Documentation/mm/slub.rst

8 SLUB can enable debugging only for selected slabs in order to avoid
13 to the kernel command line. That will enable full debugging for
18 slabs that have data in them. See "slabinfo -h" for more options when
22 	gcc -o slabinfo tools/mm/slabinfo.c
30 -------------------------------------------
35 slub_debug=<Debug-Options>
36 	Enable options for all slabs
38 slub_debug=<Debug-Options>,<slab name1>,<slab name2>,...
39 	Enable options only for select slabs (no spaces
55 	A		Enable failslab filter mark for the cache
58 	-		Switch all debugging off (useful if the kernel is
69 to only enable debugging on the dentry cache.  You may use an asterisk at the
74 	slub_debug=P,kmalloc-*,dentry
90 of options. This will enable red zoning for dentry and user tracking for
93 	slub_debug=Z,dentry;U,kmalloc-*
95 You can also enable options (e.g. sanity checks and poisoning) for all caches
98 with "-" as options::
100 	slub_debug=FZ;-,zs_handle,zspage
117 failslab file is writable, so writing 1 or 0 will enable or disable
118 the option at runtime. Write returns -EINVAL if cache is an alias.
127 ``slabinfo -a`` displays which slabs were merged together.
136 	slabinfo -v
141 In that case ``slabinfo -v`` simply tests all reachable objects. Usually
184  BUG kmalloc-8: Right Redzone overwritten
185  --------------------------------------------------------------------
187  INFO: 0xc90f6d28-0xc90f6d2b. First byte 0x00 instead of 0xcc
219  FIX kmalloc-8: Restoring Redzone 0xc90f6d28-0xc90f6d2b=0xcc
231      -----------------------------------------------
233      INFO: <corruption start>-<corruption_end> <more info>
255 	typically contain poison values. Any non-poison value shows a
301 This will be generally be enough to enable the resiliency features of slub
317 may be advisable to enable a Redzone to avoid corrupting the beginning
322 Extended slabinfo mode and plotting
325 The ``slabinfo`` tool has a special 'extended' ('-X') mode that includes:
326  - Slabcache Totals
327  - Slabs sorted by size (up to -N <num> slabs, default 1)
328  - Slabs sorted by loss (up to -N <num> slabs, default 1)
332 also available to other slabinfo modes via '-B' option) which makes
333 reporting more precise and accurate. Moreover, in some sense the `-X'
335 output can be plotted using the ``slabinfo-gnuplot.sh`` script. So it
337 to something easier -- visual analysis.
341 a) collect slabinfo extended records, for example::
343 	while [ 1 ]; do slabinfo -X >> FOO_STATS; sleep 1; done
345 b) pass stats file(-s) to ``slabinfo-gnuplot.sh`` script::
347 	slabinfo-gnuplot.sh FOO_STATS [FOO_STATS2 .. FOO_STATSN]
349    The ``slabinfo-gnuplot.sh`` script will pre-processes the collected records
350    and generates 3 png files (and 3 pre-processing cache files) per STATS
352    - Slabcache Totals: FOO_STATS-totals.png
353    - Slabs sorted by size: FOO_STATS-slabs-by-size.png
354    - Slabs sorted by loss: FOO_STATS-slabs-by-loss.png
356 Another use case, when ``slabinfo-gnuplot.sh`` can be useful, is when you
358 modification.  To help you out there, ``slabinfo-gnuplot.sh`` script
364 	while [ 1 ]; do slabinfo -X >> STATS<X>; sleep 1; done
366 b) Pre-process those STATS files::
368 	slabinfo-gnuplot.sh STATS1 STATS2 .. STATSN
370 c) Execute ``slabinfo-gnuplot.sh`` in '-t' mode, passing all of the
371    generated pre-processed \*-totals::
373 	slabinfo-gnuplot.sh -t STATS1-totals STATS2-totals .. STATSN-totals
378    can go unnoticed. To deal with that, ``slabinfo-gnuplot.sh`` has two
379    options to 'zoom-in'/'zoom-out':
381    a) ``-s %d,%d`` -- overwrites the default image width and height
382    b) ``-r %d,%d`` -- specifies a range of samples to use (for example,
383       in ``slabinfo -X >> FOO_STATS; sleep 1;`` case, using a ``-r
384       40,60`` range will plot only samples collected between 40th and
404     kmalloc objects(total/per-object), minimal/average/maximal jiffies
405     since alloc, pid range of the allocating processes, cpu mask of
410     338 pci_alloc_dev+0x2c/0xa0 waste=521872/1544 age=290837/291891/293509 pid=1 cpus=106 nodes=0-1
431     objects. The freeing traces thus come from the previous life-cycle of the
437     pid range of the freeing processes, cpu mask of freeing cpus, and stack trace.
441     1980 <not-available> age=4294912290 pid=0 cpus=0