Summary: | [BXT/SKL]gem_mmap_gtt/copy* are killiing process, out of memory | ||||||
---|---|---|---|---|---|---|---|
Product: | DRI | Reporter: | Elio <elio.martinez.monroy> | ||||
Component: | DRM/Intel | Assignee: | Intel GFX Bugs mailing list <intel-gfx-bugs> | ||||
Status: | CLOSED FIXED | QA Contact: | Intel GFX Bugs mailing list <intel-gfx-bugs> | ||||
Severity: | normal | ||||||
Priority: | medium | CC: | intel-gfx-bugs | ||||
Version: | DRI git | ||||||
Hardware: | x86-64 (AMD64) | ||||||
OS: | Linux (All) | ||||||
Whiteboard: | ReadyForDev | ||||||
i915 platform: | BXT | i915 features: | GEM/PPGTT | ||||
Attachments: |
|
Description
Elio
2017-04-05 18:57:00 UTC
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.