tmp); + vlv_dpio_write(display->drm, phy, VLV_PLL_DW3(ch), tmp); tmp |= DPIO_ENABLE_CALIBRATION; - vlv_dpio_write(dev_priv, phy, VLV_PLL_DW3(ch), tmp); + vlv_dpio_write(display->drm, phy, VLV_PLL_DW3(ch), tmp); /* Set HBR and RBR LPF coefficients */ if (crtc_state->port_clock == 162000 || intel_crtc_has_type(crtc_state, INTEL_OUTPUT_ANALOG) || intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI)) - vlv_dpio_write(dev_priv, phy, VLV_PLL_DW18(ch), - 0x009f0003); + vlv_dpio_write(display->drm, phy, VLV_PLL_DW18(ch), 0x009f0003); else - vlv_dpio_write(dev_priv, phy, VLV_PLL_DW18(ch), - 0x00d0000f); + vlv_dpio_write(display->drm, phy, VLV_PLL_DW18(ch), 0x00d0000f); if (intel_crtc_has_dp_encoder(crtc_state)) { /* Use SSC source */ if (pipe == PIPE_A) - vlv_dpio_write(dev_priv, phy, VLV_PLL_DW5(ch), - 0x0df40000); + vlv_dpio_write(display->drm, phy, VLV_PLL_DW5(ch), 0x0df40000); else - vlv_dpio_write(dev_priv, phy, VLV_PLL_DW5(ch), - 0x0df70000); + vlv_dpio_write(display->drm, phy, VLV_PLL_DW5(ch), 0x0df70000); } else { /* HDMI or VGA */ /* Use bend source */ if (pipe == PIPE_A) - vlv_dpio_write(dev_priv, phy, VLV_PLL_DW5(ch), - 0x0df70000); + vlv_dpio_write(display->drm, phy, VLV_PLL_DW5(ch), 0x0df70000); else - vlv_dpio_write(dev_priv, phy, VLV_PLL_DW5(ch), - 0x0df40000); + vlv_dpio_write(display->drm, phy, VLV_PLL_DW5(ch), 0x0df40000); } - coreclk = vlv_dpio_read(dev_priv, phy, VLV_PLL_DW7(ch)); + coreclk = vlv_dpio_read(display->drm, phy, VLV_PLL_DW7(ch)); coreclk = (coreclk & 0x0000ff00) | 0x01c00000; if (intel_crtc_has_dp_encoder(crtc_state)) coreclk |= 0x01000000; - vlv_dpio_write(dev_priv, phy, VLV_PLL_DW7(ch), coreclk); + vlv_dpio_write(display->drm, phy, VLV_PLL_DW7(ch), coreclk); - vlv_dpio_write(dev_priv, phy, VLV_PLL_DW19(ch), 0x87871000); + vlv_dpio_write(display->drm, phy, VLV_PLL_DW19(ch), 0x87871000); - vlv_dpio_put(dev_priv); + vlv_dpio_put(display->drm); } static void _vlv_enable_pll(const struct intel_crtc_state *crtc_state) @@ -2028,8 +2021,8 @@ void vlv_enable_pll(const struct intel_crtc_state *crtc_state) static void chv_prepare_pll(const struct intel_crtc_state *crtc_state) { + struct intel_display *display = to_intel_display(crtc_state); struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); - struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); const struct dpll *clock = &crtc_state->dpll; enum dpio_channel ch = vlv_pipe_to_channel(crtc->pipe); enum dpio_phy phy = vlv_pipe_to_phy(crtc->pipe); @@ -2038,44 +2031,44 @@ static void chv_prepare_pll(const struct intel_crtc_state *crtc_state) m2_frac = clock->m2 & 0x3fffff; - vlv_dpio_get(dev_priv); + vlv_dpio_get(display->drm); /* p1 and p2 divider */ - vlv_dpio_write(dev_priv, phy, CHV_CMN_DW13(ch), + vlv_dpio_write(display->drm, phy, CHV_CMN_DW13(ch), DPIO_CHV_S1_DIV(5) | DPIO_CHV_P1_DIV(clock->p1) | DPIO_CHV_P2_DIV(clock->p2) | DPIO_CHV_K_DIV(1)); /* Feedback post-divider - m2 */ - vlv_dpio_write(dev_priv, phy, CHV_PLL_DW0(ch), + vlv_dpio_write(display->drm, phy, CHV_PLL_DW0(ch), DPIO_CHV_M2_DIV(clock->m2 >> 22)); /* Feedback refclk divider - n and m1 */ - vlv_dpio_write(dev_priv, phy, CHV_PLL_DW1(ch), + vlv_dpio_write(display->drm, phy, CHV_PLL_DW1(ch), DPIO_CHV_M1_DIV(DPIO_CHV_M1_DIV_BY_2) | DPIO_CHV_N_DIV(1)); /* M2 fraction division */ - vlv_dpio_write(dev_priv, phy, CHV_PLL_DW2(ch), + vlv_dpio_write(display->drm, phy, CHV_PLL_DW2(ch), DPIO_CHV_M2_FRAC_DIV(m2_frac)); /* M2 fraction division enable */ - tmp = vlv_dpio_read(dev_priv, phy, CHV_PLL_DW3(ch)); + tmp = vlv_dpio_read(display->drm, phy, CHV_PLL_DW3(ch)); tmp &= ~(DPIO_CHV_FEEDFWD_GAIN_MASK | DPIO_CHV_FRAC_DIV_EN); tmp |= DPIO_CHV_FEEDFWD_GAIN(2); if (m2_frac) tmp |= DPIO_CHV_FRAC_DIV_EN; - vlv_dpio_write(dev_priv, phy, CHV_PLL_DW3(ch), tmp); + vlv_dpio_write(display->drm, phy, CHV_PLL_DW3(ch), tmp); /* Program digital lock detect threshold */ - tmp = vlv_dpio_read(dev_priv, phy, CHV_PLL_DW9(ch)); + tmp = vlv_dpio_read(display->drm, phy, CHV_PLL_DW9(ch)); tmp &= ~(DPIO_CHV_INT_LOCK_THRESHOLD_MASK | DPIO_CHV_INT_LOCK_THRESHOLD_SEL_COARSE); tmp |= DPIO_CHV_INT_LOCK_THRESHOLD(0x5); if (!m2_frac) tmp |= DPIO_CHV_INT_LOCK_THRESHOLD_SEL_COARSE; - vlv_dpio_write(dev_priv, phy, CHV_PLL_DW9(ch), tmp); + vlv_dpio_write(display->drm, phy, CHV_PLL_DW9(ch), tmp); /* Loop filter */ if (clock->vco == 5400000) { @@ -2100,40 +2093,39 @@ static void chv_prepare_pll(const struct intel_crtc_state *crtc_state) DPIO_CHV_GAIN_CTRL(0x3); tribuf_calcntr = 0; } - vlv_dpio_write(dev_priv, phy, CHV_PLL_DW6(ch), loopfilter); + vlv_dpio_write(display->drm, phy, CHV_PLL_DW6(ch), loopfilter); - tmp = vlv_dpio_read(dev_priv, phy, CHV_PLL_DW8(ch)); + tmp = vlv_dpio_read(display->drm, phy, CHV_PLL_DW8(ch)); tmp &= ~DPIO_CHV_TDC_TARGET_CNT_MASK; tmp |= DPIO_CHV_TDC_TARGET_CNT(tribuf_calcntr); - vlv_dpio_write(dev_priv, phy, CHV_PLL_DW8(ch), tmp); + vlv_dpio_write(display->drm, phy, CHV_PLL_DW8(ch), tmp); /* AFC Recal */ - vlv_dpio_write(dev_priv, phy, CHV_CMN_DW14(ch), - vlv_dpio_read(dev_priv, phy, CHV_CMN_DW14(ch)) | + vlv_dpio_write(display->drm, phy, CHV_CMN_DW14(ch), + vlv_dpio_read(display->drm, phy, CHV_CMN_DW14(ch)) | DPIO_AFC_RECAL); - vlv_dpio_put(dev_priv); + vlv_dpio_put(display->drm); } static void _chv_enable_pll(const struct intel_crtc_state *crtc_state) { struct intel_display *display = to_intel_display(crtc_state); struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); - struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); const struct i9xx_dpll_hw_state *hw_state = &crtc_state->dpll_hw_state.i9xx; enum dpio_channel ch = vlv_pipe_to_channel(crtc->pipe); enum dpio_phy phy = vlv_pipe_to_phy(crtc->pipe); enum pipe pipe = crtc->pipe; u32 tmp; - vlv_dpio_get(dev_priv); + vlv_dpio_get(display->drm); /* Enable back the 10bit clock to display controller */ - tmp = vlv_dpio_read(dev_priv, phy, CHV_CMN_DW14(ch)); + tmp = vlv_dpio_read(display->drm, phy, CHV_CMN_DW14(ch)); tmp |= DPIO_DCLKP_EN; - vlv_dpio_write(dev_priv, phy, CHV_CMN_DW14(ch), tmp); + vlv_dpio_write(display->drm, phy, CHV_CMN_DW14(ch), tmp); - vlv_dpio_put(dev_priv); + vlv_dpio_put(display->drm); /* * Need to wait > 100ns between dclkp clock enable bit and PLL enable. @@ -2252,7 +2244,6 @@ void vlv_disable_pll(struct intel_display *display, enum pipe pipe) void chv_disable_pll(struct intel_display *display, enum pipe pipe) { - struct drm_i915_private *dev_priv = to_i915(display->drm); enum dpio_channel ch = vlv_pipe_to_channel(pipe); enum dpio_phy phy = vlv_pipe_to_phy(pipe); u32 val; @@ -2268,14 +2259,14 @@ void chv_disable_pll(struct intel_display *display, enum pipe pipe) intel_de_write(display, DPLL(display, pipe), val); intel_de_posting_read(display, DPLL(display, pipe)); - vlv_dpio_get(dev_priv); + vlv_dpio_get(display->drm); /* Disable 10bit clock to display controller */ - val = vlv_dpio_read(dev_priv, phy, CHV_CMN_DW14(ch)); + val = vlv_dpio_read(display->drm, phy, CHV_CMN_DW14(ch)); val &= ~DPIO_DCLKP_EN; - vlv_dpio_write(dev_priv, phy, CHV_CMN_DW14(ch), val); + vlv_dpio_write(display->drm, phy, CHV_CMN_DW14(ch), val); - vlv_dpio_put(dev_priv); + vlv_dpio_put(display->drm); } void i9xx_disable_pll(const struct intel_crtc_state *crtc_state) diff --git a/drivers/gpu/drm/i915/display/intel_dpll.h b/drivers/gpu/drm/i915/display/intel_dpll.h index 280e90a57c87516522675c968db036e07b49d226..3444a2dd3166e5b73c1a23f8ebf5560ff152935c 100644 --- a/drivers/gpu/drm/i915/display/intel_dpll.h +++ b/drivers/gpu/drm/i915/display/intel_dpll.h @@ -19,8 +19,8 @@ struct intel_dpll_hw_state; void intel_dpll_init_clock_hook(struct intel_display *display); int intel_dpll_crtc_compute_clock(struct intel_atomic_state *state, struct intel_crtc *crtc); -int intel_dpll_crtc_get_shared_dpll(struct intel_atomic_state *state, - struct intel_crtc *crtc); +int intel_dpll_crtc_get_dpll(struct intel_atomic_state *state, + struct intel_crtc *crtc); int i9xx_calc_dpll_params(int refclk, struct dpll *clock); u32 i9xx_dpll_compute_fp(const struct dpll *dpll); void i9xx_dpll_get_hw_state(struct intel_crtc *crtc, diff --git a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c index 9da051a3f455020a6a53caee31ff0c9085a9022f..33e0398120c805e9889c11d573345080ef272ca3 100644 --- a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c +++ b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c @@ -27,10 +27,10 @@ #include #include "bxt_dpio_phy_regs.h" -#include "i915_reg.h" #include "i915_utils.h" #include "intel_cx0_phy.h" #include "intel_de.h" +#include "intel_display_regs.h" #include "intel_display_types.h" #include "intel_dkl_phy.h" #include "intel_dkl_phy_regs.h" @@ -52,34 +52,34 @@ * share a PLL if their configurations match. * * This file provides an abstraction over display PLLs. The function - * intel_shared_dpll_init() initializes the PLLs for the given platform. The + * intel_dpll_init() initializes the PLLs for the given platform. The * users of a PLL are tracked and that tracking is integrated with the atomic * modset interface. During an atomic operation, required PLLs can be reserved * for a given CRTC and encoder configuration by calling - * intel_reserve_shared_dplls() and previously reserved PLLs can be released - * with intel_release_shared_dplls(). + * intel_dpll_reserve() and previously reserved PLLs can be released + * with intel_dpll_release(). * Changes to the users are first staged in the atomic state, and then made - * effective by calling intel_shared_dpll_swap_state() during the atomic + * effective by calling intel_dpll_swap_state() during the atomic * commit phase. */ /* platform specific hooks for managing DPLLs */ -struct intel_shared_dpll_funcs { +struct intel_dpll_funcs { /* - * Hook for enabling the pll, called from intel_enable_shared_dpll() if + * Hook for enabling the pll, called from intel_enable_dpll() if * the pll is not already enabled. */ void (*enable)(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, const struct intel_dpll_hw_state *dpll_hw_state); /* - * Hook for disabling the pll, called from intel_disable_shared_dpll() + * Hook for disabling the pll, called from intel_disable_dpll() * only when it is safe to disable the pll, i.e., there are no more * tracked users for it. */ void (*disable)(struct intel_display *display, - struct intel_shared_dpll *pll); + struct intel_dpll *pll); /* * Hook for reading the