<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN"
"http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
<head><meta http-equiv="content-type" content="text/html; charset=utf-8" />
<title>[210496] trunk/Source</title>
</head>
<body>
<style type="text/css"><!--
#msg dl.meta { border: 1px #006 solid; background: #369; padding: 6px; color: #fff; }
#msg dl.meta dt { float: left; width: 6em; font-weight: bold; }
#msg dt:after { content:':';}
#msg dl, #msg dt, #msg ul, #msg li, #header, #footer, #logmsg { font-family: verdana,arial,helvetica,sans-serif; font-size: 10pt; }
#msg dl a { font-weight: bold}
#msg dl a:link { color:#fc3; }
#msg dl a:active { color:#ff0; }
#msg dl a:visited { color:#cc6; }
h3 { font-family: verdana,arial,helvetica,sans-serif; font-size: 10pt; font-weight: bold; }
#msg pre { overflow: auto; background: #ffc; border: 1px #fa0 solid; padding: 6px; }
#logmsg { background: #ffc; border: 1px #fa0 solid; padding: 1em 1em 0 1em; }
#logmsg p, #logmsg pre, #logmsg blockquote { margin: 0 0 1em 0; }
#logmsg p, #logmsg li, #logmsg dt, #logmsg dd { line-height: 14pt; }
#logmsg h1, #logmsg h2, #logmsg h3, #logmsg h4, #logmsg h5, #logmsg h6 { margin: .5em 0; }
#logmsg h1:first-child, #logmsg h2:first-child, #logmsg h3:first-child, #logmsg h4:first-child, #logmsg h5:first-child, #logmsg h6:first-child { margin-top: 0; }
#logmsg ul, #logmsg ol { padding: 0; list-style-position: inside; margin: 0 0 0 1em; }
#logmsg ul { text-indent: -1em; padding-left: 1em; }#logmsg ol { text-indent: -1.5em; padding-left: 1.5em; }
#logmsg > ul, #logmsg > ol { margin: 0 0 1em 0; }
#logmsg pre { background: #eee; padding: 1em; }
#logmsg blockquote { border: 1px solid #fa0; border-left-width: 10px; padding: 1em 1em 0 1em; background: white;}
#logmsg dl { margin: 0; }
#logmsg dt { font-weight: bold; }
#logmsg dd { margin: 0; padding: 0 0 0.5em 0; }
#logmsg dd:before { content:'\00bb';}
#logmsg table { border-spacing: 0px; border-collapse: collapse; border-top: 4px solid #fa0; border-bottom: 1px solid #fa0; background: #fff; }
#logmsg table th { text-align: left; font-weight: normal; padding: 0.2em 0.5em; border-top: 1px dotted #fa0; }
#logmsg table td { text-align: right; border-top: 1px dotted #fa0; padding: 0.2em 0.5em; }
#logmsg table thead th { text-align: center; border-bottom: 1px solid #fa0; }
#logmsg table th.Corner { text-align: left; }
#logmsg hr { border: none 0; border-top: 2px dashed #fa0; height: 1px; }
#header, #footer { color: #fff; background: #636; border: 1px #300 solid; padding: 6px; }
#patch { width: 100%; }
#patch h4 {font-family: verdana,arial,helvetica,sans-serif;font-size:10pt;padding:8px;background:#369;color:#fff;margin:0;}
#patch .propset h4, #patch .binary h4 {margin:0;}
#patch pre {padding:0;line-height:1.2em;margin:0;}
#patch .diff {width:100%;background:#eee;padding: 0 0 10px 0;overflow:auto;}
#patch .propset .diff, #patch .binary .diff {padding:10px 0;}
#patch span {display:block;padding:0 10px;}
#patch .modfile, #patch .addfile, #patch .delfile, #patch .propset, #patch .binary, #patch .copfile {border:1px solid #ccc;margin:10px 0;}
#patch ins {background:#dfd;text-decoration:none;display:block;padding:0 10px;}
#patch del {background:#fdd;text-decoration:none;display:block;padding:0 10px;}
#patch .lines, .info {color:#888;background:#fff;}
--></style>
<div id="msg">
<dl class="meta">
<dt>Revision</dt> <dd><a href="http://trac.webkit.org/projects/webkit/changeset/210496">210496</a></dd>
<dt>Author</dt> <dd>annulen@yandex.ru</dd>
<dt>Date</dt> <dd>2017-01-08 09:44:57 -0800 (Sun, 08 Jan 2017)</dd>
</dl>
<h3>Log Message</h3>
<pre>Introduce CPU(X86_SSE2) instead of various SSE2 checks
https://bugs.webkit.org/show_bug.cgi?id=166808
Reviewed by Michael Catanzaro.
