[webkit-changes] [155099] trunk

2013-09-05 Thread zherczeg
Title: [155099] trunk








Revision 155099
Author zherc...@webkit.org
Date 2013-09-04 23:38:55 -0700 (Wed, 04 Sep 2013)


Log Message
https://webkit.org/b/118553 Hash navigation doesn't affect history when the page is retrieved from appcache

Reviewed by Darin Adler.

Source/WebCore: 

A new flag called keepInHistory is added to allow keeping
certain pages in the history even if they have substitute data.

Test: http/tests/appcache/history-test.html

* loader/DocumentLoader.cpp:
(WebCore::DocumentLoader::urlForHistory):
* loader/SubstituteData.h:
(WebCore::SubstituteData::SubstituteData):
(WebCore::SubstituteData::shouldRevealToSessionHistory):
(WebCore::SubstituteData::setShouldRevealToSessionHistory):
* loader/appcache/ApplicationCacheHost.cpp:
(WebCore::ApplicationCacheHost::maybeLoadMainResource):

LayoutTests: 

* http/tests/appcache/history-test-expected.txt: Added.
* http/tests/appcache/history-test.html: Added.
* http/tests/appcache/resources/history-test.html: Added.
* http/tests/appcache/resources/history-test.manifest: Added.

Modified Paths

trunk/LayoutTests/ChangeLog
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/loader/DocumentLoader.cpp
trunk/Source/WebCore/loader/SubstituteData.h
trunk/Source/WebCore/loader/appcache/ApplicationCacheHost.cpp


Added Paths

trunk/LayoutTests/http/tests/appcache/history-test-expected.txt
trunk/LayoutTests/http/tests/appcache/history-test.html
trunk/LayoutTests/http/tests/appcache/resources/history-test.html
trunk/LayoutTests/http/tests/appcache/resources/history-test.manifest




Diff

Modified: trunk/LayoutTests/ChangeLog (155098 => 155099)

--- trunk/LayoutTests/ChangeLog	2013-09-05 06:34:58 UTC (rev 155098)
+++ trunk/LayoutTests/ChangeLog	2013-09-05 06:38:55 UTC (rev 155099)
@@ -1,3 +1,14 @@
+2013-09-04  Zoltan Herczeg  zherc...@webkit.org
+
+https://webkit.org/b/118553 Hash navigation doesn't affect history when the page is retrieved from appcache
+
+Reviewed by Darin Adler.
+
+* http/tests/appcache/history-test-expected.txt: Added.
+* http/tests/appcache/history-test.html: Added.
+* http/tests/appcache/resources/history-test.html: Added.
+* http/tests/appcache/resources/history-test.manifest: Added.
+
 2013-09-04  Mihai Tica  mit...@adobe.com
 
 Test background blend modes with animated gif background layers.


Added: trunk/LayoutTests/http/tests/appcache/history-test-expected.txt (0 => 155099)

--- trunk/LayoutTests/http/tests/appcache/history-test-expected.txt	(rev 0)
+++ trunk/LayoutTests/http/tests/appcache/history-test-expected.txt	2013-09-05 06:38:55 UTC (rev 155099)
@@ -0,0 +1,3 @@
+Test whether window.history.back() is supported in application cache.
+
+PASS: Returning to the original page.


Added: trunk/LayoutTests/http/tests/appcache/history-test.html (0 => 155099)

--- trunk/LayoutTests/http/tests/appcache/history-test.html	(rev 0)
+++ trunk/LayoutTests/http/tests/appcache/history-test.html	2013-09-05 06:38:55 UTC (rev 155099)
@@ -0,0 +1,11 @@
+html
+body
+script
+if (window.testRunner) {
+testRunner.waitUntilDone();
+testRunner.dumpAsText();
+}
+window.location.assign(resources/history-test.html);
+/script
+/body
+/html


Added: trunk/LayoutTests/http/tests/appcache/resources/history-test.html (0 => 155099)

--- trunk/LayoutTests/http/tests/appcache/resources/history-test.html	(rev 0)
+++ trunk/LayoutTests/http/tests/appcache/resources/history-test.html	2013-09-05 06:38:55 UTC (rev 155099)
@@ -0,0 +1,23 @@
+html manifest=history-test.manifest
+body
+pTest whether window.history.back() is supported in application cache./p
+
+script
+
+if (document.cookie) {
+document.write(pPASS: Returning to the original page./p);
+if (window.testRunner)
+testRunner.notifyDone();
+} else {
+document.write(pSTART: Redirecting to other page./p);
+document.cookie = 'cookie1=testcookie';
+
+// Give time to the history manager to cache this page.
+setTimeout(function () {
+window.location.assign(/navigation/resources/go-back.html);
+}, 200);
+}
+
+/script
+/body
+/html


Added: trunk/LayoutTests/http/tests/appcache/resources/history-test.manifest (0 => 155099)

--- trunk/LayoutTests/http/tests/appcache/resources/history-test.manifest	(rev 0)
+++ trunk/LayoutTests/http/tests/appcache/resources/history-test.manifest	2013-09-05 06:38:55 UTC (rev 155099)
@@ -0,0 +1,5 @@
+CACHE MANIFEST
+history-test.html
+
+NETWORK:
+*


Modified: trunk/Source/WebCore/ChangeLog (155098 => 155099)

--- trunk/Source/WebCore/ChangeLog	2013-09-05 06:34:58 UTC (rev 155098)
+++ trunk/Source/WebCore/ChangeLog	2013-09-05 06:38:55 UTC (rev 155099)
@@ -1,3 +1,23 @@
+2013-09-04  Zoltan Herczeg  zherc...@webkit.org
+
+https://webkit.org/b/118553 Hash navigation doesn't affect history when the page is retrieved from appcache
+
+Reviewed by Darin Adler.
+
+A new 

[webkit-changes] [146396] trunk/Source/JavaScriptCore

2013-03-20 Thread zherczeg
Title: [146396] trunk/Source/_javascript_Core








Revision 146396
Author zherc...@webkit.org
Date 2013-03-20 15:07:23 -0700 (Wed, 20 Mar 2013)


Log Message
ARMv7 replaceWithJump ASSERT failure after r135330.
https://bugs.webkit.org/show_bug.cgi?id=103146

Reviewed by Filip Pizlo.

On Linux, the 24 bit distance range of jumps sometimes does not
enough to cover all targets addresses. This patch supports jumps
outside of this range using a mov/movt/bx 10 byte long sequence.

* assembler/ARMv7Assembler.h:
(ARMv7Assembler):
(JSC::ARMv7Assembler::revertJumpTo_movT3movtcmpT2):
(JSC::ARMv7Assembler::nopw):
(JSC::ARMv7Assembler::label):
(JSC::ARMv7Assembler::replaceWithJump):
(JSC::ARMv7Assembler::maxJumpReplacementSize):
* assembler/MacroAssemblerARMv7.h:
(JSC::MacroAssemblerARMv7::revertJumpReplacementToBranchPtrWithPatch):

Modified Paths

trunk/Source/_javascript_Core/ChangeLog
trunk/Source/_javascript_Core/assembler/ARMv7Assembler.h
trunk/Source/_javascript_Core/assembler/MacroAssemblerARMv7.h




Diff

Modified: trunk/Source/_javascript_Core/ChangeLog (146395 => 146396)

--- trunk/Source/_javascript_Core/ChangeLog	2013-03-20 22:05:08 UTC (rev 146395)
+++ trunk/Source/_javascript_Core/ChangeLog	2013-03-20 22:07:23 UTC (rev 146396)
@@ -1,3 +1,24 @@
+2013-03-20  Zoltan Herczeg  zherc...@webkit.org
+
+ARMv7 replaceWithJump ASSERT failure after r135330.
+https://bugs.webkit.org/show_bug.cgi?id=103146
+
+Reviewed by Filip Pizlo.
+
+On Linux, the 24 bit distance range of jumps sometimes does not
+enough to cover all targets addresses. This patch supports jumps
+outside of this range using a mov/movt/bx 10 byte long sequence.
+
+* assembler/ARMv7Assembler.h:
+(ARMv7Assembler):
+(JSC::ARMv7Assembler::revertJumpTo_movT3movtcmpT2):
+(JSC::ARMv7Assembler::nopw):
+(JSC::ARMv7Assembler::label):
+(JSC::ARMv7Assembler::replaceWithJump):
+(JSC::ARMv7Assembler::maxJumpReplacementSize):
+* assembler/MacroAssemblerARMv7.h:
+(JSC::MacroAssemblerARMv7::revertJumpReplacementToBranchPtrWithPatch):
+
 2013-03-20  Mark Hahnenberg  mhahnenb...@apple.com
 
 Objective-C API: Fix over-releasing in allocateConstructorAndPrototypeWithSuperClassInfo:


Modified: trunk/Source/_javascript_Core/assembler/ARMv7Assembler.h (146395 => 146396)

--- trunk/Source/_javascript_Core/assembler/ARMv7Assembler.h	2013-03-20 22:05:08 UTC (rev 146395)
+++ trunk/Source/_javascript_Core/assembler/ARMv7Assembler.h	2013-03-20 22:07:23 UTC (rev 146396)
@@ -1266,6 +1266,20 @@
 m_formatter.twoWordOp5i6Imm4Reg4EncodedImm(OP_MOV_imm_T3, imm.m_value.imm4, rd, imm);
 }
 
+#if OS(LINUX)
+static void revertJumpTo_movT3movtcmpT2(void* instructionStart, RegisterID left, RegisterID right, uintptr_t imm)
+{
+uint16_t* address = static_castuint16_t*(instructionStart);
+ARMThumbImmediate lo16 = ARMThumbImmediate::makeUInt16(static_castuint16_t(imm));
+ARMThumbImmediate hi16 = ARMThumbImmediate::makeUInt16(static_castuint16_t(imm  16));
+address[0] = twoWordOp5i6Imm4Reg4EncodedImmFirst(OP_MOV_imm_T3, lo16);
+address[1] = twoWordOp5i6Imm4Reg4EncodedImmSecond(right, lo16);
+address[2] = twoWordOp5i6Imm4Reg4EncodedImmFirst(OP_MOVT, hi16);
+address[3] = twoWordOp5i6Imm4Reg4EncodedImmSecond(right, hi16);
+address[4] = OP_CMP_reg_T2 | left;
+cacheFlush(address, sizeof(uint16_t) * 5);
+}
+#else
 static void revertJumpTo_movT3(void* instructionStart, RegisterID rd, ARMThumbImmediate imm)
 {
 ASSERT(imm.isValid());
@@ -1277,6 +1291,7 @@
 address[1] = twoWordOp5i6Imm4Reg4EncodedImmSecond(rd, imm);
 cacheFlush(address, sizeof(uint16_t) * 2);
 }
+#endif
 
 ALWAYS_INLINE void mov(RegisterID rd, ARMThumbImmediate imm)
 {
@@ -1882,7 +1897,12 @@
 {
 m_formatter.oneWordOp8Imm8(OP_NOP_T1, 0);
 }
-
+
+void nopw()
+{
+m_formatter.twoWordOp16Op16(OP_NOP_T2a, OP_NOP_T2b);
+}
+
 AssemblerLabel labelIgnoringWatchpoints()
 {
 return m_formatter.label();
@@ -1902,7 +1922,10 @@
 {
 AssemblerLabel result = m_formatter.label();
 while (UNLIKELY(static_castint(result.m_offset)  m_indexOfTailOfLastWatchpoint)) {
-nop();
+if (UNLIKELY(static_castint(result.m_offset) + 4 = m_indexOfTailOfLastWatchpoint))
+nopw();
+else
+nop();
 result = m_formatter.label();
 }
 return result;
@@ -2160,15 +2183,31 @@
 {
 ASSERT(!(bitwise_castuintptr_t(instructionStart)  1));
 ASSERT(!(bitwise_castuintptr_t(to)  1));
+
+#if OS(LINUX)
+if (canBeJumpT4(reinterpret_castuint16_t*(instructionStart), to)) {
+uint16_t* ptr = reinterpret_castuint16_t*(instructionStart) + 2;
+linkJumpT4(ptr, to);
+cacheFlush(ptr - 2, sizeof(uint16_t) 

[webkit-changes] [143346] trunk/Source/JavaScriptCore

2013-02-19 Thread zherczeg
Title: [143346] trunk/Source/_javascript_Core








Revision 143346
Author zherc...@webkit.org
Date 2013-02-19 09:12:41 -0800 (Tue, 19 Feb 2013)


Log Message
revertBranchPtrWithPatch is incorrect on ARM traditional
https://bugs.webkit.org/show_bug.cgi?id=110201

Reviewed by Oliver Hunt.

Revert two instructions back to their original value.

