November 2008 - MPlayer-dev-eng - lists.mplayerhq.hu

Re: [MPlayer-dev-eng] [PATCH] yadif SSE2/SSSE3 optimization
by Zhou, Zongyi 25 Nov '08

25 Nov '08

>cosmetic patch: >- "paddw %%mm3, "%%mm2 \n\t"\ >+ "paddw "MM"3, "MM"2 \n\t"\ >... >+#define MM "%%mm" >... >+FILTER_LINE_ROUTINE(mmx2) > > >SSE2 adding patch: >+#undef MM >+#define MM "%%xmm" >... >+FILTER_LINE_ROUTINE(sse2) This is the cosmetics patch. I will make the SSE2 adding patch when this gets applied. ZZ

2 1

Re: [MPlayer-dev-eng] [PATCH] yadif SSE2/SSSE3 optimization
by Zhou, Zongyi 25 Nov '08

25 Nov '08

Hi Michael, Plz review the new patch. All duplicated codes are now factorized in macros. Regards, ZZ

4 4

[PATCH] Direct3D vo performance increase
by Georgi Petrov 25 Nov '08

25 Nov '08

I post the performance patch from Jim here. It's modified with the new coding style. Just one question - should me eliminated the "goto"? Jim - please find a way to preserve the thread as Reimar suggested. I would like this performance patch's discussion to remain here :)

3 19

stream seek eof handling
by Laurent 24 Nov '08

24 Nov '08

Hi, I've been tracking an issue with mp4 file header parsing that created an infinite loop in libavformat (in mov.c). The infinite loop has been prevented in libavformat, though, the root cause of this issue seems to come from MPlayer, as ffmpeg is not impacted by it. I have been pointed to the bug #1267, that may be the issue I encountered. Basically, stream/stream.c:stream_seek_long() behavior is inconsistent regarding eof handling. In my case, stream_seek_long() fails to move the stream position because eof has been reached in a previous stream_fill_buffer() call. As the stream position never moves forward, libavformat reads in an infinite loop. This issue comes from the handling of position at or after EOF in streams. Sometimes, it is considered as a valid position, sometimes not. Sometimes, eof flag is cleared during a seek, sometimes not. Does someone knows what should be the right behavior in stream/stream.c: * Shall the stream eof flag be cleared when seek succeeds? * Shall EOF be a valid position for seek? * Shall positions past EOF be valid for seek? As a comparison, the system lseek has such behavior. Thanks, Laurent

2 4

[PATCH] enable compilation of vo_macosx.m in 64-bits mode
by Guillaume Poirier 24 Nov '08

24 Nov '08

Hello, Attached patch, extracted from thread: from Gregor Riepl <onitake(a)gmail.com> date Fri, Oct 3, 2008 at 4:49 PM subject Re: [MPlayer-dev-eng] [PATCH] Replaced deprecated QuickDraw calls in vo_quartz allows vo_macosx.m to be compiled in 64-bits mode. It is really small, so I plan to merge it to SVN in a 2-3 days. Guillaume

3 4

[PATCH] Direct3D panscan draw leftover fix
by Georgi Petrov 24 Nov '08

24 Nov '08

>Thing is, it seems to happen also when just switching to fullscreen, >without panscan, and there is already a ColorFill. >Though i just realize that the ColorFill is probably outside the >BeginScene, that might cause it... >Maybe a needs_clear flag and then clearing in the draw_frame if it is >set would be better... I attach the patch that I believe fixes this problem in the best possible way.

2 4

[PATCH] consistent indentation of libvo/vo_quartz.c
by Guillaume Poirier 24 Nov '08

24 Nov '08

Hello, Attached patch makes indentation of libvo/vo_quartz.c consistent throughout the file. I also converted tabs to 4 spaces since IMHO tabs makes it too hard to get a reproducible good indentation. Last but not least, I converted the file header to doxygen syntax. Guillaume

3 2

[PATCH] separate Tremor from libvorbis in console output
by Diego Biurrun 24 Nov '08

24 Nov '08

Here is a patch that makes MPlayer show different output depending on whether it is using Tremor or libvorbis. It's not a beauty, but it does the job. Being unable to differentiate them from the console output is a long-standing usability issue. Any better ideas? Otherwise I will apply after the weekend. Diego

1 1

[PATCH] Fix sparc build
by Alexis Ballier 24 Nov '08

24 Nov '08

Hi all, Building current trunk (rev 28004) on sparc fails with undefined references to mpeg2_idct_copy_mvi & friends. Enclosed patch fixes this, assuming this is a typo and should have been MVI. I'm not sure how libmpeg2_changes.diff should be updated but that patch takes care of it. Alexis.

2 1

[PATCH] Direct3D draw_slice coding style fix
by Georgi Petrov 24 Nov '08

24 Nov '08

Fix for the coding style inside draw_slice().

2 3