Now copyLCharsFromUCharSource can use SSE2 implementation on non-Darwin
OSes, and all SSE2 code paths are available for MSVC on x86 if /arch:SSE2
or higher is enabled, and for MSVC on x86_64.
Source/WebCore:
No new tests needed.
* platform/audio/SincResampler.cpp:
(WebCore::SincResampler::process):
* platform/audio/VectorMath.cpp:
(WebCore::VectorMath::vsma):
(WebCore::VectorMath::vsmul):
(WebCore::VectorMath::vadd):
(WebCore::VectorMath::vmul):
(WebCore::VectorMath::zvmul):
(WebCore::VectorMath::vsvesq):
(WebCore::VectorMath::vmaxmgv):
Source/WTF:
* wtf/Platform.h:
* wtf/text/ASCIIFastPath.h:
(WTF::copyLCharsFromUCharSource):</pre>
<h3>Modified Paths</h3>
<ul>
<li><a href="#trunkSourceWTFChangeLog">trunk/Source/WTF/ChangeLog</a></li>
<li><a href="#trunkSourceWTFwtfPlatformh">trunk/Source/WTF/wtf/Platform.h</a></li>
<li><a href="#trunkSourceWTFwtftextASCIIFastPathh">trunk/Source/WTF/wtf/text/ASCIIFastPath.h</a></li>
<li><a href="#trunkSourceWebCoreChangeLog">trunk/Source/WebCore/ChangeLog</a></li>
<li><a href="#trunkSourceWebCoreplatformaudioSincResamplercpp">trunk/Source/WebCore/platform/audio/SincResampler.cpp</a></li>
<li><a href="#trunkSourceWebCoreplatformaudioVectorMathcpp">trunk/Source/WebCore/platform/audio/VectorMath.cpp</a></li>
</ul>
</div>
<div id="patch">
<h3>Diff</h3>
<a id="trunkSourceWTFChangeLog"></a>
<div class="modfile"><h4>Modified: trunk/Source/WTF/ChangeLog (210495 => 210496)</h4>
<pre class="diff"><span>
<span class="info">--- trunk/Source/WTF/ChangeLog 2017-01-08 06:02:25 UTC (rev 210495)
+++ trunk/Source/WTF/ChangeLog 2017-01-08 17:44:57 UTC (rev 210496)
</span><span class="lines">@@ -1,3 +1,18 @@
</span><ins>+2017-01-08 Konstantin Tokarev <annulen@yandex.ru>
+
+ Introduce CPU(X86_SSE2) instead of various SSE2 checks
+ https://bugs.webkit.org/show_bug.cgi?id=166808
+
+ Reviewed by Michael Catanzaro.
+
+ Now copyLCharsFromUCharSource can use SSE2 implementation on non-Darwin
+ OSes, and all SSE2 code paths are available for MSVC on x86 if /arch:SSE2
+ or higher is enabled, and for MSVC on x86_64.