values currently programmed to the DPLL @@ -87,7 +87,7 @@ struct intel_shared_dpll_funcs { * verification after a mode set. */ bool (*get_hw_state)(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, struct intel_dpll_hw_state *dpll_hw_state); /* @@ -95,7 +95,7 @@ struct intel_shared_dpll_funcs { * in state. */ int (*get_freq)(struct intel_display *i915, - const struct intel_shared_dpll *pll, + const struct intel_dpll *pll, const struct intel_dpll_hw_state *dpll_hw_state); }; @@ -122,18 +122,18 @@ struct intel_dpll_mgr { static void intel_atomic_duplicate_dpll_state(struct intel_display *display, - struct intel_shared_dpll_state *shared_dpll) + struct intel_dpll_state *dpll_state) { - struct intel_shared_dpll *pll; + struct intel_dpll *pll; int i; - /* Copy shared dpll state */ - for_each_shared_dpll(display, pll, i) - shared_dpll[pll->index] = pll->state; + /* Copy dpll state */ + for_each_dpll(display, pll, i) + dpll_state[pll->index] = pll->state; } -static struct intel_shared_dpll_state * -intel_atomic_get_shared_dpll_state(struct drm_atomic_state *s) +static struct intel_dpll_state * +intel_atomic_get_dpll_state(struct drm_atomic_state *s) { struct intel_atomic_state *state = to_intel_atomic_state(s); struct intel_display *display = to_intel_display(state); @@ -144,28 +144,28 @@ intel_atomic_get_shared_dpll_state(struct drm_atomic_state *s) state->dpll_set = true; intel_atomic_duplicate_dpll_state(display, - state->shared_dpll); + state->dpll_state); } - return state->shared_dpll; + return state->dpll_state; } /** - * intel_get_shared_dpll_by_id - get a DPLL given its id + * intel_get_dpll_by_id - get a DPLL given its id * @display: intel_display device instance * @id: pll id * * Returns: * A pointer to the DPLL with @id */ -struct intel_shared_dpll * -intel_get_shared_dpll_by_id(struct intel_display *display, - enum intel_dpll_id id) +struct intel_dpll * +intel_get_dpll_by_id(struct intel_display *display, + enum intel_dpll_id id) { - struct intel_shared_dpll *pll; + struct intel_dpll *pll; int i; - for_each_shared_dpll(display, pll, i) { + for_each_dpll(display, pll, i) { if (pll->info->id == id) return pll; } @@ -175,9 +175,9 @@ intel_get_shared_dpll_by_id(struct intel_display *display, } /* For ILK+ */ -void assert_shared_dpll(struct intel_display *display, - struct intel_shared_dpll *pll, - bool state) +void assert_dpll(struct intel_display *display, + struct intel_dpll *pll, + bool state) { bool cur_state; struct intel_dpll_hw_state hw_state; @@ -205,7 +205,7 @@ enum intel_dpll_id icl_tc_port_to_pll_id(enum tc_port tc_port) static i915_reg_t intel_combo_pll_enable_reg(struct intel_display *display, - struct intel_shared_dpll *pll) + struct intel_dpll *pll) { if (display->platform.dg1) return DG1_DPLL_ENABLE(pll->info->id); @@ -218,7 +218,7 @@ intel_combo_pll_enable_reg(struct intel_display *display, static i915_reg_t intel_tc_pll_enable_reg(struct intel_display *display, - struct intel_shared_dpll *pll) + struct intel_dpll *pll) { const enum intel_dpll_id id = pll->info->id; enum tc_port tc_port = icl_pll_id_to_tc_port(id); @@ -230,7 +230,7 @@ intel_tc_pll_enable_reg(struct intel_display *display, } static void _intel_enable_shared_dpll(struct intel_display *display, - struct intel_shared_dpll *pll) + struct intel_dpll *pll) { if (pll->info->power_domain) pll->wakeref = intel_display_power_get(display, pll->info->power_domain); @@ -240,7 +240,7 @@ static void _intel_enable_shared_dpll(struct intel_display *display, } static void _intel_disable_shared_dpll(struct intel_display *display, - struct intel_shared_dpll *pll) + struct intel_dpll *pll) { pll->info->funcs->disable(display, pll); pll->on = false; @@ -250,16 +250,16 @@ static void _intel_disable_shared_dpll(struct intel_display *display, } /** - * intel_enable_shared_dpll - enable a CRTC's shared DPLL - * @crtc_state: CRTC, and its state, which has a shared DPLL + * intel_dpll_enable - enable a CRTC's DPLL + * @crtc_state: CRTC, and its state, which has a DPLL * - * Enable the shared DPLL used by @crtc. + * Enable DPLL used by @crtc. */ -void intel_enable_shared_dpll(const struct intel_crtc_state *crtc_state) +void intel_dpll_enable(const struct intel_crtc_state *crtc_state) { struct intel_display *display = to_intel_display(crtc_state); struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); - struct intel_shared_dpll *pll = crtc_state->shared_dpll; + struct intel_dpll *pll = crtc_state->intel_dpll; unsigned int pipe_mask = intel_crtc_joined_pipe_mask(crtc_state); unsigned int old_mask; @@ -282,7 +282,7 @@ void intel_enable_shared_dpll(const struct intel_crtc_state *crtc_state) if (old_mask) { drm_WARN_ON(display->drm, !pll->on); - assert_shared_dpll_enabled(display, pll); + assert_dpll_enabled(display, pll); goto out; } drm_WARN_ON(display->drm, pll->on); @@ -296,16 +296,16 @@ out: } /** - * intel_disable_shared_dpll - disable a CRTC's shared DPLL + * intel_dpll_disable - disable a CRTC's shared DPLL * @crtc_state: CRTC, and its state, which has a shared DPLL * - * Disable the shared DPLL used by @crtc. + * Disable DPLL used by @crtc. */ -void intel_disable_shared_dpll(const struct intel_crtc_state *crtc_state) +void intel_dpll_disable(const struct intel_crtc_state *crtc_state) { struct intel_display *display = to_intel_display(crtc_state); struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); - struct intel_shared_dpll *pll = crtc_state->shared_dpll; + struct intel_dpll *pll = crtc_state->intel_dpll; unsigned int pipe_mask = intel_crtc_joined_pipe_mask(crtc_state); /* PCH only available on ILK+ */ @@ -326,7 +326,7 @@ void intel_disable_shared_dpll(const struct intel_crtc_state *crtc_state) pll->info->name, pll->active_mask, pll->on, crtc->base.base.id, crtc->base.name); - assert_shared_dpll_enabled(display, pll); + assert_dpll_enabled(display, pll); drm_WARN_ON(display->drm, !pll->on); pll->active_mask &= ~pipe_mask; @@ -344,11 +344,11 @@ out: static unsigned long intel_dpll_mask_all(struct intel_display *display) { - struct intel_shared_dpll *pll; + struct intel_dpll *pll; unsigned long dpll_mask = 0; int i; - for_each_shared_dpll(display, pll, i) { + for_each_dpll(display, pll, i) { drm_WARN_ON(display->drm, dpll_mask & BIT(pll->info->id)); dpll_mask |= BIT(pll->info->id); @@ -357,44 +357,44 @@ intel_dpll_mask_all(struct intel_display *display) return dpll_mask; } -static struct intel_shared_dpll * -intel_find_shared_dpll(struct intel_atomic_state *state, - const struct intel_crtc *crtc, - const struct intel_dpll_hw_state *dpll_hw_state, - unsigned long dpll_mask) +static struct intel_dpll * +intel_find_dpll(struct intel_atomic_state *state, + const struct intel_crtc *crtc, + const struct intel_dpll_hw_state *dpll_hw_state, + unsigned long dpll_mask) { struct intel_display *display = to_intel_display(crtc); unsigned long dpll_mask_all = intel_dpll_mask_all(display); - struct intel_shared_dpll_state *shared_dpll; - struct intel_shared_dpll *unused_pll = NULL; + struct intel_dpll_state *dpll_state; + struct intel_dpll *unused_pll = NULL; enum intel_dpll_id id; - shared_dpll = intel_atomic_get_shared_dpll_state(&state->base); + dpll_state = intel_atomic_get_dpll_state(&state->base); drm_WARN_ON(display->drm, dpll_mask & ~dpll_mask_all); for_each_set_bit(id, &dpll_mask, fls(dpll_mask_all)) { - struct intel_shared_dpll *pll; + struct intel_dpll *pll; - pll = intel_get_shared_dpll_by_id(display, id); + pll = intel_get_dpll_by_id(display, id); if (!pll) continue; /* Only want to check enabled timings first */ - if (shared_dpll[pll->index].pipe_mask == 0) { + if (dpll_state[pll->index].pipe_mask == 0) { if (!unused_pll) unused_pll = pll; continue; } if (memcmp(dpll_hw_state, - &shared_dpll[pll->index].hw_state, + &dpll_state[pll->index].hw_state, sizeof(*dpll_hw_state)) == 0) { drm_dbg_kms(display->drm, "[CRTC:%d:%s] sharing existing %s (pipe mask 0x%x, active 0x%x)\n", crtc->base.base.id, crtc->base.name, pll->info->name, - shared_dpll[pll->index].pipe_mask, + dpll_state[pll->index].pipe_mask, pll->active_mask); return pll; } @@ -412,76 +412,76 @@ intel_find_shared_dpll(struct intel_atomic_state *state, } /** - * intel_reference_shared_dpll_crtc - Get a DPLL reference for a CRTC + * intel_dpll_crtc_get - Get a DPLL reference for a CRTC * @crtc: CRTC on which behalf the reference is taken * @pll: DPLL for which the reference is taken - * @shared_dpll_state: the DPLL atomic state in which the reference is tracked + * @dpll_state: the DPLL atomic state in which the reference is tracked * * Take a reference for @pll tracking the use of it by @crtc. */ static void -intel_reference_shared_dpll_crtc(const struct intel_crtc *crtc, - const struct intel_shared_dpll *pll, - struct intel_shared_dpll_state *shared_dpll_state) +intel_dpll_crtc_get(const struct intel_crtc *crtc, + const struct intel_dpll *pll, + struct intel_dpll_state *dpll_state) { struct intel_display *display = to_intel_display(crtc); - drm_WARN_ON(display->drm, (shared_dpll_state->pipe_mask & BIT(crtc->pipe)) != 0); + drm_WARN_ON(display->drm, (dpll_state->pipe_mask & BIT(crtc->pipe)) != 0); - shared_dpll_state->pipe_mask |= BIT(crtc->pipe); + dpll_state->pipe_mask |= BIT(crtc->pipe); drm_dbg_kms(display->drm, "[CRTC:%d:%s] reserving %s\n", crtc->base.base.id, crtc->base.name, pll->info->name); } static void -intel_reference_shared_dpll(struct intel_atomic_state *state, - const struct intel_crtc *crtc, - const struct intel_shared_dpll *pll, - const struct intel_dpll_hw_state *dpll_hw_state) +intel_reference_dpll(struct intel_atomic_state *state, + const struct intel_crtc *crtc, + const struct intel_dpll *pll, + const struct intel_dpll_hw_state *dpll_hw_state) { - struct intel_shared_dpll_state *shared_dpll; + struct intel_dpll_state *dpll_state; - shared_dpll = intel_atomic_get_shared_dpll_state(&state->base); + dpll_state = intel_atomic_get_dpll_state(&state->base); - if (shared_dpll[pll->index].pipe_mask == 0) - shared_dpll[pll->index].hw_state = *dpll_hw_state; + if (dpll_state[pll->index].pipe_mask == 0) + dpll_state[pll->index].hw_state = *dpll_hw_state; - intel_reference_shared_dpll_crtc(crtc, pll, &shared_dpll[pll->index]); + intel_dpll_crtc_get(crtc, pll, &dpll_state[pll->index]); } /** - * intel_unreference_shared_dpll_crtc - Drop a DPLL reference for a CRTC + * intel_dpll_crtc_put - Drop a DPLL reference for a CRTC * @crtc: CRTC on which behalf the reference is dropped * @pll: DPLL for which the reference is dropped - * @shared_dpll_state: the DPLL atomic state in which the reference is tracked + * @dpll_state: the DPLL atomic state in which the reference is tracked * * Drop a reference for @pll tracking the end of use of it by @crtc. */ void -intel_unreference_shared_dpll_crtc(const struct intel_crtc *crtc, - const struct intel_shared_dpll *pll, - struct intel_shared_dpll_state *shared_dpll_state) +intel_dpll_crtc_put(const struct intel_crtc *crtc, + const struct intel_dpll *pll, + struct intel_dpll_state *dpll_state) { struct intel_display *display = to_intel_display(crtc); - drm_WARN_ON(display->drm, (shared_dpll_state->pipe_mask & BIT(crtc->pipe)) == 0); + drm_WARN_ON(display->drm, (dpll_state->pipe_mask & BIT(crtc->pipe)) == 0); - shared_dpll_state->pipe_mask &= ~BIT(crtc->pipe); + dpll_state->pipe_mask &= ~BIT(crtc->pipe); drm_dbg_kms(display->drm, "[CRTC:%d:%s] releasing %s\n", crtc->base.base.id, crtc->base.name, pll->info->name); } -static void intel_unreference_shared_dpll(struct intel_atomic_state *state, - const struct intel_crtc *crtc, - const struct intel_shared_dpll *pll) +static void intel_unreference_dpll(struct intel_atomic_state *state, + const struct intel_crtc *crtc, + const struct intel_dpll *pll) { - struct intel_shared_dpll_state *shared_dpll; + struct intel_dpll_state *dpll_state; - shared_dpll = intel_atomic_get_shared_dpll_state(&state->base); + dpll_state = intel_atomic_get_dpll_state(&state->base); - intel_unreference_shared_dpll_crtc(crtc, pll, &shared_dpll[pll->index]); + intel_dpll_crtc_put(crtc, pll, &dpll_state[pll->index]); } static void intel_put_dpll(struct intel_atomic_state *state, @@ -492,16 +492,16 @@ static void intel_put_dpll(struct intel_atomic_state *state, struct intel_crtc_state *new_crtc_state = intel_atomic_get_new_crtc_state(state, crtc); - new_crtc_state->shared_dpll = NULL; + new_crtc_state->intel_dpll = NULL; - if (!old_crtc_state->shared_dpll) + if (!old_crtc_state->intel_dpll) return; - intel_unreference_shared_dpll(state, crtc, old_crtc_state->shared_dpll); + intel_unreference_dpll(state, crtc, old_crtc_state->intel_dpll); } /** - * intel_shared_dpll_swap_state - make atomic DPLL configuration effective + * intel_dpll_swap_state - make atomic DPLL configuration effective * @state: atomic state * * This is the dpll version of drm_atomic_helper_swap_state() since the @@ -511,22 +511,22 @@ static void intel_put_dpll(struct intel_atomic_state *state, * i.e. it also puts the current state into @state, even though there is no * need for that at this moment. */ -void intel_shared_dpll_swap_state(struct intel_atomic_state *state) +void intel_dpll_swap_state(struct intel_atomic_state *state) { struct intel_display *display = to_intel_display(state); - struct intel_shared_dpll_state *shared_dpll = state->shared_dpll; - struct intel_shared_dpll *pll; + struct intel_dpll_state *dpll_state = state->dpll_state; + struct intel_dpll *pll; int i; if (!state->dpll_set) return; - for_each_shared_dpll(display, pll, i) - swap(pll->state, shared_dpll[pll->index]); + for_each_dpll(display, pll, i) + swap(pll->state, dpll_state[pll->index]); } static bool ibx_pch_dpll_get_hw_state(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, struct intel_dpll_hw_state *dpll_hw_state) { struct i9xx_dpll_hw_state *hw_state = &dpll_hw_state->i9xx; @@ -562,7 +562,7 @@ static void ibx_assert_pch_refclk_enabled(struct intel_display *display) } static void ibx_pch_dpll_enable(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, const struct intel_dpll_hw_state *dpll_hw_state) { const struct i9xx_dpll_hw_state *hw_state = &dpll_hw_state->i9xx; @@ -591,7 +591,7 @@ static void ibx_pch_dpll_enable(struct intel_display *display, } static void ibx_pch_dpll_disable(struct intel_display *display, - struct intel_shared_dpll *pll) + struct intel_dpll *pll) { const enum intel_dpll_id id = pll->info->id; @@ -614,33 +614,33 @@ static int ibx_get_dpll(struct intel_atomic_state *state, struct intel_display *display = to_intel_display(state); struct intel_crtc_state *crtc_state = intel_atomic_get_new_crtc_state(state, crtc); - struct intel_shared_dpll *pll; + struct intel_dpll *pll; enum intel_dpll_id id; if (HAS_PCH_IBX(display)) { /* Ironlake PCH has a fixed PLL->PCH pipe mapping. */ id = (enum intel_dpll_id) crtc->pipe; - pll = intel_get_shared_dpll_by_id(display, id); + pll = intel_get_dpll_by_id(display, id); drm_dbg_kms(display->drm, "[CRTC:%d:%s] using pre-allocated %s\n", crtc->base.base.id, crtc->base.name, pll->info->name); } else { - pll = intel_find_shared_dpll(state, crtc, - &crtc_state->dpll_hw_state, - BIT(DPLL_ID_PCH_PLL_B) | - BIT(DPLL_ID_PCH_PLL_A)); + pll = intel_find_dpll(state, crtc, + &crtc_state->dpll_hw_state, + BIT(DPLL_ID_PCH_PLL_B) | + BIT(DPLL_ID_PCH_PLL_A)); } if (!pll) return -EINVAL; /* reference the pll */ - intel_reference_shared_dpll(state, crtc, - pll, &crtc_state->dpll_hw_state); + intel_reference_dpll(state, crtc, + pll, &crtc_state->dpll_hw_state); - crtc_state->shared_dpll = pll; + crtc_state->intel_dpll = pll; return 0; } @@ -670,7 +670,7 @@ static bool ibx_compare_hw_state(const struct intel_dpll_hw_state *_a, a->fp1 == b->fp1; } -static const struct intel_shared_dpll_funcs ibx_pch_dpll_funcs = { +static const struct intel_dpll_funcs ibx_pch_dpll_funcs = { .enable = ibx_pch_dpll_enable, .disable = ibx_pch_dpll_disable, .get_hw_state = ibx_pch_dpll_get_hw_state, @@ -692,7 +692,7 @@ static const struct intel_dpll_mgr pch_pll_mgr = { }; static void hsw_ddi_wrpll_enable(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, const struct intel_dpll_hw_state *dpll_hw_state) { const struct hsw_dpll_hw_state *hw_state = &dpll_hw_state->hsw; @@ -704,7 +704,7 @@ static void hsw_ddi_wrpll_enable(struct intel_display *display, } static void hsw_ddi_spll_enable(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, const struct intel_dpll_hw_state *dpll_hw_state) { const struct hsw_dpll_hw_state *hw_state = &dpll_hw_state->hsw; @@ -715,7 +715,7 @@ static void hsw_ddi_spll_enable(struct intel_display *display, } static void hsw_ddi_wrpll_disable(struct intel_display *display, - struct intel_shared_dpll *pll) + struct intel_dpll *pll) { const enum intel_dpll_id id = pll->info->id; @@ -731,7 +731,7 @@ static void hsw_ddi_wrpll_disable(struct intel_display *display, } static void hsw_ddi_spll_disable(struct intel_display *display, - struct intel_shared_dpll *pll) + struct intel_dpll *pll) { enum intel_dpll_id id = pll->info->id; @@ -747,7 +747,7 @@ static void hsw_ddi_spll_disable(struct intel_display *display, } static bool hsw_ddi_wrpll_get_hw_state(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, struct intel_dpll_hw_state *dpll_hw_state) { struct hsw_dpll_hw_state *hw_state = &dpll_hw_state->hsw; @@ -769,7 +769,7 @@ static bool hsw_ddi_wrpll_get_hw_state(struct intel_display *display, } static bool hsw_ddi_spll_get_hw_state(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, struct intel_dpll_hw_state *dpll_hw_state) { struct hsw_dpll_hw_state *hw_state = &dpll_hw_state->hsw; @@ -996,7 +996,7 @@ hsw_ddi_calculate_wrpll(int clock /* in Hz */, } static int hsw_ddi_wrpll_get_freq(struct intel_display *display, - const struct intel_shared_dpll *pll, + const struct intel_dpll *pll, const struct intel_dpll_hw_state *dpll_hw_state) { const struct hsw_dpll_hw_state *hw_state = &dpll_hw_state->hsw; @@ -1059,14 +1059,14 @@ hsw_ddi_wrpll_compute_dpll(struct intel_atomic_state *state, return 0; } -static struct intel_shared_dpll * +static struct intel_dpll * hsw_ddi_wrpll_get_dpll(struct intel_atomic_state *state, struct intel_crtc *crtc) { struct intel_crtc_state *crtc_state = intel_atomic_get_new_crtc_state(state, crtc); - return intel_find_shared_dpll(state, crtc, + return intel_find_dpll(state, crtc, &crtc_state->dpll_hw_state, BIT(DPLL_ID_WRPLL2) | BIT(DPLL_ID_WRPLL1)); @@ -1090,11 +1090,11 @@ hsw_ddi_lcpll_compute_dpll(struct intel_crtc_state *crtc_state) } } -static struct intel_shared_dpll * +static struct intel_dpll * hsw_ddi_lcpll_get_dpll(struct intel_crtc_state *crtc_state) { struct intel_display *display = to_intel_display(crtc_state); - struct intel_shared_dpll *pll; + struct intel_dpll *pll; enum intel_dpll_id pll_id; int clock = crtc_state->port_clock; @@ -1113,7 +1113,7 @@ hsw_ddi_lcpll_get_dpll(struct intel_crtc_state *crtc_state) return NULL; } - pll = intel_get_shared_dpll_by_id(display, pll_id); + pll = intel_get_dpll_by_id(display, pll_id); if (!pll) return NULL; @@ -1122,7 +1122,7 @@ hsw_ddi_lcpll_get_dpll(struct intel_crtc_state *crtc_state) } static int hsw_ddi_lcpll_get_freq(struct intel_display *display, - const struct intel_shared_dpll *pll, + const struct intel_dpll *pll, const struct intel_dpll_hw_state *dpll_hw_state) { int link_clock = 0; @@ -1162,19 +1162,19 @@ hsw_ddi_spll_compute_dpll(struct intel_atomic_state *state, return 0; } -static struct intel_shared_dpll * +static struct intel_dpll * hsw_ddi_spll_get_dpll(struct intel_atomic_state *state, struct intel_crtc *crtc) { struct intel_crtc_state *crtc_state = intel_atomic_get_new_crtc_state(state, crtc); - return intel_find_shared_dpll(state, crtc, &crtc_state->dpll_hw_state, + return intel_find_dpll(state, crtc, &crtc_state->dpll_hw_state, BIT(DPLL_ID_SPLL)); } static int hsw_ddi_spll_get_freq(struct intel_display *display, - const struct intel_shared_dpll *pll, + const struct intel_dpll *pll, const struct intel_dpll_hw_state *dpll_hw_state) { const struct hsw_dpll_hw_state *hw_state = &dpll_hw_state->hsw; @@ -1221,7 +1221,7 @@ static int hsw_get_dpll(struct intel_atomic_state *state, { struct intel_crtc_state *crtc_state = intel_atomic_get_new_crtc_state(state, crtc); - struct intel_shared_dpll *pll = NULL; + struct intel_dpll *pll = NULL; if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI)) pll = hsw_ddi_wrpll_get_dpll(state, crtc); @@ -1233,10 +1233,10 @@ static int hsw_get_dpll(struct intel_atomic_state *state, if (!pll) return -EINVAL; - intel_reference_shared_dpll(state, crtc, - pll, &crtc_state->dpll_hw_state); + intel_reference_dpll(state, crtc, + pll, &crtc_state->dpll_hw_state); - crtc_state->shared_dpll = pll; + crtc_state->intel_dpll = pll; return 0; } @@ -1270,14 +1270,14 @@ static bool hsw_compare_hw_state(const struct intel_dpll_hw_state *_a, a->spll == b->spll; } -static const struct intel_shared_dpll_funcs hsw_ddi_wrpll_funcs = { +static const struct intel_dpll_funcs hsw_ddi_wrpll_funcs = { .enable = hsw_ddi_wrpll_enable, .disable = hsw_ddi_wrpll_disable, .get_hw_state = hsw_ddi_wrpll_get_hw_state, .get_freq = hsw_ddi_wrpll_get_freq, }; -static const struct intel_shared_dpll_funcs hsw_ddi_spll_funcs = { +static const struct intel_dpll_funcs hsw_ddi_spll_funcs = { .enable = hsw_ddi_spll_enable, .disable = hsw_ddi_spll_disable, .get_hw_state = hsw_ddi_spll_get_hw_state, @@ -1285,24 +1285,24 @@ static const struct intel_shared_dpll_funcs hsw_ddi_spll_funcs = { }; static void hsw_ddi_lcpll_enable(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, const struct intel_dpll_hw_state *hw_state) { } static void hsw_ddi_lcpll_disable(struct intel_display *display, - struct intel_shared_dpll *pll) + struct intel_dpll *pll) { } static bool hsw_ddi_lcpll_get_hw_state(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, struct intel_dpll_hw_state *dpll_hw_state) { return true; } -static const struct intel_shared_dpll_funcs hsw_ddi_lcpll_funcs = { +static const struct intel_dpll_funcs hsw_ddi_lcpll_funcs = { .enable = hsw_ddi_lcpll_enable, .disable = hsw_ddi_lcpll_disable, .get_hw_state = hsw_ddi_lcpll_get_hw_state, @@ -1364,7 +1364,7 @@ static const struct skl_dpll_regs skl_dpll_regs[4] = { }; static void skl_ddi_pll_write_ctrl1(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, const struct skl_dpll_hw_state *hw_state) { const enum intel_dpll_id id = pll->info->id; @@ -1378,7 +1378,7 @@ static void skl_ddi_pll_write_ctrl1(struct intel_display *display, } static void skl_ddi_pll_enable(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, const struct intel_dpll_hw_state *dpll_hw_state) { const struct skl_dpll_hw_state *hw_state = &dpll_hw_state->skl; @@ -1400,7 +1400,7 @@ static void skl_ddi_pll_enable(struct intel_display *display, } static void skl_ddi_dpll0_enable(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, const struct intel_dpll_hw_state *dpll_hw_state) { const struct skl_dpll_hw_state *hw_state = &dpll_hw_state->skl; @@ -1409,7 +1409,7 @@ static void skl_ddi_dpll0_enable(struct intel_display *display, } static void skl_ddi_pll_disable(struct intel_display *display, - struct intel_shared_dpll *pll) + struct intel_dpll *pll) { const struct skl_dpll_regs *regs = skl_dpll_regs; const enum intel_dpll_id id = pll->info->id; @@ -1420,12 +1420,12 @@ static void skl_ddi_pll_disable(struct intel_display *display, } static void skl_ddi_dpll0_disable(struct intel_display *display, - struct intel_shared_dpll *pll) + struct intel_dpll *pll) { } static bool skl_ddi_pll_get_hw_state(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, struct intel_dpll_hw_state *dpll_hw_state) { struct skl_dpll_hw_state *hw_state = &dpll_hw_state->skl; @@ -1463,7 +1463,7 @@ out: } static bool skl_ddi_dpll0_get_hw_state(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, struct intel_dpll_hw_state *dpll_hw_state) { struct skl_dpll_hw_state *hw_state = &dpll_hw_state->skl; @@ -1736,7 +1736,7 @@ skip_remaining_dividers: } static int skl_ddi_wrpll_get_freq(struct intel_display *display, - const struct intel_shared_dpll *pll, + const struct intel_dpll *pll, const struct intel_dpll_hw_state *dpll_hw_state) { const struct skl_dpll_hw_state *hw_state = &dpll_hw_state->skl; @@ -1884,7 +1884,7 @@ skl_ddi_dp_set_dpll_hw_state(struct intel_crtc_state *crtc_state) } static int skl_ddi_lcpll_get_freq(struct intel_display *display, - const struct intel_shared_dpll *pll, + const struct intel_dpll *pll, const struct intel_dpll_hw_state *dpll_hw_state) { const struct skl_dpll_hw_state *hw_state = &dpll_hw_state->skl; @@ -1939,31 +1939,31 @@ static int skl_get_dpll(struct intel_atomic_state *state, { struct intel_crtc_state *crtc_state = intel_atomic_get_new_crtc_state(state, crtc); - struct intel_shared_dpll *pll; + struct intel_dpll *pll; if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_EDP)) - pll = intel_find_shared_dpll(state, crtc, - &crtc_state->dpll_hw_state, - BIT(DPLL_ID_SKL_DPLL0)); + pll = intel_find_dpll(state, crtc, + &crtc_state->dpll_hw_state, + BIT(DPLL_ID_SKL_DPLL0)); else - pll = intel_find_shared_dpll(state, crtc, - &crtc_state->dpll_hw_state, - BIT(DPLL_ID_SKL_DPLL3) | - BIT(DPLL_ID_SKL_DPLL2) | - BIT(DPLL_ID_SKL_DPLL1)); + pll = intel_find_dpll(state, crtc, + &crtc_state->dpll_hw_state, + BIT(DPLL_ID_SKL_DPLL3) | + BIT(DPLL_ID_SKL_DPLL2) | + BIT(DPLL_ID_SKL_DPLL1)); if (!pll) return -EINVAL; - intel_reference_shared_dpll(state, crtc, - pll, &crtc_state->dpll_hw_state); + intel_reference_dpll(state, crtc, + pll, &crtc_state->dpll_hw_state); - crtc_state->shared_dpll = pll; + crtc_state->intel_dpll = pll; return 0; } static int skl_ddi_pll_get_freq(struct intel_display *display, - const struct intel_shared_dpll *pll, + const struct intel_dpll *pll, const struct intel_dpll_hw_state *dpll_hw_state) { const struct skl_dpll_hw_state *hw_state = &dpll_hw_state->skl; @@ -2004,14 +2004,14 @@ static bool skl_compare_hw_state(const struct intel_dpll_hw_state *_a, a->cfgcr2 == b->cfgcr2; } -static const struct intel_shared_dpll_funcs skl_ddi_pll_funcs = { +static const struct intel_dpll_funcs skl_ddi_pll_funcs = { .enable = skl_ddi_pll_enable, .disable = skl_ddi_pll_disable, .get_hw_state = skl_ddi_pll_get_hw_state, .get_freq = skl_ddi_pll_get_freq, }; -static const struct intel_shared_dpll_funcs skl_ddi_dpll0_funcs = { +static const struct intel_dpll_funcs skl_ddi_dpll0_funcs = { .enable = skl_ddi_dpll0_enable, .disable = skl_ddi_dpll0_disable, .get_hw_state = skl_ddi_dpll0_get_hw_state, @@ -2038,7 +2038,7 @@ static const struct intel_dpll_mgr skl_pll_mgr = { }; static void bxt_ddi_pll_enable(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, const struct intel_dpll_hw_state *dpll_hw_state) { const struct bxt_dpll_hw_state *hw_state = &dpll_hw_state->bxt; @@ -2141,7 +2141,7 @@ static void bxt_ddi_pll_enable(struct intel_display *display, } static void bxt_ddi_pll_disable(struct intel_display *display, - struct intel_shared_dpll *pll) + struct intel_dpll *pll) { enum port port = (enum port)pll->info->id; /* 1:1 port->PLL mapping */ @@ -2160,7 +2160,7 @@ static void bxt_ddi_pll_disable(struct intel_display *display, } static bool bxt_ddi_pll_get_hw_state(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, struct intel_dpll_hw_state *dpll_hw_state) { struct bxt_dpll_hw_state *hw_state = &dpll_hw_state->bxt; @@ -2360,7 +2360,7 @@ static int bxt_ddi_set_dpll_hw_state(struct intel_crtc_state *crtc_state, } static int bxt_ddi_pll_get_freq(struct intel_display *display, - const struct intel_shared_dpll *pll, + const struct intel_dpll *pll, const struct intel_dpll_hw_state *dpll_hw_state) { const struct bxt_dpll_hw_state *hw_state = &dpll_hw_state->bxt; @@ -2429,20 +2429,20 @@ static int bxt_get_dpll(struct intel_atomic_state *state, struct intel_display *display = to_intel_display(state); struct intel_crtc_state *crtc_state = intel_atomic_get_new_crtc_state(state, crtc); - struct intel_shared_dpll *pll; + struct intel_dpll *pll; enum intel_dpll_id id; /* 1:1 mapping between ports and PLLs */ id = (enum intel_dpll_id) encoder->port; - pll = intel_get_shared_dpll_by_id(display, id); + pll = intel_get_dpll_by_id(display, id); drm_dbg_kms(display->drm, "[CRTC:%d:%s] using pre-allocated %s\n", crtc->base.base.id, crtc->base.name, pll->info->name); - intel_reference_shared_dpll(state, crtc, - pll, &crtc_state->dpll_hw_state); + intel_reference_dpll(state, crtc, + pll, &crtc_state->dpll_hw_state); - crtc_state->shared_dpll = pll; + crtc_state->intel_dpll = pll; return 0; } @@ -2486,7 +2486,7 @@ static bool bxt_compare_hw_state(const struct intel_dpll_hw_state *_a, a->pcsdw12 == b->pcsdw12; } -static const struct intel_shared_dpll_funcs bxt_ddi_pll_funcs = { +static const struct intel_dpll_funcs bxt_ddi_pll_funcs = { .enable = bxt_ddi_pll_enable, .disable = bxt_ddi_pll_disable, .get_hw_state = bxt_ddi_pll_get_hw_state, @@ -2755,7 +2755,7 @@ static int icl_calc_tbt_pll(struct intel_crtc_state *crtc_state, } static int icl_ddi_tbt_pll_get_freq(struct intel_display *display, - const struct intel_shared_dpll *pll, + const struct intel_dpll *pll, const struct intel_dpll_hw_state *dpll_hw_state) { /* @@ -2826,7 +2826,7 @@ icl_calc_wrpll(struct intel_crtc_state *crtc_state, } static int icl_ddi_combo_pll_get_freq(struct intel_display *display, - const struct intel_shared_dpll *pll, + const struct intel_dpll *pll, const struct intel_dpll_hw_state *dpll_hw_state) { const struct icl_dpll_hw_state *hw_state = &dpll_hw_state->icl; @@ -3199,7 +3199,7 @@ static int icl_calc_mg_pll_state(struct intel_crtc_state *crtc_state, } static int icl_ddi_mg_pll_get_freq(struct intel_display *display, - const struct intel_shared_dpll *pll, + const struct intel_dpll *pll, const struct intel_dpll_hw_state *dpll_hw_state) { const struct icl_dpll_hw_state *hw_state = &dpll_hw_state->icl; @@ -3285,7 +3285,7 @@ void icl_set_active_port_dpll(struct intel_crtc_state *crtc_state, struct icl_port_dpll *port_dpll = &crtc_state->icl_port_dplls[port_dpll_id]; - crtc_state->shared_dpll = port_dpll->pll; + crtc_state->intel_dpll = port_dpll->pll; crtc_state->dpll_hw_state = port_dpll->hw_state; } @@ -3388,14 +3388,14 @@ static int icl_get_combo_phy_dpll(struct intel_atomic_state *state, /* Eliminate DPLLs from consideration if reserved by HTI */ dpll_mask &= ~intel_hti_dpll_mask(display); - port_dpll->pll = intel_find_shared_dpll(state, crtc, - &port_dpll->hw_state, - dpll_mask); + port_dpll->pll = intel_find_dpll(state, crtc, + &port_dpll->hw_state, + dpll_mask); if (!port_dpll->pll) return -EINVAL; - intel_reference_shared_dpll(state, crtc, - port_dpll->pll, &port_dpll->hw_state); + intel_reference_dpll(state, crtc, + port_dpll->pll, &port_dpll->hw_state); icl_update_active_dpll(state, crtc, encoder); @@ -3428,8 +3428,8 @@ static int icl_compute_tc_phy_dplls(struct intel_atomic_state *state, return ret; /* this is mainly for the fastset check */ - if (old_crtc_state->shared_dpll && - old_crtc_state->shared_dpll->info->id == DPLL_ID_ICL_TBTPLL) + if (old_crtc_state->intel_dpll && + old_crtc_state->intel_dpll->info->id == DPLL_ID_ICL_TBTPLL) icl_set_active_port_dpll(crtc_state, ICL_PORT_DPLL_DEFAULT); else icl_set_active_port_dpll(crtc_state, ICL_PORT_DPLL_MG_PHY); @@ -3452,26 +3452,25 @@ static int icl_get_tc_phy_dplls(struct intel_atomic_state *state, int ret; port_dpll = &crtc_state->icl_port_dplls[ICL_PORT_DPLL_DEFAULT]; - port_dpll->pll = intel_find_shared_dpll(state, crtc, - &port_dpll->hw_state, - BIT(DPLL_ID_ICL_TBTPLL)); + port_dpll->pll = intel_find_dpll(state, crtc, + &port_dpll->hw_state, + BIT(DPLL_ID_ICL_TBTPLL)); if (!port_dpll->pll) return -EINVAL; - intel_reference_shared_dpll(state, crtc, - port_dpll->pll, &port_dpll->hw_state); - + intel_reference_dpll(state, crtc, + port_dpll->pll, &port_dpll->hw_state); port_dpll = &crtc_state->icl_port_dplls[ICL_PORT_DPLL_MG_PHY]; dpll_id = icl_tc_port_to_pll_id(intel_encoder_to_tc(encoder)); - port_dpll->pll = intel_find_shared_dpll(state, crtc, - &port_dpll->hw_state, - BIT(dpll_id)); + port_dpll->pll = intel_find_dpll(state, crtc, + &port_dpll->hw_state, + BIT(dpll_id)); if (!port_dpll->pll) { ret = -EINVAL; goto err_unreference_tbt_pll; } - intel_reference_shared_dpll(state, crtc, - port_dpll->pll, &port_dpll->hw_state); + intel_reference_dpll(state, crtc, + port_dpll->pll, &port_dpll->hw_state); icl_update_active_dpll(state, crtc, encoder); @@ -3479,7 +3478,7 @@ static int icl_get_tc_phy_dplls(struct intel_atomic_state *state, err_unreference_tbt_pll: port_dpll = &crtc_state->icl_port_dplls[ICL_PORT_DPLL_DEFAULT]; - intel_unreference_shared_dpll(state, crtc, port_dpll->pll); + intel_unreference_dpll(state, crtc, port_dpll->pll); return ret; } @@ -3521,7 +3520,7 @@ static void icl_put_dplls(struct intel_atomic_state *state, intel_atomic_get_new_crtc_state(state, crtc); enum icl_port_dpll_id id; - new_crtc_state->shared_dpll = NULL; + new_crtc_state->intel_dpll = NULL; for (id = ICL_PORT_DPLL_DEFAULT; id < ICL_PORT_DPLL_COUNT; id++) { const struct icl_port_dpll *old_port_dpll = @@ -3534,12 +3533,12 @@ static void icl_put_dplls(struct intel_atomic_state *state, if (!old_port_dpll->pll) continue; - intel_unreference_shared_dpll(state, crtc, old_port_dpll->pll); + intel_unreference_dpll(state, crtc, old_port_dpll->pll); } } static bool mg_pll_get_hw_state(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, struct intel_dpll_hw_state *dpll_hw_state) { struct icl_dpll_hw_state *hw_state = &dpll_hw_state->icl; @@ -3606,7 +3605,7 @@ out: } static bool dkl_pll_get_hw_state(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, struct intel_dpll_hw_state *dpll_hw_state) { struct icl_dpll_hw_state *hw_state = &dpll_hw_state->icl; @@ -3678,7 +3677,7 @@ out: } static bool icl_pll_get_hw_state(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, struct intel_dpll_hw_state *dpll_hw_state, i915_reg_t enable_reg) { @@ -3739,7 +3738,7 @@ out: } static bool combo_pll_get_hw_state(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, struct intel_dpll_hw_state *dpll_hw_state) { i915_reg_t enable_reg = intel_combo_pll_enable_reg(display, pll); @@ -3748,14 +3747,14 @@ static bool combo_pll_get_hw_state(struct intel_display *display, } static bool tbt_pll_get_hw_state(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, struct intel_dpll_hw_state *dpll_hw_state) { return icl_pll_get_hw_state(display, pll, dpll_hw_state, TBT_PLL_ENABLE); } static void icl_dpll_write(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, const struct icl_dpll_hw_state *hw_state) { const enum intel_dpll_id id = pll->info->id; @@ -3797,7 +3796,7 @@ static void icl_dpll_write(struct intel_display *display, } static void icl_mg_pll_write(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, const struct icl_dpll_hw_state *hw_state) { enum tc_port tc_port = icl_pll_id_to_tc_port(pll->info->id); @@ -3840,7 +3839,7 @@ static void icl_mg_pll_write(struct intel_display *display, } static void dkl_pll_write(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, const struct icl_dpll_hw_state *hw_state) { enum tc_port tc_port = icl_pll_id_to_tc_port(pll->info->id); @@ -3905,7 +3904,7 @@ static void dkl_pll_write(struct intel_display *display, } static void icl_pll_power_enable(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, i915_reg_t enable_reg) { intel_de_rmw(display, enable_reg, 0, PLL_POWER_ENABLE); @@ -3920,7 +3919,7 @@ static void icl_pll_power_enable(struct intel_display *display, } static void icl_pll_enable(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, i915_reg_t enable_reg) { intel_de_rmw(display, enable_reg, 0, PLL_ENABLE); @@ -3930,7 +3929,7 @@ static void icl_pll_enable(struct intel_display *display, drm_err(display->drm, "PLL %d not locked\n", pll->info->id); } -static void adlp_cmtg_clock_gating_wa(struct intel_display *display, struct intel_shared_dpll *pll) +static void adlp_cmtg_clock_gating_wa(struct intel_display *display, struct intel_dpll *pll) { u32 val; @@ -3955,7 +3954,7 @@ static void adlp_cmtg_clock_gating_wa(struct intel_display *display, struct inte } static void combo_pll_enable(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, const struct intel_dpll_hw_state *dpll_hw_state) { const struct icl_dpll_hw_state *hw_state = &dpll_hw_state->icl; @@ -3979,7 +3978,7 @@ static void combo_pll_enable(struct intel_display *display, } static void tbt_pll_enable(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, const struct intel_dpll_hw_state *dpll_hw_state) { const struct icl_dpll_hw_state *hw_state = &dpll_hw_state->icl; @@ -4000,7 +3999,7 @@ static void tbt_pll_enable(struct intel_display *display, } static void mg_pll_enable(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, const struct intel_dpll_hw_state *dpll_hw_state) { const struct icl_dpll_hw_state *hw_state = &dpll_hw_state->icl; @@ -4025,7 +4024,7 @@ static void mg_pll_enable(struct intel_display *display, } static void icl_pll_disable(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, i915_reg_t enable_reg) { /* The first steps are done by intel_ddi_post_disable(). */ @@ -4056,7 +4055,7 @@ static void icl_pll_disable(struct intel_display *display, } static void combo_pll_disable(struct intel_display *display, - struct intel_shared_dpll *pll) + struct intel_dpll *pll) { i915_reg_t enable_reg = intel_combo_pll_enable_reg(display, pll); @@ -4064,13 +4063,13 @@ static void combo_pll_disable(struct intel_display *display, } static void tbt_pll_disable(struct intel_display *display, - struct intel_shared_dpll *pll) + struct intel_dpll *pll) { icl_pll_disable(display, pll, TBT_PLL_ENABLE); } static void mg_pll_disable(struct intel_display *display, - struct intel_shared_dpll *pll) + struct intel_dpll *pll) { i915_reg_t enable_reg = intel_tc_pll_enable_reg(display, pll); @@ -4129,21 +4128,21 @@ static bool icl_compare_hw_state(const struct intel_dpll_hw_state *_a, a->mg_pll_tdc_coldst_bias == b->mg_pll_tdc_coldst_bias; } -static const struct intel_shared_dpll_funcs combo_pll_funcs = { +static const struct intel_dpll_funcs combo_pll_funcs = { .enable = combo_pll_enable, .disable = combo_pll_disable, .get_hw_state = combo_pll_get_hw_state, .get_freq = icl_ddi_combo_pll_get_freq, }; -static const struct intel_shared_dpll_funcs tbt_pll_funcs = { +static const struct intel_dpll_funcs tbt_pll_funcs = { .enable = tbt_pll_enable, .disable = tbt_pll_disable, .get_hw_state = tbt_pll_get_hw_state, .get_freq = icl_ddi_tbt_pll_get_freq, }; -static const struct intel_shared_dpll_funcs mg_pll_funcs = { +static const struct intel_dpll_funcs mg_pll_funcs = { .enable = mg_pll_enable, .disable = mg_pll_disable, .get_hw_state = mg_pll_get_hw_state, @@ -4191,7 +4190,7 @@ static const struct intel_dpll_mgr ehl_pll_mgr = { .compare_hw_state = icl_compare_hw_state, }; -static const struct intel_shared_dpll_funcs dkl_pll_funcs = { +static const struct intel_dpll_funcs dkl_pll_funcs = { .enable = mg_pll_enable, .disable = mg_pll_disable, .get_hw_state = dkl_pll_get_hw_state, @@ -4300,12 +4299,12 @@ static const struct intel_dpll_mgr adlp_pll_mgr = { }; /** - * intel_shared_dpll_init - Initialize shared DPLLs + * intel_dpll_init - Initialize DPLLs * @display: intel_display device * - * Initialize shared DPLLs for @display. + * Initialize DPLLs for @display. */ -void intel_shared_dpll_init(struct intel_display *display) +void intel_dpll_init(struct intel_display *display) { const struct intel_dpll_mgr *dpll_mgr = NULL; const struct dpll_info *dpll_info; @@ -4346,23 +4345,23 @@ void intel_shared_dpll_init(struct intel_display *display) for (i = 0; dpll_info[i].name; i++) { if (drm_WARN_ON(display->drm, - i >= ARRAY_SIZE(display->dpll.shared_dplls))) + i >= ARRAY_SIZE(display->dpll.dplls))) break; /* must fit into unsigned long bitmask on 32bit */ if (drm_WARN_ON(display->drm, dpll_info[i].id >= 32)) break; - display->dpll.shared_dplls[i].info = &dpll_info[i]; - display->dpll.shared_dplls[i].index = i; + display->dpll.dplls[i].info = &dpll_info[i]; + display->dpll.dplls[i].index = i; } display->dpll.mgr = dpll_mgr; - display->dpll.num_shared_dpll = i; + display->dpll.num_dpll = i; } /** - * intel_compute_shared_dplls - compute DPLL state CRTC and encoder combination + * intel_dpll_compute - compute DPLL state CRTC and encoder combination * @state: atomic state * @crtc: CRTC to compute DPLLs for * @encoder: encoder @@ -4370,14 +4369,14 @@ void intel_shared_dpll_init(struct intel_display *display) * This function computes the DPLL state for the given CRTC and encoder. * * The new configuration in the atomic commit @state is made effective by - * calling intel_shared_dpll_swap_state(). + * calling intel_dpll_swap_state(). * * Returns: * 0 on success, negative error code on failure. */ -int intel_compute_shared_dplls(struct intel_atomic_state *state, - struct intel_crtc *crtc, - struct intel_encoder *encoder) +int intel_dpll_compute(struct intel_atomic_state *state, + struct intel_crtc *crtc, + struct intel_encoder *encoder) { struct intel_display *display = to_intel_display(state); const struct intel_dpll_mgr *dpll_mgr = display->dpll.mgr; @@ -4389,7 +4388,7 @@ int intel_compute_shared_dplls(struct intel_atomic_state *state, } /** - * intel_reserve_shared_dplls - reserve DPLLs for CRTC and encoder combination + * intel_dpll_reserve - reserve DPLLs for CRTC and encoder combination * @state: atomic state * @crtc: CRTC to reserve DPLLs for * @encoder: encoder @@ -4399,18 +4398,18 @@ int intel_compute_shared_dplls(struct intel_atomic_state *state, * state. * * The new configuration in the atomic commit @state is made effective by - * calling intel_shared_dpll_swap_state(). + * calling intel_dpll_swap_state(). * * The reserved DPLLs should be released by calling - * intel_release_shared_dplls(). + * intel_dpll_release(). * * Returns: * 0 if all required DPLLs were successfully reserved, * negative error code otherwise. */ -int intel_reserve_shared_dplls(struct intel_atomic_state *state, - struct intel_crtc *crtc, - struct intel_encoder *encoder) +int intel_dpll_reserve(struct intel_atomic_state *state, + struct intel_crtc *crtc, + struct intel_encoder *encoder) { struct intel_display *display = to_intel_display(state); const struct intel_dpll_mgr *dpll_mgr = display->dpll.mgr; @@ -4422,18 +4421,18 @@ int intel_reserve_shared_dplls(struct intel_atomic_state *state, } /** - * intel_release_shared_dplls - end use of DPLLs by CRTC in atomic state + * intel_dpll_release - end use of DPLLs by CRTC in atomic state * @state: atomic state * @crtc: crtc from which the DPLLs are to be released * - * This function releases all DPLLs reserved by intel_reserve_shared_dplls() + * This function releases all DPLLs reserved by intel_dpll_reserve() * from the current atomic commit @state and the old @crtc atomic state. * * The new configuration in the atomic commit @state is made effective by - * calling intel_shared_dpll_swap_state(). + * calling intel_dpll_swap_state(). */ -void intel_release_shared_dplls(struct intel_atomic_state *state, - struct intel_crtc *crtc) +void intel_dpll_release(struct intel_atomic_state *state, + struct intel_crtc *crtc) { struct intel_display *display = to_intel_display(state); const struct intel_dpll_mgr *dpll_mgr = display->dpll.mgr; @@ -4441,7 +4440,7 @@ void intel_release_shared_dplls(struct intel_atomic_state *state, /* * FIXME: this function is called for every platform having a * compute_clock hook, even though the platform doesn't yet support - * the shared DPLL framework and intel_reserve_shared_dplls() is not + * the DPLL framework and intel_dpll_reserve() is not * called on those. */ if (!dpll_mgr) @@ -4451,16 +4450,16 @@ void intel_release_shared_dplls(struct intel_atomic_state *state, } /** - * intel_update_active_dpll - update the active DPLL for a CRTC/encoder + * intel_dpll_update_active - update the active DPLL for a CRTC/encoder * @state: atomic state * @crtc: the CRTC for which to update the active DPLL * @encoder: encoder determining the type of port DPLL * * Update the active DPLL for the given @crtc/@encoder in @crtc's atomic state, - * from the port DPLLs reserved previously by intel_reserve_shared_dplls(). The + * from the port DPLLs reserved previously by intel_dpll_reserve(). The * DPLL selected will be based on the current mode of the encoder's port. */ -void intel_update_active_dpll(struct intel_atomic_state *state, +void intel_dpll_update_active(struct intel_atomic_state *state, struct intel_crtc *crtc, struct intel_encoder *encoder) { @@ -4482,7 +4481,7 @@ void intel_update_active_dpll(struct intel_atomic_state *state, * Return the output frequency corresponding to @pll's passed in @dpll_hw_state. */ int intel_dpll_get_freq(struct intel_display *display, - const struct intel_shared_dpll *pll, + const struct intel_dpll *pll, const struct intel_dpll_hw_state *dpll_hw_state) { if (drm_WARN_ON(display->drm, !pll->info->funcs->get_freq)) @@ -4500,14 +4499,14 @@ int intel_dpll_get_freq(struct intel_display *display, * Read out @pll's hardware state into @dpll_hw_state. */ bool intel_dpll_get_hw_state(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, struct intel_dpll_hw_state *dpll_hw_state) { return pll->info->funcs->get_hw_state(display, pll, dpll_hw_state); } static void readout_dpll_hw_state(struct intel_display *display, - struct intel_shared_dpll *pll) + struct intel_dpll *pll) { struct intel_crtc *crtc; @@ -4521,8 +4520,8 @@ static void readout_dpll_hw_state(struct intel_display *display, struct intel_crtc_state *crtc_state = to_intel_crtc_state(crtc->base.state); - if (crtc_state->hw.active && crtc_state->shared_dpll == pll) - intel_reference_shared_dpll_crtc(crtc, pll, &pll->state); + if (crtc_state->hw.active && crtc_state->intel_dpll == pll) + intel_dpll_crtc_get(crtc, pll, &pll->state); } pll->active_mask = pll->state.pipe_mask; @@ -4539,15 +4538,15 @@ void intel_dpll_update_ref_clks(struct intel_display *display) void intel_dpll_readout_hw_state(struct intel_display *display) { - struct intel_shared_dpll *pll; + struct intel_dpll *pll; int i; - for_each_shared_dpll(display, pll, i) + for_each_dpll(display, pll, i) readout_dpll_hw_state(display, pll); } static void sanitize_dpll_state(struct intel_display *display, - struct intel_shared_dpll *pll) + struct intel_dpll *pll) { if (!pll->on) return; @@ -4566,12 +4565,12 @@ static void sanitize_dpll_state(struct intel_display *display, void intel_dpll_sanitize_state(struct intel_display *display) { - struct intel_shared_dpll *pll; + struct intel_dpll *pll; int i; intel_cx0_pll_power_save_wa(display); - for_each_shared_dpll(display, pll, i) + for_each_dpll(display, pll, i) sanitize_dpll_state(display, pll); } @@ -4623,7 +4622,7 @@ bool intel_dpll_compare_hw_state(struct intel_display *display, static void verify_single_dpll_state(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, struct intel_crtc *crtc, const struct intel_crtc_state *new_crtc_state) { @@ -4676,15 +4675,15 @@ verify_single_dpll_state(struct intel_display *display, pll->info->name); } -static bool has_alt_port_dpll(const struct intel_shared_dpll *old_pll, - const struct intel_shared_dpll *new_pll) +static bool has_alt_port_dpll(const struct intel_dpll *old_pll, + const struct intel_dpll *new_pll) { return old_pll && new_pll && old_pll != new_pll && (old_pll->info->is_alt_port_dpll || new_pll->info->is_alt_port_dpll); } -void intel_shared_dpll_state_verify(struct intel_atomic_state *state, - struct intel_crtc *crtc) +void intel_dpll_state_verify(struct intel_atomic_state *state, + struct intel_crtc *crtc) { struct intel_display *display = to_intel_display(state); const struct intel_crtc_state *old_crtc_state = @@ -4692,34 +4691,34 @@ void intel_shared_dpll_state_verify(struct intel_atomic_state *state, const struct intel_crtc_state *new_crtc_state = intel_atomic_get_new_crtc_state(state, crtc); - if (new_crtc_state->shared_dpll) - verify_single_dpll_state(display, new_crtc_state->shared_dpll, + if (new_crtc_state->intel_dpll) + verify_single_dpll_state(display, new_crtc_state->intel_dpll, crtc, new_crtc_state); - if (old_crtc_state->shared_dpll && - old_crtc_state->shared_dpll != new_crtc_state->shared_dpll) { + if (old_crtc_state->intel_dpll && + old_crtc_state->intel_dpll != new_crtc_state->intel_dpll) { u8 pipe_mask = BIT(crtc->pipe); - struct intel_shared_dpll *pll = old_crtc_state->shared_dpll; + struct intel_dpll *pll = old_crtc_state->intel_dpll; INTEL_DISPLAY_STATE_WARN(display, pll->active_mask & pipe_mask, "%s: pll active mismatch (didn't expect pipe %c in active mask (0x%x))\n", pll->info->name, pipe_name(crtc->pipe), pll->active_mask); /* TC ports have both MG/TC and TBT PLL referenced simultaneously */ - INTEL_DISPLAY_STATE_WARN(display, !has_alt_port_dpll(old_crtc_state->shared_dpll, - new_crtc_state->shared_dpll) && + INTEL_DISPLAY_STATE_WARN(display, !has_alt_port_dpll(old_crtc_state->intel_dpll, + new_crtc_state->intel_dpll) && pll->state.pipe_mask & pipe_mask, "%s: pll enabled crtcs mismatch (found pipe %c in enabled mask (0x%x))\n", pll->info->name, pipe_name(crtc->pipe), pll->state.pipe_mask); } } -void intel_shared_dpll_verify_disabled(struct intel_atomic_state *state) +void intel_dpll_verify_disabled(struct intel_atomic_state *state) { struct intel_display *display = to_intel_display(state); - struct intel_shared_dpll *pll; + struct intel_dpll *pll; int i; - for_each_shared_dpll(display, pll, i) + for_each_dpll(display, pll, i) verify_single_dpll_state(display, pll, NULL, NULL); } diff --git a/drivers/gpu/drm/i915/display/intel_dpll_mgr.h b/drivers/gpu/drm/i915/display/intel_dpll_mgr.h index caffb084830ce430613a4ce2628b3c51a98dec1b..f131bdd1c975de2f126e88d0fdabf73f9e7d9a87 100644 --- a/drivers/gpu/drm/i915/display/intel_dpll_mgr.h +++ b/drivers/gpu/drm/i915/display/intel_dpll_mgr.h @@ -30,18 +30,18 @@ #include "intel_display_power.h" #include "intel_wakeref.h" -#define for_each_shared_dpll(__display, __pll, __i) \ - for ((__i) = 0; (__i) < (__display)->dpll.num_shared_dpll && \ - ((__pll) = &(__display)->dpll.shared_dplls[(__i)]) ; (__i)++) +#define for_each_dpll(__display, __pll, __i) \ + for ((__i) = 0; (__i) < (__display)->dpll.num_dpll && \ + ((__pll) = &(__display)->dpll.dplls[(__i)]) ; (__i)++) enum tc_port; struct drm_printer; struct intel_atomic_state; struct intel_crtc; struct intel_crtc_state; +struct intel_dpll_funcs; struct intel_encoder; struct intel_shared_dpll; -struct intel_shared_dpll_funcs; /** * enum intel_dpll_id - possible DPLL ids @@ -280,7 +280,7 @@ struct intel_dpll_hw_state { }; /** - * struct intel_shared_dpll_state - hold the DPLL atomic state + * struct intel_dpll_state - hold the DPLL atomic state * * This structure holds an atomic state for the DPLL, that can represent * either its current state (in struct &intel_shared_dpll) or a desired @@ -289,7 +289,7 @@ struct intel_dpll_hw_state { * * See also intel_reserve_shared_dplls() and intel_release_shared_dplls(). */ -struct intel_shared_dpll_state { +struct intel_dpll_state { /** * @pipe_mask: mask of pipes using this DPLL, active or not */ @@ -314,7 +314,7 @@ struct dpll_info { /** * @funcs: platform specific hooks */ - const struct intel_shared_dpll_funcs *funcs; + const struct intel_dpll_funcs *funcs; /** * @id: unique identifier for this DPLL @@ -344,16 +344,16 @@ struct dpll_info { }; /** - * struct intel_shared_dpll - display PLL with tracked state and users + * struct intel_dpll - display PLL with tracked state and users */ -struct intel_shared_dpll { +struct intel_dpll { /** * @state: * * Store the state for the pll, including its hw state * and CRTCs using it. */ - struct intel_shared_dpll_state state; + struct intel_dpll_state state; /** * @index: index for atomic state @@ -387,41 +387,41 @@ struct