Environment: -------------- libva: commit a78128ac9a52b7567296e076c3cd9e5b7ea640ad Author: Jonathan Bian <jonathan.bian@intel.com> Date: Sun Sep 16 21:26:00 2012 -0700 intel-driver: commit eb5f7f88fbd9085c3346a6b00698cef091e2ece2 Author: Gwenole Beauchesne <gwenole.beauchesne@intel.com> Date: Wed Aug 29 18:37:25 2012 +0200 gstreamer-vaapi : commit 49dd2800459af462c608ef3bfabd6fb6093711e0 Author: Wind Yuan <feng.yuan@intel.com> Date: Fri Sep 21 17:03:35 2012 +0800 Bug Info: -------------- There are 90 frames with the CAPA1_TOSHIBA_B.264, decoding process is as below: Setting pipeline to PAUSED ... libva info: VA-API version 0.34.0 libva info: va_getDriverName() returns 0 libva info: Trying to open /opt/X11R7/lib/dri/i965_drv_video.so libva info: Found init function __vaDriverInit_0_34 libva info: va_openDriver() returns 0 Pipeline is PREROLLING ... decode: 1, 99:99:99.999999999 size:0 sink: 1, 99:99:99.999999999 Pipeline is PREROLLED ... Setting pipeline to PLAYING ... sink: 2, 99:99:99.999999999 New clock: GstSystemClock Got EOS from element "playbin20". Execution ended after 290380 ns. Setting pipeline to PAUSED ... Setting pipeline to READY ... Setting pipeline to NULL ... Freeing pipeline ... Reproduce steps: ---------------- 1. xinit & 2. gst-launch-0.10 playbin2 video-sink="vaapisink sync=false" uri=file:///home/gst_AVC_conformance/Base_Ext_Main_profile/CAPA1_TOSHIBA_B.264 Issue can be reproduced with the below files: CAPA1_TOSHIBA_B.264 CAWP5_TOSHIBA_E/CAWP5_TOSHIBA_E.264 CAMANL1_TOSHIBA_B.264 CVMA1_TOSHIBA_B.264 CVWP2_TOSHIBA_E/CVWP2_TOSHIBA_E.264 CABA3_TOSHIBA_E.264 CVWP3_TOSHIBA_E/CVWP3_TOSHIBA_E.264 CVWP1_TOSHIBA_E/CVWP1_TOSHIBA_E.264 FM1_FT_E.264 CVWP5_TOSHIBA_E/CVWP5_TOSHIBA_E.264 CAWP1_TOSHIBA_E/CAWP1_TOSHIBA_E.264 CVMANL1_TOSHIBA_B.264 CVMAQP2_Sony_G.jsv CAMASL3_Sony_B.jsv CVFC1_Sony_C.jsv CVSE3_Sony_H.jsv CVCANLMA2_Sony_C.jsv CABA2_Sony_E.jsv CVSEFDFT3_Sony_E.jsv CANL2_Sony_E.jsv CAMACI3_Sony_C.jsv CVNLFI2_Sony_H.jsv CVNLFI1_Sony_C.jsv CABACI3_Sony_B.jsv CAPM3_Sony_D.jsv BA2_Sony_F.jsv CVFI2_Sony_H.jsv CVFI1_Sony_D.jsv CVMAPAQP3_Sony_E.jsv CVMAQP3_Sony_D.jsv CACQP3_Sony_D.jsv CVSE2_Sony_B.jsv CANLMA3_Sony_C.jsv cama3_vtc_b.avc cama2_vtc_b.avc cama1_vtc_c.avc You can download the files from http://tinderbox.sh.intel.com/media/FrameWork/AVC_conformance/Base_Ext_Main_profile/
Issue still can be reproduced libva: commit 86484495155e65fd8ac33ed3ede43fb42defd966 Author: Gwenole Beauchesne <gwenole.beauchesne@intel.com> Date: Tue Oct 9 06:21:59 2012 +0200 intel driver: commit f557dd6ad06c31bcf787468e804c948ecc4cf39b Author: Zhao Halley <halley.zhao@intel.com> Date: Fri Sep 21 10:30:55 2012 +0800 gstreamer-vaapi: commit 49dd2800459af462c608ef3bfabd6fb6093711e0 Author: Wind Yuan <feng.yuan@intel.com> Date: Fri Sep 21 17:03:35 2012 +0800
Hi, this is an interlaced stream.
Libva (staging) c59c512df262cdc44fb6e9101621f246eda176df Inter-driver (staging) 3bd6f3b3611dd9a39a3075f82cf938fa08e22884 Gst-vaapi (qa) e7d517323fd98608918de1211dc42f795b48765f Platfrom: IVB Issue can be reproduced.
CAPA1_TOSHIBA_B.264 can now be decoded correctly on IVB, from gst-vaapi git master branch. Please try it out on SNB and ILK platforms too. Thanks.
Retest in in gstream-vaapi master branch(cb0da4ed7acbb51f2d5dba34c0b37b2c5ad8bff4) and got below error: X Error of failed request: BadValue (integer parameter out of range for operation) Major opcode of failed request: 12 (X_ConfigureWindow) Value in failed request: 0x0 Serial number of failed request: 34 Current serial number in output stream: 35
platform: SNB IVB libva: (staging) 38c94cd922473095814ed9a9f99ad98fcc9c285d inter-driver: (staging) c0ef9d99df37ae45589fecb898727be495e50304 Test with the latest gstream-vaapi code in qa branch, Issue can be reproduced. but when using gst-launch-0.10 filesrc location=/home/AVC_conformance/Base_Ext_Main_profile/CAPA1_TOSHIBA_B.264 ! h264parse ! vaapidecode ! vaapisink sync=false file can be decoded correctly.
Cong, please add comments and update the progress on this issue
add compiling option of gstreamer-vaapi, make codec parser invisible
this bug is not important,so I suggest that QA could only use manual pipeline to test "sync=false"
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.