+
+ * wtf/Platform.h:
+ * wtf/text/ASCIIFastPath.h:
+ (WTF::copyLCharsFromUCharSource):
+
</ins><span class="cx"> 2017-01-05 Myles C. Maxfield <mmaxfield@apple.com>
</span><span class="cx">
</span><span class="cx"> Carets can split up marriages and families
</span></span></pre></div>
<a id="trunkSourceWTFwtfPlatformh"></a>
<div class="modfile"><h4>Modified: trunk/Source/WTF/wtf/Platform.h (210495 => 210496)</h4>
<pre class="diff"><span>
<span class="info">--- trunk/Source/WTF/wtf/Platform.h 2017-01-08 06:02:25 UTC (rev 210495)
+++ trunk/Source/WTF/wtf/Platform.h 2017-01-08 17:44:57 UTC (rev 210496)
</span><span class="lines">@@ -162,12 +162,18 @@
</span><span class="cx"> || defined(_X86_) \
</span><span class="cx"> || defined(__THW_INTEL)
</span><span class="cx"> #define WTF_CPU_X86 1
</span><ins>+
+#if defined(__SSE2__) || (defined(_M_IX86_FP) && _M_IX86_FP >= 2)
+#define WTF_CPU_X86_SSE2 1
</ins><span class="cx"> #endif
</span><span class="cx">
</span><ins>+#endif
+
</ins><span class="cx"> /* CPU(X86_64) - AMD64 / Intel64 / x86_64 64-bit */
</span><span class="cx"> #if defined(__x86_64__) \
</span><span class="cx"> || defined(_M_X64)
</span><span class="cx"> #define WTF_CPU_X86_64 1
</span><ins>+#define WTF_CPU_X86_SSE2 1
</ins><span class="cx"> #endif
</span><span class="cx">
</span><span class="cx"> /* CPU(ARM64) - Apple */
</span></span></pre></div>
<a id="trunkSourceWTFwtftextASCIIFastPathh"></a>
<div class="modfile"><h4>Modified: trunk/Source/WTF/wtf/text/ASCIIFastPath.h (210495 => 210496)</h4>
<pre class="diff"><span>
<span class="info">--- trunk/Source/WTF/wtf/text/ASCIIFastPath.h 2017-01-08 06:02:25 UTC (rev 210495)
+++ trunk/Source/WTF/wtf/text/ASCIIFastPath.h 2017-01-08 17:44:57 UTC (rev 210496)
</span><span class="lines">@@ -27,7 +27,7 @@
</span><span class="cx"> #include <wtf/StdLibExtras.h>
</span><span class="cx"> #include <wtf/text/LChar.h>
</span><span class="cx">
</span><del>-#if OS(DARWIN) && (CPU(X86) || CPU(X86_64))
</del><ins>+#if CPU(X86_SSE2)
</ins><span class="cx"> #include <emmintrin.h>
</span><span class="cx"> #endif
</span><span class="cx">
</span><span class="lines">@@ -109,7 +109,7 @@
</span><span class="cx">
</span><span class="cx"> inline void copyLCharsFromUCharSource(LChar* destination, const UChar* source, size_t length)
</span><span class="cx"> {
</span><del>-#if OS(DARWIN) && (CPU(X86) || CPU(X86_64))
</del><ins>+#if CPU(X86_SSE2)
</ins><span class="cx"> const uintptr_t memoryAccessSize = 16; // Memory accesses on 16 byte (128 bit) alignment
</span><span class="cx"> const uintptr_t memoryAccessMask = memoryAccessSize - 1;
</span><span class="cx">
</span></span></pre></div>
<a id="trunkSourceWebCoreChangeLog"></a>
<div class="modfile"><h4>Modified: trunk/Source/WebCore/ChangeLog (210495 => 210496)</h4>
<pre class="diff"><span>
<span class="info">--- trunk/Source/WebCore/ChangeLog 2017-01-08 06:02:25 UTC (rev 210495)
+++ trunk/Source/WebCore/ChangeLog 2017-01-08 17:44:57 UTC (rev 210496)
</span><span class="lines">@@ -1,3 +1,27 @@
</span><ins>+2017-01-08 Konstantin Tokarev <annulen@yandex.ru>
+
+ Introduce CPU(X86_SSE2) instead of various SSE2 checks
+ https://bugs.webkit.org/show_bug.cgi?id=166808
+
+ Reviewed by Michael Catanzaro.
+
+ Now copyLCharsFromUCharSource can use SSE2 implementation on non-Darwin
+ OSes, and all SSE2 code paths are available for MSVC on x86 if /arch:SSE2
+ or higher is enabled, and for MSVC on x86_64.
