System Environment: -------------------------- Platform: Baytrail Kernel(drm-intel-nightly)1be8f2b4dd6d3db00af24d4891c82d2650bd282d Bug detailed description: ------------------------- gem_evict_alignment subtest minor-interruptible 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. output: IGT-Version: 1.5-g06189c6 (x86_64) (Linux: 3.13.0_drm-intel-nightly_1be8f2_20140220+ x86_64) Killed dmesg: [ 163.745888] [drm:valleyview_set_rps], GPU freq request from 167 MHz (191) to 708 MHz (217) [ 166.044325] sendmail invoked oom-killer: gfp_mask=0x201da, order=0, oom_score_adj=0 [ 166.044425] sendmail cpuset=/ mems_allowed=0 [ 166.044479] CPU: 1 PID: 3171 Comm: sendmail Not tainted 3.13.0_drm-intel-nightly_1be8f2_20140220+ #218 [ 166.044583] 0000000000000000 00000000000201da ffffffff8171193b ffff8800024886e0 [ 166.044683] ffffffff8170df1a 0000000000000752 0000000000000010 000000000000074d [ 166.044782] ffff8800721644c0 0000000000000206 ffffffff812cf77a ffff880002489b80 [ 166.044881] Call Trace: [ 166.044921] [<ffffffff8171193b>] ? dump_stack+0x41/0x51 [ 166.044985] [<ffffffff8170df1a>] ? dump_header.isra.8+0x69/0x191 [ 166.045057] [<ffffffff812cf77a>] ? ___ratelimit+0xae/0xc8 [ 166.045123] [<ffffffff810a33b8>] ? oom_kill_process+0x76/0x2f8 [ 166.045192] [<ffffffff810a3b4e>] ? out_of_memory+0x3b2/0x3e5 [ 166.045260] [<ffffffff810a7258>] ? __alloc_pages_nodemask+0x651/0x75e [ 166.045358] [<ffffffff810d07c3>] ? alloc_pages_current+0xbf/0xdc [ 166.045431] [<ffffffff810a236c>] ? filemap_fault+0x25c/0x381 [ 166.045502] [<ffffffff810b8116>] ? __do_fault+0xac/0x3bf [ 166.045567] [<ffffffff810bb872>] ? handle_mm_fault+0x1e7/0x7e2 [ 166.045637] [<ffffffff81719cab>] ? __do_page_fault+0x41c/0x469 [ 166.045708] [<ffffffff810361a1>] ? timespec_add_safe+0x1e/0x4d [ 166.045778] [<ffffffff810ec585>] ? dput+0x16/0xeb [ 166.045836] [<ffffffff8106dff5>] ? ktime_get_ts+0x49/0xab [ 166.045900] [<ffffffff810e99ae>] ? poll_select_copy_remaining+0xe0/0xf6 [ 166.045978] [<ffffffff81717272>] ? page_fault+0x22/0x30 [ 166.046039] Mem-Info: [ 166.046068] Node 0 DMA per-cpu: [ 166.046111] CPU 0: hi: 0, btch: 1 usd: 0 [ 166.046166] CPU 1: hi: 0, btch: 1 usd: 0 [ 166.046221] Node 0 DMA32 per-cpu: [ 166.046265] CPU 0: hi: 186, btch: 31 usd: 0 [ 166.046329] CPU 1: hi: 186, btch: 31 usd: 0 [ 166.046390] active_anon:11076 inactive_anon:456684 isolated_anon:32 Reproduce steps: ------------------------- 1. ./gem_evict_alignment --run-subtest minor-interruptible
Created attachment 94409 [details] dmesg
Bug 75138 is triggered by the same test case and results in an OOM too, so holding off on this until that's resolved.
commit ceabbba524fb43989875f66a6c06d7ce0410fe5c Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Tue Mar 25 13:23:04 2014 +0000 drm/i915: Include bound and active pages in the count of shrinkable objects When the machine is under a lot of memory pressure and being stressed by multiple GPU threads, we quite often report fewer than shrinker->batch (i.e. SHRINK_BATCH) pages to be freed. This causes the shrink_control to skip calling into i915.ko to release pages, despite the GPU holding onto most of the physical pages in its active lists. References: https://bugs.freedesktop.org/show_bug.cgi?id=72742 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Robert Beckett <robert.beckett@intel.com> Reviewed-by: Rafael Barbalho <rafael.barbalho@intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Created attachment 99463 [details] dmesg(ceabbb) It still fails with OOM killer. output: IGT-Version: 1.6-g5e83f04 (x86_64) (Linux: 3.14.0_kcloud_ceabbb_20140521+ x86_64 Killed
Wrong kernel.
Created attachment 99845 [details] dmesg(c37f30) Test on latest -nightly kernel.It still exists. output: IGT-Version: 1.6-gff3c122 (x86_64) (Linux: 3.15.0-rc6_drm-intel-nightly_c37f30_20140526+ x86_64) Killed kernel config: https://bugs.freedesktop.org/show_bug.cgi?id=69247#c71
Test on latest -nightly kernel.It still exists.
skip instead of oom killer. IGT-Version: 1.6-g18d2130 (x86_64) (Linux: 3.15.0-rc8_drm-intel-nightly_969b3c_20140608+ x86_64) Estimated that we need 3222798336 bytes for the test, but only have 1649410048 bytes available (RAM) Test requirement not met in function minor_evictions, file gem_evict_alignment.c:137: Last errno: 0, Success Test requirement: (!(intel_check_memory(2*count, size, 0x1))) Subtest minor-normal: SKIP Estimated that we need 6442455040 bytes for the test, but only have 1649410048 bytes available (RAM) Test requirement not met in function major_evictions, file gem_evict_alignment.c:167: Last errno: 0, Success Test requirement: (!(intel_check_memory(count, size, 0x1))) Subtest major-normal: SKIP Estimated that we need 3222798336 bytes for the test, but only have 1649410048 bytes available (RAM) Test requirement not met in function minor_evictions, file gem_evict_alignment.c:137: Last errno: 0, Success Test requirement: (!(intel_check_memory(2*count, size, 0x1))) Subtest minor-interruptible: SKIP Estimated that we need 6442455040 bytes for the test, but only have 1648361472 bytes available (RAM) Test requirement not met in function major_evictions, file gem_evict_alignment.c:167: Last errno: 0, Success Test requirement: (!(intel_check_memory(count, size, 0x1))) Subtest major-interruptible: SKIP
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.