Bug 44364 - [SNB] mplayer -vo gl wedges gpu in a couple of seconds
Summary: [SNB] mplayer -vo gl wedges gpu in a couple of seconds
Status: RESOLVED FIXED
Alias: None
Product: Mesa
Classification: Unclassified
Component: Drivers/DRI/i965 (show other bugs)
Version: git
Hardware: x86-64 (AMD64) Linux (All)
: medium critical
Assignee: Kenneth Graunke
QA Contact:
URL:
Whiteboard:
Keywords:
: 44091 (view as bug list)
Depends on:
Blocks:
 
Reported: 2012-01-01 04:39 UTC by Sami Farin
Modified: 2012-11-16 23:50 UTC (History)
3 users (show)

See Also:
i915 platform:
i915 features:


Attachments
intel_reg_dumper.intel-770a953ff0--mesa-2cd7e5b737.txt (11.31 KB, text/plain)
2012-01-01 04:40 UTC, Sami Farin
Details
i915_error_state.intel-770a953ff0--mesa-2cd7e5b737.txt (2.14 MB, text/plain)
2012-01-01 04:40 UTC, Sami Farin
Details
intel_reg_dumper.intel-307276d7-mesa-f82d40d4b1.txt (11.31 KB, text/plain)
2012-01-05 11:31 UTC, Sami Farin
Details
i915_error_state.intel-307276d7-mesa-f82d40d4b1.txt (2.03 MB, text/plain)
2012-01-05 11:32 UTC, Sami Farin
Details
i915_error_state--3.4.0-rc1-6c216ec6--mesa-129d5138d636.txt (2.14 MB, text/plain)
2012-04-05 09:43 UTC, Sami Farin
Details
regs--3.4.0-rc1-6c216ec6--mesa-129d5138d636.txt (12.03 KB, text/plain)
2012-04-05 09:43 UTC, Sami Farin
Details
intel_reg_purgeable-f7b00693175.txt (12.03 KB, text/plain)
2012-04-29 13:38 UTC, Sami Farin
Details

Description Sami Farin 2012-01-01 04:39:38 UTC
I have tried several xf86-video-intel and mesa versions, and none of them are stable.

This happens with intel-git-770a953ff0 and mesa-git-2cd7e5b737.

[  215.602434] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[  215.602501] [drm] capturing error event; look for more information in /debug/dri/0/i915_error_state
[  215.605075] [drm:i915_wait_request] *ERROR* i915_wait_request returns -11 (awaiting 7936 at 7935, next 7938)
[  222.029945] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[  222.029963] [drm:i915_wait_request] *ERROR* i915_wait_request returns -11 (awaiting 7970 at 7935, next 8088)
[  223.840683] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[  223.840700] [drm:i915_wait_request] *ERROR* i915_wait_request returns -11 (awaiting 8196 at 7935, next 8238)
[  223.841264] [drm:i915_reset] *ERROR* GPU hanging too fast, declaring wedged!
[  223.841266] [drm:i915_reset] *ERROR* Failed to reset chip.
Comment 1 Sami Farin 2012-01-01 04:40:23 UTC
Created attachment 55016 [details]
intel_reg_dumper.intel-770a953ff0--mesa-2cd7e5b737.txt
Comment 2 Sami Farin 2012-01-01 04:40:45 UTC
Created attachment 55017 [details]
i915_error_state.intel-770a953ff0--mesa-2cd7e5b737.txt
Comment 3 Sami Farin 2012-01-01 04:45:27 UTC
System environment: 
-- chipset: Intel(R) Sandybridge Desktop (GT2)
-- system architecture: 64-bit
-- xf86-video-intel: 770a953ff03bb8328c3f29e274d225528840f30c
-- xserver: e7df42ab68e30588a5e32ed543b0711821daf009
-- mesa: 2cd7e5b737e1384bb6d27f3cc2c8524b07230d56
-- libdrm: ef20301a11afae50bfe127002913dbd0b81ddccc
-- kernel: 3.0.14
-- Linux distribution: Fedora
-- Machine or mobo model: Asus P8Z68-V PRO GEN3
-- Display connector: DVI
-- CPU: Intel Core i5-2500K