+
+ No new tests needed.
+
+ * platform/audio/SincResampler.cpp:
+ (WebCore::SincResampler::process):
+ * platform/audio/VectorMath.cpp:
+ (WebCore::VectorMath::vsma):
+ (WebCore::VectorMath::vsmul):
+ (WebCore::VectorMath::vadd):
+ (WebCore::VectorMath::vmul):
+ (WebCore::VectorMath::zvmul):
+ (WebCore::VectorMath::vsvesq):
+ (WebCore::VectorMath::vmaxmgv):
+
</ins><span class="cx"> 2017-01-07 Simon Fraser <simon.fraser@apple.com>
</span><span class="cx">
</span><span class="cx"> Avoid triggering rebuilds for minor changes of CSSProperties.json
</span></span></pre></div>
<a id="trunkSourceWebCoreplatformaudioSincResamplercpp"></a>
<div class="modfile"><h4>Modified: trunk/Source/WebCore/platform/audio/SincResampler.cpp (210495 => 210496)</h4>
<pre class="diff"><span>
<span class="info">--- trunk/Source/WebCore/platform/audio/SincResampler.cpp 2017-01-08 06:02:25 UTC (rev 210495)
+++ trunk/Source/WebCore/platform/audio/SincResampler.cpp 2017-01-08 17:44:57 UTC (rev 210496)
</span><span class="lines">@@ -35,7 +35,7 @@
</span><span class="cx"> #include "AudioBus.h"
</span><span class="cx"> #include <wtf/MathExtras.h>
</span><span class="cx">
</span><del>-#ifdef __SSE2__
</del><ins>+#if CPU(X86_SSE2)
</ins><span class="cx"> #include <emmintrin.h>
</span><span class="cx"> #endif
</span><span class="cx">
</span><span class="lines">@@ -260,7 +260,7 @@
</span><span class="cx"> {
</span><span class="cx"> float input;
</span><span class="cx">
</span><del>-#ifdef __SSE2__
</del><ins>+#if CPU(X86_SSE2)
</ins><span class="cx"> // If the sourceP address is not 16-byte aligned, the first several frames (at most three) should be processed seperately.
</span><span class="cx"> while ((reinterpret_cast<uintptr_t>(inputP) & 0x0F) && n) {
</span><span class="cx"> CONVOLVE_ONE_SAMPLE
</span></span></pre></div>
<a id="trunkSourceWebCoreplatformaudioVectorMathcpp"></a>
<div class="modfile"><h4>Modified: trunk/Source/WebCore/platform/audio/VectorMath.cpp (210495 => 210496)</h4>
<pre class="diff"><span>
<span class="info">--- trunk/Source/WebCore/platform/audio/VectorMath.cpp 2017-01-08 06:02:25 UTC (rev 210495)
+++ trunk/Source/WebCore/platform/audio/VectorMath.cpp 2017-01-08 17:44:57 UTC (rev 210496)
</span><span class="lines">@@ -32,7 +32,7 @@
</span><span class="cx"> #include <Accelerate/Accelerate.h>
</span><span class="cx"> #endif
</span><span class="cx">
</span><del>-#ifdef __SSE2__
</del><ins>+#if CPU(X86_SSE2)
</ins><span class="cx"> #include <emmintrin.h>
</span><span class="cx"> #endif
</span><span class="cx">
</span><span class="lines">@@ -134,7 +134,7 @@
</span><span class="cx"> {
</span><span class="cx"> int n = framesToProcess;
</span><span class="cx">
</span><del>-#ifdef __SSE2__
</del><ins>+#if CPU(X86_SSE2)
</ins><span class="cx"> if ((sourceStride == 1) && (destStride == 1)) {
</span><span class="cx"> float k = *scale;
</span><span class="cx">
</span><span class="lines">@@ -207,7 +207,7 @@
</span><span class="cx"> {
</span><span class="cx"> int n = framesToProcess;
</span><span class="cx">
</span><del>-#ifdef __SSE2__
</del><ins>+#if CPU(X86_SSE2)
</ins><span class="cx"> if ((sourceStride == 1) && (destStride == 1)) {
</span><span class="cx"> float k = *scale;
</span><span class="cx">
</span><span class="lines">@@ -278,7 +278,7 @@
</span><span class="cx"> sourceP += sourceStride;
</span><span class="cx"> destP += destStride;
</span><span class="cx"> }
</span><del>-#ifdef __SSE2__
</del><ins>+#if CPU(X86_SSE2)
</ins><span class="cx"> }
</span><span class="cx"> #endif
</span><span class="cx"> }
</span><span class="lines">@@ -287,7 +287,7 @@
</span><span class="cx"> {
</span><span class="cx"> int n = framesToProcess;
</span><span class="cx">
</span><del>-#ifdef __SSE2__
</del><ins>+#if CPU(X86_SSE2)
</ins><span class="cx"> if ((sourceStride1 ==1) && (sourceStride2 == 1) && (destStride == 1)) {
</span><span class="cx"> // If the sourceP address is not 16-byte aligned, the first several frames (at most three) should be processed separately.
