Bug 72506

Summary: [IVB/HSW/BYT] igt/gem_ctx_exec/eviction fails with OOM killer
Product: DRI Reporter: lu hua <huax.lu>
Component: DRM/IntelAssignee: Intel GFX Bugs mailing list <intel-gfx-bugs>
Status: CLOSED DUPLICATE QA Contact: Intel GFX Bugs mailing list <intel-gfx-bugs>
Severity: normal    
Priority: medium CC: intel-gfx-bugs
Version: unspecified   
Hardware: All   
OS: Linux (All)   
Whiteboard:
i915 platform: i915 features:
Attachments:
Description Flags
dmesg
none
BYT dmesg none

Description lu hua 2013-12-09 07:18:45 UTC
Created attachment 90490 [details]
dmesg

System Environment:
--------------------------
Arch:             x86_64
Platform:         ivybridge 
Kernel:      (drm-intel-nightly)ddd0a9e2da6087c7aac74aca34fcdc90645cbdb4

Bug detailed description:
---------------------------
It fails with OOM Killer on ivybridge and haswell with -nightly and -queued kernel. It fails on -fixes kernel.
It's a new case.

output:
Killed

Reproduce steps:
-------------------------
1. ./gem_ctx_exec --run-subtest eviction
Comment 1 Daniel Vetter 2013-12-09 10:18:46 UTC
Since the OOM killer is a regression in -nightly compared to -fixes (where the test only fails) can you please try to bisect which commit in -nightly causes the OOM?

The test checks for available system ram, so OOM is an additional bug (and not what I've expected).

Functional failure exercised by the tests is tracked in bug #72507 (and this is what I expected to see actually).
Comment 2 lu hua 2013-12-10 07:43:27 UTC
Retest on latest -nightly kernel. It only fails like Bug 72507.
Retest on latest -queued kernel, It has another Bug 72557.
Comment 3 Daniel Vetter 2013-12-10 09:49:15 UTC
Please don't file a new bug report when you have the requested bisect result ...

*** This bug has been marked as a duplicate of bug 72557 ***
Comment 4 lu hua 2013-12-19 05:47:51 UTC
Created attachment 90968 [details]
BYT dmesg

It also fails on BYT with OOM killer.
Kernel: (drm-intel-nightly) 02905525c3310ba673676f8041d0ff6be4da0e24
Call Trace:
[ 3245.610793]  [<ffffffff81711fd2>] ? dump_stack+0x41/0x51
[ 3245.610870]  [<ffffffff8170e5f7>] ? dump_header.isra.8+0x69/0x191
[ 3245.610948]  [<ffffffff8106decd>] ? ktime_get_ts+0x49/0xab
[ 3245.611030]  [<ffffffff812cf45e>] ? ___ratelimit+0xae/0xc8
[ 3245.611109]  [<ffffffff810a3208>] ? oom_kill_process+0x76/0x2f8
[ 3245.611183]  [<ffffffff810a3112>] ? oom_scan_process_thread+0xd/0x8d
[ 3245.611270]  [<ffffffff810a399e>] ? out_of_memory+0x3b2/0x3e5
[ 3245.611352]  [<ffffffff810a70e8>] ? __alloc_pages_nodemask+0x664/0x771
[ 3245.611434]  [<ffffffff810d0623>] ? alloc_pages_current+0xbf/0xdc
[ 3245.611521]  [<ffffffff810a21bc>] ? filemap_fault+0x25c/0x381
[ 3245.611630]  [<ffffffff810b7f9e>] ? __do_fault+0xac/0x3bf
[ 3245.611702]  [<ffffffff810bb6fa>] ? handle_mm_fault+0x1e7/0x7e2
[ 3245.611793]  [<ffffffff8171a32c>] ? __do_page_fault+0x41c/0x469
[ 3245.611881]  [<ffffffff810360fd>] ? timespec_add_safe+0x1e/0x4d
[ 3245.611966]  [<ffffffff810ec3bf>] ? dput+0x16/0xeb
[ 3245.612028]  [<ffffffff8106decd>] ? ktime_get_ts+0x49/0xab
[ 3245.612107]  [<ffffffff810e97f6>] ? poll_select_copy_remaining+0xe0/0xf6
[ 3245.612191]  [<ffffffff817178f2>] ? page_fault+0x22/0x30

