| Summary: | [IVB] GPU HANG: ecode 7:0:0x85fffefa, in glmark2 [8458], reason: Ring hung, action: reset | ||
|---|---|---|---|
| Product: | Mesa | Reporter: | it.is.for.spam |
| Component: | Drivers/DRI/i965 | Assignee: | Intel 3D Bugs Mailing List <intel-3d-bugs> |
| Status: | RESOLVED FIXED | QA Contact: | Intel 3D Bugs Mailing List <intel-3d-bugs> |
| Severity: | normal | ||
| Priority: | medium | CC: | intel-gfx-bugs |
| Version: | unspecified | ||
| Hardware: | x86-64 (AMD64) | ||
| OS: | Linux (All) | ||
| Whiteboard: | |||
| i915 platform: | IVB | i915 features: | GPU hang |
| Attachments: |
/sys/class/drm/card0/error
full dmesg dmesg new /sys/class/drm/card0/error new |
||
Created attachment 129228 [details]
full dmesg
Hardware: Intel Celeron G1610 (integrated video) on ASUS P8H61-I R2.0 motherboard with DVI monitor According to mesa engineers, mesa only emits 3DSTATE_VERTEX_ELEMENTS on-demand right before 3DPRIMITIVE. So I recommend to update to latest version of xf86-video-intel since fix should be already in and mark as REOPENED if you can reproduce (and attach fresh gpu error dump) and RESOLVED/* if you cannot reproduce. In addition, there are constant improvements pushed in kernel and Mesa that will benefit to your system, so please re-test with latest kernel & Mesa (12 or 13). In either case (still or not occurring), please confirm you config. *Details: - Kernel: 4.4.0-59-generic - Platform: IvyBridge (PCI ID: 0x0152, PCI Revision: 0x09, PCI Subsystem: 1043:844d) - Mesa: [Please confirm your version] - xf86-video-intel: [Please confirm your version] From this error dump, hung is happening in render ring batch with active head at 0x7cc98930, with 0x7a000003 (PIPE_CONTROL) as IPEHR. Batch extract (around 0x7cc98930): 0x7cc988ec: 0x78090003: 3DSTATE_VERTEX_ELEMENTS 0x7cc988f0: 0x02400000: buffer 0: valid, type 0x0040, src offset 0x0000 bytes 0x7cc988f4: 0x11130000: (X, Y, Z, 1.0), dst offset 0x00 bytes 0x7cc988f8: 0x06400000: buffer 1: valid, type 0x0040, src offset 0x0000 bytes 0x7cc988fc: 0x11130000: (X, Y, Z, 1.0), dst offset 0x00 bytes 0x7cc98900: 0x7b000005: 3DPRIMITIVE: 0x7cc98904: 0x00000004: tri list sequential 0x7cc98908: 0x0000540c: vertex count 0x7cc9890c: 0x00000000: start vertex 0x7cc98910: 0x00000001: instance count 0x7cc98914: 0x00000000: start instance 0x7cc98918: 0x00000000: index bias 0x7cc9891c: 0x7a000003: PIPE_CONTROL 0x7cc98920: 0x00101001: no write, cs stall, render target cache flush, depth cache flush, 0x7cc98924: 0x00000000: destination address 0x7cc98928: 0x00000000: immediate dword low 0x7cc9892c: 0x00000000: immediate dword high 0x7cc98930: 0x7a000003: PIPE_CONTROL 0x7cc98934: 0x00000408: no write, texture cache invalidate, constant cache invalidate, 0x7cc98938: 0x00000000: destination address 0x7cc9893c: 0x00000000: immediate dword low 0x7cc98940: 0x00000000: immediate dword high This bug report corresponds to following versions: Mesa: 17.0.0~git20170126 commit d283ec0a xserver-xorg-video-intel: 2:2.99.917+git20161127 commit bde94605 (In reply to it.is.for.spam from comment #4) > This bug report corresponds to following versions: > > Mesa: 17.0.0~git20170126 commit d283ec0a > xserver-xorg-video-intel: 2:2.99.917+git20161127 commit bde94605 Can you try with latest version of xf86-video-intel? Chris pushed a fix on 4acd4a7d3d2f41227022fa7581cfb85a0b124eae (https://cgit.freedesktop.org/xorg/driver/xf86-video-intel/commit/?id=4acd4a7d3d2f41227022fa7581cfb85a0b124eae) on December 2016 (more recent than yours). Moreover, if this is still occurring, to confirm what is causing it, can you also try with modesetting driver (https://cgit.freedesktop.org/xorg/driver/xf86-video-modesetting) rather than Intel DDX and let us know the status? Chris, since this is gen7 here, do you think that SNA fix (4acd4a7d3d2f41227022fa7581cfb85a0b124eae) may be also be done for other gen? Chris, when you change to Mesa, would you mind to add a comment, even brief, giving the reason? thanks I built intel_drv.so from the master branch of git repo https://cgit.freedesktop.org/xorg/driver/xf86-video-intel/ commit 2d6f2e8 But the error still persists (i attached new full dmesg and /sys/class/drm/card0/error) Then I built modesetting_drv.so from master of https://cgit.freedesktop.org/xorg/driver/xf86-video-modesetting With it the error is gone and system does not freeze anymore, but all opengl apps are very slow Created attachment 129256 [details]
dmesg new
Created attachment 129257 [details]
/sys/class/drm/card0/error new
Sorry to bother you, but is there a chance this problem will be fixed or should i consider buing more up-to-date CPU ? :_) The bug disappeared after I updated my bios firmware |
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.
Created attachment 129227 [details] /sys/class/drm/card0/error When I launch any app that uses opengl systems starts freezing periodically. Finally after some time application crashes. I tried to install different distributive versions, but the error persist: I tried Debian 8.7.1, Ubuntu 14.04, Ubuntu 14.04.5, Ubuntu 16.04.1, Ubuntu 16.04.1 with latest drivers from xorg-edgers. I suspect this started after upgrade to mesa 10.0, because Ubuntu 13.10 used to work fine. Ubuntu 16.04.1 4.4.0-59-generic x86_64 [drm] stuck on render ring [drm] GPU HANG: ecode 7:0:0x85fffefa, in glmark2 [8458], reason: Ring hung, action: reset [drm] GPU crash dump saved to /sys/class/drm/card0/error drm/i915: Resetting chip after gpu hang [drm] stuck on render ring [drm] GPU HANG: ecode 7:0:0x85ff7cfa, in glmark2 [8458], reason: Ring hung, action: reset