2018-03-21 22:22:37 +03:00
|
|
|
.. _slub:
|
|
|
|
|
|
|
|
==========================
|
2007-05-07 01:49:47 +04:00
|
|
|
Short users guide for SLUB
|
2018-03-21 22:22:37 +03:00
|
|
|
==========================
|
2007-05-07 01:49:47 +04:00
|
|
|
|
|
|
|
The basic philosophy of SLUB is very different from SLAB. SLAB
|
|
|
|
requires rebuilding the kernel to activate debug options for all
|
2007-05-31 11:40:47 +04:00
|
|
|
slab caches. SLUB always includes full debugging but it is off by default.
|
2007-05-07 01:49:47 +04:00
|
|
|
SLUB can enable debugging only for selected slabs in order to avoid
|
|
|
|
an impact on overall system performance which may make a bug more
|
|
|
|
difficult to find.
|
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
In order to switch debugging on one can add an option ``slub_debug``
|
2007-05-07 01:49:47 +04:00
|
|
|
to the kernel command line. That will enable full debugging for
|
|
|
|
all slabs.
|
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
Typically one would then use the ``slabinfo`` command to get statistical
|
|
|
|
data and perform operation on the slabs. By default ``slabinfo`` only lists
|
2007-05-07 01:49:47 +04:00
|
|
|
slabs that have data in them. See "slabinfo -h" for more options when
|
2018-03-21 22:22:37 +03:00
|
|
|
running the command. ``slabinfo`` can be compiled with
|
|
|
|
::
|
2007-05-07 01:49:47 +04:00
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
gcc -o slabinfo tools/vm/slabinfo.c
|
2007-05-07 01:49:47 +04:00
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
Some of the modes of operation of ``slabinfo`` require that slub debugging
|
2007-05-07 01:49:47 +04:00
|
|
|
be enabled on the command line. F.e. no tracking information will be
|
|
|
|
available without debugging on and validation can only partially
|
|
|
|
be performed if debugging was not switched on.
|
|
|
|
|
|
|
|
Some more sophisticated uses of slub_debug:
|
|
|
|
-------------------------------------------
|
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
Parameters may be given to ``slub_debug``. If none is specified then full
|
2007-05-07 01:49:47 +04:00
|
|
|
debugging is enabled. Format:
|
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
slub_debug=<Debug-Options>
|
|
|
|
Enable options for all slabs
|
|
|
|
|
2018-10-27 01:03:15 +03:00
|
|
|
slub_debug=<Debug-Options>,<slab name1>,<slab name2>,...
|
|
|
|
Enable options only for select slabs (no spaces
|
|
|
|
after a comma)
|
2018-03-21 22:22:37 +03:00
|
|
|
|
2020-08-07 09:18:35 +03:00
|
|
|
Multiple blocks of options for all slabs or selected slabs can be given, with
|
|
|
|
blocks of options delimited by ';'. The last of "all slabs" blocks is applied
|
|
|
|
to all slabs except those that match one of the "select slabs" block. Options
|
|
|
|
of the first "select slabs" blocks that matches the slab's name are applied.
