Question about pixman,x11perf and xserver
hi,all I'am using ubuntu running on our chip of cortex-a9 with neon.And there is no 2D/3D hardware in our chip.There are several questions during my trying to improve the user experience. 1.I recompile the pixman with enable-neon,but when I use the x11perf for testing ,I can‘t see the significant difference between the pixman with neon and pixman without neon.That's why? 2.xserver use which drawing library to resizing the window,scroll the text window ,and so on? 3.which library excepts pixman could I optimize using neon to improve the performance of xserver? Best Regards, David ___ xorg@lists.freedesktop.org: X.Org support Archives: http://lists.freedesktop.org/archives/xorg Info: http://lists.freedesktop.org/mailman/listinfo/xorg Your subscription address: arch...@mail-archive.com
Re: 8k resolution video causes X Error
On 05/12/2011 2:03 AM, Maarten Maathuis wrote: On Sun, Dec 4, 2011 at 6:11 PM, Donald McLachlan donald.mclach...@crc.ca wrote: Hi, I don't know where to start to resolve this problem and guessed maybe this is a good place to start. If not, please point me in the right direction. Our ultimate goal is to stream 8k resolution video using sage (see www.sagecommons.org). - We first used ffmpeg to convert a 4k resolution video file to yuv format, and we were able to view it with ffplay, mplayer, and crcview (an in house program). - We then used ffmpeg to convert/resample the same 4k resolution video file to yuv/8k resolution; the conversion completed without error. - When trying to view the resulting yuv/8k resolution file all three viewer programs failed with the same X Error. For example, here is the output from ffplay: ffplay -i Lupe.8k.yuv -s 8192x4320 -pix_fmt yuv420p -x 1920 -y 1080 ffplay version 0.8, Copyright (c) 2003-2011 the FFmpeg developers built on Nov 30 2011 13:01:22 with gcc 4.5.1 20101208 [gcc-4_5-branch revision 167585] configuration: libavutil51. 9. 1 / 51. 9. 1 libavcodec 53. 7. 0 / 53. 7. 0 libavformat 53. 4. 0 / 53. 4. 0 libavdevice 53. 1. 1 / 53. 1. 1 libavfilter 2. 23. 0 / 2. 23. 0 libswscale2. 0. 0 / 2. 0. 0 [rawvideo @ 0x129d740] Estimating duration from bitrate, this may be inaccurate Input #0, rawvideo, from 'Lupe.8k.yuv': Duration: N/A, start: 0.00, bitrate: N/A Stream #0.0: Video: rawvideo, yuv420p, 8192x4320, 25 tbr, 25 tbn, 25 tbc X Error of failed request: BadLength (poly request too large or internal Xlib length error) Major opcode of failed request: 132 (XVideo) Minor opcode of failed request: 18 () Serial number of failed request: 23 Current serial number in output stream: 24 In case it matters, we are using openSuse 11.4 64 bit linux, on an ASUS P6T7 WS Supercomputer motherboard, with 12 G RAM, and a ASUS GTX590 video card. My guess is the 8k resolution video format is exceeding a buffer size limit somewhere, either in software, or maybe on the video card. Is there a way to find out what buffers are affected and is there a way to overcome these limits? Thanks for any assistance you can provide, Don ___ xorg@lists.freedesktop.org: X.Org support Archives: http://lists.freedesktop.org/archives/xorg Info: http://lists.freedesktop.org/mailman/listinfo/xorg Your subscription address: If this is using the nouveau driver (check lsmod or xorg log), i see that for some reason that it's limited to 4096x4096 for xvideo. See this line: http://cgit.freedesktop.org/nouveau/xf86-video-nouveau/tree/src/nouveau_xv.c#n2031 And then check the contents of DummyEncodingTex and you'll find it refers to the maximum sizes. The command xvinfo confirms this. NV50 and higher (everything starting geforce 8) are able to do 8192x8192, it should just be a matter of making a NV50 specific DummyEncodingTex structure. Hi Maaten, I believe we replaced the nouveau driver with the nvidia driver, but I will double check. I will run the xvinfo command to see what it says limits are. - If need be I guess we could revert to the nouveau driver and modify the DummyEncodingTex structure. - Does anyone know if there something similar we can do with the nvidia driver to enable 8k? (I guess maybe Nvidia are the ones to ask. :-) ) Thanks, and I'll let you know how it goes, Don ___ xorg@lists.freedesktop.org: X.Org support Archives: http://lists.freedesktop.org/archives/xorg Info: http://lists.freedesktop.org/mailman/listinfo/xorg Your subscription address: arch...@mail-archive.com