* assembler/ARMAssembler.h:
(JSC::ARMAssembler::revertBranchPtrWithPatch):
(ARMAssembler):
* assembler/MacroAssemblerARM.h:
(JSC::MacroAssemblerARM::branchPtrWithPatch):
(JSC::MacroAssemblerARM::revertJumpReplacementToBranchPtrWithPatch):

Modified Paths

trunk/Source/_javascript_Core/ChangeLog
trunk/Source/_javascript_Core/assembler/ARMAssembler.h
trunk/Source/_javascript_Core/assembler/MacroAssemblerARM.h




Diff

Modified: trunk/Source/_javascript_Core/ChangeLog (143345 => 143346)

--- trunk/Source/_javascript_Core/ChangeLog	2013-02-19 17:09:31 UTC (rev 143345)
+++ trunk/Source/_javascript_Core/ChangeLog	2013-02-19 17:12:41 UTC (rev 143346)
@@ -1,3 +1,19 @@
+2013-02-19  Zoltan Herczeg  zherc...@webkit.org
+
+revertBranchPtrWithPatch is incorrect on ARM traditional
+https://bugs.webkit.org/show_bug.cgi?id=110201
+
+Reviewed by Oliver Hunt.
+
+Revert two instructions back to their original value.
+
+* assembler/ARMAssembler.h:
+(JSC::ARMAssembler::revertBranchPtrWithPatch):
+(ARMAssembler):
+* assembler/MacroAssemblerARM.h:
+(JSC::MacroAssemblerARM::branchPtrWithPatch):
+(JSC::MacroAssemblerARM::revertJumpReplacementToBranchPtrWithPatch):
+
 2013-02-19  Filip Pizlo  fpi...@apple.com
 
 REGRESSION(r143241): It made 27 layout tests crash on 32 bit platforms


Modified: trunk/Source/_javascript_Core/assembler/ARMAssembler.h (143345 => 143346)

--- trunk/Source/_javascript_Core/assembler/ARMAssembler.h	2013-02-19 17:09:31 UTC (rev 143345)
+++ trunk/Source/_javascript_Core/assembler/ARMAssembler.h	2013-02-19 17:12:41 UTC (rev 143346)
@@ -952,6 +952,17 @@
 }
 }
 
+static void revertBranchPtrWithPatch(void* instructionStart, RegisterID rn, ARMWord imm)
+{
+ARMWord* instruction = reinterpret_castARMWord*(instructionStart);
+
+ASSERT((instruction[2]  LdrPcImmediateInstructionMask) == LdrPcImmediateInstruction);
+instruction[0] = toARMWord(AL) | ((instruction[2]  0x0fff0fff) + sizeof(ARMWord)) | RD(ARMRegisters::S1);
+*getLdrImmAddress(instruction) = imm;
+instruction[1] = toARMWord(AL) | CMP | SetConditionalCodes | RN(rn) | RM(ARMRegisters::S1);
+cacheFlush(instruction, 2 * sizeof(ARMWord));
+}
+
 // Address operations
 
 static void* getRelocatedAddress(void* code, AssemblerLabel label)


Modified: trunk/Source/_javascript_Core/assembler/MacroAssemblerARM.h (143345 => 143346)

--- trunk/Source/_javascript_Core/assembler/MacroAssemblerARM.h	2013-02-19 17:09:31 UTC (rev 143345)
+++ trunk/Source/_javascript_Core/assembler/MacroAssemblerARM.h	2013-02-19 17:12:41 UTC (rev 143346)
@@ -980,6 +980,7 @@
 
 Jump branchPtrWithPatch(RelationalCondition cond, RegisterID left, DataLabelPtr dataLabel, TrustedImmPtr initialRightValue = TrustedImmPtr(0))
 {
+ensureSpace(3 * sizeof(ARMWord), 2 * sizeof(ARMWord));
 dataLabel = moveWithPatch(initialRightValue, ARMRegisters::S1);
 Jump jump = branch32(cond, left, ARMRegisters::S1, true);
 return jump;
@@ -988,6 +989,7 @@
 Jump branchPtrWithPatch(RelationalCondition cond, Address left, DataLabelPtr dataLabel, TrustedImmPtr initialRightValue = TrustedImmPtr(0))
 {
 load32(left, ARMRegisters::S1);
+ensureSpace(3 * sizeof(ARMWord), 2 * sizeof(ARMWord));
 dataLabel = moveWithPatch(initialRightValue, ARMRegisters::S0);
 Jump jump = branch32(cond, ARMRegisters::S0, ARMRegisters::S1, true);
 return jump;
@@ -1310,7 +1312,7 @@
 
 static void revertJumpReplacementToBranchPtrWithPatch(CodeLocationLabel instructionStart, RegisterID reg, void* initialValue)
 {
-ARMAssembler::revertJump(instructionStart.dataLocation(), reg, reinterpret_castuintptr_t(initialValue)  0x);
+ARMAssembler::revertBranchPtrWithPatch(instructionStart.dataLocation(), reg, reinterpret_castuintptr_t(initialValue)  0x);
 }
 
 static void revertJumpReplacementToPatchableBranchPtrWithPatch(CodeLocationLabel instructionStart, Address, void* initialValue)






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes


[webkit-changes] [142970] trunk/Source/JavaScriptCore/ChangeLog

2013-02-15 Thread zherczeg
Title: [142970] trunk/Source/_javascript_Core/ChangeLog








Revision 142970
Author zherc...@webkit.org
Date 2013-02-15 01:00:51 -0800 (Fri, 15 Feb 2013)


Log Message
ChangeLog fix for bug 109689.
https://bugs.webkit.org/show_bug.cgi?id=109689

Modified Paths

trunk/Source/_javascript_Core/ChangeLog




Diff

Modified: trunk/Source/_javascript_Core/ChangeLog (142969 => 142970)

--- trunk/Source/_javascript_Core/ChangeLog	2013-02-15 07:45:16 UTC (rev 142969)
+++ trunk/Source/_javascript_Core/ChangeLog	2013-02-15 09:00:51 UTC (rev 142970)
@@ -323,7 +323,7 @@
 replaceWithJump should not decrease the offset by 1 on ARM traditional.
 https://bugs.webkit.org/show_bug.cgi?id=109689
 
-Reviewed by Zoltan Herczeg.
+Reviewed by Oliver Hunt.
 
 * assembler/ARMAssembler.h:
 (JSC::ARMAssembler::replaceWithJump):






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes


[webkit-changes] [142751] trunk/Source/JavaScriptCore

2013-02-13 Thread zherczeg
Title: [142751] trunk/Source/_javascript_Core








Revision 142751
Author zherc...@webkit.org
Date 2013-02-13 08:23:29 -0800 (Wed, 13 Feb 2013)


Log Message
replaceWithJump should not decrease the offset by 1 on ARM traditional.
https://bugs.webkit.org/show_bug.cgi?id=109689

Reviewed by Zoltan Herczeg.

* assembler/ARMAssembler.h:
(JSC::ARMAssembler::replaceWithJump):

Modified Paths

trunk/Source/_javascript_Core/ChangeLog
trunk/Source/_javascript_Core/assembler/ARMAssembler.h




Diff

Modified: trunk/Source/_javascript_Core/ChangeLog (142750 => 142751)

--- trunk/Source/_javascript_Core/ChangeLog	2013-02-13 16:16:45 UTC (rev 142750)
+++ trunk/Source/_javascript_Core/ChangeLog	2013-02-13 16:23:29 UTC (rev 142751)
@@ -1,3 +1,13 @@
+2013-02-13  Zoltan Herczeg  zherc...@webkit.org
+
+replaceWithJump should not decrease the offset by 1 on ARM traditional.
+https://bugs.webkit.org/show_bug.cgi?id=109689
+
+Reviewed by Zoltan Herczeg.
+
+* assembler/ARMAssembler.h:
+(JSC::ARMAssembler::replaceWithJump):
+
 2013-02-12  Joseph Pecoraro  pecor...@apple.com
 
 [iOS] Enable PAGE_VISIBILITY_API


Modified: trunk/Source/_javascript_Core/assembler/ARMAssembler.h (142750 => 142751)

--- trunk/Source/_javascript_Core/assembler/ARMAssembler.h	2013-02-13 16:16:45 UTC (rev 142750)
+++ trunk/Source/_javascript_Core/assembler/ARMAssembler.h	2013-02-13 16:23:29 UTC (rev 142751)
@@ -904,7 +904,7 @@
 
 static void replaceWithJump(void* instructionStart, void* to)
 {
-ARMWord* instruction = reinterpret_castARMWord*(instructionStart) - 1;
+ARMWord* instruction = reinterpret_castARMWord*(instructionStart);
 intptr_t difference = reinterpret_castintptr_t(to) - (reinterpret_castintptr_t(instruction) + DefaultPrefetchOffset * sizeof(ARMWord));
 
 if (!(difference  1)) {






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes


[webkit-changes] [142146] trunk/Source/JavaScriptCore

2013-02-07 Thread zherczeg
Title: [142146] trunk/Source/_javascript_Core








Revision 142146
Author zherc...@webkit.org
Date 2013-02-07 10:04:04 -0800 (Thu, 07 Feb 2013)


Log Message
Invalid code is generated for storing constants with baseindex addressing modes on ARM traditional.
https://bugs.webkit.org/show_bug.cgi?id=109050

Reviewed by Oliver Hunt.

The S! scratch register is reused, but it should contain the constant value.

* assembler/ARMAssembler.cpp:
(JSC::ARMAssembler::baseIndexTransfer32):
(JSC::ARMAssembler::baseIndexTransfer16):

Modified Paths

trunk/Source/_javascript_Core/ChangeLog
trunk/Source/_javascript_Core/assembler/ARMAssembler.cpp




Diff

Modified: trunk/Source/_javascript_Core/ChangeLog (142145 => 142146)

--- trunk/Source/_javascript_Core/ChangeLog	2013-02-07 18:03:17 UTC (rev 142145)
+++ trunk/Source/_javascript_Core/ChangeLog	2013-02-07 18:04:04 UTC (rev 142146)
@@ -1,3 +1,16 @@
+2013-02-07  Zoltan Herczeg  zherc...@webkit.org
+
+Invalid code is generated for storing constants with baseindex addressing modes on ARM traditional.
+https://bugs.webkit.org/show_bug.cgi?id=109050
+
+Reviewed by Oliver Hunt.
+
+The S! scratch register is reused, but it should contain the constant value.
+
+* assembler/ARMAssembler.cpp:
+(JSC::ARMAssembler::baseIndexTransfer32):
+(JSC::ARMAssembler::baseIndexTransfer16):
+
 2013-02-07  Andras Becsi  andras.be...@digia.com
 
 [Qt] Use GNU ar's thin archive format for intermediate static libs


Modified: trunk/Source/_javascript_Core/assembler/ARMAssembler.cpp (142145 => 142146)

--- trunk/Source/_javascript_Core/assembler/ARMAssembler.cpp	2013-02-07 18:03:17 UTC (rev 142145)
+++ trunk/Source/_javascript_Core/assembler/ARMAssembler.cpp	2013-02-07 18:04:04 UTC (rev 142146)
@@ -297,8 +297,15 @@
 return;
 }
 
-add(ARMRegisters::S1, base, op2);
-dataTransfer32(transferType, srcDst, ARMRegisters::S1, offset);
+if (offset = 0xf  offset = -0xf) {
+add(ARMRegisters::S0, base, op2);
+dataTransfer32(transferType, srcDst, ARMRegisters::S0, offset);
+return;
+}
+
+moveImm(offset, ARMRegisters::S0);
+add(ARMRegisters::S0, ARMRegisters::S0, op2);
+dtrUpRegister(transferType, srcDst, base, ARMRegisters::S0);
 }
 
 void ARMAssembler::dataTransfer16(DataTransferTypeB transferType, RegisterID srcDst, RegisterID base, int32_t offset)
@@ -333,8 +340,17 @@
 return;
 }
 
-add(ARMRegisters::S1, base, lsl(index, scale));
-dataTransfer16(transferType, srcDst, ARMRegisters::S1, offset);
+ARMWord op2 = lsl(index, scale);
+
+if (offset = 0x  offset = -0x) {
+add(ARMRegisters::S0, base, op2);
+dataTransfer16(transferType, srcDst, ARMRegisters::S0, offset);
+return;
+}
+
+moveImm(offset, ARMRegisters::S0);
+add(ARMRegisters::S0, ARMRegisters::S0, op2);
+halfDtrUpRegister(transferType, srcDst, base, ARMRegisters::S0);
 }
 
 void ARMAssembler::dataTransferFloat(DataTransferTypeFloat transferType, FPRegisterID srcDst, RegisterID base, int32_t offset)






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
https://lists.webkit.org/mailman/listinfo/webkit-changes


[webkit-changes] [136661] trunk/Source/WebCore

2012-12-05 Thread zherczeg
Title: [136661] trunk/Source/WebCore








Revision 136661
Author zherc...@webkit.org
Date 2012-12-05 02:35:59 -0800 (Wed, 05 Dec 2012)


Log Message
Optimize ColorMatrix filter
https://bugs.webkit.org/show_bug.cgi?id=103728

Reviewed by Dirk Schulze.