other reports in bug #44091
Comment 4 Sami Farin 2012-01-05 11:31:01 UTC
mesa-f82d40d4b1 crashes the same way
Comment 5 Sami Farin 2012-01-05 11:31:41 UTC
Created attachment 55182 [details]
intel_reg_dumper.intel-307276d7-mesa-f82d40d4b1.txt
Comment 6 Sami Farin 2012-01-05 11:32:16 UTC
Created attachment 55183 [details]
i915_error_state.intel-307276d7-mesa-f82d40d4b1.txt
Comment 7 Chris Wilson 2012-02-24 09:15:16 UTC
*** Bug 44091 has been marked as a duplicate of this bug. ***
Comment 8 Sami Farin 2012-04-05 09:43:12 UTC
Created attachment 59527 [details]
i915_error_state--3.4.0-rc1-6c216ec6--mesa-129d5138d636.txt
Comment 9 Sami Farin 2012-04-05 09:43:31 UTC
Created attachment 59528 [details]
regs--3.4.0-rc1-6c216ec6--mesa-129d5138d636.txt
Comment 10 Sami Farin 2012-04-05 09:44:12 UTC
is this thing supported, or should I buy a real graphics card?
Comment 11 Sami Farin 2012-04-05 09:50:30 UTC
didn't notice a WARN_ON logged:
WARN_ON(dev_priv->fence_regs[obj->fence_reg].pin_count);

[ 2358.819474] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[ 2358.819536] [drm] capturing error event; look for more information in /debug/dri/0/i915_error_state
[ 2365.131049] [drm:i915_hangcheck_elapsed] *ERROR* Hangcheck timer elapsed... GPU hung
[ 2365.499807] usb 2-1.4: unlink qh1-0e01/ffff8803bf5bd280 start 0 [1/3 us]
[ 2372.712011] ehci_hcd 0000:00:1d.0: reused qh ffff8803bf5bd280 schedule
[ 2372.712015] usb 2-1.4: link qh1-0e01/ffff8803bf5bd280 start 0 [1/3 us]
[ 2577.368174] ------------[ cut here ]------------
[ 2577.368195] WARNING: at drivers/gpu/drm/i915/i915_gem.c:2410 i915_gem_object_put_fence+0x99/0xb0 [i915]()
[ 2577.368197] Hardware name: System Product Name
[ 2577.368198] Modules linked in: xt_NOTRACK xt_CLASSIFY ipt_ECN xt_connmark xt_length xt_connlimit xt_set xt_multiport ip_set_bitmap_port ip_set_hash_net sch_sfb nf_conntrack_netlink snd_usb_audio snd_usbmidi_lib snd_hwdep snd_rawmidi snd_seq_device arptable_filter arp_tables ip6t_rt nf_conntrack_ipv6 nf_defrag_ipv6 ipt_REJECT xt_limit xt_LOG xt_hashlimit xt_owner nf_conntrack_ipv4 nf_defrag_ipv4 xt_state xt_tcpudp ip6table_filter ip6table_mangle ip6_tables iptable_filter iptable_mangle iptable_raw ip_tables bridge stp llc snd_hda_intel ip_set nfnetlink sch_netem sch_hfsc sch_htb sch_sfq cls_fw cls_u32 cls_route sch_ingress sch_red sch_tbf sch_teql sch_prio sch_gred cls_rsvp cls_rsvp6 cls_tcindex sch_cbq sch_dsmark at24 at25 uvcvideo videobuf2_vmalloc videobuf2_memops videobuf2_core videodev pps_ldisc pps_core pl2303 af_key xfrm4_tunnel esp4 ah4 ipcomp xfrm_ipcomp ipip tun udf vfat fat oprofile nf_conntrack x_tables dccp_diag dccp cmtp kernelcapi loop ppdev parport_pc lp parport i2c_dev ppp_async ppp_generic slhc ftdi_sio usbserial bnep w83627ehf hwmon_vid snd_hda_codec_hdmi snd_hda_codec_realtek snd_hda_codec snd_pcm coretemp snd_page_alloc snd_timer snd hwmon btusb soundcore bluetooth mxm_wmi rfkill wmi i2c_i801 iTCO_wdt iTCO_vendor_support e1000e rtc_cmos kvm binfmt_misc tcp_cubic autofs4 firewire_ohci firewire_core i915 drm_kms_helper video button [last unloaded: pcspkr]
[ 2577.368272] Pid: 24024, comm: X Not tainted 3.4.0-rc1-6c216ec6+ #5
[ 2577.368273] Call Trace:
[ 2577.368279]  [<ffffffff810432f0>] warn_slowpath_common+0x6f/0x9e
[ 2577.368282]  [<ffffffff81043334>] warn_slowpath_null+0x15/0x17
[ 2577.368291]  [<ffffffffa002c068>] i915_gem_object_put_fence+0x99/0xb0 [i915]
[ 2577.368299]  [<ffffffffa002d58f>] i915_gem_object_unbind+0x79/0x1ef [i915]
[ 2577.368307]  [<ffffffffa002d725>] i915_gem_free_object_tail+0x20/0x108 [i915]
[ 2577.368315]  [<ffffffffa002b856>] ? i915_gem_retire_requests_ring+0x11e/0x250 [i915]
[ 2577.368324]  [<ffffffffa003012c>] i915_gem_free_object+0x4b/0x50 [i915]
[ 2577.368328]  [<ffffffff813d2f63>] drm_gem_object_free+0x25/0x29
[ 2577.368336]  [<ffffffffa002a1e1>] i915_gem_object_move_to_inactive+0x10e/0x114 [i915]
[ 2577.368344]  [<ffffffffa002b936>] i915_gem_retire_requests_ring+0x1fe/0x250 [i915]
[ 2577.368351]  [<ffffffffa002bb8d>] i915_wait_request+0x205/0x462 [i915]
[ 2577.368355]  [<ffffffff81061bb9>] ? __init_waitqueue_head+0x51/0x51
[ 2577.368363]  [<ffffffffa002be1a>] i915_gem_object_wait_rendering+0x30/0x34 [i915]
[ 2577.368371]  [<ffffffffa0031a8f>] i915_gem_do_execbuffer.isra.7+0x96f/0x17bb [i915]
[ 2577.368375]  [<ffffffff810889cb>] ? lock_release_holdtime.part.7+0xf/0x150
[ 2577.368383]  [<ffffffffa0032df4>] i915_gem_execbuffer2+0x92/0x27e [i915]

