Created attachment 130706 [details] dmesg log The following test cases are following showing "out of memory" gme_mmap_gtt@swap-copy gem_mmap_gtt@swap-copy-odd gem_mmap_gtt@swap-copy-xy Steps to reproduce: 1. install IGT 2. execute $ sudo -E ./gem_mmap_gtt --run-subtest copy copy-xy copy-odd Expected result: Test should pass Actual results: Fail Output: IGT-Version: 1.18-g56741ce (x86_64) (Linux: 4.11.0-rc5-drm-tip-qa-ww14-commit-5bc82ec+ x86_64) Writing a 100% child 0 died with signal 9, Killed Subtest swap-copy failed. **** DEBUG **** (gem_mmap_gtt:1552) intel-os-DEBUG: Checking 2 surfaces of size 969,412,608 bytes (total 1,938,829,312) against RAM + swap (gem_mmap_gtt:1552) drmtest-DEBUG: Test requirement passed: !(fd<0) (gem_mmap_gtt:1552) igt-debugfs-DEBUG: Opening debugfs directory '/sys/kernel/debug/dri/0' (gem_mmap_gtt:1552) intel-os-DEBUG: Test requirement passed: __intel_check_memory(count, size, mode, &required, &total) (gem_mmap_gtt:1552) igt-core-DEBUG: Test requirement passed: !igt_run_in_simulation() **** END **** Subtest swap-copy: FAIL (1.894s) Configuration: Configuration: Kernel version: 4.11.0-rc5 commit-5bc82ec Component : drm url : http://cgit.freedesktop.org/mesa/drm tag : libdrm-2.4.76-16-g6312017 commit : 6312017 author : Emil Velikov <emil.l.velikov@gmail.com> age : Mon Apr 3 18:01:49 2017 +0100 7 hours ago comment : configure.ac: bring back pthread-stubs check Component : mesa url : http://cgit.freedesktop.org/mesa/mesa tag : 17.0-branchpoint-2397-g405ef7b commit : 405ef7b author : Jason Ekstrand <jason@jlekstrand.net> age : Mon Apr 3 16:58:35 2017 -0700 30 minutes ago comment : intel/vec4: Add some fall through comments Component : xf86-video-intel url : http://cgit.freedesktop.org/xorg/driver/xf86-video-intel tag : 2.99.917-770-gcb6ba2d commit : cb6ba2d author : Chris Wilson <chris@chris-wilson.co.uk> age : Sat Mar 25 01:21:46 2017 +0000 10 days ago Component : libva url : http://cgit.freedesktop.org/libva tag : libva-1.7.3.pre1-85-gefc164d commit : efc164d author : Xiang Haihao <haihao.xiang@intel.com> age : Tue Mar 7 23:42:43 2017 +0800 4 weeks ago comment : Bump libva to 1.8.1.pre1 for development Component : intel-driver url : http://cgit.freedesktop.org/vaapi/intel-driver tag : 1.7.3-359-g437cbe0 commit : 437cbe0 author : Víctor Manuel Jáquez Leal <vjaquez@igalia.com> age : Wed Mar 29 08:29:46 2017 +0800 6 days ago comment : gen8: accept P010 as valid format Component : cairo url : http://cgit.freedesktop.org/cairo tag : 1.15.4-11-gcffa452 commit : cffa452 author : Debarshi Ray <debarshir@freedesktop.org> age : Wed Mar 15 20:26:22 2017 -0700 3 weeks ago comment : doc: Clarify when the device scale is inherited and when Component : xserver url : http://cgit.freedesktop.org/xorg/xserver tag : xorg-server-1.19.0-184-ge4d0757 commit : e4d0757 author : Adam Jackson <ajax@redhat.com> age : Thu Mar 30 11:32:02 2017 -0400 4 days ago comment : xfree86: Remove driver entity hooks and private Component : macros url : https://cgit.freedesktop.org/xorg/util/macros tag : util-macros-1.19.1-2-g39f07f7 commit : 39f07f7 author : Emil Velikov <emil.veliko@collabora.com> age : Mon Feb 20 10:16:40 2017 +1000 6 weeks ago comment : Rework INSTALL_CMD to touch/echo >&2 only as needed Component : intel-gpu-tools url : https://cgit.freedesktop.org/xorg/app/intel-gpu-tools tag : intel-gpu-tools-1.18-56-g56741ce commit : 56741ce author : Chris Wilson <chris@chris-wilson.co.uk> age : Mon Apr 3 19:19:42 2017 +0100 6 hours ago comment : tests/gem_media_fill: Fixup typo Component : piglit url : https://cgit.freedesktop.org/piglit tag : piglit-v1 commit : 3d1cbd9 author : Vinson Lee <vlee@freedesktop.org> age : Mon Apr 3 15:46:33 2017 -0700 2 hours ago comment : glslparsertest: Add test case for FDO bug #100438.
