summaryrefslogtreecommitdiffstats
path: root/libavutil/arm
Commit message (Collapse)AuthorAgeFilesLines
* ARM: optimised integer clip functionsMåns Rullgård2010-07-071-0/+45
| | | | Originally committed as revision 24089 to svn://svn.ffmpeg.org/ffmpeg/trunk
* ARM: intmath.h cosmeticsMåns Rullgård2010-07-071-3/+7
| | | | Originally committed as revision 24088 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Replace many includes of libavutil/common.h with what is actually neededMåns Rullgård2010-03-092-2/+2
| | | | | | | This reduces the number of false dependencies on header files and speeds up compilation. Originally committed as revision 22407 to svn://svn.ffmpeg.org/ffmpeg/trunk
* ARM: add some missing includesMåns Rullgård2010-03-081-0/+1
| | | | Originally committed as revision 22340 to svn://svn.ffmpeg.org/ffmpeg/trunk
* ARM: change argument/return type of bswap_16() to unsigned 32-bitMåns Rullgård2010-02-221-2/+2
| | | | | | This avoids unnecessary masking otherwise added by the compilers. Originally committed as revision 21953 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Move FASTDIV macro to intmath.hMåns Rullgård2010-01-191-0/+54
| | | | Originally committed as revision 21335 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Mark all intreadwrite functions av_always_inlineMåns Rullgård2010-01-181-6/+6
| | | | Originally committed as revision 21278 to svn://svn.ffmpeg.org/ffmpeg/trunk
* ARM: first value loaded in AV_RN64 needs to be early-clobberMåns Rullgård2009-08-161-1/+1
| | | | Originally committed as revision 19656 to svn://svn.ffmpeg.org/ffmpeg/trunk
* ARM: use gcc inline asm in bswap.h only when availableMåns Rullgård2009-07-011-1/+1
| | | | Originally committed as revision 19311 to svn://svn.ffmpeg.org/ffmpeg/trunk
* ARM asm for AV_RN*()Måns Rullgård2009-04-181-0/+78
| | | | | | | | | | | | | | | ARMv6 and later support unaligned loads and stores for single word/halfword but not double/multiple. GCC is ignorant of this and will always use bytewise accesses for unaligned data. Casting to an int32_t pointer is dangerous since a load/store double or multiple instruction might be used (this happens with some code in FFmpeg). Implementing the AV_[RW]* macros with inline asm using only supported instructions gives fast and safe unaligned accesses. ARM RVCT does the right thing with generic code. This gives an overall speedup of up to 10%. Originally committed as revision 18601 to svn://svn.ffmpeg.org/ffmpeg/trunk
* ARM: START/STOP_TIMER support for ARMv7Måns Rullgård2009-04-011-0/+39
| | | | Originally committed as revision 18305 to svn://svn.ffmpeg.org/ffmpeg/trunk
* ARM: armcc versions of bswap_16/32Måns Rullgård2009-03-051-0/+21
| | | | Originally committed as revision 17830 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Change semantic of CONFIG_*, HAVE_* and ARCH_*.Aurelien Jacobs2009-01-131-2/+2
| | | | | | They are now always defined to either 0 or 1. Originally committed as revision 16590 to svn://svn.ffmpeg.org/ffmpeg/trunk
* split bswap.h into per-arch filesMåns Rullgård2008-10-211-0/+51
Originally committed as revision 15663 to svn://svn.ffmpeg.org/ffmpeg/trunk
OpenPOWER on IntegriCloud