[x265] [PATCH] sao: add comment for the reason of two versions of sao primitives

[chen]

just C pointer, no review necessary

At 2015-04-27 21:48:09,"Divya Manivannan" <divya at multicorewareinc.com> wrote:
># HG changeset patch
># User Divya Manivannan <divya at multicorewareinc.com>
># Date 1430142169 -19800
>#      Mon Apr 27 19:12:49 2015 +0530
># Node ID 063c929c30fd7f1e2af3ac1b0fd1d647d690f781
># Parent  ffe98e3a23af5a05d5cedafb876db1b4960c71fc
>sao: add comment for the reason of two versions of sao primitives
>
>diff -r ffe98e3a23af -r 063c929c30fd source/common/primitives.h
>--- a/source/common/primitives.h Mon Apr 27 18:07:45 2015 +0530
>+++ b/source/common/primitives.h Mon Apr 27 19:12:49 2015 +0530
>@@ -275,8 +275,17 @@
> 
>     sign_t                sign;
>     saoCuOrgE0_t          saoCuOrgE0;
>+
>+    /* To avoid the overhead in avx2 optimization in handling width=16, SAO_E0_1 is split
>+     * into two parts: saoCuOrgE1, saoCuOrgE1_2Rows */
>     saoCuOrgE1_t          saoCuOrgE1, saoCuOrgE1_2Rows;
>+
>+    // saoCuOrgE2[0] is used for width<=16 and saoCuOrgE2[1] is used for width > 16.
>     saoCuOrgE2_t          saoCuOrgE2[2];
>+
>+    /* In avx2 optimization, two rows cannot be handled simultaneously since it requires 
>+     * a pixel from the previous row. So, saoCuOrgE3[0] is used for width<=16 and 
>+     * saoCuOrgE3[1] is used for width > 16. */
>     saoCuOrgE3_t          saoCuOrgE3[2];
>     saoCuOrgB0_t          saoCuOrgB0;
> 
>_______________________________________________
>x265-devel mailing list
>x265-devel at videolan.org
>https://mailman.videolan.org/listinfo/x265-devel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/x265-devel/attachments/20150428/533f3af6/attachment.html>