|
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
Possible debug options are::
|
2007-05-07 01:49:47 +04:00
|
|
|
|
2016-03-16 00:55:06 +03:00
|
|
|
F Sanity checks on (enables SLAB_DEBUG_CONSISTENCY_CHECKS
|
|
|
|
Sorry SLAB legacy issues)
|
2007-05-07 01:49:47 +04:00
|
|
|
Z Red zoning
|
|
|
|
P Poisoning (object and padding)
|
|
|
|
U User tracking (free and alloc)
|
|
|
|
T Trace (please only use on single slabs)
|
2020-06-02 07:46:00 +03:00
|
|
|
A Enable failslab filter mark for the cache
|
2009-07-07 11:14:14 +04:00
|
|
|
O Switch debugging off for caches that would have
|
|
|
|
caused higher minimum slab orders
|
2007-07-16 10:38:14 +04:00
|
|
|
- Switch all debugging off (useful if the kernel is
|
|
|
|
configured with CONFIG_SLUB_DEBUG_ON)
|
2007-05-07 01:49:47 +04:00
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
F.e. in order to boot just with sanity checks and red zoning one would specify::
|
2007-05-07 01:49:47 +04:00
|
|
|
|
|
|
|
slub_debug=FZ
|
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
Trying to find an issue in the dentry cache? Try::
|
2007-05-07 01:49:47 +04:00
|
|
|
|
2008-03-06 02:07:30 +03:00
|
|
|
slub_debug=,dentry
|
2007-05-07 01:49:47 +04:00
|
|
|
|
2018-10-27 01:03:15 +03:00
|
|
|
to only enable debugging on the dentry cache. You may use an asterisk at the
|
|
|
|
end of the slab name, in order to cover all slabs with the same prefix. For
|
|
|
|
example, here's how you can poison the dentry cache as well as all kmalloc
|
2019-01-31 07:06:22 +03:00
|
|
|
slabs::
|
2018-10-27 01:03:15 +03:00
|
|
|
|
|
|
|
slub_debug=P,kmalloc-*,dentry
|
2007-05-07 01:49:47 +04:00
|
|
|
|
|
|
|
Red zoning and tracking may realign the slab. We can just apply sanity checks
|
2018-03-21 22:22:37 +03:00
|
|
|
to the dentry cache with::
|
2007-05-07 01:49:47 +04:00
|
|
|
|
2008-03-06 02:07:30 +03:00
|
|
|
slub_debug=F,dentry
|
2007-05-07 01:49:47 +04:00
|
|
|
|
2009-07-07 11:14:14 +04:00
|
|
|
Debugging options may require the minimum possible slab order to increase as
|
|
|
|
a result of storing the metadata (for example, caches with PAGE_SIZE object
|
|
|
|
sizes). This has a higher liklihood of resulting in slab allocation errors
|
|
|
|
in low memory situations or if there's high fragmentation of memory. To
|
2018-03-21 22:22:37 +03:00
|
|
|
switch off debugging for such caches by default, use::
|
2009-07-07 11:14:14 +04:00
|
|
|
|
|
|
|
slub_debug=O
|
|
|
|
|
2020-08-07 09:18:35 +03:00
|
|
|
You can apply different options to different list of slab names, using blocks
|
|
|
|
of options. This will enable red zoning for dentry and user tracking for
|
|
|
|
kmalloc. All other slabs will not get any debugging enabled::
|
|
|
|
|
|
|
|
slub_debug=Z,dentry;U,kmalloc-*
|
|
|
|
|
|
|
|
You can also enable options (e.g. sanity checks and poisoning) for all caches
|
|
|
|
except some that are deemed too performance critical and don't need to be
|
|
|
|
debugged by specifying global debug options followed by a list of slab names
|
|
|
|
with "-" as options::
|
|
|
|
|
|
|
|
slub_debug=FZ;-,zs_handle,zspage
|
|
|
|
|
mm, slub: make some slub_debug related attributes read-only
SLUB_DEBUG creates several files under /sys/kernel/slab/<cache>/ that can
be read to check if the respective debugging options are enabled for given
cache. The options can be also toggled at runtime by writing into the
files. Some of those, namely red_zone, poison, and store_user can be
toggled only when no objects yet exist in the cache.
Vijayanand reports [1] that there is a problem with freelist randomization
if changing the debugging option's state results in different number of
objects per page, and the random sequence cache needs thus needs to be
recomputed.
However, another problem is that the check for "no objects yet exist in
the cache" is racy, as noted by Jann [2] and fixing that would add
overhead or otherwise complicate the allocation/freeing paths. Thus it
would be much simpler just to remove the runtime toggling support. The
documentation describes it's "In case you forgot to enable debugging on
the kernel command line", but the neccessity of having no objects limits
its usefulness anyway for many caches.
Vijayanand describes an use case [3] where debugging is enabled for all
but zram caches for memory overhead reasons, and using the runtime toggles
was the only way to achieve such configuration. After the previous patch
it's now possible to do that directly from the kernel boot option, so we
can remove the dangerous runtime toggles by making the /sys attribute
files read-only.
While updating it, also improve the documentation of the debugging /sys files.
