Re: [v3.17-rc8] LTP oom testsuite produces OOPS

From: Sedat Dilek
Date: Tue Mar 25 2014 - 16:32:02 EST


On Tue, Mar 25, 2014 at 4:56 PM, Andrew Morton
<akpm@xxxxxxxxxxxxxxxxxxxx> wrote:
> On Tue, 25 Mar 2014 08:45:34 +0100 Sedat Dilek <sedat.dilek@xxxxxxxxx> wrote:
>
>> Hi,
>>
>> as reported in [1] in my post-scriptum I see several OOPs when running
>> LTP and OOM tests (here: oom3).
>> Linus requested to send you mm-folks my bug-report.
>>
>> # cd /opt/ltp/
>>
>> # cat Version
>> 20140115
>>
>> # ./testcases/bin/oom03
>>
>> I have tested with latest LTP (15-Jan-2014).
>>
>> If you need additional information, please let me know.
>
> I don't actually see any oopses there. There are some stack traces
> associated with the oom-killing events:
>
>
> [ 104.383349] Memory cgroup out of memory: Kill process 2518 (oom03) score 777 or sacrifice child
> [ 104.383352] Killed process 2518 (oom03) total-vm:3152196kB, anon-rss:1048444kB, file-rss:192kB
> [ 107.946908] oom03 invoked oom-killer: gfp_mask=0xd0, order=0, oom_score_adj=0
> [ 107.946912] oom03 cpuset=/ mems_allowed=0
> [ 107.946915] CPU: 0 PID: 2521 Comm: oom03 Not tainted 3.14.0-rc8-1-iniza-small #1
> [ 107.946917] Hardware name: SAMSUNG ELECTRONICS CO., LTD. 530U3BI/530U4BI/530U4BH/530U3BI/530U4BI/530U4BH, BIOS 13XK 03/28/2013
> [ 107.946919] ffff880108e58400 ffff880064125be8 ffffffff8170fb45 0000000000000007
> [ 107.946923] ffff880108e7a190 ffff880064125c68 ffffffff8170c6e5 ffff880064125cd8
> [ 107.946925] ffff88010b686858 ffff88011fdf9d80 ffff88010b6863d0 0000000000000000
> [ 107.946928] Call Trace:
> [ 107.946934] [<ffffffff8170fb45>] dump_stack+0x46/0x58
> [ 107.946937] [<ffffffff8170c6e5>] dump_header+0x7e/0x1c3
> [ 107.946942] [<ffffffff813881e0>] ? ___ratelimit+0xa0/0x120
> [ 107.946946] [<ffffffff811580e4>] oom_kill_process+0x214/0x370
> [ 107.946949] [<ffffffff81075ce5>] ? has_ns_capability_noaudit+0x15/0x20
> ...
>
> But these aren't oopses - mm/oom_kill.c:dump_header() deliberately
> performs a dump_stack() while reporting on the event.
>

Indicated by the lines '...expected victim is...' and according to you
'expected'?

# cd /opt/ltp/

# ./testcases/bin/oom03
oom03 0 TINFO : set overcommit_memory to 1
oom03 0 TINFO : start normal OOM testing.
oom03 0 TINFO : expected victim is 2834.
oom03 0 TINFO : allocating 3221225472 bytes.
oom03 0 TINFO : start OOM testing for mlocked pages.
oom03 0 TINFO : expected victim is 2835.
oom03 0 TINFO : allocating 3221225472 bytes.
oom03 0 TINFO : start OOM testing for KSM pages.
oom03 0 TINFO : expected victim is 2836.
oom03 0 TINFO : allocating 3221225472 bytes.
oom03 1 TCONF : memcg swap accounting is disabled
oom03 0 TINFO : set overcommit_memory to 0

# dmesg | egrep '2834|2835|2836'
[ 1220.478440] CPU: 0 PID: 2834 Comm: oom03 Not tainted
3.14.0-rc8-1-iniza-small #1
[ 1220.478552] [ 2834] 0 2834 788049 248262 623 65524
0 oom03
[ 1220.478554] Memory cgroup out of memory: Kill process 2834 (oom03)
score 930 or sacrifice child
[ 1220.478558] Killed process 2834 (oom03) total-vm:3152196kB,
anon-rss:992852kB, file-rss:196kB
[ 1221.544018] CPU: 0 PID: 2835 Comm: oom03 Not tainted
3.14.0-rc8-1-iniza-small #1
[ 1221.544151] [ 2835] 0 2835 788049 262161 522 13
0 oom03
[ 1221.544152] Memory cgroup out of memory: Kill process 2835 (oom03)
score 777 or sacrifice child
[ 1221.544155] Killed process 2835 (oom03) total-vm:3152196kB,
anon-rss:1048452kB, file-rss:192kB
[ 1225.016810] CPU: 0 PID: 2836 Comm: oom03 Not tainted
3.14.0-rc8-1-iniza-small #1
[ 1225.016941] [ 2836] 0 2836 788049 250693 628 65535
0 oom03
[ 1225.016943] Memory cgroup out of memory: Kill process 2836 (oom03)
score 937 or sacrifice child
[ 1225.016945] Killed process 2836 (oom03) total-vm:3152196kB,
anon-rss:1002564kB, file-rss:208kB

- Sedat -
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/