Bug 42580 - Decoding abort with KDE
Decoding abort with KDE
Status: RESOLVED NOTOURBUG
Product: libva
Classification: Unclassified
Component: intel
unspecified
x86 (IA32) Linux (All)
: medium major
Assigned To: haihao
:
Depends on:
Blocks:
  Show dependency treegraph
 
Reported: 2011-11-03 23:48 UTC by cc
Modified: 2011-11-15 16:53 UTC (History)
1 user (show)

See Also:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description cc 2011-11-03 23:48:31 UTC
I try to decoding video-file(H264/MPEG2/VC1) wtih KDE on. But the mplayer abort when I did full and not full screen repeat. I pasted the error info as follows.

Environment:
Arch:		x86_64
Platform:	SandyBridge and Ivb	
OSD:		Fedora release 15 (Lovelock)
Libdrm:		(master)2.4.27-1-g961bf9b5c2866ccb4fedf2b45b29fb688519d0db
Mesa:		(master)9f7b6a39f6ebc070ff5020578cea2d299b21d476

libva:
commit 824a28b40129a1f5467fa87f0442d1d6c118965a
Author: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Date:   Thu Nov 3 10:41:59 2011 +0100


Error info:
[VD_FFMPEG] XVMC-accelerated MPEG-2.
XIO:  fatal IO error 104 (Connection reset by peer) on X server ":0.0"
      after 282 requests (282 known processed) with 0 events remaining.
==========================
Backtrace:
0: X (xorg_backtrace+0x28) [0x565268]
1: X (0x400000+0x1691f9) [0x5691f9]
2: /lib64/libpthread.so.0 (0x3d75c00000+0xf3c0) [0x3d75c0f3c0]
3: /lib64/libc.so.6 (0x3d75800000+0x131d20) [0x3d75931d20]
4: /opt/X11R7/lib/xorg/modules/drivers/intel_drv.so (0x7f8651001000+0x11056) [0x7f8651012056]
5: /opt/X11R7/lib/xorg/modules/drivers/intel_drv.so (0x7f8651001000+0x11855) [0x7f8651012855]
6: /opt/X11R7/lib/xorg/modules/drivers/intel_drv.so (0x7f8651001000+0x29533) [0x7f865102a533]
7: X (miCopyRegion+0xa5) [0x543e35]
8: X (miDoCopy+0x44a) [0x5444da]
9: /opt/X11R7/lib/xorg/modules/drivers/intel_drv.so (0x7f8651001000+0x28c0e) [0x7f8651029c0e]
10: X (0x400000+0xf5a7a) [0x4f5a7a]
11: /opt/X11R7/lib/xorg/modules/drivers/intel_drv.so (0x7f8651001000+0x251f5) [0x7f86510261f5]
12: /opt/X11R7/lib/xorg/modules/drivers/intel_drv.so (0x7f8651001000+0x25c76) [0x7f8651026c76]
13: /opt/X11R7/lib/libdrm.so.2 (drmHandleEvent+0xdb) [0x7f865166583b]
14: X (WakeupHandler+0x5d) [0x435d5d]
15: X (WaitForSomething+0x1ef) [0x562a7f]
16: X (0x400000+0x31fb2) [0x431fb2]
17: X (0x400000+0x211ea) [0x4211ea]
18: /lib64/libc.so.6 (__libc_start_main+0xed) [0x3d7582143d]
19: X (0x400000+0x20d79) [0x420d79]
Segmentation fault at address 0x1b14ab0

Fatal server error:
Caught signal 11 (Segmentation fault). Server aborting

Reproduce steps:
1. xinit&
2. startkde
3. mplayer -vo vaapi -va vaapi MPEG-2.mpeg
4. Full Screen --> quit Full Screen  and repeat ,then the mplayer will abort.
Comment 1 haihao 2011-11-08 17:47:49 UTC
Hi, lan
   Could you reproduce this issue with KDE?
Comment 2 Hai 2011-11-08 21:08:03 UTC
This issue only happens with KDE
Comment 3 cc 2011-11-08 23:26:15 UTC
I  reproduced this issue today again. with KDE on and keep desktop effects on.
The desktop effects is open by default.
Comment 4 cc 2011-11-10 23:10:47 UTC
I reproduced the issue and the backtrace info as follows:

Program received signal SIGSEGV, Segmentation fault.
0xb75ae867 in drm_intel_bo_flink () from /opt/X11R7/lib/libdrm_intel.so.1
(gdb) bt
#0  0xb75ae867 in drm_intel_bo_flink () from /opt/X11R7/lib/libdrm_intel.so.1
#1  0xb75da346 in pixmap_flink (pixmap=<optimized out>) at intel_dri.c:88
#2  I830DRI2CreateBuffer (drawable=0x9d96080, attachment=0, format=0) at intel_dri.c:417
#3  0xb7602011 in allocate_or_reuse_buffer (pDraw=0x9d96080, ds=0x967ea10, pPriv=0x9dbd8c8, attachment=0, 
    format=0, dimensions_match=0, buffer=0x9d9b09c) at dri2.c:349
#4  0xb7603390 in do_get_buffers (pDraw=0x9d96080, width=0xbfb954f8, height=0xbfb95500, 
    attachments=<optimized out>, count=1, out_count=0xbfb954f0, has_format=0) at dri2.c:456
#5  0xb7604db8 in ProcDRI2GetBuffers (client=0x9d96248) at dri2ext.c:276
#6  ProcDRI2Dispatch (client=0x9d96248) at dri2ext.c:563
#7  0x080741f7 in Dispatch () at dispatch.c:432
#8  0x080621c5 in main (argc=2, argv=0xbfb95664, envp=0xbfb95670) at main.c:287
Comment 5 haihao 2011-11-14 18:17:42 UTC
I can reproduce this issue with -vo gl on some machines as well, so this isn't a VAAPI specific issue.  I will mark this issue as NOTOURBUG.  Please file a new bug about this issue.  If you still experience this issue with libva after fixing the new bug, feel free to reopen this bug.