[1] https://lkml.kernel.org/r/1580379523-32272-1-git-send-email-vjitta@codeaurora.org
[2] https://lore.kernel.org/r/CAG48ez31PP--h6_FzVyfJ4H86QYczAFPdxtJHUEEan+7VJETAQ@mail.gmail.com
[3] https://lore.kernel.org/r/1383cd32-1ddc-4dac-b5f8-9c42282fa81c@codeaurora.org
Reported-by: Vijayanand Jitta <vjitta@codeaurora.org>
Reported-by: Jann Horn <jannh@google.com>
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Kees Cook <keescook@chromium.org>
Acked-by: Roman Gushchin <guro@fb.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Pekka Enberg <penberg@kernel.org>
Link: http://lkml.kernel.org/r/20200610163135.17364-3-vbabka@suse.cz
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-08-07 09:18:38 +03:00
|
|
|
The state of each debug option for a slab can be found in the respective files
|
|
|
|
under::
|
2007-05-07 01:49:47 +04:00
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
/sys/kernel/slab/<slab name>/
|
2007-05-07 01:49:47 +04:00
|
|
|
|
mm, slub: make some slub_debug related attributes read-only
SLUB_DEBUG creates several files under /sys/kernel/slab/<cache>/ that can
be read to check if the respective debugging options are enabled for given
cache. The options can be also toggled at runtime by writing into the
files. Some of those, namely red_zone, poison, and store_user can be
toggled only when no objects yet exist in the cache.
Vijayanand reports [1] that there is a problem with freelist randomization
if changing the debugging option's state results in different number of
objects per page, and the random sequence cache needs thus needs to be
recomputed.
However, another problem is that the check for "no objects yet exist in
the cache" is racy, as noted by Jann [2] and fixing that would add
overhead or otherwise complicate the allocation/freeing paths. Thus it
would be much simpler just to remove the runtime toggling support. The
documentation describes it's "In case you forgot to enable debugging on
the kernel command line", but the neccessity of having no objects limits
its usefulness anyway for many caches.
Vijayanand describes an use case [3] where debugging is enabled for all
but zram caches for memory overhead reasons, and using the runtime toggles
was the only way to achieve such configuration. After the previous patch
it's now possible to do that directly from the kernel boot option, so we
can remove the dangerous runtime toggles by making the /sys attribute
files read-only.
While updating it, also improve the documentation of the debugging /sys files.
[1] https://lkml.kernel.org/r/1580379523-32272-1-git-send-email-vjitta@codeaurora.org
[2] https://lore.kernel.org/r/CAG48ez31PP--h6_FzVyfJ4H86QYczAFPdxtJHUEEan+7VJETAQ@mail.gmail.com
[3] https://lore.kernel.org/r/1383cd32-1ddc-4dac-b5f8-9c42282fa81c@codeaurora.org
Reported-by: Vijayanand Jitta <vjitta@codeaurora.org>
Reported-by: Jann Horn <jannh@google.com>
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Kees Cook <keescook@chromium.org>
Acked-by: Roman Gushchin <guro@fb.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Pekka Enberg <penberg@kernel.org>
Link: http://lkml.kernel.org/r/20200610163135.17364-3-vbabka@suse.cz
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-08-07 09:18:38 +03:00
|
|
|
If the file contains 1, the option is enabled, 0 means disabled. The debug
|
|
|
|
options from the ``slub_debug`` parameter translate to the following files::
|
|
|
|
|
|
|
|
F sanity_checks
|
|
|
|
Z red_zone
|
|
|
|
P poison
|
|
|
|
U store_user
|
|
|
|
T trace
|
|
|
|
A failslab
|
|
|
|
|
2020-08-07 09:18:45 +03:00
|
|
|
Careful with tracing: It may spew out lots of information and never stop if
|
|
|
|
used on the wrong slab.
|
2007-05-07 01:49:47 +04:00
|
|
|
|
2007-05-31 11:40:47 +04:00
|
|
|
Slab merging
|
2018-03-21 22:22:37 +03:00
|
|
|
============
|
2007-05-07 01:49:47 +04:00
|
|
|
|
2007-05-31 11:40:47 +04:00
|
|
|
If no debug options are specified then SLUB may merge similar slabs together
|
2007-05-07 01:49:47 +04:00
|
|
|
in order to reduce overhead and increase cache hotness of objects.
|
2018-03-21 22:22:37 +03:00
|
|
|
``slabinfo -a`` displays which slabs were merged together.