ColorMatrix filter recalculates several constants for
every pixel. It is enough to do these calculations only
once during the initialization. Style issues also fixed.

Existing tests cover this feature.

* platform/graphics/filters/FEColorMatrix.cpp:
(WebCore::matrix):
(WebCore::saturateAndHueRotate):
(WebCore::effectType):

Modified Paths

trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/platform/graphics/filters/FEColorMatrix.cpp




Diff

Modified: trunk/Source/WebCore/ChangeLog (136660 => 136661)

--- trunk/Source/WebCore/ChangeLog	2012-12-05 10:24:55 UTC (rev 136660)
+++ trunk/Source/WebCore/ChangeLog	2012-12-05 10:35:59 UTC (rev 136661)
@@ -1,3 +1,21 @@
+2012-12-05  Zoltan Herczeg  zherc...@webkit.org
+
+Optimize ColorMatrix filter
+https://bugs.webkit.org/show_bug.cgi?id=103728
+
+Reviewed by Dirk Schulze.
+
+ColorMatrix filter recalculates several constants for
+every pixel. It is enough to do these calculations only
+once during the initialization. Style issues also fixed.
+
+Existing tests cover this feature.
+
+* platform/graphics/filters/FEColorMatrix.cpp:
+(WebCore::matrix):
+(WebCore::saturateAndHueRotate):
+(WebCore::effectType):
+
 2012-12-05  Carlos Garcia Campos  cgar...@igalia.com
 
 Flex item auto margins in the cross direction should safe center


Modified: trunk/Source/WebCore/platform/graphics/filters/FEColorMatrix.cpp (136660 => 136661)

--- trunk/Source/WebCore/platform/graphics/filters/FEColorMatrix.cpp	2012-12-05 10:24:55 UTC (rev 136660)
+++ trunk/Source/WebCore/platform/graphics/filters/FEColorMatrix.cpp	2012-12-05 10:35:59 UTC (rev 136661)
@@ -75,9 +75,9 @@
 
 inline void matrix(float red, float green, float blue, float alpha, const Vectorfloat values)
 {
-float r = values[0]  * red + values[1]  * green + values[2]  * blue + values[3]  * alpha + values[4] * 255;
-float g = values[5]  * red + values[6]  * green + values[7]  * blue + values[8]  * alpha + values[9] * 255;
-float b = values[10]  * red + values[11]  * green + values[12] * blue + values[13] * alpha + values[14] * 255;
+float r = values[0] * red + values[1] * green + values[2] * blue + values[3] * alpha + values[4] * 255;
+float g = values[5] * red + values[6] * green + values[7] * blue + values[8] * alpha + values[9] * 255;
+float b = values[10] * red + values[11] * green + values[12] * blue + values[13] * alpha + values[14] * 255;
 float a = values[15] * red + values[16] * green + values[17] * blue + values[18] * alpha + values[19] * 255;
 
 red = r;
@@ -86,36 +86,17 @@
 alpha = a;
 }
 
-inline void saturate(float red, float green, float blue, const float s)
+inline void saturateAndHueRotate(float red, float green, float blue, const float* components)
 {
-float r = (0.213 + 0.787 * s) * red + (0.715 - 0.715 * s) * green + (0.072 - 0.072 * s) * blue;
-float g = (0.213 - 0.213 * s) * red + (0.715 + 0.285 * s) * green + (0.072 - 0.072 * s) * blue;
-float b = (0.213 - 0.213 * s) * red + (0.715 - 0.715 * s) * green + (0.072 + 0.928 * s) * blue;
+float r = red * components[0] + green * components[1] + blue * components[2];
+float g = red * components[3] + green * components[4] + blue * components[5];
+float b = red * components[6] + green * components[7] + blue * components[8];
 
 red = r;
 green = g;
 blue = b;
 }
 
-inline void huerotate(float red, float green, float blue, const float hue)
-{
-float cosHue = cos(hue * piDouble / 180); 
-float sinHue = sin(hue * piDouble / 180); 
-float r = red   * (0.213 + cosHue * 0.787 - sinHue * 0.213) +
-   green * (0.715 - cosHue * 0.715 - sinHue * 0.715) +
-   blue  * (0.072 - cosHue * 0.072 + sinHue * 0.928);
-float g = red   * (0.213 - cosHue * 0.213 + sinHue * 0.143) +
-   green * (0.715 + cosHue * 0.285 + sinHue * 0.140) +
-   blue  * (0.072 - cosHue * 0.072 - sinHue * 0.283);
-float b = red   * (0.213 - cosHue * 0.213 - sinHue * 0.787) +
-   green * (0.715 - cosHue * 0.715 + sinHue * 0.715) +
-   blue  * (0.072 + cosHue * 0.928 + sinHue * 0.072);
-
-red = r;
-green = g;
-blue = b;
-}
-
 inline void luminance(float red, float green, float blue, float alpha)
 {
 alpha = 0.2125 * red + 0.7154 * green + 0.0721 * blue;
@@ -124,10 +105,45 @@
 blue = 0;
 }
 
+inline void calculateSaturateComponents(float* components, float value)
+{
+components[0] = (0.213 + 0.787 * value);
+components[1] = (0.715 - 0.715 * value);
+components[2] = (0.072 - 0.072 * value);
+components[3] = (0.213 - 0.213 * value);
+components[4] = (0.715 + 0.285 * 

[webkit-changes] [129224] trunk/LayoutTests

2012-09-21 Thread zherczeg
Title: [129224] trunk/LayoutTests








Revision 129224
Author zherc...@webkit.org
Date 2012-09-21 07:05:36 -0700 (Fri, 21 Sep 2012)


Log Message
[Qt] Unreviewed gardening.

Skip a test because layoutTestController.setBackingScaleFactor() is missing
on Qt. This test timeouts at the moment, which considerably increase the testing time.

* platform/qt/Skipped:

Modified Paths

trunk/LayoutTests/ChangeLog
trunk/LayoutTests/platform/qt/Skipped




Diff

Modified: trunk/LayoutTests/ChangeLog (129223 => 129224)

--- trunk/LayoutTests/ChangeLog	2012-09-21 13:51:35 UTC (rev 129223)
+++ trunk/LayoutTests/ChangeLog	2012-09-21 14:05:36 UTC (rev 129224)
@@ -1,3 +1,12 @@
+2012-09-21  Zoltan Herczeg  zherc...@webkit.org
+
+[Qt] Unreviewed gardening.
+
+Skip a test because layoutTestController.setBackingScaleFactor() is missing
+on Qt. This test timeouts at the moment, which considerably increase the testing time.
+
+* platform/qt/Skipped:
+
 2012-09-21  Christophe Dumez  christophe.du...@intel.com
 
 [WK2][WKTR] EventSender needs to implement scheduleAsynchronousClick


Modified: trunk/LayoutTests/platform/qt/Skipped (129223 => 129224)

--- trunk/LayoutTests/platform/qt/Skipped	2012-09-21 13:51:35 UTC (rev 129223)
+++ trunk/LayoutTests/platform/qt/Skipped	2012-09-21 14:05:36 UTC (rev 129224)
@@ -1231,6 +1231,7 @@
 # Needs layoutTestController.setBackingScaleFactor()
 fast/canvas/2d.backingStorePixelRatio.html
 fast/canvas/2d.imageDataHD.html
+svg/as-image/svg-as-image-canvas.html
 
 # This requires didClearWindowObjectForFrameInIsolatedWorld foo in FrameLoaderClient
 http/tests/security/isolatedWorld/didClearWindowObject.html






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
http://lists.webkit.org/mailman/listinfo/webkit-changes


[webkit-changes] [123849] trunk/LayoutTests

2012-07-27 Thread zherczeg
Title: [123849] trunk/LayoutTests








Revision 123849
Author zherc...@webkit.org
Date 2012-07-27 01:28:51 -0700 (Fri, 27 Jul 2012)


Log Message
[Qt] Unreviewed, rebaseline for tests introduced in r123790

Patch by János Badics jbad...@inf.u-szeged.hu on 2012-07-27
Rubber stamped by Zoltan Herczeg.

* platform/qt/css3/filters/effect-brightness-clamping-expected.png: Added.
* platform/qt/css3/filters/effect-brightness-clamping-expected.txt: Added.
* platform/qt/css3/filters/effect-brightness-clamping-hw-expected.png: Added.
* platform/qt/css3/filters/effect-brightness-clamping-hw-expected.txt: Added.

Modified Paths

trunk/LayoutTests/ChangeLog


Added Paths

trunk/LayoutTests/platform/qt/css3/filters/effect-brightness-clamping-expected.png
trunk/LayoutTests/platform/qt/css3/filters/effect-brightness-clamping-expected.txt
trunk/LayoutTests/platform/qt/css3/filters/effect-brightness-clamping-hw-expected.png
trunk/LayoutTests/platform/qt/css3/filters/effect-brightness-clamping-hw-expected.txt




Diff

Modified: trunk/LayoutTests/ChangeLog (123848 => 123849)

--- trunk/LayoutTests/ChangeLog	2012-07-27 08:25:36 UTC (rev 123848)
+++ trunk/LayoutTests/ChangeLog	2012-07-27 08:28:51 UTC (rev 123849)
@@ -1,3 +1,14 @@
+2012-07-27  János Badics  jbad...@inf.u-szeged.hu
+
+[Qt] Unreviewed, rebaseline for tests introduced in r123790
+
+Rubber stamped by Zoltan Herczeg.
+
+* platform/qt/css3/filters/effect-brightness-clamping-expected.png: Added.
+* platform/qt/css3/filters/effect-brightness-clamping-expected.txt: Added.
+* platform/qt/css3/filters/effect-brightness-clamping-hw-expected.png: Added.
+* platform/qt/css3/filters/effect-brightness-clamping-hw-expected.txt: Added.
+
 2012-07-27  Alec Flett  alecfl...@chromium.org
 
 IndexedDB: inject index keys on cursor/objectstore/index get success handlers


Added: trunk/LayoutTests/platform/qt/css3/filters/effect-brightness-clamping-expected.png

(Binary files differ)

Property changes on: trunk/LayoutTests/platform/qt/css3/filters/effect-brightness-clamping-expected.png
___

Added: svn:mime-type

Added: trunk/LayoutTests/platform/qt/css3/filters/effect-brightness-clamping-expected.txt (0 => 123849)

--- trunk/LayoutTests/platform/qt/css3/filters/effect-brightness-clamping-expected.txt	(rev 0)
+++ trunk/LayoutTests/platform/qt/css3/filters/effect-brightness-clamping-expected.txt	2012-07-27 08:28:51 UTC (rev 123849)
@@ -0,0 +1,52 @@
+layer at (0,0) size 800x600
+  RenderView at (0,0) size 800x600
+layer at (0,0) size 800x600
+  RenderBlock {HTML} at (0,0) size 800x600
+RenderBody {BODY} at (8,8) size 784x584
+  RenderText {#text} at (160,75) size 4x19
+text run at (160,75) width 4:  
+  RenderText {#text} at (324,75) size 4x19
+text run at (324,75) width 4:  
+  RenderText {#text} at (488,75) size 4x19
+text run at (488,75) width 4:  
+  RenderText {#text} at (652,75) size 4x19
+text run at (652,75) width 4:  
+  RenderText {#text} at (160,169) size 4x19
+text run at (160,169) width 4:  
+  RenderText {#text} at (324,169) size 4x19
+text run at (324,169) width 4:  
+  RenderText {#text} at (488,169) size 4x19
+text run at (488,169) width 4:  
+  RenderText {#text} at (652,169) size 4x19
+text run at (652,169) width 4:  
+  RenderText {#text} at (160,263) size 4x19
+text run at (160,263) width 4:  
+  RenderText {#text} at (324,263) size 4x19
+text run at (324,263) width 4:  
+  RenderText {#text} at (488,263) size 4x19
+text run at (488,263) width 4:  
+  RenderText {#text} at (0,0) size 0x0
+layer at (8,8) size 160x90
+  RenderImage {IMG} at (0,0) size 160x90
+layer at (172,8) size 160x90
+  RenderImage {IMG} at (164,0) size 160x90
+layer at (336,8) size 160x90
+  RenderImage {IMG} at (328,0) size 160x90
+layer at (500,8) size 160x90
+  RenderImage {IMG} at (492,0) size 160x90
+layer at (8,102) size 160x90
+  RenderImage {IMG} at (0,94) size 160x90
+layer at (172,102) size 160x90
+  RenderImage {IMG} at (164,94) size 160x90
+layer at (336,102) size 160x90
+  RenderImage {IMG} at (328,94) size 160x90
+layer at (500,102) size 160x90
+  RenderImage {IMG} at (492,94) size 160x90
+layer at (8,196) size 160x90
+  RenderImage {IMG} at (0,188) size 160x90
+layer at (172,196) size 160x90
+  RenderImage {IMG} at (164,188) size 160x90
+layer at (336,196) size 160x90
+  RenderImage {IMG} at (328,188) size 160x90
+layer at (500,196) size 160x90
+  RenderImage {IMG} at (492,188) size 160x90


Added: trunk/LayoutTests/platform/qt/css3/filters/effect-brightness-clamping-hw-expected.png

(Binary files differ)

Property changes on: trunk/LayoutTests/platform/qt/css3/filters/effect-brightness-clamping-hw-expected.png
___

Added: svn:mime-type


[webkit-changes] [123868] trunk/LayoutTests

2012-07-27 Thread zherczeg
Title: [123868] trunk/LayoutTests








Revision 123868
Author zherc...@webkit.org
Date 2012-07-27 05:51:14 -0700 (Fri, 27 Jul 2012)


Log Message
[Qt] Gardening after r123786. It made 3 fast/animation tests fail
https://bugs.webkit.org/show_bug.cgi?id=92490

Patch by János Badics jbad...@inf.u-szeged.hu on 2012-07-27
Rubber stamped by Zoltan Herczeg.