intel_shared_dpll { #define SKL_DPLL2 2 #define SKL_DPLL3 3 -/* shared dpll functions */ -struct intel_shared_dpll * -intel_get_shared_dpll_by_id(struct intel_display *display, - enum intel_dpll_id id); -void assert_shared_dpll(struct intel_display *display, - struct intel_shared_dpll *pll, - bool state); -#define assert_shared_dpll_enabled(d, p) assert_shared_dpll(d, p, true) -#define assert_shared_dpll_disabled(d, p) assert_shared_dpll(d, p, false) -int intel_compute_shared_dplls(struct intel_atomic_state *state, - struct intel_crtc *crtc, - struct intel_encoder *encoder); -int intel_reserve_shared_dplls(struct intel_atomic_state *state, - struct intel_crtc *crtc, - struct intel_encoder *encoder); -void intel_release_shared_dplls(struct intel_atomic_state *state, - struct intel_crtc *crtc); -void intel_unreference_shared_dpll_crtc(const struct intel_crtc *crtc, - const struct intel_shared_dpll *pll, - struct intel_shared_dpll_state *shared_dpll_state); +/* dpll functions */ +struct intel_dpll * +intel_get_dpll_by_id(struct intel_display *display, + enum intel_dpll_id id); +void assert_dpll(struct intel_display *display, + struct intel_dpll *pll, + bool state); +#define assert_dpll_enabled(d, p) assert_dpll(d, p, true) +#define assert_dpll_disabled(d, p) assert_dpll(d, p, false) +int intel_dpll_compute(struct intel_atomic_state *state, + struct intel_crtc *crtc, + struct intel_encoder *encoder); +int intel_dpll_reserve(struct intel_atomic_state *state, + struct intel_crtc *crtc, + struct intel_encoder *encoder); +void intel_dpll_release(struct intel_atomic_state *state, + struct intel_crtc *crtc); +void intel_dpll_crtc_put(const struct intel_crtc *crtc, + const struct intel_dpll *pll, + struct intel_dpll_state *shared_dpll_state); void icl_set_active_port_dpll(struct intel_crtc_state *crtc_state, enum icl_port_dpll_id port_dpll_id); -void intel_update_active_dpll(struct intel_atomic_state *state, +void intel_dpll_update_active(struct intel_atomic_state *state, struct intel_crtc *crtc, struct intel_encoder *encoder); int intel_dpll_get_freq(struct intel_display *display, - const struct intel_shared_dpll *pll, + const struct intel_dpll *pll, const struct intel_dpll_hw_state *dpll_hw_state); bool intel_dpll_get_hw_state(struct intel_display *display, - struct intel_shared_dpll *pll, + struct intel_dpll *pll, struct intel_dpll_hw_state *dpll_hw_state); -void intel_enable_shared_dpll(const struct intel_crtc_state *crtc_state); -void intel_disable_shared_dpll(const struct intel_crtc_state *crtc_state); -void intel_shared_dpll_swap_state(struct intel_atomic_state *state); -void intel_shared_dpll_init(struct intel_display *display); +void intel_dpll_enable(const struct intel_crtc_state *crtc_state); +void intel_dpll_disable(const struct intel_crtc_state *crtc_state); +void intel_dpll_swap_state(struct intel_atomic_state *state); +void intel_dpll_init(struct intel_display *display); void intel_dpll_update_ref_clks(struct intel_display *display); void intel_dpll_readout_hw_state(struct intel_display *display); void intel_dpll_sanitize_state(struct intel_display *display); @@ -435,8 +435,8 @@ bool intel_dpll_compare_hw_state(struct intel_display *display, enum intel_dpll_id icl_tc_port_to_pll_id(enum tc_port tc_port); bool intel_dpll_is_combophy(enum intel_dpll_id id); -void intel_shared_dpll_state_verify(struct intel_atomic_state *state, - struct intel_crtc *crtc); -void intel_shared_dpll_verify_disabled(struct intel_atomic_state *state); +void intel_dpll_state_verify(struct intel_atomic_state *state, + struct intel_crtc *crtc); +void intel_dpll_verify_disabled(struct intel_atomic_state *state); #endif /* _INTEL_DPLL_MGR_H_ */ diff --git a/drivers/gpu/drm/i915/display/intel_dpt.c b/drivers/gpu/drm/i915/display/intel_dpt.c index 43bd97e4f589897dff85aad3112c30f98b45155b..aea249e2699f3fde2fb779569335672c47802167 100644 --- a/drivers/gpu/drm/i915/display/intel_dpt.c +++ b/drivers/gpu/drm/i915/display/intel_dpt.c @@ -9,6 +9,7 @@ #include "gt/gen8_ppgtt.h" #include "i915_drv.h" +#include "intel_display_core.h" #include "intel_display_rpm.h" #include "intel_display_types.h" #include "intel_dpt.h" @@ -126,7 +127,7 @@ struct i915_vma *intel_dpt_pin_to_ggtt(struct i915_address_space *vm, unsigned int alignment) { struct drm_i915_private *i915 = vm->i915; - struct intel_display *display = &i915->display; + struct intel_display *display = i915->display; struct i915_dpt *dpt = i915_vm_to_dpt(vm); struct ref_tracker *wakeref; struct i915_vma *vma; diff --git a/drivers/gpu/drm/i915/display/intel_dpt_common.c b/drivers/gpu/drm/i915/display/intel_dpt_common.c index ce5aa0ca0fa5356db645a4f7581865d9c42d9b4e..5eb88d51dba1bec89055aa2e3b45c2778a7b32fc 100644 --- a/drivers/gpu/drm/i915/display/intel_dpt_common.c +++ b/drivers/gpu/drm/i915/display/intel_dpt_common.c @@ -3,8 +3,8 @@ * Copyright © 2023 Intel Corporation */ -#include "i915_reg.h" #include "intel_de.h" +#include "intel_display_regs.h" #include "intel_display_types.h" #include "intel_dpt_common.h" #include "skl_universal_plane_regs.h" diff --git a/drivers/gpu/drm/i915/display/intel_drrs.c b/drivers/gpu/drm/i915/display/intel_drrs.c index 05cd0f6e6d71bfd8388f49e679b3309fa8b1f58f..0fdb32ef241cde4920e687f1d50e1118c1890d85 100644 --- a/drivers/gpu/drm/i915/display/intel_drrs.c +++ b/drivers/gpu/drm/i915/display/intel_drrs.c @@ -5,10 +5,11 @@ #include -#include "i915_drv.h" -#include "i915_reg.h" +#include + #include "intel_atomic.h" #include "intel_de.h" +#include "intel_display_regs.h" #include "intel_display_types.h" #include "intel_drrs.h" #include "intel_frontbuffer.h" @@ -123,9 +124,9 @@ static void intel_drrs_set_state(struct intel_crtc *crtc, static void intel_drrs_schedule_work(struct intel_crtc *crtc) { - struct drm_i915_private *i915 = to_i915(crtc->base.dev); + struct intel_display *display = to_intel_display(crtc); - mod_delayed_work(i915->unordered_wq, &crtc->drrs.work, msecs_to_jiffies(1000)); + mod_delayed_work(display->wq.unordered, &crtc->drrs.work, msecs_to_jiffies(1000)); } static unsigned int intel_drrs_frontbuffer_bits(const struct intel_crtc_state *crtc_state) diff --git a/drivers/gpu/drm/i915/display/intel_dsb.c b/drivers/gpu/drm/i915/display/intel_dsb.c index 481488d1fe67793b413cd57e56eb86d3c8b10f6d..53d8ae3a70e90d73a7613c8ad1f4712cfea92bcd 100644 --- a/drivers/gpu/drm/i915/display/intel_dsb.c +++ b/drivers/gpu/drm/i915/display/intel_dsb.c @@ -7,11 +7,10 @@ #include #include -#include "i915_irq.h" -#include "i915_reg.h" #include "i915_utils.h" #include "intel_crtc.h" #include "intel_de.h" +#include "intel_display_regs.h" #include "intel_display_rpm.h" #include "intel_display_types.h" #include "intel_dsb.h" @@ -94,6 +93,10 @@ struct intel_dsb { /* see DSB_REG_VALUE_MASK */ #define DSB_OPCODE_POLL 0xA /* see DSB_REG_VALUE_MASK */ +#define DSB_OPCODE_GOSUB 0xC /* ptl+ */ +#define DSB_GOSUB_HEAD_SHIFT 26 +#define DSB_GOSUB_TAIL_SHIFT 0 +#define DSB_GOSUB_CONVERT_ADDR(x) ((x) >> 6) static bool pre_commit_is_vrr_active(struct intel_atomic_state *state, struct intel_crtc *crtc) @@ -205,6 +208,15 @@ static bool assert_dsb_has_room(struct intel_dsb *dsb) crtc->base.base.id, crtc->base.name, dsb->id); } +static bool assert_dsb_tail_is_aligned(struct intel_dsb *dsb) +{ + struct intel_crtc *crtc = dsb->crtc; + struct intel_display *display = to_intel_display(crtc->base.dev); + + return !drm_WARN_ON(display->drm, + !IS_ALIGNED(dsb->free_pos * 4, CACHELINE_BYTES)); +} + static void intel_dsb_dump(struct intel_dsb *dsb) { struct intel_crtc *crtc = dsb->crtc; @@ -229,13 +241,40 @@ static bool is_dsb_busy(struct intel_display *display, enum pipe pipe, return intel_de_read_fw(display, DSB_CTRL(pipe, dsb_id)) & DSB_STATUS_BUSY; } +unsigned int intel_dsb_size(struct intel_dsb *dsb) +{ + return dsb->free_pos * 4; +} + +unsigned int intel_dsb_head(struct intel_dsb *dsb) +{ + return intel_dsb_buffer_ggtt_offset(&dsb->dsb_buf); +} + +static unsigned int intel_dsb_tail(struct intel_dsb *dsb) +{ + return intel_dsb_buffer_ggtt_offset(&dsb->dsb_buf) + intel_dsb_size(dsb); +} + +static void intel_dsb_ins_align(struct intel_dsb *dsb) +{ + /* + * Every instruction should be 8 byte aligned. + * + * The only way to get unaligned free_pos is via + * intel_dsb_reg_write_indexed() which already + * makes sure the next dword is zeroed, so no need + * to clear it here. + */ + dsb->free_pos = ALIGN(dsb->free_pos, 2); +} + static void intel_dsb_emit(struct intel_dsb *dsb, u32 ldw, u32 udw) { if (!assert_dsb_has_room(dsb)) return; - /* Every instruction should be 8 byte aligned. */ - dsb->free_pos = ALIGN(dsb->free_pos, 2); + intel_dsb_ins_align(dsb); dsb->ins_start_offset = dsb->free_pos; dsb->ins[0] = ldw; @@ -493,6 +532,8 @@ static void intel_dsb_align_tail(struct intel_dsb *dsb) { u32 aligned_tail, tail; + intel_dsb_ins_align(dsb); + tail = dsb->free_pos * 4; aligned_tail = ALIGN(tail, CACHELINE_BYTES); @@ -503,20 +544,90 @@ static void intel_dsb_align_tail(struct intel_dsb *dsb) dsb->free_pos = aligned_tail / 4; } -void intel_dsb_finish(struct intel_dsb *dsb) +static void intel_dsb_gosub_align(struct intel_dsb *dsb) +{ + u32 aligned_tail, tail; + + intel_dsb_ins_align(dsb); + + tail = dsb->free_pos * 4; + aligned_tail = ALIGN(tail, CACHELINE_BYTES); + + /* + * Wa_16024917128 + * "Ensure GOSUB is not placed in cacheline QW slot 6 or 7 (numbered 0-7)" + */ + if (aligned_tail - tail <= 2 * 8) + intel_dsb_buffer_memset(&dsb->dsb_buf, dsb->free_pos, 0, + aligned_tail - tail); + + dsb->free_pos = aligned_tail / 4; +} + +void intel_dsb_gosub(struct intel_dsb *dsb, + struct intel_dsb *sub_dsb) { struct intel_crtc *crtc = dsb->crtc; + struct intel_display *display = to_intel_display(crtc->base.dev); + unsigned int head, tail; + u64 head_tail; + + if (drm_WARN_ON(display->drm, dsb->id != sub_dsb->id)) + return; + + if (!assert_dsb_tail_is_aligned(sub_dsb)) + return; + + intel_dsb_gosub_align(dsb); + + head = intel_dsb_head(sub_dsb); + tail = intel_dsb_tail(sub_dsb); /* - * DSB_FORCE_DEWAKE remains active even after DSB is - * disabled, so make sure to clear it (if set during - * intel_dsb_commit()). And clear DSB_ENABLE_DEWAKE as - * well for good measure. + * The GOSUB instruction has the following memory layout. + * + * +------------------------------------------------------------+ + * | Opcode | Rsvd | Head Ptr | Tail