|
2007-05-07 01:49:47 +04:00
|
|
|
|
2007-05-31 11:40:47 +04:00
|
|
|
Slab validation
|
2018-03-21 22:22:37 +03:00
|
|
|
===============
|
2007-05-31 11:40:47 +04:00
|
|
|
|
|
|
|
SLUB can validate all object if the kernel was booted with slub_debug. In
|
2018-03-21 22:22:37 +03:00
|
|
|
order to do so you must have the ``slabinfo`` tool. Then you can do
|
|
|
|
::
|
2007-05-31 11:40:47 +04:00
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
slabinfo -v
|
2007-05-31 11:40:47 +04:00
|
|
|
|
|
|
|
which will test all objects. Output will be generated to the syslog.
|
|
|
|
|
|
|
|
This also works in a more limited way if boot was without slab debug.
|
2018-03-21 22:22:37 +03:00
|
|
|
In that case ``slabinfo -v`` simply tests all reachable objects. Usually
|
2007-05-31 11:40:47 +04:00
|
|
|
these are in the cpu slabs and the partial slabs. Full slabs are not
|
|
|
|
tracked by SLUB in a non debug situation.
|
|
|
|
|
2007-05-07 01:49:47 +04:00
|
|
|
Getting more performance
|
2018-03-21 22:22:37 +03:00
|
|
|
========================
|
2007-05-07 01:49:47 +04:00
|
|
|
|
|
|
|
To some degree SLUB's performance is limited by the need to take the
|
|
|
|
list_lock once in a while to deal with partial slabs. That overhead is
|
|
|
|
governed by the order of the allocation for each slab. The allocations
|
|
|
|
can be influenced by kernel parameters:
|
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
.. slub_min_objects=x (default 4)
|
|
|
|
.. slub_min_order=x (default 0)
|
|
|
|
.. slub_max_order=x (default 3 (PAGE_ALLOC_COSTLY_ORDER))
|
|
|
|
|
|
|
|
``slub_min_objects``
|
|
|
|
allows to specify how many objects must at least fit into one
|
|
|
|
slab in order for the allocation order to be acceptable. In
|
|
|
|
general slub will be able to perform this number of
|
|
|
|
allocations on a slab without consulting centralized resources
|
|
|
|
(list_lock) where contention may occur.
|
|
|
|
|
|
|
|
``slub_min_order``
|
2019-01-31 07:06:21 +03:00
|
|
|
specifies a minimum order of slabs. A similar effect like
|
2018-03-21 22:22:37 +03:00
|
|
|
``slub_min_objects``.
|
|
|
|
|
|
|
|
``slub_max_order``
|
|
|
|
specified the order at which ``slub_min_objects`` should no
|
|
|
|
longer be checked. This is useful to avoid SLUB trying to
|
|
|
|
generate super large order pages to fit ``slub_min_objects``
|
|
|
|
of a slab cache with large object sizes into one high order
|
|
|
|
page. Setting command line parameter
|
|
|
|
``debug_guardpage_minorder=N`` (N > 0), forces setting
|
|
|
|
``slub_max_order`` to 0, what cause minimum possible order of
|
|
|
|
slabs allocation.
|
2007-05-07 01:49:47 +04:00
|
|
|
|
2007-05-31 11:40:47 +04:00
|
|
|
SLUB Debug output
|
2018-03-21 22:22:37 +03:00
|
|
|
=================
|
|
|
|
|
|
|
|
Here is a sample of slub debug output::
|
|
|
|
|
|
|
|
====================================================================
|
|
|
|
BUG kmalloc-8: Redzone overwritten
|
|
|
|
--------------------------------------------------------------------
|
|
|
|
|
|
|
|
INFO: 0xc90f6d28-0xc90f6d2b. First byte 0x00 instead of 0xcc
|
|
|
|
INFO: Slab 0xc528c530 flags=0x400000c3 inuse=61 fp=0xc90f6d58
|
|
|
|
INFO: Object 0xc90f6d20 @offset=3360 fp=0xc90f6d58
|
|
|
|
INFO: Allocated in get_modalias+0x61/0xf5 age=53 cpu=1 pid=554
|
|
|
|
|
|
|
|
Bytes b4 0xc90f6d10: 00 00 00 00 00 00 00 00 5a 5a 5a 5a 5a 5a 5a 5a ........ZZZZZZZZ
|
|
|
|
Object 0xc90f6d20: 31 30 31 39 2e 30 30 35 1019.005
|
|
|
|
Redzone 0xc90f6d28: 00 cc cc cc .