* platform/qt/Skipped:

Modified Paths

trunk/LayoutTests/ChangeLog
trunk/LayoutTests/platform/qt/Skipped




Diff

Modified: trunk/LayoutTests/ChangeLog (123867 => 123868)

--- trunk/LayoutTests/ChangeLog	2012-07-27 12:27:47 UTC (rev 123867)
+++ trunk/LayoutTests/ChangeLog	2012-07-27 12:51:14 UTC (rev 123868)
@@ -1,3 +1,12 @@
+2012-07-27  János Badics  jbad...@inf.u-szeged.hu
+
+[Qt] Gardening after r123786. It made 3 fast/animation tests fail
+https://bugs.webkit.org/show_bug.cgi?id=92490
+
+Rubber stamped by Zoltan Herczeg.
+
+* platform/qt/Skipped:
+
 2012-07-27  Ádám Kallai  ka...@inf.u-szeged.hu
 
 [Qt] platform/qt/editing/pasteboard tests needs update after font change.


Modified: trunk/LayoutTests/platform/qt/Skipped (123867 => 123868)

--- trunk/LayoutTests/platform/qt/Skipped	2012-07-27 12:27:47 UTC (rev 123867)
+++ trunk/LayoutTests/platform/qt/Skipped	2012-07-27 12:51:14 UTC (rev 123868)
@@ -1961,6 +1961,12 @@
 # https://bugs.webkit.org/show_bug.cgi?id=84013
 fast/repaint/line-flow-with-floats-in-regions.html
 
+# [Qt]REGRESSION(r123786) It made 3 fast/animation tests fail
+# https://bugs.webkit.org/show_bug.cgi?id=92490
+fast/animation/request-animation-frame-cancel2.html
+fast/animation/request-animation-frame-detach-element.html
+fast/animation/request-animation-frame-during-modal.html
+
 # [Qt] fast/block/positioning/offsetLeft-offsetTop-multicolumn.html is failing
 # https://bugs.webkit.org/show_bug.cgi?id=86130
 fast/block/positioning/offsetLeft-offsetTop-multicolumn.html






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
http://lists.webkit.org/mailman/listinfo/webkit-changes


[webkit-changes] [123876] trunk/LayoutTests

2012-07-27 Thread zherczeg
Title: [123876] trunk/LayoutTests








Revision 123876
Author zherc...@webkit.org
Date 2012-07-27 07:15:24 -0700 (Fri, 27 Jul 2012)


Log Message
[Qt][WK2] Gardening after r123786. It made 5 fast/animation tests fail
https://bugs.webkit.org/show_bug.cgi?id=92490

Patch by Balazs Ankes b...@inf.u-szeged.hu on 2012-07-27
Rubber stamped by Zoltan Herczeg.

* platform/qt-5.0-wk2/Skipped:

Modified Paths

trunk/LayoutTests/ChangeLog
trunk/LayoutTests/platform/qt-5.0-wk2/Skipped




Diff

Modified: trunk/LayoutTests/ChangeLog (123875 => 123876)

--- trunk/LayoutTests/ChangeLog	2012-07-27 14:14:43 UTC (rev 123875)
+++ trunk/LayoutTests/ChangeLog	2012-07-27 14:15:24 UTC (rev 123876)
@@ -1,3 +1,12 @@
+2012-07-27  Balazs Ankes  b...@inf.u-szeged.hu
+
+[Qt][WK2] Gardening after r123786. It made 5 fast/animation tests fail
+https://bugs.webkit.org/show_bug.cgi?id=92490
+
+Rubber stamped by Zoltan Herczeg.
+
+* platform/qt-5.0-wk2/Skipped:
+
 2012-07-27  Dominik Röttsches  dominik.rottsc...@intel.com
 
 [EFL] Rebaseline after harfbuzz-ng complex font support is enabled


Modified: trunk/LayoutTests/platform/qt-5.0-wk2/Skipped (123875 => 123876)

--- trunk/LayoutTests/platform/qt-5.0-wk2/Skipped	2012-07-27 14:14:43 UTC (rev 123875)
+++ trunk/LayoutTests/platform/qt-5.0-wk2/Skipped	2012-07-27 14:15:24 UTC (rev 123876)
@@ -550,3 +550,11 @@
 svg/carto.net/scrollbar.svg
 svg/carto.net/selectionlist.svg
 tables/mozilla_expected_failures/bugs/bug85016.html
+
+# [Qt]REGRESSION(r123786): It made 3 fast/animation tests fail
+# https://bugs.webkit.org/show_bug.cgi?id=92490
+fast/animation/request-animation-frame-iframe.html
+fast/animation/request-animation-frame-timestamps-advance.html
+fast/animation/request-animation-frame-timestamps.html
+fast/animation/request-animation-frame-within-callback.html
+fast/animation/request-animation-frame.html






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
http://lists.webkit.org/mailman/listinfo/webkit-changes


[webkit-changes] [123735] trunk/Source/JavaScriptCore

2012-07-26 Thread zherczeg
Title: [123735] trunk/Source/_javascript_Core








Revision 123735
Author zherc...@webkit.org
Date 2012-07-26 05:29:10 -0700 (Thu, 26 Jul 2012)


Log Message
[Qt][ARM]ARMAssembler needs buildfix afert r123417
https://bugs.webkit.org/show_bug.cgi?id=92086

Reviewed by Csaba Osztrogonác.

The ARM implementation of this should be optimized code path
is covered by a non-optimized code path. This patch fixes this,
and adds a new function which returns with the offset range.

* assembler/ARMAssembler.h:
(JSC::ARMAssembler::readPointer):
(ARMAssembler):
(JSC::ARMAssembler::repatchInt32):
(JSC::ARMAssembler::repatchCompact):
* assembler/MacroAssemblerARM.h:
(MacroAssemblerARM):
(JSC::MacroAssemblerARM::isCompactPtrAlignedAddressOffset):
(JSC::MacroAssemblerARM::load32WithCompactAddressOffsetPatch):

Modified Paths

trunk/Source/_javascript_Core/ChangeLog
trunk/Source/_javascript_Core/assembler/ARMAssembler.h
trunk/Source/_javascript_Core/assembler/MacroAssemblerARM.h




Diff

Modified: trunk/Source/_javascript_Core/ChangeLog (123734 => 123735)

--- trunk/Source/_javascript_Core/ChangeLog	2012-07-26 12:21:09 UTC (rev 123734)
+++ trunk/Source/_javascript_Core/ChangeLog	2012-07-26 12:29:10 UTC (rev 123735)
@@ -1,3 +1,24 @@
+2012-07-26  Zoltan Herczeg  zherc...@webkit.org
+
+[Qt][ARM]ARMAssembler needs buildfix afert r123417
+https://bugs.webkit.org/show_bug.cgi?id=92086
+
+Reviewed by Csaba Osztrogonác.
+
+The ARM implementation of this should be optimized code path
+is covered by a non-optimized code path. This patch fixes this,
+and adds a new function which returns with the offset range.
+
+* assembler/ARMAssembler.h:
+(JSC::ARMAssembler::readPointer):
+(ARMAssembler):
+(JSC::ARMAssembler::repatchInt32):
+(JSC::ARMAssembler::repatchCompact):
+* assembler/MacroAssemblerARM.h:
+(MacroAssemblerARM):
+(JSC::MacroAssemblerARM::isCompactPtrAlignedAddressOffset):
+(JSC::MacroAssemblerARM::load32WithCompactAddressOffsetPatch):
+
 2012-07-25  Mark Hahnenberg  mhahnenb...@apple.com
 
 Build fix for 32-bit after r123682


Modified: trunk/Source/_javascript_Core/assembler/ARMAssembler.h (123734 => 123735)

--- trunk/Source/_javascript_Core/assembler/ARMAssembler.h	2012-07-26 12:21:09 UTC (rev 123734)
+++ trunk/Source/_javascript_Core/assembler/ARMAssembler.h	2012-07-26 12:29:10 UTC (rev 123735)
@@ -817,11 +817,11 @@
 // Read pointers
 static void* readPointer(void* from)
 {
-ARMWord* insn = reinterpret_castARMWord*(from);
-ARMWord* addr = getLdrImmAddress(insn);
-return *reinterpret_castvoid**(addr);
+ARMWord* instruction = reinterpret_castARMWord*(from);
+ARMWord* address = getLdrImmAddress(instruction);
+return *reinterpret_castvoid**(address);
 }
-
+
 // Patch pointers
 
 static void linkPointer(void* code, AssemblerLabel from, void* to)
@@ -829,14 +829,20 @@
 patchPointerInternal(reinterpret_castintptr_t(code) + from.m_offset, to);
 }
 
-static void repatchInt32(void* from, int32_t to)
+static void repatchInt32(void* where, int32_t to)
 {
-patchPointerInternal(reinterpret_castintptr_t(from), reinterpret_castvoid*(to));
+patchPointerInternal(reinterpret_castintptr_t(where), reinterpret_castvoid*(to));
 }
-
+
 static void repatchCompact(void* where, int32_t value)
 {
-repatchInt32(where, value);
+ARMWord* instruction = reinterpret_castARMWord*(where);
+ASSERT((*instruction  0x0f70) == LoadUint32);
+if (value = 0)
+*instruction = (*instruction  0xff7ff000) | DT_UP | value;
+else
+*instruction = (*instruction  0xff7ff000) | -value;
+cacheFlush(instruction, sizeof(ARMWord));
 }
 
 static void repatchPointer(void* from, void* to)


Modified: trunk/Source/_javascript_Core/assembler/MacroAssemblerARM.h (123734 => 123735)

--- trunk/Source/_javascript_Core/assembler/MacroAssemblerARM.h	2012-07-26 12:21:09 UTC (rev 123734)
+++ trunk/Source/_javascript_Core/assembler/MacroAssemblerARM.h	2012-07-26 12:29:10 UTC (rev 123735)
@@ -41,7 +41,6 @@
 COMPILE_ASSERT(!(DoubleConditionBitSpecial  DoubleConditionMask), DoubleConditionBitSpecial_should_not_interfere_with_ARMAssembler_Condition_codes);
 public:
 typedef ARMRegisters::FPRegisterID FPRegisterID;
-static const int MaximumCompactPtrAlignedAddressOffset = 0x7FFF;
 
 enum RelationalCondition {
 Equal = ARMAssembler::EQ,
@@ -408,11 +407,20 @@
 m_assembler.dtr_ur(ARMAssembler::LoadUint32, dest, address.base, ARMRegisters::S0);
 return dataLabel;
 }
-
+
+static bool isCompactPtrAlignedAddressOffset(ptrdiff_t value)
+{
+return value = -4095  value 

[webkit-changes] [122060] trunk/Source/JavaScriptCore

2012-07-08 Thread zherczeg
Title: [122060] trunk/Source/_javascript_Core








Revision 122060
Author zherc...@webkit.org
Date 2012-07-08 03:00:04 -0700 (Sun, 08 Jul 2012)


Log Message
[Qt][ARM] Implementing missing macro assembler instructions after r121925
https://bugs.webkit.org/show_bug.cgi?id=90657

Reviewed by Csaba Osztrogonác.

Implementing convertibleLoadPtr, replaceWithLoad and
replaceWithAddressComputation.

