As of commit c4c54dd1 (kernel/cpu.c: change type of cpu_possible_bits
and friends), cpu_present_bits no longer exists. Add new method of finding
number of cpus and also change other functions to use the get_num_cpus
API instead of defining it again.

Change-Id: I6d38d505ba7674b94e8481e6679226803eba977e

Add check to avoid "object has no attribute 'kaslr_addr'" on boards
where kaslr_addr is not defined.

Change-Id: I2691b313f6a0ab75b216cd678245e97cf22ce4e7

Kaslr-offset in stored in imem by the kernel driver. This patch
enables the parser to extract the required offset before parsing.

Change-Id: Ie773a2e52989b0f2a83c19d39ffdba7f77911e39

When kernel address space layout randomization is enabled by kernel,
a new kaslr_offset will be added to the kimage_vaddress. This patch
adds this offset to get the correct kimage_vaddress.

Change-Id: Ia19c5b4d47674e2753ba167e5a50295902720457

The startup scripts used with T32 is not generated properly for use with
Linux. Fix this:

* Make T32 use absolute paths on both Windows and Linux
* Update T32 config scripts to work with Linux

Change-Id: If96dd762d8cb5a9419d149d370ae21cf222ed93b

the paths used by the T32 launcher used the incorrect directory when the
autodump option was passed.

Additionally, fix an issue where not all of the files detected by autodump
are printed out.

Change-Id: Id96500e6a59961aa1d06af515560b367a42f2e5b

Customers normally share extracted logs from RAMDUMP because of
various reasons including security and to protect personal
information. Add support to extract WLAN module debug details so
that customer issues can be triaged easily.

CRs-fixed: 1074311
Change-Id: If7a3f2c6ed990d9203ea93c8e0688c5c44f13c53

   File creation is failing in some filers if it is created
in a non-existing folder.

Change-Id: Ieaacc3f1992b38683c52df6225727ab88944d65a

Latest kernel introduces KIMAGE_VADDR which allows the
virtual base of the kernel region to be below PAGE_OFFSET.

Change-Id: I9d063f882e220d0795eca45a618ad72580dff552

The board number is intended to be a human readable identification value.
Allowing this to be a string is more flexible.

CRs-Fixed: 1013327
Change-Id: If523acafc54f5026680d18c95e4899fb0436599d

Calling read_byte() on windows may return the empty string '' prior
to the end of the file without opening the file in 'rb' mode.

The 'rb' mode was removed by
commit 5743b613
("lrdp-v2: ramdump: Fix autoparse for some targets.")

Change-Id: Ibea608112400c06a9adc6f1fd3b4842a87717846

   Add memory files used for some targets.

Change-Id: I64df5549568a97f907b0e493f3d4cac81a98ddf9

Add support for parsing dump_info.txt and load.cmm to properly set phys
offset and find the base addresses of memory files.

This change fixed autoparse on a few targets, though may not apply to all
platforms and situations.

Changes based heavily on those made from a prior change.
See Change-Id: I4be54a48a659ac1e6346d9178d19a8b40e2e1b6d

CRs-Fixed: 1038185
Change-Id: If98a4073e9afc98fd3903e606539b9e4e9ca2c51

Debug image parsing currently includes QDSS ETM and ETR binary data
parsing by default. However, these two parsing operations can take a
significant amount of time if either of their respective data buffers
are very large. Furthermore, this binary data is not needed for most
common debug image parsing use cases.

Add an argument to optionally disable QDSS ETM and ETR binary data
parsing when debug image parsing has also been selected.

Change-Id: Iab25c15d00dde1bb5dfafb3149e6e55760fd6291

Add module to get information from MDP dumps, which can occur under
situations where MDP cannot recover.

Change-Id: I8ecdced37206947c434c7ddf4d7d61979c9cf92f

Currently we split each config line on the equals sign, assuming that
there's only one equals sign there.  This isn't a valid assumption for
cases like:

    CONFIG_CMDLINE="console=ttyAMA0"

Fix this by splitting at the first equals sign, rather than splitting
at *all* equals signs.

Change-Id: I01e44fe2e9a0f09a8bebdd76715002f6b12cf0b5

Add '--kaslr-offset' option.  This specifies the offset for loading symbols
for a KASLR-enabled kernel.

Change-Id: I208d964ea3ca58536a0b80a32b4e334c73aae534

Some of our module documentation is a bit lacking.  Fix this by filling
in some docstrings and add then adding some pointers to existing
modules/functions in the docs.

Change-Id: Ibf714225f1fc185be916e514861540fd98cb2de5

Most read functions that take a virtual address also understand string
variable names, for convenience.  Add this support to the `hexdump'
function as well.

Change-Id: I773992ba3793a68226b49525b66141b021924387

Most of the time 100 is a good default max length when reading any old
C-string.  Use it.

Change-Id: I1805b0b4b2b9bc18c06f4ca0e7cd6cd70b854620

GdbMI.symbol_at returns a string, not a GdbSymbol object, so the extra
field traversal is unnecessary.

