]> git.sesse.net Git - vlc/commit
projects / vlc / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 24c25e9) | patch
copy: minor speed enhancement to USWC copy
authorJean-Yves Avenard <jyavenard@mythtv.org>
Sun, 15 Jun 2014 09:04:37 +0000 (09:04 +0000)
committerJean-Baptiste Kempf <jb@videolan.org>
Mon, 16 Jun 2014 07:50:24 +0000 (09:50 +0200)
commitb50a07a3ee8728c602d88432365aae67abc81b28
tree7d1be70249e5a82a036cf89549c740ac2c8bbd77tree | snapshot
parent24c25e9999f6dd90efe22ad8c1af744f9b3452b0commit | diff
copy: minor speed enhancement to USWC copy

Adopt suggestions as per https://software.intel.com/en-us/articles/copying-accelerated-video-decode-frame-buffers:
- 64kB aligned memory buffer
- Call to mfence only around copy to cache routine.

Use a single instruction to copy non-aligned memory at startup.
Remove compilation warning on mac

This result in a small 1.3% speed increase on an i7-4650U

Signed-off-by: Jean-Baptiste Kempf <jb@videolan.org>
modules/video_chroma/copy.c diff | blob | history
Unnamed repository; edit this file 'description' to name the repository.