diff options
| author | Frank Barchard <fbarchard@google.com> | 2018-02-16 09:30:12 -0800 |
|---|---|---|
| committer | Commit Bot <commit-bot@chromium.org> | 2018-02-16 18:55:38 +0000 |
| commit | 3d6b5658d7c6b63c482b168b958e4844f122b0b7 (patch) | |
| tree | b895a74d619adbacdd9012c9d2614cab333c09b9 /tools_libyuv | |
| parent | 9c9215b2187f16ddfc38312c994ffa881e7b50f9 (diff) | |
| download | libyuv-3d6b5658d7c6b63c482b168b958e4844f122b0b7.tar.gz | |
AR30ToARGB using shifts and masking to vectorize
AR30ToARGB will vectorize if the output is masked
together as an int instead of 4 byte stores.
Performance is 2x faster
Was AR30ToARGB_Opt (1585 ms)
Now AR30ToARGB_Opt (746 ms)
Bug: libyuv:777
Test:LibYUVConvertTest.AR30ToARGB_Opt
Change-Id: Idd47ae599d5d125207bb53e618d6d7e784d4a37c
Reviewed-on: https://chromium-review.googlesource.com/923169
Reviewed-by: Miguel Casas <mcasas@chromium.org>
Commit-Queue: Frank Barchard <fbarchard@chromium.org>
Diffstat (limited to 'tools_libyuv')
0 files changed, 0 insertions, 0 deletions