</span><span class="cx"> while ((reinterpret_cast<size_t>(source1P) & 0x0F) && n) {
</span><span class="lines">@@ -390,7 +390,7 @@
</span><span class="cx"> source2P += sourceStride2;
</span><span class="cx"> destP += destStride;
</span><span class="cx"> }
</span><del>-#ifdef __SSE2__
</del><ins>+#if CPU(X86_SSE2)
</ins><span class="cx"> }
</span><span class="cx"> #endif
</span><span class="cx"> }
</span><span class="lines">@@ -400,7 +400,7 @@
</span><span class="cx">
</span><span class="cx"> int n = framesToProcess;
</span><span class="cx">
</span><del>-#ifdef __SSE2__
</del><ins>+#if CPU(X86_SSE2)
</ins><span class="cx"> if ((sourceStride1 == 1) && (sourceStride2 == 1) && (destStride == 1)) {
</span><span class="cx"> // If the source1P address is not 16-byte aligned, the first several frames (at most three) should be processed separately.
</span><span class="cx"> while ((reinterpret_cast<uintptr_t>(source1P) & 0x0F) && n) {
</span><span class="lines">@@ -473,7 +473,7 @@
</span><span class="cx"> void zvmul(const float* real1P, const float* imag1P, const float* real2P, const float* imag2P, float* realDestP, float* imagDestP, size_t framesToProcess)
</span><span class="cx"> {
</span><span class="cx"> unsigned i = 0;
</span><del>-#ifdef __SSE2__
</del><ins>+#if CPU(X86_SSE2)
</ins><span class="cx"> // Only use the SSE optimization in the very common case that all addresses are 16-byte aligned.
</span><span class="cx"> // Otherwise, fall through to the scalar code below.
</span><span class="cx"> if (!(reinterpret_cast<uintptr_t>(real1P) & 0x0F)
</span><span class="lines">@@ -531,7 +531,7 @@
</span><span class="cx"> int n = framesToProcess;
</span><span class="cx"> float sum = 0;
</span><span class="cx">
</span><del>-#ifdef __SSE2__
</del><ins>+#if CPU(X86_SSE2)
</ins><span class="cx"> if (sourceStride == 1) {
</span><span class="cx"> // If the sourceP address is not 16-byte aligned, the first several frames (at most three) should be processed separately.
</span><span class="cx"> while ((reinterpret_cast<uintptr_t>(sourceP) & 0x0F) && n) {
</span><span class="lines">@@ -596,7 +596,7 @@
</span><span class="cx"> int n = framesToProcess;
</span><span class="cx"> float max = 0;
</span><span class="cx">
</span><del>-#ifdef __SSE2__
</del><ins>+#if CPU(X86_SSE2)
</ins><span class="cx"> if (sourceStride == 1) {
</span><span class="cx"> // If the sourceP address is not 16-byte aligned, the first several frames (at most three) should be processed separately.
</span><span class="cx"> while ((reinterpret_cast<uintptr_t>(sourceP) & 0x0F) && n) {
</span></span></pre>
</div>
</div>
</body>
</html>