[ 2577.368391]  [<ffffffffa0032d62>] ? i915_gem_execbuffer+0x487/0x487 [i915]
[ 2577.368394]  [<ffffffff813d1520>] drm_ioctl+0x3b4/0x43e
[ 2577.368399]  [<ffffffff81126127>] do_vfs_ioctl+0x8a/0x4f9
[ 2577.368402]  [<ffffffff811265d0>] sys_ioctl+0x3a/0x7a
[ 2577.368406]  [<ffffffff816b4662>] system_call_fastpath+0x16/0x1b
[ 2577.368408] ---[ end trace fcae78d230edb007 ]---
Comment 12 Sami Farin 2012-04-29 13:37:34 UTC
now I tried mplayer with kernel-3a69ddd6f872180b6f61fda87152b37202118fbc and 
mesa-b328949a37.

[  456.495796] [drm:i915_gem_object_bind_to_gtt] *ERROR* Attempting to bind a purgeable object

(this is a new error not seen earlier)
It is not wedged, fortunately.  mplayer exited when it got that error.

# intel_reg_read 0x2120
0x2120 : 0x6800
Comment 13 Sami Farin 2012-04-29 13:38:13 UTC
Created attachment 60779 [details]
intel_reg_purgeable-f7b00693175.txt
Comment 14 Kenneth Graunke 2012-08-10 17:46:08 UTC
I've tried mplayer -vo gl on any number of videos, across several different versions of software, and I've never had any GPU hangs.  It all seems to be working fine.  This is also on Sandybridge GT2.

I almost wonder if the reporter has a hardware problem, considering that he also gets GPU hangs playing simple games like Neverball.
Comment 15 JS 2012-11-16 22:33:00 UTC
Does
Option "SwapbuffersWait" "false"
in Section "Device" of xorg.conf make any difference?
Comment 16 Sami Farin 2012-11-16 22:43:46 UTC
I haven't had hangs since July.
My hardware is still the same, so it was a software problem.
Comment 17 Kenneth Graunke 2012-11-16 23:50:44 UTC
Must have been fixed at some point then.  Closing.


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.