Bug 65481 - Page allocation failure errors on Trinity when playing a HL2 mod (i386 code on 64 bit kernel )
Summary: Page allocation failure errors on Trinity when playing a HL2 mod (i386 code o...
Status: RESOLVED WILL_NOT_FIX
Alias: None
Product: Drivers
Classification: Unclassified
Component: Video(DRI - non Intel) (show other bugs)
Hardware: x86-64 Linux
: P1 normal
Assignee: drivers_video-dri
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2013-11-22 13:49 UTC by Kertesz Laszlo
Modified: 2013-11-26 23:10 UTC (History)
2 users (show)

See Also:
Kernel Version: 3.12.0 git
Subsystem:
Regression: No
Bisected commit-id:


Attachments
full dmesg showing the errors (117.45 KB, text/plain)
2013-11-22 13:49 UTC, Kertesz Laszlo
Details

Description Kertesz Laszlo 2013-11-22 13:49:45 UTC
Created attachment 115641 [details]
full dmesg showing the errors

I use a 64 bit Debian system with kernel (mainline, now 3.12.0-10928-g527d151), mesa, llvm, drm, xf86-ati compiled from git on an A8-5500 APU (radeon/r600 driver, .
While i played a Half Life 2 mod called No more room in hell on Steam (32-bit application) i noticed that dmesg is full of errors like:

hl2_linux: page allocation failure: order:4, mode:0x2000d0
CPU: 0 PID: 17112 Comm: hl2_linux Not tainted 3.12.0-10928-g527d151-dirty #4
Hardware name: Gigabyte Technology Co., Ltd. To be filled by O.E.M./F2A85X-D3H, BIOS F1 10/09/2012
 ffff880136a03988 ffffffff8143ae44 00000000002000d0 ffffffff81100853
 0000000000000000 0000000000000002 0000000000000000 ffff88013eff8b00
 0000000000000001 ffffffff8111610d ffff88013eff8b08 0000000000000002
Call Trace:
 [<ffffffff8143ae44>] ? dump_stack+0x41/0x51
 [<ffffffff81100853>] ? warn_alloc_failed+0xe3/0x130
 [<ffffffff8111610d>] ? next_online_pgdat+0x1d/0x50
 [<ffffffff811048c0>] ? __alloc_pages_nodemask+0x860/0xa30
 [<ffffffff8114461f>] ? kmem_getpages+0x5f/0x1a0
 [<ffffffff811452fc>] ? fallback_alloc+0x16c/0x250
 [<ffffffff81146f76>] ? __kmalloc+0x2b6/0x490
 [<ffffffffa033707e>] ? radeon_cs_ioctl+0x1ee/0x990 [radeon]
 [<ffffffffa033707e>] ? radeon_cs_ioctl+0x1ee/0x990 [radeon]
 [<ffffffffa026a22d>] ? drm_ioctl+0x46d/0x570 [drm]
 [<ffffffffa0306044>] ? radeon_drm_ioctl+0x44/0x80 [radeon]
 [<ffffffffa03cedfb>] ? radeon_kms_compat_ioctl+0xb/0x20 [radeon]
 [<ffffffff811a94bf>] ? compat_sys_ioctl+0xbf/0x1110
 [<ffffffff810a7c29>] ? compat_SyS_futex+0x69/0x140
 [<ffffffff814490d5>] ? cstar_dispatch+0x7/0x1a
Mem-Info:
Node 0 DMA per-cpu:
CPU    0: hi:    0, btch:   1 usd:   0
CPU    1: hi:    0, btch:   1 usd:   0
CPU    2: hi:    0, btch:   1 usd:   0
CPU    3: hi:    0, btch:   1 usd:   0
Node 0 DMA32 per-cpu:
CPU    0: hi:  186, btch:  31 usd:   0
CPU    1: hi:  186, btch:  31 usd:   0
CPU    2: hi:  186, btch:  31 usd:   0
CPU    3: hi:  186, btch:  31 usd:   0
Node 0 Normal per-cpu:
CPU    0: hi:  186, btch:  31 usd:   0
CPU    1: hi:  186, btch:  31 usd:   0
CPU    2: hi:  186, btch:  31 usd:   0
CPU    3: hi:  186, btch:  31 usd:   0
active_anon:323486 inactive_anon:159407 isolated_anon:0
 active_file:127844 inactive_file:121901 isolated_file:0
 unevictable:12 dirty:2 writeback:0 unstable:0
 free:42704 slab_reclaimable:13044 slab_unreclaimable:11651
 mapped:28143 shmem:1782 pagetables:6574 bounce:0
 free_cma:0
Node 0 DMA free:13916kB min:308kB low:384kB high:460kB active_anon:440kB inactive_anon:520kB active_file:4kB inactive_file:4kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:15988kB managed:15904kB mlocked:0kB dirty:0kB writeback:0kB mapped:8kB shmem:0kB slab_reclaimable:24kB slab_unreclaimable:712kB kernel_stack:16kB pagetables:92kB unstable:0kB bounce:0kB free_cma:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no
lowmem_reserve[]: 0 2433 3389 3389
Node 0 DMA32 free:116396kB min:48284kB low:60352kB high:72424kB active_anon:1020080kB inactive_anon:363328kB active_file:385068kB inactive_file:366056kB unevictable:40kB isolated(anon):0kB isolated(file):0kB present:2567696kB managed:2494688kB mlocked:40kB dirty:8kB writeback:0kB mapped:80552kB shmem:5240kB slab_reclaimable:34832kB slab_unreclaimable:26316kB kernel_stack:2312kB pagetables:18104kB unstable:0kB bounce:0kB free_cma:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no
lowmem_reserve[]: 0 0 956 956
Node 0 Normal free:40504kB min:18988kB low:23732kB high:28480kB active_anon:273424kB inactive_anon:273780kB active_file:126304kB inactive_file:121544kB unevictable:8kB isolated(anon):0kB isolated(file):0kB present:1032188kB managed:979784kB mlocked:8kB dirty:0kB writeback:0kB mapped:32012kB shmem:1888kB slab_reclaimable:17320kB slab_unreclaimable:19576kB kernel_stack:944kB pagetables:8100kB unstable:0kB bounce:0kB free_cma:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no
lowmem_reserve[]: 0 0 0 0
Node 0 DMA: 45*4kB (UEM) 43*8kB (UEM) 17*16kB (UEM) 6*32kB (UM) 4*64kB (UEM) 3*128kB (UEM) 2*256kB (UE) 1*512kB (E) 1*1024kB (E) 3*2048kB (EMR) 1*4096kB (M) = 13916kB
Node 0 DMA32: 10239*4kB (UEM) 7669*8kB (UEM) 640*16kB (UEM) 103*32kB (UM) 11*64kB (UM) 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 116548kB
Node 0 Normal: 8793*4kB (UEM) 217*8kB (UEM) 220*16kB (UM) 9*32kB (M) 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 40716kB
Node 0 hugepages_total=0 hugepages_free=0 hugepages_surp=0 hugepages_size=2048kB
251617 total pagecache pages
92 pages in swap cache
Swap cache stats: add 6593, delete 6501, find 231/244
Free swap  = 6802856kB
Total swap = 6828756kB
903968 pages RAM
0 pages HighMem/MovableOnly
13101 pages reserved
SLAB: Unable to allocate memory on node 0 (gfp=0xd0)
  cache: kmalloc-65536, object size: 65536, order: 4
  node 0: slabs: 4/4, objs: 4/4, free: 0