* assembler/ARMAssembler.h:
(JSC::ARMAssembler::replaceWithLoad):
(ARMAssembler):
(JSC::ARMAssembler::replaceWithAddressComputation):
* assembler/MacroAssemblerARM.h:
(JSC::MacroAssemblerARM::convertibleLoadPtr):
(MacroAssemblerARM):

Modified Paths

trunk/Source/_javascript_Core/ChangeLog
trunk/Source/_javascript_Core/assembler/ARMAssembler.h
trunk/Source/_javascript_Core/assembler/MacroAssemblerARM.h




Diff

Modified: trunk/Source/_javascript_Core/ChangeLog (122059 => 122060)

--- trunk/Source/_javascript_Core/ChangeLog	2012-07-08 09:04:05 UTC (rev 122059)
+++ trunk/Source/_javascript_Core/ChangeLog	2012-07-08 10:00:04 UTC (rev 122060)
@@ -1,3 +1,21 @@
+2012-07-08  Zoltan Herczeg  zherc...@webkit.org
+
+[Qt][ARM] Implementing missing macro assembler instructions after r121925
+https://bugs.webkit.org/show_bug.cgi?id=90657
+
+Reviewed by Csaba Osztrogonác.
+
+Implementing convertibleLoadPtr, replaceWithLoad and
+replaceWithAddressComputation.
+
+* assembler/ARMAssembler.h:
+(JSC::ARMAssembler::replaceWithLoad):
+(ARMAssembler):
+(JSC::ARMAssembler::replaceWithAddressComputation):
+* assembler/MacroAssemblerARM.h:
+(JSC::MacroAssemblerARM::convertibleLoadPtr):
+(MacroAssemblerARM):
+
 2012-07-06  Filip Pizlo  fpi...@apple.com
 
 WebKit Version 5.1.7 (6534.57.2, r121935): Double-click no longer works on OpenStreetMap


Modified: trunk/Source/_javascript_Core/assembler/ARMAssembler.h (122059 => 122060)

--- trunk/Source/_javascript_Core/assembler/ARMAssembler.h	2012-07-08 09:04:05 UTC (rev 122059)
+++ trunk/Source/_javascript_Core/assembler/ARMAssembler.h	2012-07-08 10:00:04 UTC (rev 122060)
@@ -908,6 +908,30 @@
 return sizeof(ARMWord) * 2;
 }
 
+static void replaceWithLoad(void* instructionStart)
+{
+ARMWord* instruction = reinterpret_castARMWord*(instructionStart);
+cacheFlush(instruction, sizeof(ARMWord));
+
+ASSERT((*instruction  0x0ff0) == 0x0280 || (*instruction  0x0ff0) == 0x0590);
+if ((*instruction  0x0ff0) == 0x0280) {
+ *instruction = (*instruction  0xf00f) | 0x0590;
+ cacheFlush(instruction, sizeof(ARMWord));
+}
+}
+
+static void replaceWithAddressComputation(void* instructionStart)
+{
+ARMWord* instruction = reinterpret_castARMWord*(instructionStart);
+cacheFlush(instruction, sizeof(ARMWord));
+
+ASSERT((*instruction  0x0ff0) == 0x0280 || (*instruction  0x0ff0) == 0x0590);
+if ((*instruction  0x0ff0) == 0x0590) {
+ *instruction = (*instruction  0xf00f) | 0x0280;
+ cacheFlush(instruction, sizeof(ARMWord));
+}
+}
+
 // Address operations
 
 static void* getRelocatedAddress(void* code, AssemblerLabel label)


Modified: trunk/Source/_javascript_Core/assembler/MacroAssemblerARM.h (122059 => 122060)

--- trunk/Source/_javascript_Core/assembler/MacroAssemblerARM.h	2012-07-08 09:04:05 UTC (rev 122059)
+++ trunk/Source/_javascript_Core/assembler/MacroAssemblerARM.h	2012-07-08 10:00:04 UTC (rev 122060)
@@ -386,6 +386,14 @@
 load16(address, dest);
 }
 
+ConvertibleLoadLabel convertibleLoadPtr(Address address, RegisterID dest)
+{
+ConvertibleLoadLabel result(this);
+ASSERT(address.offset = 0  address.offset = 255);
+m_assembler.dtr_u(ARMAssembler::LoadUint32, dest, address.base, address.offset);
+return result;
+}
+
 DataLabel32 load32WithAddressOffsetPatch(Address address, RegisterID dest)
 {
 DataLabel32 dataLabel(this);






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes


[webkit-changes] [121948] trunk/Source/JavaScriptCore

2012-07-06 Thread zherczeg
Title: [121948] trunk/Source/_javascript_Core








Revision 121948
Author zherc...@webkit.org
Date 2012-07-06 01:42:29 -0700 (Fri, 06 Jul 2012)


Log Message
[Qt][ARM] REGRESSION(r121885): It broke 30 jsc tests, 500+ layout tests
https://bugs.webkit.org/show_bug.cgi?id=90656

Reviewed by Csaba Osztrogonác.

Typo fixes.

* assembler/MacroAssemblerARM.cpp:
(JSC::MacroAssemblerARM::load32WithUnalignedHalfWords):
Rename getOp2Byte() - getOp2Half()
* assembler/MacroAssemblerARMv7.h:
(JSC::MacroAssemblerARMv7::convertibleLoadPtr):
Add a necessary space.
* jit/JITStubs.cpp:
(JSC):
Revert INLINE_ARM_FUNCTION macro.

Modified Paths

trunk/Source/_javascript_Core/ChangeLog
trunk/Source/_javascript_Core/assembler/MacroAssemblerARM.cpp
trunk/Source/_javascript_Core/assembler/MacroAssemblerARMv7.h
trunk/Source/_javascript_Core/jit/JITStubs.cpp




Diff

Modified: trunk/Source/_javascript_Core/ChangeLog (121947 => 121948)

--- trunk/Source/_javascript_Core/ChangeLog	2012-07-06 07:17:22 UTC (rev 121947)
+++ trunk/Source/_javascript_Core/ChangeLog	2012-07-06 08:42:29 UTC (rev 121948)
@@ -1,3 +1,22 @@
+2012-07-06  Zoltan Herczeg  zherc...@webkit.org
+
+[Qt][ARM] REGRESSION(r121885): It broke 30 jsc tests, 500+ layout tests
+https://bugs.webkit.org/show_bug.cgi?id=90656
+
+Reviewed by Csaba Osztrogonác.
+
+Typo fixes.
+
+* assembler/MacroAssemblerARM.cpp:
+(JSC::MacroAssemblerARM::load32WithUnalignedHalfWords):
+Rename getOp2Byte() - getOp2Half()
+* assembler/MacroAssemblerARMv7.h:
+(JSC::MacroAssemblerARMv7::convertibleLoadPtr):
+Add a necessary space.
+* jit/JITStubs.cpp:
+(JSC):
+Revert INLINE_ARM_FUNCTION macro.
+
 2012-07-05  Filip Pizlo  fpi...@apple.com
 
 REGRESSION(r121925): It broke 5 sputnik tests on x86 platforms


Modified: trunk/Source/_javascript_Core/assembler/MacroAssemblerARM.cpp (121947 => 121948)

--- trunk/Source/_javascript_Core/assembler/MacroAssemblerARM.cpp	2012-07-06 07:17:22 UTC (rev 121947)
+++ trunk/Source/_javascript_Core/assembler/MacroAssemblerARM.cpp	2012-07-06 08:42:29 UTC (rev 121948)
@@ -81,8 +81,8 @@
 m_assembler.dtrh_u(ARMAssembler::LoadUint16, ARMRegisters::S0, ARMRegisters::S0, ARMAssembler::getOp2Half(address.offset + 0x2));
 } else if (address.offset  0  address.offset = -0xff) {
 m_assembler.add_r(ARMRegisters::S0, address.base, op2);
-m_assembler.dtrh_d(ARMAssembler::LoadUint16, dest, ARMRegisters::S0, ARMAssembler::getOp2Byte(-address.offset));
-m_assembler.dtrh_d(ARMAssembler::LoadUint16, ARMRegisters::S0, ARMRegisters::S0, ARMAssembler::getOp2Byte(-address.offset - 0x2));
+m_assembler.dtrh_d(ARMAssembler::LoadUint16, dest, ARMRegisters::S0, ARMAssembler::getOp2Half(-address.offset));
+m_assembler.dtrh_d(ARMAssembler::LoadUint16, ARMRegisters::S0, ARMRegisters::S0, ARMAssembler::getOp2Half(-address.offset - 0x2));
 } else {
 m_assembler.moveImm(address.offset, ARMRegisters::S0);
 m_assembler.add_r(ARMRegisters::S0, ARMRegisters::S0, op2);


Modified: trunk/Source/_javascript_Core/assembler/MacroAssemblerARMv7.h (121947 => 121948)

--- trunk/Source/_javascript_Core/assembler/MacroAssemblerARMv7.h	2012-07-06 07:17:22 UTC (rev 121947)
+++ trunk/Source/_javascript_Core/assembler/MacroAssemblerARMv7.h	2012-07-06 08:42:29 UTC (rev 121948)
@@ -604,7 +604,7 @@
 ConvertibleLoadLabel convertibleLoadPtr(Address address, RegisterID dest)
 {
 ConvertibleLoadLabel result(this);
-ASSERT(address.offset =0  address.offset = 255);
+ASSERT(address.offset = 0  address.offset = 255);
 m_assembler.ldrWide8BitImmediate(dest, address.base, address.offset);
 return result;
 }


Modified: trunk/Source/_javascript_Core/jit/JITStubs.cpp (121947 => 121948)

--- trunk/Source/_javascript_Core/jit/JITStubs.cpp	2012-07-06 07:17:22 UTC (rev 121947)
+++ trunk/Source/_javascript_Core/jit/JITStubs.cpp	2012-07-06 08:42:29 UTC (rev 121948)
@@ -1086,7 +1086,8 @@
 .align 2 \n \
 .globl  SYMBOL_STRING(cti_##op) \n \
 HIDE_SYMBOL(cti_##op) \n \
-INLINE_ARM_FUNCTION(cti_##op) \
+.thumb \n \
+.thumb_func  THUMB_FUNC_PARAM(cti_##op) \n \
 SYMBOL_STRING(cti_##op) : \n \
 str lr, [sp, # STRINGIZE_VALUE_OF(THUNK_RETURN_ADDRESS_OFFSET) ] \n \
 bl  SYMBOL_STRING(JITStubThunked_##op) \n \






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes


[webkit-changes] [121887] trunk/Source/JavaScriptCore

2012-07-05 Thread zherczeg
Title: [121887] trunk/Source/_javascript_Core








Revision 121887
Author zherc...@webkit.org
Date 2012-07-05 00:53:51 -0700 (Thu, 05 Jul 2012)


Log Message
[Qt] Unreviewed buildfix after r121886. Typo fix.

* assembler/MacroAssemblerARM.cpp:
(JSC::MacroAssemblerARM::load32WithUnalignedHalfWords):

Modified Paths

trunk/Source/_javascript_Core/ChangeLog
trunk/Source/_javascript_Core/assembler/MacroAssemblerARM.cpp




Diff

Modified: trunk/Source/_javascript_Core/ChangeLog (121886 => 121887)

--- trunk/Source/_javascript_Core/ChangeLog	2012-07-05 07:24:21 UTC (rev 121886)
+++ trunk/Source/_javascript_Core/ChangeLog	2012-07-05 07:53:51 UTC (rev 121887)
@@ -1,5 +1,12 @@
 2012-07-05  Zoltan Herczeg  zherc...@webkit.org
 
+[Qt] Unreviewed buildfix after r121886. Typo fix.
+
+* assembler/MacroAssemblerARM.cpp:
+(JSC::MacroAssemblerARM::load32WithUnalignedHalfWords):
+
+2012-07-05  Zoltan Herczeg  zherc...@webkit.org
+
 Port DFG JIT to traditional ARM
 https://bugs.webkit.org/show_bug.cgi?id=90198
 


Modified: trunk/Source/_javascript_Core/assembler/MacroAssemblerARM.cpp (121886 => 121887)

