diff --git a/linux-ramdump-parser-v2/parsers/slabsummary.py b/linux-ramdump-parser-v2/parsers/slabsummary.py
new file mode 100644
index 0000000000000000000000000000000000000000..34b43dc496f544a81a855ea6370c0cdba9d955a8
--- /dev/null
+++ b/linux-ramdump-parser-v2/parsers/slabsummary.py
@@ -0,0 +1,142 @@
+# Copyright (c) 2012-2016, The Linux Foundation. All rights reserved.
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License version 2 and
+# only version 2 as published by the Free Software Foundation.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+
+import math
+
+from mm import pfn_to_page
+from parser_util import register_parser, RamParser
+
+# /kernel/msm-4.4/mm/slub.c
+OO_SHIFT = 16
+PAGE_SHIFT = 12
+
+
+@register_parser('--slabsummary', 'print summary of slab', optional=True)
+class Slabinfo_summary(RamParser):
+
+ def cal_free_pages(
+ self, ramdump,
+ start, slab_lru_offset,
+ max_page):
+ page = self.ramdump.read_word(start)
+ if page == 0:
+ return
+ seen = []
+ totalfree = 0
+ mapcount = 0
+ total_objects = 0
+ inuse = 0
+ while page != start:
+ if page is None:
+ return
+ if page in seen:
+ return
+ if page > max_page:
+ return
+ seen.append(page)
+ page = page - slab_lru_offset
+ mapcount = self.ramdump.read_structure_field(
+ page, 'struct page', '_mapcount')
+ inuse = mapcount & 0x0000FFFF
+ total_objects = (mapcount >> 16) & 0x00007FFF
+ freeobj = total_objects - inuse
+ totalfree = totalfree + freeobj
+ page = self.ramdump.read_word(page + slab_lru_offset)
+ return totalfree
+
+ # Currently with assumption there is only one numa node
+ def print_slab_summary(self, slab_out):
+ total_freeobjects = 0
+ original_slab = self.ramdump.address_of('slab_caches')
+ cpu_present_bits_addr = self.ramdump.address_of('cpu_present_bits')
+ cpu_present_bits = self.ramdump.read_word(cpu_present_bits_addr)
+ cpus = bin(cpu_present_bits).count('1')
+ slab_list_offset = self.ramdump.field_offset(
+ 'struct kmem_cache', 'list')
+ slab_name_offset = self.ramdump.field_offset(
+ 'struct kmem_cache', 'name')
+ slab_node_offset = self.ramdump.field_offset(
+ 'struct kmem_cache', 'node')
+ cpu_cache_page_offset = self.ramdump.field_offset(
+ 'struct kmem_cache_cpu', 'page')
+ cpu_slab_offset = self.ramdump.field_offset(
+ 'struct kmem_cache', 'cpu_slab')
+ slab_partial_offset = self.ramdump.field_offset(
+ 'struct kmem_cache_node', 'partial')
+ slab = self.ramdump.read_word(original_slab)
+ slab_lru_offset = self.ramdump.field_offset(
+ 'struct page', 'lru')
+ max_pfn_addr = self.ramdump.address_of('max_pfn')
+ max_pfn = self.ramdump.read_word(max_pfn_addr)
+ max_page = pfn_to_page(self.ramdump, max_pfn)
+ format_string = '\n{0:35} {1:9} {2:10} {3:10} {4:8}K {5:10} {6:10}K'
+ slab_out.write(
+ '{0:35} {1:9} {2:10} {3:10} {4:8} {5:10} {6:10}'.format(
+ "NAME", "OBJSIZE", "ALLOCATED",
+ "TOTAL", "TOTAL*SIZE", "SLABS",
+ "SSIZE"))
+
+ while slab != original_slab:
+ total_freeobjects = 0
+ slab = slab - slab_list_offset
+ slab_name_addr = self.ramdump.read_word(
+ slab + slab_name_offset)
+ # actually an array but again, no numa
+ slab_node_addr = self.ramdump.read_word(
+ slab + slab_node_offset)
+ slab_name = self.ramdump.read_cstring(
+ slab_name_addr, 48)
+ cpu_slab_addr = self.ramdump.read_word(
+ slab + cpu_slab_offset)
+ oo = self.ramdump.read_structure_field(
+ slab, 'struct kmem_cache', 'oo')
+ obj_size = self.ramdump.read_structure_field(
+ slab, 'struct kmem_cache', 'object_size')
+ objsize_w_metadata = self.ramdump.read_structure_field(
+ slab, 'struct kmem_cache', 'size')
+ nr_total_objects = self.ramdump.read_structure_field(
+ slab_node_addr,
+ 'struct kmem_cache_node', 'total_objects')
+ num_slabs = self.ramdump.read_structure_field(
+ slab_node_addr,
+ 'struct kmem_cache_node', 'nr_slabs')
+ # per cpu slab
+ for i in range(0, cpus):
+ cpu_slabn_addr = self.ramdump.read_word(
+ cpu_slab_addr, cpu=i)
+ if cpu_slabn_addr == 0 or None:
+ break
+ total_freeobjects = total_freeobjects + self.cal_free_pages(
+ self.ramdump,
+ (cpu_slabn_addr + cpu_cache_page_offset),
+ slab_lru_offset,
+ max_page)
+
+ total_freeobjects = total_freeobjects + self.cal_free_pages(
+ self.ramdump,
+ slab_node_addr + slab_partial_offset,
+ slab_lru_offset, max_page)
+
+ total_allocated = nr_total_objects - total_freeobjects
+ page_order = oo >> OO_SHIFT
+ slab_size = int(math.pow(2, page_order + PAGE_SHIFT))
+ slab_size = slab_size / 1024
+ slab = self.ramdump.read_word(slab + slab_list_offset)
+ slab_out.write(format_string.format(
+ slab_name, obj_size, total_allocated,
+ nr_total_objects,
+ (objsize_w_metadata * nr_total_objects)/1024,
+ num_slabs, slab_size))
+
+ def parse(self):
+ slab_out = self.ramdump.open_file('slabsummary.txt')
+ self.print_slab_summary(slab_out)
+ slab_out.close()