|
|
|
|
Padding 0xc90f6d50: 5a 5a 5a 5a 5a 5a 5a 5a ZZZZZZZZ
|
|
|
|
|
|
|
|
[<c010523d>] dump_trace+0x63/0x1eb
|
|
|
|
[<c01053df>] show_trace_log_lvl+0x1a/0x2f
|
|
|
|
[<c010601d>] show_trace+0x12/0x14
|
|
|
|
[<c0106035>] dump_stack+0x16/0x18
|
|
|
|
[<c017e0fa>] object_err+0x143/0x14b
|
|
|
|
[<c017e2cc>] check_object+0x66/0x234
|
|
|
|
[<c017eb43>] __slab_free+0x239/0x384
|
|
|
|
[<c017f446>] kfree+0xa6/0xc6
|
|
|
|
[<c02e2335>] get_modalias+0xb9/0xf5
|
|
|
|
[<c02e23b7>] dmi_dev_uevent+0x27/0x3c
|
|
|
|
[<c027866a>] dev_uevent+0x1ad/0x1da
|
|
|
|
[<c0205024>] kobject_uevent_env+0x20a/0x45b
|
|
|
|
[<c020527f>] kobject_uevent+0xa/0xf
|
|
|
|
[<c02779f1>] store_uevent+0x4f/0x58
|
|
|
|
[<c027758e>] dev_attr_store+0x29/0x2f
|
|
|
|
[<c01bec4f>] sysfs_write_file+0x16e/0x19c
|
|
|
|
[<c0183ba7>] vfs_write+0xd1/0x15a
|
|
|
|
[<c01841d7>] sys_write+0x3d/0x72
|
|
|
|
[<c0104112>] sysenter_past_esp+0x5f/0x99
|
|
|
|
[<b7f7b410>] 0xb7f7b410
|
|
|
|
=======================
|
|
|
|
|
|
|
|
FIX kmalloc-8: Restoring Redzone 0xc90f6d28-0xc90f6d2b=0xcc
|
2007-05-31 11:40:47 +04:00
|
|
|
|
2007-07-17 15:03:18 +04:00
|
|
|
If SLUB encounters a corrupted object (full detection requires the kernel
|
|
|
|
to be booted with slub_debug) then the following output will be dumped
|
|
|
|
into the syslog:
|
2007-05-31 11:40:47 +04:00
|
|
|
|
2007-07-17 15:03:18 +04:00
|
|
|
1. Description of the problem encountered
|
2007-05-31 11:40:47 +04:00
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
This will be a message in the system log starting with::
|
2007-05-31 11:40:47 +04:00
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
===============================================
|
|
|
|
BUG <slab cache affected>: <What went wrong>
|
|
|
|
-----------------------------------------------
|
2007-05-31 11:40:47 +04:00
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
INFO: <corruption start>-<corruption_end> <more info>
|
|
|
|
INFO: Slab <address> <slab information>
|
|
|
|
INFO: Object <address> <object information>
|
|
|
|
INFO: Allocated in <kernel function> age=<jiffies since alloc> cpu=<allocated by
|
2007-07-17 15:03:18 +04:00
|
|
|
cpu> pid=<pid of the process>
|
2018-03-21 22:22:37 +03:00
|
|
|
INFO: Freed in <kernel function> age=<jiffies since free> cpu=<freed by cpu>
|
|
|
|
pid=<pid of the process>
|
2007-05-31 11:40:47 +04:00
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
(Object allocation / free information is only available if SLAB_STORE_USER is
|
|
|
|
set for the slab. slub_debug sets that option)
|
2007-05-31 11:40:47 +04:00
|
|
|
|
2007-07-17 15:03:18 +04:00
|
|
|
2. The object contents if an object was involved.
|
2007-05-31 11:40:47 +04:00
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
Various types of lines can follow the BUG SLUB line:
|
2007-05-31 11:40:47 +04:00
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
Bytes b4 <address> : <bytes>
|
2007-07-17 15:03:18 +04:00
|
|
|
Shows a few bytes before the object where the problem was detected.
|
2007-05-31 11:40:47 +04:00
|
|
|
Can be useful if the corruption does not stop with the start of the
|
|
|
|
object.
|
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
Object <address> : <bytes>
|
2007-05-31 11:40:47 +04:00
|
|
|
The bytes of the object. If the object is inactive then the bytes
|
2007-07-17 15:03:18 +04:00
|
|
|
typically contain poison values. Any non-poison value shows a
|
2007-05-31 11:40:47 +04:00
|
|
|
corruption by a write after free.