Change-Id: I85acd83657e4b6055bd1f7945da73dcf9d4839f1

The name symbol_at gives a better idea of what the function does, and it
more closely mirrors the corresponding gdbmi function.

Change-Id: If2f3df505bc129a3b963ea363e4c544599626948

In lieu of a more generic way of reading fields from structures,
introduce another specialized structure field reader for reading C
strings.

Change-Id: I6aba392f59179fa70a065a08f19a694e87f683dd

Rename addr_lookup to address_of to match common terminology more
closely.

Change-Id: I7fc744ce88d1f8880f13dfbf83fb0efcbe9ba0f2

Pointers can be either 32-bits or 64-bits, depending on the
architecture.  Add an accessor function to abstract this away.

Change-Id: I64e2c18c85ec3d9ca7304163dc5a42edc83f1443

A few reader functions currently assume that a virtual address (a
number) was passed in, but those functions also support reading variable
names.  Resolve the address or name to a virtual address before doing
arithmetic on it.

Change-Id: I55aa649ff203926565f8eaf3cb234677ef6b6104

Some of the return sequences in the memory accessor code are
unnecessarily complex.  Simplify them.

Change-Id: Iefe489b3ce4a4591574a2cc8d51aa23ce425b2b1

Currently there are "trace" print statements scattered throughout the
code.  These make the code harder to follow and aren't terribly useful.
Remove them in favor of on-demand instrumentation or more sophisticated
debugging methods (like a debugger).

Change-Id: I70a91b8dca34566d48ed1f4902cba5b9f9b4de40

The memory accessor functions that have documentation are using comments
rather than docstrings.  Fix this.  Also rename some variables for
improved self-documentation.

Change-Id: I6a196ff27665c5b0c96b3c5095b39690015b07e1

It can be useful to look up the values of config items.  Store a
dictionary mapping config options to their values.

Change-Id: I99f7615b2c5926558dfba100fa0c45b6e29ecdb5

Presently auto-dump option finds DDR base addresses from
statically defined per target meta data in boards.py. This
leads to auto-dump feature fail on different targets, say
with different DDR base and size.

Add support to parse DDR base addresses from load.cmm,
dump_info.txt or ramdump_msg.txt which ever is available in
that order. Fall-back to conventional static populated data
in boards.py if none of them is available.

Change-Id: Ie4ac4f281e92c9b527de17fa665c9cd6c3d5d384

The ram_dump.kernel_version variable is defined as a list and
incorrectly compared with tuple. This comparison always fails.
Fix it by changing ram_dump.kernel_version variable to tuple.

Change-Id: I12d4556693101fccfad795e5ea737b42e96acac7

Ram dump parser does not support unwinding the stack for NON
HLOS images. In certain cases, we get below exception when
parser tries to dump the context for NON HLOS address. Check
for NON HLOS address and return from the function without
unwinding the frame pointer.

Exception: Invalid stage 1 first- or second-level translation
descriptor: (value: 0x0 {dtype[1:0]=>0x0})
addr: (value: 0x21812010 {base[47:12]=>0x21812, offset[11:3]=>0x2})

Change-Id: I6cacec18574737d0e8261ba1f39dfe2b9b302304

Add support to parse DCC captured data from crash dumps and produce
human readable result.

Change-Id: I898b404b074fc037e339bf535c27bf190619cc9d

`major' is a bit of a misnomer, since the variable also includes the
minor version and patch version.  Rename it to `kernel_version'.

Change-Id: Ic567131c7f12ab03c9dea2716242e9053c17372f

In 3.18 kernel there is a slight change in radix_tree_node structure.
Fix irq parser to read radix_tree_node according to kernel version.

Change-Id: Ib9e67d137994f4b7424acd2b44676dcdcdb2d0a4

Integrate the IPC Logging extraction script module into
linux-ramdump-parser-v2.

Move ipc_logging.py from into the linux-ramdump-parser-v2/parsers
directory, and create a relative symbolic link in the original location.
Add a new file, ipc_logging_ramparse.py, which performs the actual
integration.

Change-Id: If0db4f82814a644094af7ebd0e4290dd145f94ac

The constructor currently used in ramparse.py to initialize the RamDump
object has a very long argument list. This list can be shortened by
passing the options object to the constructor, rather than enumerating
every member of the options object as a separate argument.

Pass the options object to the RamDump constructor instead of passing
each member as a separate argument.

Change-Id: I7cbf173232838dd8f9fdbdad0e2b7336ad8fc07b

Do not use hardcoded value for TTBR1. Compute TTBR1 value using
swapper_pg_dir_addr and phys_offset variables dynamically.

Change-Id: Ia9b4704b07e933c686e366ed0af8230be1d97e9c

Running launch_t32.bat with newer T32 gives syntax errors without newlines
before "SCREEN=" and "PRINTER=WINDOWS" strings in t32_config.t32. Add
newlines before these strings to avoid T32 syntax errors.

Change-Id: I53ffd02d4c86959804a89bd744bf1ce7f42f163c