Re: 8k resolution video causes X Error
On 05/12/2011 9:38 AM, Donald McLachlan wrote: On 05/12/2011 2:03 AM, Maarten Maathuis wrote: On Sun, Dec 4, 2011 at 6:11 PM, Donald McLachlan donald.mclach...@crc.ca wrote: Hi, I don't know where to start to resolve this problem and guessed maybe this is a good place to start. If not, please point me in the right direction. Our ultimate goal is to stream 8k resolution video using sage (see www.sagecommons.org). - We first used ffmpeg to convert a 4k resolution video file to yuv format, and we were able to view it with ffplay, mplayer, and crcview (an in house program). - We then used ffmpeg to convert/resample the same 4k resolution video file to yuv/8k resolution; the conversion completed without error. - When trying to view the resulting yuv/8k resolution file all three viewer programs failed with the same X Error. For example, here is the output from ffplay: ffplay -i Lupe.8k.yuv -s 8192x4320 -pix_fmt yuv420p -x 1920 -y 1080 ffplay version 0.8, Copyright (c) 2003-2011 the FFmpeg developers built on Nov 30 2011 13:01:22 with gcc 4.5.1 20101208 [gcc-4_5-branch revision 167585] configuration: libavutil51. 9. 1 / 51. 9. 1 libavcodec 53. 7. 0 / 53. 7. 0 libavformat 53. 4. 0 / 53. 4. 0 libavdevice 53. 1. 1 / 53. 1. 1 libavfilter 2. 23. 0 / 2. 23. 0 libswscale2. 0. 0 / 2. 0. 0 [rawvideo @ 0x129d740] Estimating duration from bitrate, this may be inaccurate Input #0, rawvideo, from 'Lupe.8k.yuv': Duration: N/A, start: 0.00, bitrate: N/A Stream #0.0: Video: rawvideo, yuv420p, 8192x4320, 25 tbr, 25 tbn, 25 tbc X Error of failed request: BadLength (poly request too large or internal Xlib length error) Major opcode of failed request: 132 (XVideo) Minor opcode of failed request: 18 () Serial number of failed request: 23 Current serial number in output stream: 24 In case it matters, we are using openSuse 11.4 64 bit linux, on an ASUS P6T7 WS Supercomputer motherboard, with 12 G RAM, and a ASUS GTX590 video card. My guess is the 8k resolution video format is exceeding a buffer size limit somewhere, either in software, or maybe on the video card. Is there a way to find out what buffers are affected and is there a way to overcome these limits? Thanks for any assistance you can provide, Don ___ xorg@lists.freedesktop.org: X.Org support Archives: http://lists.freedesktop.org/archives/xorg Info: http://lists.freedesktop.org/mailman/listinfo/xorg Your subscription address: If this is using the nouveau driver (check lsmod or xorg log), i see that for some reason that it's limited to 4096x4096 for xvideo. See this line: http://cgit.freedesktop.org/nouveau/xf86-video-nouveau/tree/src/nouveau_xv.c#n2031 And then check the contents of DummyEncodingTex and you'll find it refers to the maximum sizes. The command xvinfo confirms this. NV50 and higher (everything starting geforce 8) are able to do 8192x8192, it should just be a matter of making a NV50 specific DummyEncodingTex structure. Hi Maaten, I believe we replaced the nouveau driver with the nvidia driver, but I will double check. I will run the xvinfo command to see what it says limits are. - If need be I guess we could revert to the nouveau driver and modify the DummyEncodingTex structure. - Does anyone know if there something similar we can do with the nvidia driver to enable 8k? (I guess maybe Nvidia are the ones to ask. :-) ) Thanks, and I'll let you know how it goes, Don Hi Maarten, (sorry for the typo on your name last time.) lsmod shows: nvidia 11909611 44 and Xorg.0.log shows: /usr/lib64/xorg/modules/drivers/nvidia_drv.so [ 180.902] (II) Module nvidia: vendor=NVIDIA Corporation [ 180.902] compiled for 4.0.2, module version = 1.0.0 [ 180.902] Module class: X.Org Video Driver [ 180.902] (II) NVIDIA dlloader X Driver 285.05.09 Fri Sep 23 17:33:35 PDT 2011 [ 180.902] (II) NVIDIA Unified Driver for all Supported NVIDIA GPUs So I take it we are running the nvidia driver, not the nouveau driver. Also, xvinfo shows maximum XvImage size: 16384 x 16384 Does that cover video, or just static images? In case it helps, I've attached a file with the text output of the 3 commands. Thanks, Don Script started on Mon 05 Dec 2011 10:00:57 AM EST ]2;crc@crc-fsmanager:~]1;crc-fsmanagercrc@crc-fsmanager:~ lsmod ; xvinfo ; cat /var/log/Xorg.0.log Module Size Used by nls_iso8859_1 4633 0 nls_cp437 6319 0 vfat 10387 0 fat52270 1 vfat iptable_filter 1722 0 ip_tables 18968 1 iptable_filter ip6table_filter 1695 0 ip6_tables 19130 1 ip6table_filter x_tables 24840 4 iptable_filter,ip_tables,ip6table_filter,ip6_tables fuse 69275 3 mpt2sas 129590 2
