| Summary: | I830DRI2GetMSC: get vblank counter failed: Invalid argument" | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Product: | DRI | Reporter: | fangxun <xunx.fang> | ||||||||
| Component: | DRM/Intel | Assignee: | Chris Wilson <chris> | ||||||||
| Status: | CLOSED FIXED | QA Contact: | |||||||||
| Severity: | critical | ||||||||||
| Priority: | high | CC: | bo.b.wang, brian, bryce, jbarnes, theholyettlz | ||||||||
| Version: | unspecified | ||||||||||
| Hardware: | All | ||||||||||
| OS: | Linux (All) | ||||||||||
| Whiteboard: | |||||||||||
| i915 platform: | i915 features: | ||||||||||
| Attachments: |
|
||||||||||
Created attachment 41366 [details]
xorg log
"(WW) intel(0): get vblank counter failed: Invalid argument" Doesn't look to be a kernel regression, but that the driver is not correctly handling an error condition inside a vblank race on client termination that we have seen before. This issue still happens with following commits,e.g. 3D games:openarena,urbanterror. ------------------------------------------------------------------- Libdrm: (master)2.4.24-6-g3b04c73650b5e9bbcb602fdb8cea0b16ad82d0c0 Mesa: (master)6538b5824e298eaebede2d9686c7607c44ab446 Xserver: (master)xorg-server-1.10.0 Xf86_video_intel: (master)2.14.901 Kernel: (drm-intel-fixes) 91355834646328e7edc6bd25176ae44bcd7386c7 *** Bug 35891 has been marked as a duplicate of this bug. *** From bug 35891: "I830DRI2GetMSC:1145 get vblank counter failed" *** Bug 34322 has been marked as a duplicate of this bug. *** There is same issue in "et" game. "I830DRI2GetMSC:1145 get vblank counter failed" There is same issue in "quake3", "quake4", "ETQW-demo" game Created attachment 45467 [details] [review] Clarify EINVALs Sounds like you can almost trigger this at will now. Have you discovered a pattern? The attached patch adds some debug messages to the kernel which should help identify which error we are triggering. It works good when runnig openarena on Pineview with commit c94249d2a. But ut2004 still hangs when change resolution. I'm raising priority as it impacts many games. Does this issue still occur with the recent DDX fixes? This issue doesn't occur with recent kernel. Verified with drm-intel-next commit da3cc9202697a44057c1bd3ad685689375f1fe0c and drm-intel-fixes commit 2fb4e61d9471867677c97bf11dba8f1e9dfa7f7c. 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.
Created attachment 41365 [details] dmesg with drm.debug=0xf System Environment: -------------------------- Arch: i686 Platform: pineview Libdrm: (master)2.4.23 Mesa: (7.10)0e50c21e247b6d4246fcc2b583563a8f44bc4249 Xserver: (server-1.9-branch)xorg-server-1.9.2.902 Xf86_video_intel: (master)2.13.902-1-g71af40a75fbdd1054b1111e8cbe67ad1f97e6613 Kernel: (drm-intel-fixes)63abf3edaf42d0b9f278df90fe41c7ed4796b6b1 Bug detailed description: ------------------------- On gnome-desktop with compiz enabled, start openarena and change resolution in "system setup", the game hangs. It's kernel regression. The last known good commit is Kernel:(drm-intel-fixes) 6939a5aca7cfada279a24c307e772f33104fca20. Backtrace: #0 0xb780d424 in __kernel_vsyscall () #1 0x4aaffe46 in poll () from /lib/libc.so.6 #2 0xb764504b in _xcb_conn_wait (c=0xaf6add8, cond=0xaf6adfc, vector=0x0, count=0x0) at xcb_conn.c:306 #3 0xb76470c7 in xcb_wait_for_event (c=0xaf6add8) at xcb_in.c:437 #4 0xb76abbe7 in _XReadEvents (dpy=0xaf9cc88) at xcb_io.c:342 #5 0xb769969d in XMaskEvent (dpy=0xaf9cc88, mask=131072, event=0xbfa9328c) at MaskEvent.c:75 #6 0x4f94e67d in ?? () from /usr/lib/libSDL-1.2.so.0 #7 0x4f9559ad in ?? () from /usr/lib/libSDL-1.2.so.0 #8 0x4f959c39 in ?? () from /usr/lib/libSDL-1.2.so.0 #9 0x4f959f3d in ?? () from /usr/lib/libSDL-1.2.so.0 #10 0x4f943ec0 in SDL_SetVideoMode () from /usr/lib/libSDL-1.2.so.0 #11 0x081d9b76 in GLimp_StartDriverAndSetMode () #12 0x00000000 in ?? () Reproduce steps: ---------------- 1. xinit& 2. start openarena 3. change resolution in game "system setup"