Bug 75247 - [BYT]igt/gem_render_linear_blits causes OOM killer
Summary: [BYT]igt/gem_render_linear_blits causes OOM killer
Status: CLOSED FIXED
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/Intel (show other bugs)
Version: unspecified
Hardware: All Linux (All)
: high major
Assignee: Jani Nikula
QA Contact: Intel GFX Bugs mailing list
URL:
Whiteboard:
Keywords:
Depends on: 72742
Blocks:
  Show dependency treegraph
 
Reported: 2014-02-20 07:01 UTC by lu hua
Modified: 2017-10-06 14:39 UTC (History)
1 user (show)

See Also:
i915 platform:
i915 features:


Attachments
dmesg (120.78 KB, text/plain)
2014-02-20 07:01 UTC, lu hua
no flags Details

Description lu hua 2014-02-20 07:01:42 UTC
Created attachment 94411 [details]
dmesg

System Environment:
--------------------------
Platform: Baytrail
Kernel(drm-intel-nightly)1be8f2b4dd6d3db00af24d4891c82d2650bd282d

Bug detailed description:
------------------------- 
gem_render_linear_blits causes OOM Killer.
Run it on Baytrail with -nightly kernel and -queued kernel.
It always fails with OOM killer, run earlier kernel, It also has this issue.

case gem_render_tiled_blits also causes OOM killer.

output:
IGT-Version: 1.5-g06189c6 (x86_64) (Linux: 3.13.0_drm-intel-nightly_1be8f2_20140220+ x86_64)
Using 3072 1MiB buffers
Verifying initialisation...
Cyclic blits, forward...
Cyclic blits, backward...
Killed

[  245.094187] Call Trace:
[  245.094264]  [<ffffffff8171193b>] ? dump_stack+0x41/0x51
[  245.094334]  [<ffffffff8170df1a>] ? dump_header.isra.8+0x69/0x191
[  245.094406]  [<ffffffff8106dff5>] ? ktime_get_ts+0x49/0xab
[  245.094472]  [<ffffffff812cf77a>] ? ___ratelimit+0xae/0xc8
[  245.094538]  [<ffffffff810a33b8>] ? oom_kill_process+0x76/0x2f8
[  245.094607]  [<ffffffff810a3b4e>] ? out_of_memory+0x3b2/0x3e5
[  245.094675]  [<ffffffff810a7258>] ? __alloc_pages_nodemask+0x651/0x75e
[  245.094753]  [<ffffffff810d07c3>] ? alloc_pages_current+0xbf/0xdc
[  245.094824]  [<ffffffff810a236c>] ? filemap_fault+0x25c/0x381
[  245.094891]  [<ffffffff810b8116>] ? __do_fault+0xac/0x3bf
[  245.094955]  [<ffffffff810bb872>] ? handle_mm_fault+0x1e7/0x7e2
[  245.095025]  [<ffffffff81719cab>] ? __do_page_fault+0x41c/0x469
[  245.095095]  [<ffffffff810361a1>] ? timespec_add_safe+0x1e/0x4d
[  245.095165]  [<ffffffff810ec585>] ? dput+0x16/0xeb
[  245.095235]  [<ffffffff8106dff5>] ? ktime_get_ts+0x49/0xab
[  245.095302]  [<ffffffff810e99ae>] ? poll_select_copy_remaining+0xe0/0xf6
[  245.095380]  [<ffffffff81717272>] ? page_fault+0x22/0x30
[  245.095442] Mem-Info:
[  245.095471] Node 0 DMA per-cpu:
[  245.095515] CPU    0: hi:    0, btch:   1 usd:   0
[  245.095571] CPU    1: hi:    0, btch:   1 usd:   0
[  245.095625] Node 0 DMA32 per-cpu:
[  245.095669] CPU    0: hi:  186, btch:  31 usd:   0
[  245.095725] CPU    1: hi:  186, btch:  31 usd:   0


Reproduce steps:
-------------------------
1. ./gem_render_linear_blits
Comment 1 Chris Wilson 2014-02-21 11:47:10 UTC
It's hard to tell from the incomplete dmesg how much RAM is installed on that byt. I would guess you only have 2GiB, correct?

If so the question is then why didn't the test scale down?
Comment 2 lu hua 2014-02-24 05:17:27 UTC
(In reply to comment #1)
> It's hard to tell from the incomplete dmesg how much RAM is installed on
> that byt. I would guess you only have 2GiB, correct?
> 
> If so the question is then why didn't the test scale down?

Yes, the memory is 2 GB.
Comment 3 Chris Wilson 2014-02-24 08:58:35 UTC
commit 81ade89d6444ed4159c8cafb0512254f02ec34e0
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Mon Feb 24 08:52:28 2014 +0000

    gem_render_tiled_blits: Trim usage to fit within RAM
    
    This is a render correctness test, the intention is not to exercise the
    swapper (but preferrably some eviction code).
    
    Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=75247
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

commit 4550a18a571611197d3958ab3c1071ef69ec3838
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Mon Feb 24 08:52:28 2014 +0000

    gem_render_linear_blits: Trim usage to fit within RAM
    
    This is a render correctness test, the intention is not to exercise the
    swapper (but preferrably some eviction code).
    
    Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=75247
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Comment 4 lu hua 2014-03-28 07:23:34 UTC
Verified.Fixed.
Comment 5 Elizabeth 2017-10-06 14:39:40 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.