From 7a61759f78e37113221cfe7c40c522aa505280af Mon Sep 17 00:00:00 2001 From: Frank Barchard Date: Mon, 4 May 2020 12:32:28 -0700 Subject: NV12Mirror and MirrorUVPlane functions added HalfMergeUV AVX2 version Skylake Xeon performance for 1280x720 NV12Mirror_Any (109 ms) NV12Mirror_Unaligned (113 ms) NV12Mirror_Invert (107 ms) NV12Mirror_Opt (108 ms) NV12Mirror_NullY (19 ms) Slightly faster than comparable I420Mirror I420Mirror_Any (113 ms) I420Mirror_Unaligned (110 ms) I420Mirror_Invert (109 ms) I420Mirror_Opt (110 ms) BUG=libyuv:840, libyuv:858 Change-Id: I686b1b778383bfa10ecd1655e986bdc99e76d132 Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/2176066 Commit-Queue: Frank Barchard Reviewed-by: richard winterton --- docs/formats.md | 1 + docs/getting_started.md | 2 +- 2 files changed, 2 insertions(+), 1 deletion(-) (limited to 'docs') diff --git a/docs/formats.md b/docs/formats.md index 260dd731..771fb460 100644 --- a/docs/formats.md +++ b/docs/formats.md @@ -166,3 +166,4 @@ The 12 in NV12 refers to 12 bits per pixel. NV12 has a half width and half height chroma channel, and therefore is a 420 subsampling. NV16 is 16 bits per pixel, with half width and full height. aka 422. NV24 is 24 bits per pixel with full sized chroma channel. aka 444. +Most NV12 functions allow the destination Y pointer to be NULL. diff --git a/docs/getting_started.md b/docs/getting_started.md index 4426b606..3e339712 100644 --- a/docs/getting_started.md +++ b/docs/getting_started.md @@ -190,7 +190,7 @@ mips make V=1 -f linux.mk make V=1 -f linux.mk clean - make V=1 -f linux.mk CXX=clang++ + make V=1 -f linux.mk CXX=clang++ CC=clang ## Building the library with cmake -- cgit v1.2.3