|
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
Redzone <address> : <bytes>
|
2007-07-17 15:03:18 +04:00
|
|
|
The Redzone following the object. The Redzone is used to detect
|
2007-05-31 11:40:47 +04:00
|
|
|
writes after the object. All bytes should always have the same
|
|
|
|
value. If there is any deviation then it is due to a write after
|
|
|
|
the object boundary.
|
|
|
|
|
2007-07-17 15:03:18 +04:00
|
|
|
(Redzone information is only available if SLAB_RED_ZONE is set.
|
|
|
|
slub_debug sets that option)
|
2007-05-31 11:40:47 +04:00
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
Padding <address> : <bytes>
|
2007-05-31 11:40:47 +04:00
|
|
|
Unused data to fill up the space in order to get the next object
|
|
|
|
properly aligned. In the debug case we make sure that there are
|
2007-07-17 15:03:18 +04:00
|
|
|
at least 4 bytes of padding. This allows the detection of writes
|
2007-05-31 11:40:47 +04:00
|
|
|
before the object.
|
|
|
|
|
2007-07-17 15:03:18 +04:00
|
|
|
3. A stackdump
|
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
The stackdump describes the location where the error was detected. The cause
|
|
|
|
of the corruption is may be more likely found by looking at the function that
|
|
|
|
allocated or freed the object.
|
2007-07-17 15:03:18 +04:00
|
|
|
|
|
|
|
4. Report on how the problem was dealt with in order to ensure the continued
|
2018-03-21 22:22:37 +03:00
|
|
|
operation of the system.
|
2007-07-17 15:03:18 +04:00
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
These are messages in the system log beginning with::
|
2007-07-17 15:03:18 +04:00
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
FIX <slab cache affected>: <corrective action taken>
|
2007-07-17 15:03:18 +04:00
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
In the above sample SLUB found that the Redzone of an active object has
|
|
|
|
been overwritten. Here a string of 8 characters was written into a slab that
|
|
|
|
has the length of 8 characters. However, a 8 character string needs a
|
|
|
|
terminating 0. That zero has overwritten the first byte of the Redzone field.
|
|
|
|
After reporting the details of the issue encountered the FIX SLUB message
|
|
|
|
tells us that SLUB has restored the Redzone to its proper value and then
|
|
|
|
system operations continue.
|
2007-07-17 15:03:18 +04:00
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
Emergency operations
|
|
|
|
====================
|
2007-07-17 15:03:18 +04:00
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
Minimal debugging (sanity checks alone) can be enabled by booting with::
|
2007-07-17 15:03:18 +04:00
|
|
|
|
|
|
|
slub_debug=F
|
|
|
|
|
|
|
|
This will be generally be enough to enable the resiliency features of slub
|
|
|
|
which will keep the system running even if a bad kernel component will
|
|
|
|
keep corrupting objects. This may be important for production systems.
|
|
|
|
Performance will be impacted by the sanity checks and there will be a
|
|
|
|
continual stream of error messages to the syslog but no additional memory
|
|
|
|
will be used (unlike full debugging).
|
|
|
|
|
|
|
|
No guarantees. The kernel component still needs to be fixed. Performance
|
|
|
|
may be optimized further by locating the slab that experiences corruption
|
|
|
|
and enabling debugging only for that cache
|
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
I.e.::
|
2007-07-17 15:03:18 +04:00
|
|
|
|
|
|
|
slub_debug=F,dentry
|
|
|
|
|
|
|
|
If the corruption occurs by writing after the end of the object then it
|
|
|
|
may be advisable to enable a Redzone to avoid corrupting the beginning
|
2018-03-21 22:22:37 +03:00
|
|
|
of other objects::
|
2007-07-17 15:03:18 +04:00
|
|
|
|
|
|
|
slub_debug=FZ,dentry
|
2007-05-31 11:40:47 +04:00
|
|
|
|
2015-10-23 02:51:45 +03:00
|
|
|
Extended slabinfo mode and plotting
|
2018-03-21 22:22:37 +03:00
|
|
|
===================================
|
2015-10-23 02:51:45 +03:00
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
The ``slabinfo`` tool has a special 'extended' ('-X') mode that includes:
|
2015-10-23 02:51:45 +03:00
|
|
|
- Slabcache Totals
|
|
|
|
- Slabs sorted by size (up to -N <num> slabs, default 1)
|
|
|
|
- Slabs sorted by loss (up to -N <num> slabs, default 1)
|
|
|
|
|
2018-03-21 22:22:37 +03:00
|
|
|
Additionally, in this mode ``slabinfo`` does not dynamically scale
|
|
|
|
sizes (G/M/K) and reports everything in bytes (this functionality is
|
|
|
|
also available to other slabinfo modes via '-B' option) which makes
|
|
|
|
reporting more precise and accurate. Moreover, in some sense the `-X'
|
|
|
|
mode also simplifies the analysis of slabs' behaviour, because its
|
|
|
|
output can be plotted using the ``slabinfo-gnuplot.sh`` script. So it
|
|
|
|
pushes the analysis from looking through the numbers (tons of numbers)
|
|
|
|
to something easier -- visual analysis.