Repeated many times. And the last error messagesss were only:

[drm:radeon_cs_ioctl] *ERROR* Failed to parse relocation -12!

I didnt observe any adverse effects on system stability.
This is some type of memory allocation issue, maybe related to 32 bit code running on 64 bit kernel?
Comment 1 Kertesz Laszlo 2013-11-23 14:13:51 UTC
After geting the 3.13 rc1 kernel via git, this doesnt happen anymore. Closed.
Comment 2 Kertesz Laszlo 2013-11-24 01:25:36 UTC
It seems i was too quick to close it. It still happens with 3.13 rc1 kernel too.
The thing is that these errors appear only after a few hours of play (which i dont do often).
Comment 3 Alan 2013-11-26 21:47:25 UTC
Not actually a bug - it's just the machine warning a lot that its struggling to find enough memory immediately. I guess the game is stretching it to the limit. Its dumped as a log warning as for server setups its often very useful to know and adjust workload parameters at that point.

The DRM one looks like a graphics out of memory case so I don't think this is actually a bug.
Comment 4 Alex Deucher 2013-11-26 22:33:07 UTC
(In reply to Alan from comment #3)
> The DRM one looks like a graphics out of memory case so I don't think this
> is actually a bug.

Yes, the drm one is also an out of memory case.

Note You need to log in before you can comment on or make changes to this bug.