Also seen on SKL with the below configuration ============== SOFTWARE ========================= Component : drm tag : libdrm-2.4.76-16-g6312017 commit : 6312017 Component : cairo tag : 1.15.4-11-gcffa452 commit : cffa452 Component : intel-gpu-tools tag : intel-gpu-tools-1.18-56-g56741ce commit : 56741ce ============== KERNEL INFO ========================= commit bf30bc2a70b83a77ba63436023f3550083715c56 Author: Daniel Stone <daniels@collabora.com> Date: Tue Apr 4 21:01:18 2017 +0100 drm-tip: 2017y-04m-04d-20h-00m-56s UTC integration manifest Kernel version : 4.11.0-rc5-bf30bc2 Architecture : source amd64 all
commit b268d9fe0f10544f5f7a1b7015e2b97075e6215d Author: Chris Wilson <chris@chris-wilson.co.uk> Date: Wed Apr 5 23:15:14 2017 +0100 drm/i915: Use the right mapping_gfp_mask for final shmem allocation Many sightings report the greater prevalence of allocation failures. This is all due to the incorrect use of mapping_gfp_constraint(), so remove it in favour of just querying the mapping_gfp_mask() which are the exact gfp_t we wanted in the first place. We still do expect a higher chance of reporting ENOMEM, as that is the intention of using __GFP_NORETRY -- to fail rather than oom after having reclaimed from our bo caches, and having done a direct|kswapd reclaim pass.
(In reply to Chris Wilson from comment #2) > commit b268d9fe0f10544f5f7a1b7015e2b97075e6215d > Author: Chris Wilson <chris@chris-wilson.co.uk> > Date: Wed Apr 5 23:15:14 2017 +0100 > > drm/i915: Use the right mapping_gfp_mask for final shmem allocation > > Many sightings report the greater prevalence of allocation failures. > This is all due to the incorrect use of mapping_gfp_constraint(), so > remove it in favour of just querying the mapping_gfp_mask() which are > the exact gfp_t we wanted in the first place. > > We still do expect a higher chance of reporting ENOMEM, as that is the > intention of using __GFP_NORETRY -- to fail rather than oom after having > reclaimed from our bo caches, and having done a direct|kswapd reclaim > pass. Hi Chris Wilson, in the latest commits from drm-tip this patch is not integrated kernel (4.12)
(In reply to Humberto Israel Perez Rodriguez from comment #3) > (In reply to Chris Wilson from comment #2) > > commit b268d9fe0f10544f5f7a1b7015e2b97075e6215d > > Author: Chris Wilson <chris@chris-wilson.co.uk> > > Date: Wed Apr 5 23:15:14 2017 +0100 > > > > drm/i915: Use the right mapping_gfp_mask for final shmem allocation > > > > Many sightings report the greater prevalence of allocation failures. > > This is all due to the incorrect use of mapping_gfp_constraint(), so > > remove it in favour of just querying the mapping_gfp_mask() which are > > the exact gfp_t we wanted in the first place. > > > > We still do expect a higher chance of reporting ENOMEM, as that is the > > intention of using __GFP_NORETRY -- to fail rather than oom after having > > reclaimed from our bo caches, and having done a direct|kswapd reclaim > > pass. > > Hi Chris Wilson, in the latest commits from drm-tip this patch is not > integrated kernel (4.12) Pardon?
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.