--- trunk/Source/_javascript_Core/assembler/MacroAssemblerARM.cpp	2012-07-05 07:24:21 UTC (rev 121886)
+++ trunk/Source/_javascript_Core/assembler/MacroAssemblerARM.cpp	2012-07-05 07:53:51 UTC (rev 121887)
@@ -77,18 +77,18 @@
 
 if (address.offset = 0  address.offset + 0x2 = 0xff) {
 m_assembler.add_r(ARMRegisters::S0, address.base, op2);
-m_assembler.dtrh_u(LoadUint16, dest, ARMRegisters::S0, ARMAssembler::getOp2Half(address.offset));
-m_assembler.dtrh_u(LoadUint16, ARMRegisters::S0, ARMRegisters::S0, ARMAssembler::getOp2Half(address.offset + 0x2));
+m_assembler.dtrh_u(ARMAssembler::LoadUint16, dest, ARMRegisters::S0, ARMAssembler::getOp2Half(address.offset));
+m_assembler.dtrh_u(ARMAssembler::LoadUint16, ARMRegisters::S0, ARMRegisters::S0, ARMAssembler::getOp2Half(address.offset + 0x2));
 } else if (address.offset  0  address.offset = -0xff) {
 m_assembler.add_r(ARMRegisters::S0, address.base, op2);
-m_assembler.dtrh_d(LoadUint16, dest, ARMRegisters::S0, ARMAssembler::getOp2Byte(-address.offset));
-m_assembler.dtrh_d(LoadUint16, ARMRegisters::S0, ARMRegisters::S0, ARMAssembler::getOp2Byte(-address.offset - 0x2));
+m_assembler.dtrh_d(ARMAssembler::LoadUint16, dest, ARMRegisters::S0, ARMAssembler::getOp2Byte(-address.offset));
+m_assembler.dtrh_d(ARMAssembler::LoadUint16, ARMRegisters::S0, ARMRegisters::S0, ARMAssembler::getOp2Byte(-address.offset - 0x2));
 } else {
 m_assembler.moveImm(address.offset, ARMRegisters::S0);
 m_assembler.add_r(ARMRegisters::S0, ARMRegisters::S0, op2);
-m_assembler.dtrh_r(true, TYPE_UH, dest, address.base, ARMRegisters::S0);
+m_assembler.dtrh_ur(ARMAssembler::LoadUint16, dest, address.base, ARMRegisters::S0);
 m_assembler.add_r(ARMRegisters::S0, ARMRegisters::S0, ARMAssembler::OP2_IMM | 0x2);
-m_assembler.dtrh_r(true, TYPE_UH, ARMRegisters::S0, address.base, ARMRegisters::S0);
+m_assembler.dtrh_ur(ARMAssembler::LoadUint16, ARMRegisters::S0, address.base, ARMRegisters::S0);
 }
 m_assembler.orr_r(dest, dest, m_assembler.lsl(ARMRegisters::S0, 16));
 }






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes


[webkit-changes] [119584] trunk/Source/WebCore

2012-06-06 Thread zherczeg
Title: [119584] trunk/Source/WebCore








Revision 119584
Author zherc...@webkit.org
Date 2012-06-06 04:34:45 -0700 (Wed, 06 Jun 2012)


Log Message
[Qt] REGRESSION(r118616): It made all tests crash in debug mode
https://bugs.webkit.org/show_bug.cgi?id=87701

Reviewed by Csaba Osztrogonác.

Replace WriteBarrierJSObject to WeakJSObject by introducing a new
QtWeakObjectReference sub class. Rewrite all affected type definitions.

All existing test in debug mode cover this issue.

* bridge/qt/qt_class.cpp:
(JSC::Bindings::QtClass::fallbackObject):
* bridge/qt/qt_instance.cpp:
(QtRuntimeObject):
(JSC::Bindings::QtInstance::removeUnusedMethods):
* bridge/qt/qt_instance.h:
(QtInstance):
(QtWeakObjectReference):
(JSC::Bindings::QtInstance::QtWeakObjectReference::QtWeakObjectReference):
(JSC::Bindings::QtInstance::QtWeakObjectReference::operator=):
(JSC::Bindings::QtInstance::QtWeakObjectReference::get):
* bridge/qt/qt_runtime.cpp:
(JSC::Bindings::QtRuntimeMethodData::finalize):
* bridge/qt/qt_runtime_qt4.cpp:
(JSC::Bindings::QtRuntimeMethodData::finalize):

Modified Paths

trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/bridge/qt/qt_class.cpp
trunk/Source/WebCore/bridge/qt/qt_instance.cpp
trunk/Source/WebCore/bridge/qt/qt_instance.h
trunk/Source/WebCore/bridge/qt/qt_runtime.cpp
trunk/Source/WebCore/bridge/qt/qt_runtime_qt4.cpp




Diff

Modified: trunk/Source/WebCore/ChangeLog (119583 => 119584)

--- trunk/Source/WebCore/ChangeLog	2012-06-06 11:27:56 UTC (rev 119583)
+++ trunk/Source/WebCore/ChangeLog	2012-06-06 11:34:45 UTC (rev 119584)
@@ -1,3 +1,31 @@
+2012-06-06  Zoltan Herczeg  zherc...@webkit.org
+
+[Qt] REGRESSION(r118616): It made all tests crash in debug mode
+https://bugs.webkit.org/show_bug.cgi?id=87701
+
+Reviewed by Csaba Osztrogonác.
+
+Replace WriteBarrierJSObject to WeakJSObject by introducing a new
+QtWeakObjectReference sub class. Rewrite all affected type definitions.
+
+All existing test in debug mode cover this issue.
+
+* bridge/qt/qt_class.cpp:
+(JSC::Bindings::QtClass::fallbackObject):
+* bridge/qt/qt_instance.cpp:
+(QtRuntimeObject):
+(JSC::Bindings::QtInstance::removeUnusedMethods):
+* bridge/qt/qt_instance.h:
+(QtInstance):
+(QtWeakObjectReference):
+(JSC::Bindings::QtInstance::QtWeakObjectReference::QtWeakObjectReference):
+(JSC::Bindings::QtInstance::QtWeakObjectReference::operator=):
+(JSC::Bindings::QtInstance::QtWeakObjectReference::get):
+* bridge/qt/qt_runtime.cpp:
+(JSC::Bindings::QtRuntimeMethodData::finalize):
+* bridge/qt/qt_runtime_qt4.cpp:
+(JSC::Bindings::QtRuntimeMethodData::finalize):
+
 2012-06-06  Eugene Klyuchnikov  eustas@gmail.com
 
 Web Inspector: help/settings screen does not overlay console in split screen mode


Modified: trunk/Source/WebCore/bridge/qt/qt_class.cpp (119583 => 119584)

--- trunk/Source/WebCore/bridge/qt/qt_class.cpp	2012-06-06 11:27:56 UTC (rev 119583)
+++ trunk/Source/WebCore/bridge/qt/qt_class.cpp	2012-06-06 11:34:45 UTC (rev 119584)
@@ -87,7 +87,7 @@
 QMetaMethod m = m_metaObject-method(index);
 if (m.access() != QMetaMethod::Private) {
 QtRuntimeMetaMethod* val = QtRuntimeMetaMethod::create(exec, ustring, static_castQtInstance*(inst), index, normal, false);
-qtinst-m_methods.insert(name, WriteBarrierJSObject(exec-globalData(), qtinst-createRuntimeObject(exec), val));
+qtinst-m_methods.insert(name, val);
 return val;
 }
 }
@@ -110,7 +110,7 @@
 if (normal == m.name()) {
 #endif
 QtRuntimeMetaMethod* val = QtRuntimeMetaMethod::create(exec, ustring, static_castQtInstance*(inst), index, normal, false);
-qtinst-m_methods.insert(name, WriteBarrierJSObject(exec-globalData(), qtinst-createRuntimeObject(exec), val));
+qtinst-m_methods.insert(name, val);
 return val;
 }
 }


Modified: trunk/Source/WebCore/bridge/qt/qt_instance.cpp (119583 => 119584)

--- trunk/Source/WebCore/bridge/qt/qt_instance.cpp	2012-06-06 11:27:56 UTC (rev 119583)
+++ trunk/Source/WebCore/bridge/qt/qt_instance.cpp	2012-06-06 11:34:45 UTC (rev 119584)
@@ -61,15 +61,6 @@
 
 static const ClassInfo s_info;
 
-static void visitChildren(JSCell* cell, SlotVisitor visitor)
-{
-QtRuntimeObject* thisObject = jsCastQtRuntimeObject*(cell);
-RuntimeObject::visitChildren(thisObject, visitor);
-QtInstance* instance = static_castQtInstance*(thisObject-getInternalInstance());
-if (instance)
-instance-visitAggregate(visitor);
-}
-
 static Structure* createStructure(JSGlobalData globalData, JSGlobalObject* globalObject, JSValue prototype)
 {
 return Structure::create(globalData, globalObject, prototype, TypeInfo(ObjectType,  StructureFlags), s_info);
@@ -158,13 +149,13 @@
 

[webkit-changes] [117418] trunk/Source/WebCore

2012-05-17 Thread zherczeg
Title: [117418] trunk/Source/WebCore








Revision 117418
Author zherc...@webkit.org
Date 2012-05-17 02:12:54 -0700 (Thu, 17 May 2012)


Log Message
NEON intrinsic should be used with arithmetic mode of composite filter
https://bugs.webkit.org/show_bug.cgi?id=86622

Reviewed by Nikolas Zimmermann.

Rewrite hand written assembly code to increase portability and readibility
of the code. Remove the unnecessary FECompositeArithmeticNEON.cpp from the
project

Existing tests cover this issue.

* CMakeLists.txt:
* GNUmakefile.list.am:
* Target.pri:
* WebCore.gypi:
* WebCore.vcproj/WebCore.vcproj:
* WebCore.xcodeproj/project.pbxproj:
* platform/graphics/filters/FEComposite.cpp:
(WebCore):
(WebCore::computeArithmeticPixels):
(WebCore::arithmeticSoftware):
(WebCore::FEComposite::platformArithmeticSoftware):
* platform/graphics/filters/FEComposite.h:
* platform/graphics/filters/arm/FECompositeArithmeticNEON.cpp: Removed.
* platform/graphics/filters/arm/FECompositeArithmeticNEON.h:
(WebCore):
(WebCore::FEComposite::computeArithmeticPixelsNeon):
(WebCore::FEComposite::platformArithmeticNeon):

Modified Paths

trunk/Source/WebCore/CMakeLists.txt
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/GNUmakefile.list.am
trunk/Source/WebCore/Target.pri
trunk/Source/WebCore/WebCore.gypi
trunk/Source/WebCore/WebCore.vcproj/WebCore.vcproj
trunk/Source/WebCore/WebCore.xcodeproj/project.pbxproj
trunk/Source/WebCore/platform/graphics/filters/FEComposite.cpp
trunk/Source/WebCore/platform/graphics/filters/FEComposite.h
trunk/Source/WebCore/platform/graphics/filters/arm/FECompositeArithmeticNEON.h


Removed Paths

trunk/Source/WebCore/platform/graphics/filters/arm/FECompositeArithmeticNEON.cpp




Diff

Modified: trunk/Source/WebCore/CMakeLists.txt (117417 => 117418)

--- trunk/Source/WebCore/CMakeLists.txt	2012-05-17 09:03:20 UTC (rev 117417)
+++ trunk/Source/WebCore/CMakeLists.txt	2012-05-17 09:12:54 UTC (rev 117418)
@@ -1213,7 +1213,6 @@
 platform/graphics/filters/SourceAlpha.cpp
 platform/graphics/filters/SourceGraphic.cpp
 
-platform/graphics/filters/arm/FECompositeArithmeticNEON.cpp
 platform/graphics/filters/arm/FECompositeArithmeticNEON.h
 platform/graphics/filters/arm/FEGaussianBlurNEON.cpp
 platform/graphics/filters/arm/FEGaussianBlurNEON.h


Modified: trunk/Source/WebCore/ChangeLog (117417 => 117418)

--- trunk/Source/WebCore/ChangeLog	2012-05-17 09:03:20 UTC (rev 117417)
+++ trunk/Source/WebCore/ChangeLog	2012-05-17 09:12:54 UTC (rev 117418)
@@ -1,3 +1,34 @@
+2012-05-17  Zoltan Herczeg  zherc...@webkit.org
+
+NEON intrinsic should be used with arithmetic mode of composite filter
+https://bugs.webkit.org/show_bug.cgi?id=86622
+
+Reviewed by Nikolas Zimmermann.
+
+Rewrite hand written assembly code to increase portability and readibility
+of the code. Remove the unnecessary FECompositeArithmeticNEON.cpp from the
+project
+
+Existing tests cover this issue.
+
+* CMakeLists.txt:
+* GNUmakefile.list.am:
+* Target.pri:
+* WebCore.gypi:
+* WebCore.vcproj/WebCore.vcproj:
+* WebCore.xcodeproj/project.pbxproj:
+* platform/graphics/filters/FEComposite.cpp:
+(WebCore):
+(WebCore::computeArithmeticPixels):
+(WebCore::arithmeticSoftware):
+(WebCore::FEComposite::platformArithmeticSoftware):
+* platform/graphics/filters/FEComposite.h:
+* platform/graphics/filters/arm/FECompositeArithmeticNEON.cpp: Removed.
+* platform/graphics/filters/arm/FECompositeArithmeticNEON.h:
+(WebCore):
+(WebCore::FEComposite::computeArithmeticPixelsNeon):
+(WebCore::FEComposite::platformArithmeticNeon):
+
 2012-05-17  Takashi Sakamoto  ta...@google.com
 
 showNodePath will be useful for debugging purpose.


