[x265] [PATCH] copy_cnt_4: enable fast non zero coefficient count path

chen chenm003 at 163.com
Tue Sep 9 19:18:31 CEST 2014


this code disable because Xeon crash, keep it as a choice by user


At 2014-09-09 13:38:17,praveen at multicorewareinc.com wrote:
># HG changeset patch
># User Praveen Tiwari
># Date 1410241079 -19800
># Node ID 7510b362ca969d850d33b10acb37f3e5a6d6b7dc
># Parent  91e5893dc024ccfcdec04cef5e8832c80632ee0a
>copy_cnt_4: enable fast non zero coefficient count path
>
>diff -r 91e5893dc024 -r 7510b362ca96 source/common/x86/blockcopy8.asm
>--- a/source/common/x86/blockcopy8.asm	Tue Sep 09 10:46:25 2014 +0530
>+++ b/source/common/x86/blockcopy8.asm	Tue Sep 09 11:07:59 2014 +0530
>@@ -3973,7 +3973,7 @@
> 
>     ; get count
>     ; CHECK_ME: Intel documents said POPCNT is SSE4.2 instruction, but just implement after Nehalem
>-%if 0
>+%if 1
>     pmovmskb    eax, m0
>     not         ax
>     popcnt      ax, ax
>_______________________________________________
>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/20140910/9765b08b/attachment.html>


More information about the x265-devel mailing list