|
2015-10-23 02:51:45 +03:00
|
|
|
|
|
|
|
To generate plots:
|
2018-03-21 22:22:37 +03:00
|
|
|
|
|
|
|
a) collect slabinfo extended records, for example::
|
|
|
|
|
|
|
|
while [ 1 ]; do slabinfo -X >> FOO_STATS; sleep 1; done
|
|
|
|
|
|
|
|
b) pass stats file(-s) to ``slabinfo-gnuplot.sh`` script::
|
|
|
|
|
|
|
|
slabinfo-gnuplot.sh FOO_STATS [FOO_STATS2 .. FOO_STATSN]
|
|
|
|
|
|
|
|
The ``slabinfo-gnuplot.sh`` script will pre-processes the collected records
|
|
|
|
and generates 3 png files (and 3 pre-processing cache files) per STATS
|
|
|
|
file:
|
|
|
|
- Slabcache Totals: FOO_STATS-totals.png
|
|
|
|
- Slabs sorted by size: FOO_STATS-slabs-by-size.png
|
|
|
|
- Slabs sorted by loss: FOO_STATS-slabs-by-loss.png
|
|
|
|
|
|
|
|
Another use case, when ``slabinfo-gnuplot.sh`` can be useful, is when you
|
|
|
|
need to compare slabs' behaviour "prior to" and "after" some code
|
|
|
|
modification. To help you out there, ``slabinfo-gnuplot.sh`` script
|
|
|
|
can 'merge' the `Slabcache Totals` sections from different
|
|
|
|
measurements. To visually compare N plots:
|
|
|
|
|
|
|
|
a) Collect as many STATS1, STATS2, .. STATSN files as you need::
|
|
|
|
|
|
|
|
while [ 1 ]; do slabinfo -X >> STATS<X>; sleep 1; done
|
|
|
|
|
|
|
|
b) Pre-process those STATS files::
|
|
|
|
|
|
|
|
slabinfo-gnuplot.sh STATS1 STATS2 .. STATSN
|
|
|
|
|
|
|
|
c) Execute ``slabinfo-gnuplot.sh`` in '-t' mode, passing all of the
|
|
|
|
generated pre-processed \*-totals::
|
|
|
|
|
|
|
|
slabinfo-gnuplot.sh -t STATS1-totals STATS2-totals .. STATSN-totals
|
|
|
|
|
|
|
|
This will produce a single plot (png file).
|
|
|
|
|
|
|
|
Plots, expectedly, can be large so some fluctuations or small spikes
|
|
|
|
can go unnoticed. To deal with that, ``slabinfo-gnuplot.sh`` has two
|
|
|
|
options to 'zoom-in'/'zoom-out':
|
|
|
|
|
2020-10-22 17:26:53 +03:00
|
|
|
a) ``-s %d,%d`` -- overwrites the default image width and height
|
2018-03-21 22:22:37 +03:00
|
|
|
b) ``-r %d,%d`` -- specifies a range of samples to use (for example,
|
|
|
|
in ``slabinfo -X >> FOO_STATS; sleep 1;`` case, using a ``-r
|
|
|
|
40,60`` range will plot only samples collected between 40th and
|
|
|
|
60th seconds).
|
2015-10-23 02:51:45 +03:00
|
|
|
|
2008-07-04 20:59:22 +04:00
|
|
|
Christoph Lameter, May 30, 2007
|
2015-10-23 02:51:45 +03:00
|
|
|
Sergey Senozhatsky, October 23, 2015
|