[ 3245.615350] [ pid ]   uid  tgid total_vm      rss nr_ptes swapents oom_score_adj name
[ 3245.615466] [ 2226]     0  2226    79067       40     161       52             0 systemd-journal
[ 3245.615575] [ 2946]     0  2946    10344       34      22      209         -1000 systemd-udevd
[ 3245.615713] [ 2975]     0  2975    24466        0      25       49             0 lvmetad
[ 3245.615816] [ 3030]     0  3030    12231       26      24       75         -1000 auditd
[ 3245.615912] [ 3041]     0  3041    20053       13       8       19             0 audispd
[ 3245.616022] [ 3043]     0  3043     5993       14      26       31             0 sedispatch
[ 3245.620365] [ 3050]     0  3050    35163       47      39       38             0 abrtd
[ 3245.624699] [ 3052]     0  3052    34618        6      38       62             0 abrt-watch-log
[ 3245.629006] [ 3054]     0  3054     6069       15      16      136             0 smartd
[ 3245.633300] [ 3059]     0  3059    34618        1      38       66             0 abrt-watch-log
[ 3245.637535] [ 3069]     0  3069     8249       43      20       37             0 systemd-logind
[ 3245.641685] [ 3070]     0  3070    92524      226      62      193             0 NetworkManager
[ 3245.645729] [ 3071]     0  3071    65772        0      29      123             0 rsyslogd
[ 3245.649721] [ 3072]    70  3072     7018       54      30       21             0 avahi-daemon
[ 3245.653773] [ 3074]    81  3074     6258      202      17       36          -900 dbus-daemon
[ 3245.657803] [ 3082]     0  3082     1634        1      11       28             0 mcelog
[ 3245.661794] [ 3083]    70  3083     6985       17      27       37             0 avahi-daemon
[ 3245.665779] [ 3116]   999  3116   132413      461      53      767             0 polkitd
[ 3245.669764] [ 3124]     0  3124    40407       68      67      123          -900 modem-manager
[ 3245.673758] [ 3142]     0  3142    25512       45      50     3068             0 dhclient
[ 3245.677747] [ 3162]     0  3162   132763      255     141      900             0 libvirtd
[ 3245.681680] [ 3174]    32  3174     9422       13      23       81             0 rpcbind
[ 3245.685624] [ 3180]     0  3180    20104       14      41      187         -1000 sshd
[ 3245.689566] [ 3224]     0  3224    25191       90      50      394             0 sendmail
[ 3245.693574] [ 3239]     0  3239    31020       25      19      127             0 crond
[ 3245.697579] [ 3240]     0  3240     5930        0      16       47             0 atd
[ 3245.701565] [ 3249]     0  3249    27498       26      11        3             0 agetty
[ 3245.705550] [ 3259]    51  3259    21453        0      40      376             0 sendmail
[ 3245.709524] [ 3339]     0  3339    32766      284      63        2             0 sshd
[ 3245.713494] [ 3341]     0  3341    29132      383      20        0             0 bash
[ 3245.717458] [ 3464]     0  3464    18496      231      38        0             0 ssh
[ 3245.721365] [22297]     0 22297     5835       69      16        0             0 bluetoothd
[ 3245.725267] [22300]     0 22300    58549      250      83        0             0 upowerd
[ 3245.729153] [22361]     0 22361    96882      460      54        6             0 udisksd
[ 3245.733026] [22392]   998 22392    85556      361      55        6             0 colord
[ 3245.736884] [22401]     0 22401    48180      313      50        0             0 cupsd
[ 3245.740748] [22452]     0 22452    98705      585      46        0             0 accounts-daemon
[ 3245.744647] [22470]     0 22470    79908      613     149        0             0 vmtoolsd
[ 3245.748612] [17151]     0 17151    32799      315      66        0             0 sshd
[ 3245.752571] [17153]     0 17153    29133      383      20        0             0 bash
[ 3245.756524] [23765]     0 23765    27498       29      11        0             0 agetty
[ 3245.760501] [24529]     0 24529    22314     7174      59        0          1000 gem_ctx_exec
[ 3245.764469] Out of memory: Kill process 24529 (gem_ctx_exec) score 974 or sacrifice child
[ 3245.768462] Killed process 24529 (gem_ctx_exec) total-vm:89256kB, anon-rss:28692kB, file-rss:4kB
Comment 5 Daniel Vetter 2014-01-10 08:42:36 UTC
(In reply to comment #4)
> Created attachment 90968 [details]
> BYT dmesg
> 
> It also fails on BYT with OOM killer.

If this is still an isseu please either reopen bug 72557 or file a new bug (for new platforms). Thanks.
Comment 6 lu hua 2014-01-13 08:59:39 UTC
(In reply to comment #5)
> (In reply to comment #4)
> > Created attachment 90968 [details]
> > BYT dmesg
> > 
> > It also fails on BYT with OOM killer.
> 
> If this is still an isseu please either reopen bug 72557 or file a new bug
> (for new platforms). Thanks.


File new Bug 73543.
Comment 7 lu hua 2014-01-14 05:55:00 UTC
Verified.Fixed.
Comment 8 Elizabeth 2017-10-06 14:41:29 UTC
Closing old verified.

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.