| .. |
|
compare_common.cc
|
|
|
|
compare_neon64.cc
|
|
|
|
compare_neon.cc
|
|
|
|
compare_posix.cc
|
Remove alignment constraint for SSE2. Allows the optimized function to be used with unaligned memory, improving performance in that use case. Hurts performance on core2 and prior where memory was faster with movdqa instruction.
|
2014-09-30 18:53:34 +00:00 |
|
compare_win.cc
|
Remove alignment constraint for SSE2. Allows the optimized function to be used with unaligned memory, improving performance in that use case. Hurts performance on core2 and prior where memory was faster with movdqa instruction.
|
2014-09-30 18:53:34 +00:00 |
|
compare.cc
|
Remove alignment constraint for SSE2. Allows the optimized function to be used with unaligned memory, improving performance in that use case. Hurts performance on core2 and prior where memory was faster with movdqa instruction.
|
2014-09-30 18:53:34 +00:00 |
|
convert_argb.cc
|
The 'Any' versions of functions can handle any width now, so remove the check from the calling code. This has 2 advantages - less code, and less overhead in calling function when any function is NOT used. Downside is more code for case where any is used.
|
2014-10-24 23:29:31 +00:00 |
|
convert_from_argb.cc
|
The 'Any' versions of functions can handle any width now, so remove the check from the calling code. This has 2 advantages - less code, and less overhead in calling function when any function is NOT used. Downside is more code for case where any is used.
|
2014-10-24 23:29:31 +00:00 |
|
convert_from.cc
|
The 'Any' versions of functions can handle any width now, so remove the check from the calling code. This has 2 advantages - less code, and less overhead in calling function when any function is NOT used. Downside is more code for case where any is used.
|
2014-10-24 23:29:31 +00:00 |
|
convert_jpeg.cc
|
|
|
|
convert_to_argb.cc
|
|
|
|
convert_to_i420.cc
|
|
|
|
convert.cc
|
The 'Any' versions of functions can handle any width now, so remove the check from the calling code. This has 2 advantages - less code, and less overhead in calling function when any function is NOT used. Downside is more code for case where any is used.
|
2014-10-24 23:29:31 +00:00 |
|
cpu_id.cc
|
Change elif to endif and if to allow AVX2 as well as SSE2 in future changes instead of one or the other.
|
2014-10-16 20:47:22 +00:00 |
|
format_conversion.cc
|
The 'Any' versions of functions can handle any width now, so remove the check from the calling code. This has 2 advantages - less code, and less overhead in calling function when any function is NOT used. Downside is more code for case where any is used.
|
2014-10-24 23:29:31 +00:00 |
|
mjpeg_decoder.cc
|
|
|
|
mjpeg_validate.cc
|
|
|
|
planar_functions.cc
|
The 'Any' versions of functions can handle any width now, so remove the check from the calling code. This has 2 advantages - less code, and less overhead in calling function when any function is NOT used. Downside is more code for case where any is used.
|
2014-10-24 23:29:31 +00:00 |
|
rotate_argb.cc
|
Change elif to endif and if to allow AVX2 as well as SSE2 in future changes instead of one or the other.
|
2014-10-16 20:47:22 +00:00 |
|
rotate_mips.cc
|
|
|
|
rotate_neon64.cc
|
Copy width to 64 bit register to work around clang 3.4 warning
|
2014-10-13 23:26:17 +00:00 |
|
rotate_neon.cc
|
|
|
|
rotate.cc
|
Fix for r1122 to change back to elif for rotate build error on Mac.
|
2014-10-16 22:21:48 +00:00 |
|
row_any.cc
|
Port YUY2ToYRow_AVX2 and UYVYToYRow_AVX2 to gcc/NaCL from Windows AVX code.
|
2014-11-03 18:30:17 +00:00 |
|
row_common.cc
|
Remove unaligned functions, since most function support unaligned memory now. This reduces complexity and improves performance for unaligned cases because C code can be avoided, and overhead is less. Downside is old cpus (core2 and earlier) will be slower for aligned memory case. Except mips, which has alignment requirement, but remove unaligned variant.
|
2014-10-07 00:59:31 +00:00 |
|
row_mips.cc
|
Remove unaligned functions, since most function support unaligned memory now. This reduces complexity and improves performance for unaligned cases because C code can be avoided, and overhead is less. Downside is old cpus (core2 and earlier) will be slower for aligned memory case. Except mips, which has alignment requirement, but remove unaligned variant.
|
2014-10-07 00:59:31 +00:00 |
|
row_neon64.cc
|
Row AArch64 Neon implementation - Part 8
|
2014-10-03 18:10:05 +00:00 |
|
row_neon.cc
|
InterpolateRow used for scale handle unaligned memory. Remove HalfRow which is not used.
|
2014-10-03 17:37:11 +00:00 |
|
row_posix.cc
|
Port YUY2ToUV, YUY2ToUV422, UYVYToUV and UYVYToUV422 to AVX2 on GCC/Nacl.
|
2014-11-04 18:24:10 +00:00 |
|
row_win.cc
|
vzeroupper moved to just before ret. in one case it was done after ret, which is a bug that would cause a performance stall.
|
2014-10-30 19:27:21 +00:00 |
|
row_x86.asm
|
|
|
|
scale_argb.cc
|
The 'Any' versions of functions can handle any width now, so remove the check from the calling code. This has 2 advantages - less code, and less overhead in calling function when any function is NOT used. Downside is more code for case where any is used.
|
2014-10-24 23:29:31 +00:00 |
|
scale_common.cc
|
The 'Any' versions of functions can handle any width now, so remove the check from the calling code. This has 2 advantages - less code, and less overhead in calling function when any function is NOT used. Downside is more code for case where any is used.
|
2014-10-24 23:29:31 +00:00 |
|
scale_mips.cc
|
|
|
|
scale_neon64.cc
|
clear aarch64 related macro and fix bugs
|
2014-10-09 02:00:40 +00:00 |
|
scale_neon.cc
|
|
|
|
scale_posix.cc
|
now that libyuv requires newer nacl compiler, bundles can be assumed and bundle align macro can be removed. no impact on code gen.
|
2014-10-30 20:02:03 +00:00 |
|
scale_win.cc
|
pavgb with memory op requires alignment. This CL disables conversions that use pavgb, and resolves scale by 3/8 unittest for checking alignment works. The 3/8 code used a pavgb with a memory operand. tests are added for scaling and allow unaligning on purpose.
|
2014-10-07 01:57:34 +00:00 |
|
scale.cc
|
The 'Any' versions of functions can handle any width now, so remove the check from the calling code. This has 2 advantages - less code, and less overhead in calling function when any function is NOT used. Downside is more code for case where any is used.
|
2014-10-24 23:29:31 +00:00 |
|
video_common.cc
|
|
|
|
x86inc.asm
|
|
|
3f87404769