* commit 'b4dd424d96f09f9bafb88e47f37df65dc4529143':
Remove all SPARC architecture optimizations
Conflicts:
Makefile
configure
libavcodec/sparc/dsputil_vis.c
libavcodec/sparc/dsputil_vis.h
libavcodec/sparc/hpeldsp_vis.c
libavcodec/sparc/simple_idct_vis.c
libavcodec/sparc/vis.h
libswscale/sparc/yuv2rgb_vis.c
libswscale/swscale_internal.h
If someone wants to maintain these (or other) SPARC optimizations, please
contact me or ffmpeg-devel.
I am happy to revert this removial if theres someone considering to
maintain this code.
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit 'd6096a67422534918405abb46dafbbac4608cbc3':
Remove all SH4 architecture optimizations
Conflicts:
libavcodec/sh4/dsputil_sh4.c
libavcodec/sh4/dsputil_sh4.h
libavcodec/sh4/idct_sh4.c
libavcodec/sh4/sh4.h
If someone wants to maintain these (or other) SH4 optimizations, please
contact me or ffmpeg-devel.
I am happy to revert this removial if theres someone considering to
maintain this code.
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '88b240646f57a62299c6c4c4d7554f0a80c3a29a':
build: Drop stray mpc8 decoder dependency on dsputil
Merged-by: Michael Niedermayer <michaelni@gmx.at>
The function is supposed to confirm that the compiler provided enough
alignment, but in practice it is only run in certain code paths and
insufficient alignment problems are restricted to legacy compilers.
This reverts commit 691dec62011fe9993809fbc793126b40cac0c584.
The commit did not fix ticket #3215, it was fixed one commit earlier.
The revert may break other use-cases but they should be fixed differently,
the offending commit introduced too many problems.
Fixes ticket #3377.
Fixes ticket #3378.
* cehoyos/master:
Define ff_log2_run[] in libavcodec/internal.h.
Replace an incorrect av_free() in movenc.c with av_freep().
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit 'ed06e5d92b4c67b49068d538461fbbe0a53a8c5e':
hevc: Do not turn 32bit timebases into negative numbers
Conflicts:
libavcodec/hevc.c
See: bf2ce19e51fde57727a4ee33a4b3c2622f37a5c4
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* cus/stable:
libzvbi-teletextdec: split dvb packet to slices
libzvbi-teletextdec: use av_dlog where possible
Merged-by: Michael Niedermayer <michaelni@gmx.at>
Based on the aarch64 asm. CPU cycle counts on cortex-a9 compared to
gcc 4.8.2:
before: 475 decicycles in get_cabac_noinline, 67106035 runs, 2829 skips
after: 393 decicycles in get_cabac_noinline, 67106474 runs, 2390 skips
Overall speedup is above 2%. Code generated by clang 3.4 is slower on
the same hardware and the relative change is a little larger.
Based on the x86 branchless get_cabac asm. get_cabac_noinline() gets
approximately 20% faster (no cycle counts available) compared to clang
from Xcode 5.1 beta5. More than 6% faster overall. A part of the overall
speedup might be explained by additional inlining of get_cabac().
Instead of using the demux function of libzvbi to split the packet to slices
(vbi lines), lets do it ourselves.
- eliminates the 1 frame delay between page input and output
- handles non-ascending line numbers more gracefully
- enables us to return error codes on some invalid packets instead of silently
ignoring them
Signed-off-by: Marton Balint <cus@passwd.hu>
The overread avoidance fix in cbddee1cca0ebd01e8c5aa694d31228eb4de4b41
broke the computation for the last row since it prevented the safe
reading from the height+1-th row.
CC: libav-stable@libav.org