Re: Question about pixman,x11perf and xserver
On 2011-12-05 06:39, 杨帅 wrote: 1.I recompile the pixman with enable-neon,but when I use the x11perf for testing ,I can‘t see the significant difference between the pixman with neon and pixman without neon.That's why? x11perf is an older test suite, and mostly only tests core graphics. Only x11perf -compwinwin and -comppixwin will make heavy use of pixman (along with the new-style text tests, a*text, aa*text and rgb*text to some extent). The good news is, most modern applications draw using those requests. 2.xserver use which drawing library to resizing the window,scroll the text window ,and so on? Depends on the application. Either pixman, or the fb directory of the server. 3.which library excepts pixman could I optimize using neon to improve the performance of xserver? xserver/fb But first, make sure you're working on something that matters to your application. In particular, cairo-perf-trace is probably more relevant these days than x11perf. See http://cgit.freedesktop.org/cairo-traces/tree/README Peter Harris -- Open Text Connectivity Solutions Group Peter Harrishttp://connectivity.opentext.com/ Research and DevelopmentPhone: +1 905 762 6001 phar...@opentext.comToll Free: 1 877 359 4866 ___ xorg@lists.freedesktop.org: X.Org support Archives: http://lists.freedesktop.org/archives/xorg Info: http://lists.freedesktop.org/mailman/listinfo/xorg Your subscription address: arch...@mail-archive.com
Re: 8k resolution video causes X Error
/usr/lib64/xorg/modules/drivers/nvidia_drv.so [ 180.902] (II) Module nvidia: vendor=NVIDIA Corporation [ 180.902] compiled for 4.0.2, module version = 1.0.0 [ 180.902] Module class: X.Org Video Driver [ 180.902] (II) NVIDIA dlloader X Driver 285.05.09 Fri Sep 23 17:33:35 PDT 2011 [ 180.902] (II) NVIDIA Unified Driver for all Supported NVIDIA GPUs So I take it we are running the nvidia driver, not the nouveau driver. You are - so you need to discuss that case with Nvidia not Xorg if you want to make much progress on that driver. Alan ___ xorg@lists.freedesktop.org: X.Org support Archives: http://lists.freedesktop.org/archives/xorg Info: http://lists.freedesktop.org/mailman/listinfo/xorg Your subscription address: arch...@mail-archive.com
rRadeon 9000 + X.Org 1.11.2 + EnablePageFlip + opengl = crash
First, my config: :00:10.0 VGA compatible controller: ATI Technologies Inc Radeon RV250 If [Radeon 9000] (rev 01) Linux Jay 3.1.0 #3 SMP Tue Nov 1 17:58:39 CET 2011 ppc 7455, altivec supported PowerMac3,6 GNU/Linux When EnablePageFlip is enabled the machine crashes as soon as another window hides any part of the glxgears window. The screen turns off and a hard reset is needed (Rsys+B does not work). I'm not using KMS. $ xdpyinfo name of display::0 version number:11.0 vendor string:The X.Org Foundation vendor release number:11102000 X.Org version: 1.11.2 maximum request size: 16777212 bytes motion buffer size: 256 bitmap unit, bit order, padding:32, MSBFirst, 32 image byte order:MSBFirst number of supported pixmap formats:7 supported pixmap formats: depth 1, bits_per_pixel 1, scanline_pad 32 depth 4, bits_per_pixel 8, scanline_pad 32 depth 8, bits_per_pixel 8, scanline_pad 32 depth 15, bits_per_pixel 16, scanline_pad 32 depth 16, bits_per_pixel 16, scanline_pad 32 depth 24, bits_per_pixel 32, scanline_pad 32 depth 32, bits_per_pixel 32, scanline_pad 32 keycode range:minimum 8, maximum 255 focus: window 0xe2, revert to None number of extensions:28 BIG-REQUESTS Composite DAMAGE DOUBLE-BUFFER DPMS DRI2 GLX Generic Event Extension MIT-SCREEN-SAVER MIT-SHM RANDR RECORD RENDER SGI-GLX SHAPE SYNC X-Resource XC-MISC XFIXES XFree86-DGA XFree86-DRI XFree86-VidModeExtension XINERAMA XInputExtension XKEYBOARD XTEST XVideo XVideo-MotionCompensation default screen number:0 number of screens:1 