Ptr | + * | 0x0c | | | | + * +------------------------------------------------------------+ + * |<- 8bits->|<- 4bits ->|<-- 26bits -->|<-- 26bits -->| + * + * We have only 26 bits each to represent the head and tail + * pointers even though the addresses itself are of 32 bit. However, this + * is not a problem because the addresses are 64 bit aligned and therefore + * the last 6 bits are always Zero's. Therefore, we right shift the address + * by 6 before embedding it into the GOSUB instruction. */ - intel_dsb_reg_write(dsb, DSB_PMCTRL(crtc->pipe, dsb->id), 0); - intel_dsb_reg_write_masked(dsb, DSB_PMCTRL_2(crtc->pipe, dsb->id), - DSB_FORCE_DEWAKE, 0); + head_tail = ((u64)(DSB_GOSUB_CONVERT_ADDR(head)) << DSB_GOSUB_HEAD_SHIFT) | + ((u64)(DSB_GOSUB_CONVERT_ADDR(tail)) << DSB_GOSUB_TAIL_SHIFT); + + intel_dsb_emit(dsb, lower_32_bits(head_tail), + (DSB_OPCODE_GOSUB << DSB_OPCODE_SHIFT) | + upper_32_bits(head_tail)); + + /* + * "NOTE: the instructions within the cacheline + * FOLLOWING the GOSUB instruction must be NOPs." + */ + intel_dsb_align_tail(dsb); +} + +void intel_dsb_gosub_finish(struct intel_dsb *dsb) +{ + intel_dsb_align_tail(dsb); + + /* + * Wa_16024917128 + * "Ensure that all subroutines called by GOSUB end with a cacheline of NOPs" + */ + intel_dsb_noop(dsb, 8); + + intel_dsb_buffer_flush_map(&dsb->dsb_buf); +} + +void intel_dsb_finish(struct intel_dsb *dsb) +{ intel_dsb_align_tail(dsb); intel_dsb_buffer_flush_map(&dsb->dsb_buf); @@ -539,6 +650,9 @@ static u32 dsb_error_int_status(struct intel_display *display) if (DISPLAY_VER(display) >= 14) errors |= DSB_ATS_FAULT_INT_STATUS; + if (DISPLAY_VER(display) >= 30) + errors |= DSB_GOSUB_INT_STATUS; + return errors; } @@ -553,17 +667,46 @@ static u32 dsb_error_int_en(struct intel_display *display) if (DISPLAY_VER(display) >= 14) errors |= DSB_ATS_FAULT_INT_EN; + /* + * Wa_16024917128 + * "Disable nested GOSUB interrupt (DSB_INTERRUPT bit 21)" + */ + if (0 && DISPLAY_VER(display) >= 30) + errors |= DSB_GOSUB_INT_EN; + return errors; } +/* + * FIXME calibrate these sensibly, ideally compute based on + * the number of regisetrs to be written. But that requires + * measuring the actual DSB execution speed on each platform + * (and the speed also depends on CDCLK and memory clock)... + */ +static int intel_dsb_noarm_exec_time_us(void) +{ + return 80; +} + +static int intel_dsb_arm_exec_time_us(void) +{ + return 20; +} + +int intel_dsb_exec_time_us(void) +{ + return intel_dsb_noarm_exec_time_us() + + intel_dsb_arm_exec_time_us(); +} + void intel_dsb_vblank_evade(struct intel_atomic_state *state, struct intel_dsb *dsb) { struct intel_crtc *crtc = dsb->crtc; const struct intel_crtc_state *crtc_state = intel_pre_commit_crtc_state(state, crtc); - /* FIXME calibrate sensibly */ - int latency = intel_usecs_to_scanlines(&crtc_state->hw.adjusted_mode, 20); + int latency = intel_usecs_to_scanlines(&crtc_state->hw.adjusted_mode, + intel_dsb_arm_exec_time_us()); int start, end; /* @@ -605,13 +748,11 @@ static void _intel_dsb_chain(struct intel_atomic_state *state, struct intel_display *display = to_intel_display(state->base.dev); struct intel_crtc *crtc = dsb->crtc; enum pipe pipe = crtc->pipe; - u32 tail; if (drm_WARN_ON(display->drm, dsb->id == chained_dsb->id)) return; - tail = chained_dsb->free_pos * 4; - if (drm_WARN_ON(display->drm, !IS_ALIGNED(tail, CACHELINE_BYTES))) + if (!assert_dsb_tail_is_aligned(chained_dsb)) return; intel_dsb_reg_write(dsb, DSB_CTRL(pipe, chained_dsb->id), @@ -631,13 +772,15 @@ static void _intel_dsb_chain(struct intel_atomic_state *state, intel_dsb_reg_write(dsb, DSB_PMCTRL(pipe, chained_dsb->id), DSB_ENABLE_DEWAKE | DSB_SCANLINE_FOR_DEWAKE(hw_dewake_scanline)); + } else { + intel_dsb_reg_write(dsb, DSB_PMCTRL(pipe, chained_dsb->id), 0); } intel_dsb_reg_write(dsb, DSB_HEAD(pipe, chained_dsb->id), - intel_dsb_buffer_ggtt_offset(&chained_dsb->dsb_buf)); + intel_dsb_head(chained_dsb)); intel_dsb_reg_write(dsb, DSB_TAIL(pipe, chained_dsb->id), - intel_dsb_buffer_ggtt_offset(&chained_dsb->dsb_buf) + tail); + intel_dsb_tail(chained_dsb)); if (ctrl & DSB_WAIT_FOR_VBLANK) { /* @@ -652,6 +795,13 @@ static void _intel_dsb_chain(struct intel_atomic_state *state, intel_dsb_wait_scanline_out(state, dsb, dsb_dewake_scanline_start(state, crtc), dsb_dewake_scanline_end(state, crtc)); + + /* + * DSB_FORCE_DEWAKE remains active even after DSB is + * disabled, so make sure to clear it. + */ + intel_dsb_reg_write_masked(dsb, DSB_PMCTRL_2(crtc->pipe, dsb->id), + DSB_FORCE_DEWAKE, 0); } } @@ -676,16 +826,19 @@ void intel_dsb_wait_vblank_delay(struct intel_atomic_state *state, intel_dsb_wait_usec(dsb, usecs); } -static void _intel_dsb_commit(struct intel_dsb *dsb, u32 ctrl, - int hw_dewake_scanline) +/** + * intel_dsb_commit() - Trigger workload execution of DSB. + * @dsb: DSB context + * + * This function is used to do actual write to hardware using DSB. + */ +void intel_dsb_commit(struct intel_dsb *dsb) { struct intel_crtc *crtc = dsb->crtc; struct intel_display *display = to_intel_display(crtc->base.dev); enum pipe pipe = crtc->pipe; - u32 tail; - tail = dsb->free_pos * 4; - if (drm_WARN_ON(display->drm, !IS_ALIGNED(tail, CACHELINE_BYTES))) + if (!assert_dsb_tail_is_aligned(dsb)) return; if (is_dsb_busy(display, pipe, dsb->id)) { @@ -695,7 +848,7 @@ static void _intel_dsb_commit(struct intel_dsb *dsb, u32 ctrl, } intel_de_write_fw(display, DSB_CTRL(pipe, dsb->id), - ctrl | DSB_ENABLE); + DSB_ENABLE); intel_de_write_fw(display, DSB_CHICKEN(pipe, dsb->id), dsb->chicken); @@ -704,45 +857,13 @@ static void _intel_dsb_commit(struct intel_dsb *dsb, u32 ctrl, dsb_error_int_status(display) | DSB_PROG_INT_STATUS | dsb_error_int_en(display) | DSB_PROG_INT_EN); - intel_de_write_fw(display, DSB_HEAD(pipe, dsb->id), - intel_dsb_buffer_ggtt_offset(&dsb->dsb_buf)); - - if (hw_dewake_scanline >= 0) { - int diff, position; + intel_de_write_fw(display, DSB_PMCTRL(pipe, dsb->id), 0); - intel_de_write_fw(display, DSB_PMCTRL(pipe, dsb->id), - DSB_ENABLE_DEWAKE | - DSB_SCANLINE_FOR_DEWAKE(hw_dewake_scanline)); - - /* - * Force DEwake immediately if we're already past - * or close to racing past the target scanline. - */ - position = intel_de_read_fw(display, PIPEDSL(display, pipe)) & PIPEDSL_LINE_MASK; - - diff = hw_dewake_scanline - position; - intel_de_write_fw(display, DSB_PMCTRL_2(pipe, dsb->id), - (diff >= 0 && diff < 5 ? DSB_FORCE_DEWAKE : 0) | - DSB_BLOCK_DEWAKE_EXTENSION); - } + intel_de_write_fw(display, DSB_HEAD(pipe, dsb->id), + intel_dsb_head(dsb)); intel_de_write_fw(display, DSB_TAIL(pipe, dsb->id), - intel_dsb_buffer_ggtt_offset(&dsb->dsb_buf) + tail); -} - -/** - * intel_dsb_commit() - Trigger workload execution of DSB. - * @dsb: DSB context - * @wait_for_vblank: wait for vblank before executing - * - * This function is used to do actual write to hardware using DSB. - */ -void intel_dsb_commit(struct intel_dsb *dsb, - bool wait_for_vblank) -{ - _intel_dsb_commit(dsb, - wait_for_vblank ? DSB_WAIT_FOR_VBLANK : 0, - wait_for_vblank ? dsb->hw_dewake_scanline : -1); + intel_dsb_tail(dsb)); } void intel_dsb_wait(struct intel_dsb *dsb) @@ -895,4 +1016,7 @@ void intel_dsb_irq_handler(struct intel_display *display, if (errors & DSB_POLL_ERR_INT_STATUS) drm_err(display->drm, "[CRTC:%d:%s] DSB %d poll error\n", crtc->base.base.id, crtc->base.name, dsb_id); + if (errors & DSB_GOSUB_INT_STATUS) + drm_err(display->drm, "[CRTC:%d:%s] DSB %d GOSUB programming error\n", + crtc->base.base.id, crtc->base.name, dsb_id); } diff --git a/drivers/gpu/drm/i915/display/intel_dsb.h b/drivers/gpu/drm/i915/display/intel_dsb.h index e843c52bf97c2b369d98bad8f5b4e1f093170ec4..c8f4499916eb11cb812b6ea3eaf983e67612ed8a 100644 --- a/drivers/gpu/drm/i915/display/intel_dsb.h +++ b/drivers/gpu/drm/i915/display/intel_dsb.h @@ -26,12 +26,16 @@ enum intel_dsb_id { I915_MAX_DSBS, }; +unsigned int intel_dsb_size(struct intel_dsb *dsb); +unsigned int intel_dsb_head(struct intel_dsb *dsb); struct intel_dsb *intel_dsb_prepare(struct intel_atomic_state *state, struct intel_crtc *crtc, enum intel_dsb_id dsb_id, unsigned int max_cmds); void intel_dsb_finish(struct intel_dsb *dsb); +void intel_dsb_gosub_finish(struct intel_dsb *dsb); void intel_dsb_cleanup(struct intel_dsb *dsb); +int intel_dsb_exec_time_us(void); void intel_dsb_reg_write(struct intel_dsb *dsb, i915_reg_t reg, u32 val); void intel_dsb_reg_write_indexed(struct intel_dsb *dsb, @@ -57,13 +61,14 @@ void intel_dsb_vblank_evade(struct intel_atomic_state *state, void intel_dsb_poll(struct intel_dsb *dsb, i915_reg_t reg, u32 mask, u32 val, int wait_us, int count); +void intel_dsb_gosub(struct intel_dsb *dsb, + struct intel_dsb *sub_dsb); void intel_dsb_chain(struct intel_atomic_state *state, struct intel_dsb *dsb, struct intel_dsb *chained_dsb, bool wait_for_vblank); -void intel_dsb_commit(struct intel_dsb *dsb, - bool wait_for_vblank); +void intel_dsb_commit(struct intel_dsb *dsb); void intel_dsb_wait(struct intel_dsb *dsb); void intel_dsb_irq_handler(struct intel_display *display, diff --git a/drivers/gpu/drm/i915/display/intel_dsb_regs.h b/drivers/gpu/drm/i915/display/intel_dsb_regs.h index cb6e0e5624a6a50de9e43cb78b3b955c39694323..230104f36145896dd59e7da12fbd8dda36c55ea9 100644 --- a/drivers/gpu/drm/i915/display/intel_dsb_regs.h +++ b/drivers/gpu/drm/i915/display/intel_dsb_regs.h @@ -51,11 +51,13 @@ #define DSB_RESET_SM_STATE_MASK REG_GENMASK(5, 4) #define DSB_RUN_SM_STATE_MASK REG_GENMASK(2, 0) #define DSB_INTERRUPT(pipe, id) _MMIO(DSBSL_INSTANCE(pipe, id) + 0x28) +#define DSB_GOSUB_INT_EN REG_BIT(21) /* ptl+ */ #define DSB_ATS_FAULT_INT_EN REG_BIT(20) /* mtl+ */ #define DSB_GTT_FAULT_INT_EN REG_BIT(19) #define DSB_RSPTIMEOUT_INT_EN REG_BIT(18) #define DSB_POLL_ERR_INT_EN REG_BIT(17) #define DSB_PROG_INT_EN REG_BIT(16) +#define DSB_GOSUB_INT_STATUS REG_BIT(5) /* ptl+ */ #define DSB_ATS_FAULT_INT_STATUS REG_BIT(4) /* mtl+ */ #define DSB_GTT_FAULT_INT_STATUS REG_BIT(3) #define DSB_RSPTIMEOUT_INT_STATUS REG_BIT(2) diff --git a/drivers/gpu/drm/i915/display/intel_dsi_vbt.c b/drivers/gpu/drm/i915/display/intel_dsi_vbt.c index 29c9209834138e07af768ebc61faf1f78332c2f0..e6a851d276f831781aaafc9d1cfd3aa21979d6ea 100644 --- a/drivers/gpu/drm/i915/display/intel_dsi_vbt.c +++ b/drivers/gpu/drm/i915/display/intel_dsi_vbt.c @@ -36,12 +36,11 @@ #include #include #include - #include