Modified: trunk/Source/WebCore/GNUmakefile.list.am (117417 => 117418)

--- trunk/Source/WebCore/GNUmakefile.list.am	2012-05-17 09:03:20 UTC (rev 117417)
+++ trunk/Source/WebCore/GNUmakefile.list.am	2012-05-17 09:12:54 UTC (rev 117418)
@@ -3214,7 +3214,6 @@
 	Source/WebCore/platform/graphics/filters/SourceGraphic.h \
 	Source/WebCore/platform/graphics/filters/SpotLightSource.cpp \
 	Source/WebCore/platform/graphics/filters/SpotLightSource.h \
-	Source/WebCore/platform/graphics/filters/arm/FECompositeArithmeticNEON.cpp \
 	Source/WebCore/platform/graphics/filters/arm/FECompositeArithmeticNEON.h \
 	Source/WebCore/platform/graphics/filters/arm/FEGaussianBlurNEON.cpp \
 	Source/WebCore/platform/graphics/filters/arm/FEGaussianBlurNEON.h \


Modified: trunk/Source/WebCore/Target.pri (117417 => 117418)

--- trunk/Source/WebCore/Target.pri	2012-05-17 09:03:20 UTC (rev 117417)
+++ trunk/Source/WebCore/Target.pri	2012-05-17 09:12:54 UTC (rev 117418)
@@ -3405,7 +3405,6 @@
 platform/graphics/filters/SpotLightSource.cpp \
 platform/graphics/filters/SourceAlpha.cpp \
 platform/graphics/filters/SourceGraphic.cpp \
-

[webkit-changes] [117048] trunk/Source/WebCore

2012-05-15 Thread zherczeg
Title: [117048] trunk/Source/WebCore








Revision 117048
Author zherc...@webkit.org
Date 2012-05-15 03:09:31 -0700 (Tue, 15 May 2012)


Log Message
The ARM NEON optimized filters does not compile on THUMB2
https://bugs.webkit.org/show_bug.cgi?id=86457

Reviewed by Csaba Osztrogonác.

We should disable the compilation on Thumb2 until we fix the code.

* platform/graphics/filters/FEComposite.cpp:
(WebCore::FEComposite::platformArithmeticSoftware):
* platform/graphics/filters/FEGaussianBlur.cpp:
(WebCore::FEGaussianBlur::platformApplyWorker):
(WebCore::FEGaussianBlur::platformApply):
* platform/graphics/filters/FELighting.cpp:
(WebCore::FELighting::platformApply):
* platform/graphics/filters/arm/FECompositeArithmeticNEON.cpp:
* platform/graphics/filters/arm/FECompositeArithmeticNEON.h:
* platform/graphics/filters/arm/FEGaussianBlurNEON.cpp:
* platform/graphics/filters/arm/FEGaussianBlurNEON.h:
* platform/graphics/filters/arm/FELightingNEON.cpp:
* platform/graphics/filters/arm/FELightingNEON.h:

Modified Paths

trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/platform/graphics/filters/FEComposite.cpp
trunk/Source/WebCore/platform/graphics/filters/FEGaussianBlur.cpp
trunk/Source/WebCore/platform/graphics/filters/FELighting.cpp
trunk/Source/WebCore/platform/graphics/filters/arm/FECompositeArithmeticNEON.cpp
trunk/Source/WebCore/platform/graphics/filters/arm/FECompositeArithmeticNEON.h
trunk/Source/WebCore/platform/graphics/filters/arm/FEGaussianBlurNEON.cpp
trunk/Source/WebCore/platform/graphics/filters/arm/FEGaussianBlurNEON.h
trunk/Source/WebCore/platform/graphics/filters/arm/FELightingNEON.cpp
trunk/Source/WebCore/platform/graphics/filters/arm/FELightingNEON.h




Diff

Modified: trunk/Source/WebCore/ChangeLog (117047 => 117048)

--- trunk/Source/WebCore/ChangeLog	2012-05-15 10:04:49 UTC (rev 117047)
+++ trunk/Source/WebCore/ChangeLog	2012-05-15 10:09:31 UTC (rev 117048)
@@ -1,3 +1,26 @@
+2012-05-15  Zoltan Herczeg  zherc...@webkit.org
+
+The ARM NEON optimized filters does not compile on THUMB2
+https://bugs.webkit.org/show_bug.cgi?id=86457
+
+Reviewed by Csaba Osztrogonác.
+
+We should disable the compilation on Thumb2 until we fix the code.
+
+* platform/graphics/filters/FEComposite.cpp:
+(WebCore::FEComposite::platformArithmeticSoftware):
+* platform/graphics/filters/FEGaussianBlur.cpp:
+(WebCore::FEGaussianBlur::platformApplyWorker):
+(WebCore::FEGaussianBlur::platformApply):
+* platform/graphics/filters/FELighting.cpp:
+(WebCore::FELighting::platformApply):
+* platform/graphics/filters/arm/FECompositeArithmeticNEON.cpp:
+* platform/graphics/filters/arm/FECompositeArithmeticNEON.h:
+* platform/graphics/filters/arm/FEGaussianBlurNEON.cpp:
+* platform/graphics/filters/arm/FEGaussianBlurNEON.h:
+* platform/graphics/filters/arm/FELightingNEON.cpp:
+* platform/graphics/filters/arm/FELightingNEON.h:
+
 2012-05-15  MORITA Hajime  morr...@google.com
 
 Kill unnecessary RuntimeEnabledFeatures.h inclusion


Modified: trunk/Source/WebCore/platform/graphics/filters/FEComposite.cpp (117047 => 117048)

--- trunk/Source/WebCore/platform/graphics/filters/FEComposite.cpp	2012-05-15 10:04:49 UTC (rev 117047)
+++ trunk/Source/WebCore/platform/graphics/filters/FEComposite.cpp	2012-05-15 10:09:31 UTC (rev 117048)
@@ -185,7 +185,7 @@
 int length = source-length();
 ASSERT(length == static_castint(destination-length()));
 // The selection here eventually should happen dynamically.
-#if CPU(ARM_NEON)  COMPILER(GCC)
+#if CPU(ARM_NEON)  CPU(ARM_TRADITIONAL)  COMPILER(GCC)
 ASSERT(!(length  0x3));
 float coefficients[4]  = { k1, k2, k3, k4 };
 platformArithmeticNeon(source-data(), destination-data(), length, coefficients);


Modified: trunk/Source/WebCore/platform/graphics/filters/FEGaussianBlur.cpp (117047 => 117048)

--- trunk/Source/WebCore/platform/graphics/filters/FEGaussianBlur.cpp	2012-05-15 10:04:49 UTC (rev 117047)
+++ trunk/Source/WebCore/platform/graphics/filters/FEGaussianBlur.cpp	2012-05-15 10:09:31 UTC (rev 117048)
@@ -142,7 +142,7 @@
 void FEGaussianBlur::platformApplyWorker(PlatformApplyParameters* parameters)
 {
 IntSize paintSize(parameters-width, parameters-height);
-#if CPU(ARM_NEON)  COMPILER(GCC)
+#if CPU(ARM_NEON)  CPU(ARM_TRADITIONAL)  COMPILER(GCC)
 parameters-filter-platformApplyNeon(parameters-srcPixelArray.get(), parameters-dstPixelArray.get(),
 parameters-kernelSizeX, parameters-kernelSizeY, paintSize);
 #else
@@ -221,7 +221,7 @@
 }
 
 // The selection here eventually should happen dynamically on some platforms.
-#if CPU(ARM_NEON)  COMPILER(GCC)
+#if CPU(ARM_NEON)  CPU(ARM_TRADITIONAL)  COMPILER(GCC)
 platformApplyNeon(srcPixelArray, tmpPixelArray, kernelSizeX, kernelSizeY, paintSize);
 #else
 platformApplyGeneric(srcPixelArray, tmpPixelArray, kernelSizeX, kernelSizeY, paintSize);


Modified: 

[webkit-changes] [117058] trunk/Source

2012-05-15 Thread zherczeg
Title: [117058] trunk/Source








Revision 117058
Author zherc...@webkit.org
Date 2012-05-15 06:18:14 -0700 (Tue, 15 May 2012)


Log Message
NEONizing forceValidPreMultipliedPixels
https://bugs.webkit.org/show_bug.cgi?id=86468

Reviewed by Nikolas Zimmermann.

Source/WebCore: 

Optimize forceValidPreMultipliedPixels with ARM-NEON intrinsics.

Existing tests cover this feature.

* platform/graphics/filters/FilterEffect.cpp:
(WebCore::FilterEffect::forceValidPreMultipliedPixels):

Source/WTF: 

Allow to disable all intrinsics with a single macro.

* wtf/Platform.h:

Modified Paths

trunk/Source/WTF/ChangeLog
trunk/Source/WTF/wtf/Platform.h
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/platform/graphics/filters/FilterEffect.cpp




Diff

Modified: trunk/Source/WTF/ChangeLog (117057 => 117058)

--- trunk/Source/WTF/ChangeLog	2012-05-15 13:08:28 UTC (rev 117057)
+++ trunk/Source/WTF/ChangeLog	2012-05-15 13:18:14 UTC (rev 117058)
@@ -1,3 +1,14 @@
+2012-05-15  Zoltan Herczeg  zherc...@webkit.org
+
+NEONizing forceValidPreMultipliedPixels
+https://bugs.webkit.org/show_bug.cgi?id=86468
+
+Reviewed by Nikolas Zimmermann.
+
+Allow to disable all intrinsics with a single macro.
+
+* wtf/Platform.h:
+
 2012-05-14  Andy Estes  aes...@apple.com
 
 Add WTF_USE_APPKIT to differentiate platforms that use AppKit.framework from other Darwin platforms


Modified: trunk/Source/WTF/wtf/Platform.h (117057 => 117058)

--- trunk/Source/WTF/wtf/Platform.h	2012-05-15 13:08:28 UTC (rev 117057)
+++ trunk/Source/WTF/wtf/Platform.h	2012-05-15 13:18:14 UTC (rev 117058)
@@ -293,6 +293,11 @@
 #define WTF_CPU_ARM_NEON 1
 #endif
 
+#if CPU(ARM_NEON)  (!COMPILER(GCC) || GCC_VERSION_AT_LEAST(4, 7, 0))
+// All NEON intrinsics usage can be disabled by this macro.
+#define HAVE_ARM_NEON_INTRINSICS 1
+#endif
+
 #endif /* ARM */
 
 #if CPU(ARM) || CPU(MIPS) || CPU(SH4) || CPU(SPARC)


Modified: trunk/Source/WebCore/ChangeLog (117057 => 117058)

--- trunk/Source/WebCore/ChangeLog	2012-05-15 13:08:28 UTC (rev 117057)
+++ trunk/Source/WebCore/ChangeLog	2012-05-15 13:18:14 UTC (rev 117058)
@@ -1,3 +1,17 @@
+2012-05-15  Zoltan Herczeg  zherc...@webkit.org
+
+NEONizing forceValidPreMultipliedPixels
+https://bugs.webkit.org/show_bug.cgi?id=86468
+
+Reviewed by Nikolas Zimmermann.
+
+Optimize forceValidPreMultipliedPixels with ARM-NEON intrinsics.
+
+Existing tests cover this feature.
+
+* platform/graphics/filters/FilterEffect.cpp:
+(WebCore::FilterEffect::forceValidPreMultipliedPixels):
+
 2012-05-15  Yury Semikhatsky  yu...@chromium.org
 
 Web Inspector: remove unnecessary setTimeout in HeapSnapshotGridNodes.js


Modified: trunk/Source/WebCore/platform/graphics/filters/FilterEffect.cpp (117057 => 117058)

--- trunk/Source/WebCore/platform/graphics/filters/FilterEffect.cpp	2012-05-15 13:08:28 UTC (rev 117057)
+++ trunk/Source/WebCore/platform/graphics/filters/FilterEffect.cpp	2012-05-15 13:18:14 UTC (rev 117058)
@@ -29,6 +29,10 @@
 #include TextStream.h
 #include wtf/Uint8ClampedArray.h
 
+#if HAVE(ARM_NEON_INTRINSICS)
+#include arm_neon.h
+#endif
+
 namespace WebCore {
 
 FilterEffect::FilterEffect(Filter* filter)
@@ -131,6 +135,26 @@
 
 // We must have four bytes per pixel, and complete pixels
 ASSERT(!(pixelArrayLength % 4));
+
+#if HAVE(ARM_NEON_INTRINSICS)
+if (pixelArrayLength = 64) {
+unsigned char* lastPixel = pixelData + (pixelArrayLength  ~0x3f);
+do {
+// Increments pixelData by 64.
+uint8x16x4_t sixteenPixels = vld4q_u8(pixelData);
+sixteenPixels.val[0] = vminq_u8(sixteenPixels.val[0], sixteenPixels.val[3]);
+sixteenPixels.val[1] = vminq_u8(sixteenPixels.val[1], sixteenPixels.val[3]);
+sixteenPixels.val[2] = vminq_u8(sixteenPixels.val[2], sixteenPixels.val[3]);
+vst4q_u8(pixelData, sixteenPixels);
+pixelData += 64;
+} while (pixelData  lastPixel);
+
+pixelArrayLength = 0x3f;
+if (!pixelArrayLength)
+return;
+}
+#endif
+
 int numPixels = pixelArrayLength / 4;
 
 // Iterate over each pixel, checking alpha and adjusting color components if necessary






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes


[webkit-changes] [113417] trunk/LayoutTests

2012-04-06 Thread zherczeg
Title: [113417] trunk/LayoutTests








Revision 113417
Author zherc...@webkit.org
Date 2012-04-06 00:54:15 -0700 (Fri, 06 Apr 2012)


Log Message
[Qt] Unreviewed gardening after r113378. ENABLE(MUTATION_OBSERVERS) is disabled so
fast/mutation/clear-transient-without-delivery.html and
fast/mutation/transient-gc-crash.html have been skipped.

