Created attachment 119293 [details] /sys/class/drm/card0/error After some upgrades (only graphics-related part was xserver-xorg-video-intel:i386 from 2:2.99.917-2 to 2:2.99.917+git20151019-1~exp1) I sterted to recieve random gpu hangs. Sometimes display become garbage and system hangs entirely (no reactio to alt-sysrq), some times garbage appears but all continues to work. Suspend/resume was never used. Attached dump is from the "hangs but continues to work situation". It crashed at night, nothing were done with computer. Window manager is awesome with server-side composition by "xcompmgr -a". Also hardware-accelerated chromium was running. I'm using debian testing/expiremental with amd64 kernel and entirely 32-bit x86 userspace. Kernel version: Linux 4.2.0-1-amd64 #1 SMP Debian 4.2.1-2 (2015-09-27) x86_64 GNU/Linux lspci (part): 00:02.0 VGA compatible controller: Intel Corporation 82945G/GZ Integrated Graphics Controller (rev 02) Xorg.0.log (part) [ 38.982] (II) intel(0): Using Kernel Mode Setting driver: i915, version 1.6.0 20150522 [ 38.982] (II) intel(0): SNA compiled: xserver-xorg-video-intel 2:2.99.917+git20151019-1~exp1 (Timo Aaltonen <tjaalton@debian.org>) [ 38.982] (II) intel(0): SNA compiled for use with valgrind [ 39.052] (--) intel(0): Integrated Graphics Chipset: Intel(R) 945G [ 39.052] (--) intel(0): CPU: x86, sse2, sse3, ssse3; using a maximum of 2 threads [ 39.052] (II) intel(0): Creating default Display subsection in Screen section "Default Screen Section" for depth/fbbpp 24/32 [ 39.052] (==) intel(0): Depth 24, (--) framebuffer bpp 32 [ 39.052] (==) intel(0): RGB weight 888 [ 39.052] (==) intel(0): Default visual is TrueColor [ 39.052] (**) intel(0): Option "AccelMethod" "sna" [ 39.054] (II) intel(0): Output VGA1 has no monitor section [ 39.054] (II) intel(0): Enabled output VGA1 [ 39.054] (--) intel(0): Using a maximum size of 256x256 for hardware cursors [ 39.055] (II) intel(0): Output VIRTUAL1 has no monitor section [ 39.055] (II) intel(0): Enabled output VIRTUAL1 [ 39.055] (--) intel(0): Output VGA1 using initial mode 1920x1200 on pipe 0 [ 39.055] (==) intel(0): TearFree disabled [ 39.055] (==) intel(0): DPI set to (96, 96) [ 39.055] (II) Loading sub module "dri2" [ 39.056] (II) LoadModule: "dri2" [ 39.056] (II) Module "dri2" already built-in [ 39.056] (II) Loading sub module "present" [ 39.056] (II) LoadModule: "present" [ 39.056] (II) Module "present" already built-in [ 39.056] (==) Depth 24 pixmap format is 32 bpp [ 39.117] (II) intel(0): SNA initialized with Alviso (gen3) backend [ 39.117] (==) intel(0): Backing store enabled [ 39.117] (==) intel(0): Silken mouse enabled [ 39.137] (II) intel(0): HW Cursor enabled [ 39.137] (II) intel(0): RandR 1.2 enabled, ignore the following RandR disabled message. [ 39.144] (==) intel(0): DPMS enabled [ 39.144] (==) intel(0): Display hotplug detection enabled [ 39.167] (II) intel(0): [XvMC] i915_xvmc driver initialized. [ 39.168] (II) intel(0): [DRI2] Setup complete [ 39.168] (II) intel(0): [DRI2] DRI driver: i915 [ 39.168] (II) intel(0): [DRI2] VDPAU driver: va_gl [ 39.168] (II) intel(0): direct rendering: DRI2 enabled [ 39.168] (II) intel(0): hardware support for Present enabled [ 39.168] (--) RandR disabled [ 39.213] (II) SELinux: Disabled on system [ 39.779] (II) AIGLX: enabled GLX_MESA_copy_sub_buffer [ 39.779] (II) AIGLX: enabled GLX_ARB_create_context [ 39.779] (II) AIGLX: enabled GLX_ARB_create_context_profile [ 39.779] (II) AIGLX: enabled GLX_EXT_create_context_es2_profile [ 39.779] (II) AIGLX: enabled GLX_INTEL_swap_event [ 39.779] (II) AIGLX: enabled GLX_SGI_swap_control and GLX_MESA_swap_control [ 39.779] (II) AIGLX: enabled GLX_EXT_framebuffer_sRGB [ 39.779] (II) AIGLX: enabled GLX_ARB_fbconfig_float [ 39.779] (II) AIGLX: GLX_EXT_texture_from_pixmap backed by buffer objects [ 39.780] (II) AIGLX: Loaded and initialized i915 [ 39.780] (II) GLX: Initialized DRI2 GL provider for screen 0 [ 39.807] (II) intel(0): switch to mode 1920x1200@60.0 on VGA1 using pipe 0, position (0, 0), rotation normal, reflection none [ 39.808] (II) intel(0): Setting screen physical size to 507 x 317 dmesg(all messages except boot): [39710.792060] [drm] stuck on render ring [39710.796710] [drm] GPU HANG: ecode 3:0:0x00bbf88a, in Xorg [1117], reason: Ring hung, action: reset [39710.796719] [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace. [39710.796722] [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel [39710.796726] [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue. [39710.796729] [drm] The gpu crash dump is required to analyze gpu hangs, so please always attach it. [39710.796733] [drm] GPU crash dump saved to /sys/class/drm/card0/error [39710.796941] drm/i915: Resetting chip after gpu hang package versions of graphics-related apps: libdrm-intel1/testing,now 2.4.64-1 i386 libdrm2/testing,now 2.4.64-1 i386 xserver-xorg-video-intel/experimental,now 2:2.99.917+git20151019-1~exp xserver-xorg-core/now 2:1.17.2-2 i386 xcompmgr/testing,unstable,now 1.1.7-1 i386 libglapi-mesa/now 11.0.0~rc1-1 i386 chromium/now 46.0.2490.13-1 i386 linux-image-4.2.0-1-amd64/now 4.2.1-2 amd64
Created attachment 119294 [details] Xorg.0.log (has lots of GetModeLine - scrn messages, but they seems to be unrelated - presented with old version that never hangs)
Created attachment 119295 [details] full dmesg
*** This bug has been marked as a duplicate of bug 90841 ***
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.