gcc -o slabinfo tools/vm/slabinfo.c

Slab merging¶

If no debug options are specified then SLUB may merge similar slabs togetherin order to reduce overhead and increase cache hotness of objects.slabinfo-a displays which slabs were merged together.

Slab validation¶

SLUB can validate all object if the kernel was booted with slub_debug. Inorder to do so you must have theslabinfo tool. Then you can do

slabinfo -v

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.In that caseslabinfo-v simply tests all reachable objects. Usuallythese are in the cpu slabs and the partial slabs. Full slabs are nottracked by SLUB in a non debug situation.

Getting more performance¶

To some degree SLUB’s performance is limited by the need to take thelist_lock once in a while to deal with partial slabs. That overhead isgoverned by the order of the allocation for each slab. The allocationscan be influenced by kernel parameters:

slub_min_objects

allows to specify how many objects must at least fit into oneslab in order for the allocation order to be acceptable. Ingeneral slub will be able to perform this number ofallocations on a slab without consulting centralized resources(list_lock) where contention may occur.

slub_min_order

specifies a minimum order of slabs. A similar effect likeslub_min_objects.

slub_max_order

specified the order at whichslub_min_objects should nolonger be checked. This is useful to avoid SLUB trying togenerate super large order pages to fitslub_min_objectsof a slab cache with large object sizes into one high orderpage. Setting command line parameterdebug_guardpage_minorder=N (N > 0), forces settingslub_max_order to 0, what cause minimum possible order ofslabs allocation.

SLUB Debug output¶

Here is a sample of slub debug output:

====================================================================BUG kmalloc-8: Redzone overwritten--------------------------------------------------------------------INFO: 0xc90f6d28-0xc90f6d2b. First byte 0x00 instead of 0xccINFO: Slab 0xc528c530 flags=0x400000c3 inuse=61 fp=0xc90f6d58INFO: Object 0xc90f6d20 @offset=3360 fp=0xc90f6d58INFO: Allocated in get_modalias+0x61/0xf5 age=53 cpu=1 pid=554Bytes 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

If SLUB encounters a corrupted object (full detection requires the kernelto be booted with slub_debug) then the following output will be dumpedinto the syslog:

1. Description of the problem encountered

   This will be a message in the system log starting with:

   ===============================================BUG <slab cache affected>: <What went wrong>-----------------------------------------------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   cpu> pid=<pid of the process>INFO: Freed in <kernel function> age=<jiffies since free> cpu=<freed by cpu>   pid=<pid of the process>

   (Object allocation / free information is only available if SLAB_STORE_USER isset for the slab. slub_debug sets that option)

2. The object contents if an object was involved.

   Various types of lines can follow the BUG SLUB line:

   Bytes b4 <address>:<bytes>

   Shows a few bytes before the object where the problem was detected.Can be useful if the corruption does not stop with the start of theobject.

   Object <address>:<bytes>

   The bytes of the object. If the object is inactive then the bytestypically contain poison values. Any non-poison value shows acorruption by a write after free.

   Redzone <address>:<bytes>

   The Redzone following the object. The Redzone is used to detectwrites after the object. All bytes should always have the samevalue. If there is any deviation then it is due to a write afterthe object boundary.

   (Redzone information is only available if SLAB_RED_ZONE is set.slub_debug sets that option)

   Padding <address>:<bytes>

   Unused data to fill up the space in order to get the next objectproperly aligned. In the debug case we make sure that there areat least 4 bytes of padding. This allows the detection of writesbefore the object.

3. A stackdump

   The stackdump describes the location where the error was detected. The causeof the corruption is may be more likely found by looking at the function thatallocated or freed the object.

4. Report on how the problem was dealt with in order to ensure the continuedoperation of the system.

   These are messages in the system log beginning with:

   FIX <slab cache affected>: <corrective action taken>

   In the above sample SLUB found that the Redzone of an active object hasbeen overwritten. Here a string of 8 characters was written into a slab thathas the length of 8 characters. However, a 8 character string needs aterminating 0. That zero has overwritten the first byte of the Redzone field.After reporting the details of the issue encountered the FIX SLUB messagetells us that SLUB has restored the Redzone to its proper value and thensystem operations continue.

Emergency operations¶

Minimal debugging (sanity checks alone) can be enabled by booting with:

slub_debug=F

This will be generally be enough to enable the resiliency features of slubwhich will keep the system running even if a bad kernel component willkeep corrupting objects. This may be important for production systems.Performance will be impacted by the sanity checks and there will be acontinual stream of error messages to the syslog but no additional memorywill be used (unlike full debugging).

No guarantees. The kernel component still needs to be fixed. Performancemay be optimized further by locating the slab that experiences corruptionand enabling debugging only for that cache

I.e.:

slub_debug=F,dentry

If the corruption occurs by writing after the end of the object then itmay be advisable to enable a Redzone to avoid corrupting the beginningof other objects:

slub_debug=FZ,dentry

Extended slabinfo mode and plotting¶

Theslabinfo tool has a special ‘extended’ (‘-X’) mode that includes:

• Slabcache Totals
• Slabs sorted by size (up to -N <num> slabs, default 1)
• Slabs sorted by loss (up to -N <num> slabs, default 1)

Additionally, in this modeslabinfo does not dynamically scalesizes (G/M/K) and reports everything in bytes (this functionality isalso available to other slabinfo modes via ‘-B’ option) which makesreporting more precise and accurate. Moreover, in some sense the-X’mode also simplifies the analysis of slabs’ behaviour, because itsoutput can be plotted using the ``slabinfo-gnuplot.sh` script. So itpushes the analysis from looking through the numbers (tons of numbers)to something easier – visual analysis.

To generate plots:

1. collect slabinfo extended records, for example:

   while [ 1 ]; do slabinfo -X >> FOO_STATS; sleep 1; done

2. pass stats file(-s) toslabinfo-gnuplot.sh script:

   slabinfo-gnuplot.sh FOO_STATS [FOO_STATS2 .. FOO_STATSN]

   Theslabinfo-gnuplot.sh script will pre-processes the collected recordsand generates 3 png files (and 3 pre-processing cache files) per STATSfile:- 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, whenslabinfo-gnuplot.sh can be useful, is when youneed to compare slabs’ behaviour “prior to” and “after” some codemodification. To help you out there,slabinfo-gnuplot.sh scriptcan ‘merge’ theSlabcache Totals sections from differentmeasurements. To visually compare N plots:

1. Collect as many STATS1, STATS2, .. STATSN files as you need:

   while [ 1 ]; do slabinfo -X >> STATS<X>; sleep 1; done

2. Pre-process those STATS files:

   slabinfo-gnuplot.sh STATS1 STATS2 .. STATSN

3. Executeslabinfo-gnuplot.sh in ‘-t’ mode, passing all of thegenerated 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 spikescan go unnoticed. To deal with that,slabinfo-gnuplot.sh has twooptions to ‘zoom-in’/’zoom-out’:

   1. -s%d,%d – overwrites the default image width and heigh
   2. -r%d,%d – specifies a range of samples to use (for example,inslabinfo-X>>FOO_STATS;sleep1; case, using a-r40,60 range will plot only samples collected between 40th and60th seconds).

Christoph Lameter, May 30, 2007Sergey Senozhatsky, October 23, 2015