screen #0: print screen:no dimensions:1280x1024 pixels (338x270 millimeters) resolution:96x96 dots per inch depths (7):24, 1, 4, 8, 15, 16, 32 root window id:0x76 depth of root window:24 planes number of colormaps:minimum 1, maximum 1 default colormap:0x20 default number of colormap cells:256 preallocated pixels:black 0, white 16777215 options:backing-store NO, save-unders NO largest cursor:64x64 current input event mask:0xfa200c ButtonPressMask ButtonReleaseMaskButtonMotionMask StructureNotifyMask SubstructureNotifyMask SubstructureRedirectMask FocusChangeMask PropertyChangeMask ColormapChangeMask number of visuals:8 default visual id: 0x21 visual: visual id:0x21 class:TrueColor depth:24 planes available colormap entries:256 per subfield red, green, blue masks:0xff, 0xff00, 0xff significant bits in color specification:8 bits visual: visual id:0x22 class:DirectColor depth:24 planes available colormap entries:256 per subfield red, green, blue masks:0xff, 0xff00, 0xff significant bits in color specification:8 bits visual: visual id:0x70 class:TrueColor depth:24 planes available colormap entries:256 per subfield red, green, blue masks:0xff, 0xff00, 0xff significant bits in color specification:8 bits visual: visual id:0x71 class:TrueColor depth:24 planes available colormap entries:256 per subfield red, green, blue masks:0xff, 0xff00, 0xff significant bits in color specification:8 bits visual: visual id:0x72 class:DirectColor depth:24 planes available colormap entries:256 per subfield red, green, blue masks:0xff, 0xff00, 0xff significant bits in color specification:8 bits visual: visual id:0x73 class:DirectColor depth:24 planes available colormap entries:256 per subfield red, green, blue masks:0xff, 0xff00, 0xff significant bits in color specification:8 bits visual: visual id:0x74 class:DirectColor depth:24 planes available colormap entries:256 per subfield red, green, blue masks:0xff, 0xff00, 0xff significant bits in color specification:8 bits visual: visual id:0x67 class:TrueColor depth:32 planes available colormap entries:256 per subfield red, green, blue masks:0xff, 0xff00, 0xff significant bits in color specification:8 bits -- Giuliano. ___ xorg@lists.freedesktop.org: X.Org support Archives: http://lists.freedesktop.org/archives/xorg Info: http://lists.freedesktop.org/mailman/listinfo/xorg Your subscription address: arch...@mail-archive.com
Re:Re: Question about pixman,x11perf and xserver
hi,Peter Thanks for your reply~ Now I got the direction! I will try to optimize the xserver/fb and use the cairo-perf-trace for testing Best Regards, David 在 2011-12-05 23:47:57,Peter Harris phar...@opentext.com 写道: On 2011-12-05 06:39, 杨帅 wrote: 1.I recompile the pixman with enable-neon,but when I use the x11perf for testing ,I can‘t see the significant difference between the pixman with neon and pixman without neon.That's why? x11perf is an older test suite, and mostly only tests core graphics. Only x11perf -compwinwin and -comppixwin will make heavy use of pixman (along with the new-style text tests, a*text, aa*text and rgb*text to some extent). The good news is, most modern applications draw using those requests. 2.xserver use which drawing library to resizing the window,scroll the text window ,and so on? Depends on the application. Either pixman, or the fb directory of the server. 3.which library excepts pixman could I optimize using neon to improve the performance of xserver? xserver/fb But first, make sure you're working on something that matters to your application. In particular, cairo-perf-trace is probably more relevant these days than x11perf. See http://cgit.freedesktop.org/cairo-traces/tree/README Peter Harris -- Open Text Connectivity Solutions Group Peter Harrishttp://connectivity.opentext.com/ Research and DevelopmentPhone: +1 905 762 6001 phar...@opentext.comToll Free: 1 877 359 4866 ___ xorg@lists.freedesktop.org: X.Org support Archives: http://lists.freedesktop.org/archives/xorg Info: http://lists.freedesktop.org/mailman/listinfo/xorg Your subscription address: yangshuai...@126.com ___ xorg@lists.freedesktop.org: X.Org support Archives: http://lists.freedesktop.org/archives/xorg Info: http://lists.freedesktop.org/mailman/listinfo/xorg Your subscription address: arch...@mail-archive.com