<html>
<head>
<meta content="text/html; charset=windows-1252"
http-equiv="Content-Type">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<div class="moz-cite-prefix">On 03/27/2015 08:08 PM, chen wrote:<br>
</div>
<blockquote
cite="mid:7f872007.ed6.14c5e59bc3f.Coremail.chenm003@163.com"
type="cite">
<div
style="line-height:1.7;color:#000000;font-size:14px;font-family:arial">
<div> </div>
<pre>
At 2015-03-28 10:42:50,<a class="moz-txt-link-abbreviated" href="mailto:dtyx265@gmail.com">dtyx265@gmail.com</a> wrote:
># HG changeset patch
># User David T Yuen <a class="moz-txt-link-rfc2396E" href="mailto:dtyx265@gmail.com"><dtyx265@gmail.com></a>
># Date 1427509819 25200
># Node ID 1ef1d75462375830f47e45d53dcf1c33d458d15e
># Parent 89f337fa1d295b161d4800cc28674c9d02d4c6a9
>asm: intra_pred_ang4_18_sse2
>
>This is backported from sse4 code and replaces c code.
>
>64-bit
>
>./test/TestBench --testbench intrapred | grep "intra_ang_4x4\[18\]"
>intra_ang_4x4[18] 2.75x 367.44 1010.56
>
>32-bit
>
>./test/TestBench --testbench intrapred | grep "intra_ang_4x4\[18\]"
>intra_ang_4x4[18] 2.77x 472.46 1307.93
>
>diff -r 89f337fa1d29 -r 1ef1d7546237 source/common/x86/asm-primitives.cpp
>--- a/source/common/x86/asm-primitives.cpp Fri Mar 27 19:22:53 2015 -0700
>+++ b/source/common/x86/asm-primitives.cpp Fri Mar 27 19:30:19 2015 -0700
>@@ -1212,6 +1212,7 @@
> p.cu[BLOCK_4x4].intra_pred[15] = x265_intra_pred_ang4_15_sse2;
> p.cu[BLOCK_4x4].intra_pred[16] = x265_intra_pred_ang4_16_sse2;
> p.cu[BLOCK_4x4].intra_pred[17] = x265_intra_pred_ang4_17_sse2;
>+ p.cu[BLOCK_4x4].intra_pred[18] = x265_intra_pred_ang4_18_sse2;
> p.cu[BLOCK_4x4].intra_pred[26] = x265_intra_pred_ang4_26_sse2;
>
> p.cu[BLOCK_4x4].calcresidual = x265_getResidual4_sse2;
>diff -r 89f337fa1d29 -r 1ef1d7546237 source/common/x86/intrapred.h
>--- a/source/common/x86/intrapred.h Fri Mar 27 19:22:53 2015 -0700
>+++ b/source/common/x86/intrapred.h Fri Mar 27 19:30:19 2015 -0700
>@@ -63,6 +63,7 @@
> DECL_ANG(4, 15, sse2);
> DECL_ANG(4, 16, sse2);
> DECL_ANG(4, 17, sse2);
>+DECL_ANG(4, 18, sse2);
> DECL_ANG(4, 26, sse2);
>
> DECL_ANG(4, 2, ssse3);
>diff -r 89f337fa1d29 -r 1ef1d7546237 source/common/x86/intrapred8.asm
>--- a/source/common/x86/intrapred8.asm Fri Mar 27 19:22:53 2015 -0700
>+++ b/source/common/x86/intrapred8.asm Fri Mar 27 19:30:19 2015 -0700
>@@ -1751,6 +1751,27 @@
> mova m7, [r3 + 10 * 16] ; [24]
> jmp mangle(private_prefix %+ _ %+ intra_pred_ang4_3 %+ SUFFIX %+ .do_filter4x4)
>
>+cglobal intra_pred_ang4_18, 4,5,2
>+ mov r4d, [r2 + 8]
>+ mov r3b, byte [r2]
>+ mov [r2 + 8], r3b
>+ mov r3d, [r2 + 8]
</pre>
<pre>"mov r3b" here to avoid modify input buffer</pre>
</div>
</blockquote>
I will resubmit. FYI, this is taken from sse4<br>
<blockquote
cite="mid:7f872007.ed6.14c5e59bc3f.Coremail.chenm003@163.com"
type="cite">
<div
style="line-height:1.7;color:#000000;font-size:14px;font-family:arial">
<pre>>+ bswap r3d
>+ movd m0, r3d
>+
>+ movd m1, [r2 + 1]
>+ punpckldq m0, m1
>+ lea r3, [r1 * 3]
>+ movd [r0 + r3], m0
>+ psrldq m0, 1
>+ movd [r0 + r1 * 2], m0
>+ psrldq m0, 1
>+ movd [r0 + r1], m0
>+ psrldq m0, 1
>+ movd [r0], m0
>+ mov [r2 + 8], r4w
</pre>
<pre>can't modify input buffer</pre>
<pre>>+ RET
>+
> ;---------------------------------------------------------------------------------------------
> ; void intra_pred_dc(pixel* dst, intptr_t dstStride, pixel *srcPix, int dirMode, int bFilter)
> ;---------------------------------------------------------------------------------------------
>_______________________________________________
>x265-devel mailing list
><a class="moz-txt-link-abbreviated" href="mailto:x265-devel@videolan.org">x265-devel@videolan.org</a>
><a class="moz-txt-link-freetext" href="https://mailman.videolan.org/listinfo/x265-devel">https://mailman.videolan.org/listinfo/x265-devel</a>
</pre>
</div>
<br>
<fieldset class="mimeAttachmentHeader"></fieldset>
<br>
<pre wrap="">_______________________________________________
x265-devel mailing list
<a class="moz-txt-link-abbreviated" href="mailto:x265-devel@videolan.org">x265-devel@videolan.org</a>
<a class="moz-txt-link-freetext" href="https://mailman.videolan.org/listinfo/x265-devel">https://mailman.videolan.org/listinfo/x265-devel</a>
</pre>
</blockquote>
<br>
</body>
</html>