<div dir="ltr"><br><div class="gmail_extra"><br><br><div class="gmail_quote">On Tue, Nov 12, 2013 at 6:28 AM, <span dir="ltr"><<a href="mailto:praveen@multicorewareinc.com" target="_blank">praveen@multicorewareinc.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"># HG changeset patch<br>
# User Praveen Tiwari<br>
# Date 1384259293 -19800<br>
# Node ID 9836ede305fad313f4282063887cc9d028c435a8<br>
# Parent 694c0d3db925d4d0f8311a1e5a0a83b53ab99af7<br>
TComYuv.cpp, use new luma_copy_ps asm primitives where feasible<br>
<br>
diff -r 694c0d3db925 -r 9836ede305fa source/Lib/TLibCommon/TComYuv.cpp<br>
--- a/source/Lib/TLibCommon/TComYuv.cpp Tue Nov 12 17:41:21 2013 +0530<br>
+++ b/source/Lib/TLibCommon/TComYuv.cpp Tue Nov 12 17:58:13 2013 +0530<br>
@@ -281,7 +281,8 @@<br>
uint32_t srcstride = getStride();<br>
uint32_t dststride = dstPicYuv->m_width;<br>
<br>
- primitives.blockcpy_sp(width, height, dst, dststride, src, srcstride);<br>
+ int part = partitionFromSizes(width, height);<br>
+ primitives.luma_copy_ps[part](dst, dststride, src, srcstride);<br></blockquote><div><br></div><div>queued, but this is not necessarily a performance improvement. the additional function overhead should be avoided by passing in the part enum rather than width and height. Most callers should already have the partition enum.</div>
<div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
}<br>
<br>
void TComYuv::copyPartToPartChroma(TComYuv* dstPicYuv, uint32_t partIdx, uint32_t width, uint32_t height)<br>
_______________________________________________<br>
x265-devel mailing list<br>
<a href="mailto:x265-devel@videolan.org">x265-devel@videolan.org</a><br>
<a href="https://mailman.videolan.org/listinfo/x265-devel" target="_blank">https://mailman.videolan.org/listinfo/x265-devel</a><br>
</blockquote></div><br><br clear="all"><div><br></div>-- <br>Steve Borho
</div></div>