Patch by János Badics jbad...@inf.u-szeged.hu on 2012-04-06
Reviewed by Zoltan Herczeg.

* platform/qt/Skipped:

Modified Paths

trunk/LayoutTests/ChangeLog
trunk/LayoutTests/platform/qt/Skipped




Diff

Modified: trunk/LayoutTests/ChangeLog (113416 => 113417)

--- trunk/LayoutTests/ChangeLog	2012-04-06 07:45:48 UTC (rev 113416)
+++ trunk/LayoutTests/ChangeLog	2012-04-06 07:54:15 UTC (rev 113417)
@@ -1,3 +1,13 @@
+2012-04-06  János Badics  jbad...@inf.u-szeged.hu
+
+[Qt] Unreviewed gardening after r113378. ENABLE(MUTATION_OBSERVERS) is disabled so
+fast/mutation/clear-transient-without-delivery.html and
+fast/mutation/transient-gc-crash.html have been skipped.
+
+Reviewed by Zoltan Herczeg.
+
+* platform/qt/Skipped:
+
 2012-04-05  Martin Robinson  mrobin...@igalia.com
 
 Rebaseline a couple tests for GTK+.


Modified: trunk/LayoutTests/platform/qt/Skipped (113416 => 113417)

--- trunk/LayoutTests/platform/qt/Skipped	2012-04-06 07:45:48 UTC (rev 113416)
+++ trunk/LayoutTests/platform/qt/Skipped	2012-04-06 07:54:15 UTC (rev 113417)
@@ -282,6 +282,7 @@
 
 # ENABLE(MUTATION_OBSERVERS) is disabled
 fast/mutation/callback-arguments.html
+fast/mutation/clear-transient-without-delivery.html
 fast/mutation/create-during-delivery.html
 fast/mutation/cross-document.html
 fast/mutation/database-callback-delivery.html
@@ -298,6 +299,7 @@
 fast/mutation/observe-exceptions.html
 fast/mutation/observe-subtree.html
 fast/mutation/takeRecords.html
+fast/mutation/transient-gc-crash.html
 
 # USE(ZLIB) is disabled.
 http/tests/websocket/tests/hybi/compressed-control-frame.html






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes


[webkit-changes] [108779] trunk/Source/JavaScriptCore

2012-02-24 Thread zherczeg
Title: [108779] trunk/Source/_javascript_Core








Revision 108779
Author zherc...@webkit.org
Date 2012-02-24 05:38:30 -0800 (Fri, 24 Feb 2012)


Log Message
[Qt] Buildfix for Zero out CopiedBlocks on initialization.
https://bugs.webkit.org/show_bug.cgi?id=79199

Ruber stamped by Csaba Osztrogonác.

Temporary fix since the new member wastes a little space on
64 bit systems. Although it is harmless, it is only needed
for 32 bit systems.

* heap/CopiedBlock.h:
(CopiedBlock):

Modified Paths

trunk/Source/_javascript_Core/ChangeLog
trunk/Source/_javascript_Core/heap/CopiedBlock.h




Diff

Modified: trunk/Source/_javascript_Core/ChangeLog (108778 => 108779)

--- trunk/Source/_javascript_Core/ChangeLog	2012-02-24 13:33:20 UTC (rev 108778)
+++ trunk/Source/_javascript_Core/ChangeLog	2012-02-24 13:38:30 UTC (rev 108779)
@@ -1,3 +1,17 @@
+2012-02-24  Zoltan Herczeg  zherc...@webkit.org
+
+[Qt] Buildfix for Zero out CopiedBlocks on initialization.
+https://bugs.webkit.org/show_bug.cgi?id=79199
+
+Ruber stamped by Csaba Osztrogonác.
+
+Temporary fix since the new member wastes a little space on
+64 bit systems. Although it is harmless, it is only needed
+for 32 bit systems.
+
+* heap/CopiedBlock.h:
+(CopiedBlock):
+
 2012-02-24  Han Hojong  hojong@samsung.com
 
 Remove useless jump instructions for short circuit


Modified: trunk/Source/_javascript_Core/heap/CopiedBlock.h (108778 => 108779)

--- trunk/Source/_javascript_Core/heap/CopiedBlock.h	2012-02-24 13:33:20 UTC (rev 108778)
+++ trunk/Source/_javascript_Core/heap/CopiedBlock.h	2012-02-24 13:38:30 UTC (rev 108779)
@@ -58,6 +58,7 @@
 void* m_offset;
 uintptr_t m_isPinned;
 uintptr_t m_padding;
+uintptr_t m_dummy;
 char m_payload[1];
 };
 






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes


[webkit-changes] [108466] trunk/Source/WebCore

2012-02-22 Thread zherczeg
Title: [108466] trunk/Source/WebCore








Revision 108466
Author zherc...@webkit.org
Date 2012-02-22 01:56:58 -0800 (Wed, 22 Feb 2012)


Log Message
[Qt] clipToImageBuffer assertion fix in debug mode on Qt.

Rubber stamped by Csaba Osztrogonác.

* platform/graphics/qt/GraphicsContextQt.cpp:
(WebCore::GraphicsContext::restorePlatformState):

Modified Paths

trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/platform/graphics/qt/GraphicsContextQt.cpp




Diff

Modified: trunk/Source/WebCore/ChangeLog (108465 => 108466)

--- trunk/Source/WebCore/ChangeLog	2012-02-22 09:45:55 UTC (rev 108465)
+++ trunk/Source/WebCore/ChangeLog	2012-02-22 09:56:58 UTC (rev 108466)
@@ -1,3 +1,12 @@
+2012-02-22  Zoltan Herczeg  zherc...@webkit.org
+
+[Qt] clipToImageBuffer assertion fix in debug mode on Qt.
+
+Rubber stamped by Csaba Osztrogonác.
+
+* platform/graphics/qt/GraphicsContextQt.cpp:
+(WebCore::GraphicsContext::restorePlatformState):
+
 2012-02-22  Adam Barth  aba...@webkit.org
 
 [GTK] fast/frames/frame-dead-region.html crash


Modified: trunk/Source/WebCore/platform/graphics/qt/GraphicsContextQt.cpp (108465 => 108466)

--- trunk/Source/WebCore/platform/graphics/qt/GraphicsContextQt.cpp	2012-02-22 09:45:55 UTC (rev 108465)
+++ trunk/Source/WebCore/platform/graphics/qt/GraphicsContextQt.cpp	2012-02-22 09:56:58 UTC (rev 108466)
@@ -301,7 +301,7 @@
 {
 if (!m_data-layers.isEmpty()  !m_data-layers.top()-alphaMask.isNull())
 if (!--m_data-layers.top()-saveCounter)
-endTransparencyLayer();
+endPlatformTransparencyLayer();
 
 m_data-p()-restore();
 






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes


[webkit-changes] [108492] trunk/Source/WebCore

2012-02-22 Thread zherczeg
Title: [108492] trunk/Source/WebCore








Revision 108492
Author zherc...@webkit.org
Date 2012-02-22 06:18:33 -0800 (Wed, 22 Feb 2012)


Log Message
Drop clipToImageBuffer from RenderBoxModelObject
https://bugs.webkit.org/show_bug.cgi?id=79225

Reviewed by Nikolas Zimmermann.

-webkit-background-clip: text is a rarely used non-standard
feature uses clipToImageBuffer. It is replaced by
CompositeDestinationIn on a transparent layer. The new
approach has the same speed as the old one.

Existing tests cover this issue.

* rendering/RenderBoxModelObject.cpp:
(WebCore::RenderBoxModelObject::paintFillLayerExtended):

Modified Paths

trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/rendering/RenderBoxModelObject.cpp




Diff

Modified: trunk/Source/WebCore/ChangeLog (108491 => 108492)

--- trunk/Source/WebCore/ChangeLog	2012-02-22 13:56:24 UTC (rev 108491)
+++ trunk/Source/WebCore/ChangeLog	2012-02-22 14:18:33 UTC (rev 108492)
@@ -1,3 +1,20 @@
+2012-02-22  Zoltan Herczeg  zherc...@webkit.org
+
+Drop clipToImageBuffer from RenderBoxModelObject
+https://bugs.webkit.org/show_bug.cgi?id=79225
+
+Reviewed by Nikolas Zimmermann.
+
+-webkit-background-clip: text is a rarely used non-standard
+feature uses clipToImageBuffer. It is replaced by
+CompositeDestinationIn on a transparent layer. The new
+approach has the same speed as the old one.
+
+Existing tests cover this issue.
+
+* rendering/RenderBoxModelObject.cpp:
+(WebCore::RenderBoxModelObject::paintFillLayerExtended):
+
 2012-02-22  Kenneth Rohde Christiansen  kenn...@webkit.org
 
 Merge setVisibleRectTrajectoryVector and adjustVisibleRect to


Modified: trunk/Source/WebCore/rendering/RenderBoxModelObject.cpp (108491 => 108492)

--- trunk/Source/WebCore/rendering/RenderBoxModelObject.cpp	2012-02-22 13:56:24 UTC (rev 108491)
+++ trunk/Source/WebCore/rendering/RenderBoxModelObject.cpp	2012-02-22 14:18:33 UTC (rev 108492)
@@ -738,6 +738,9 @@
 }
 
 GraphicsContextStateSaver backgroundClipStateSaver(*context, false);
+OwnPtrImageBuffer maskImage;
+IntRect maskRect;
+
 if (bgLayer-clip() == PaddingFillBox || bgLayer-clip() == ContentFillBox) {
 // Clip to the padding or content boxes as necessary.
 bool includePadding = bgLayer-clip() == ContentFillBox;
@@ -751,17 +754,17 @@
 // We have to draw our text into a mask that can then be used to clip background drawing.
 // First figure out how big the mask has to be.  It should be no bigger than what we need
 // to actually render, so we should intersect the dirty rect with the border box of the background.
-IntRect maskRect = pixelSnappedIntRect(rect);
+maskRect = pixelSnappedIntRect(rect);
 maskRect.intersect(paintInfo.rect);
-
+
 // Now create the mask.
-OwnPtrImageBuffer maskImage = context-createCompatibleBuffer(maskRect.size());
+maskImage = context-createCompatibleBuffer(maskRect.size());
 if (!maskImage)
 return;
-
+
 GraphicsContext* maskImageContext = maskImage-context();
 maskImageContext-translate(-maskRect.x(), -maskRect.y());
-
+
 // Now add the text to the clip.  We do this by painting using a special paint phase that signals to
 // InlineTextBoxes that they should just add their contents to the clip.
 PaintInfo info(maskImageContext, maskRect, PaintPhaseTextClip, true, 0, paintInfo.renderRegion, 0);
@@ -772,12 +775,13 @@
 LayoutSize localOffset = isBox() ? toRenderBox(this)-locationOffset() : LayoutSize();
 paint(info, scrolledPaintRect.location() - localOffset);
 }
-
+
 // The mask has been created.  Now we just need to clip to it.
 backgroundClipStateSaver.save();
-context-clipToImageBuffer(maskImage.get(), maskRect);
+context-clip(maskRect);
+context-beginTransparencyLayer(1);
 }
-
+
 // Only fill with a base color (e.g., white) if we're the root document, since iframes/frames with
 // no background in the child document should show the parent's background.
 bool isOpaqueRoot = false;
@@ -855,6 +859,11 @@
 compositeOp, useLowQualityScaling);
 }
 }
+
+if (bgLayer-clip() == TextFillBox) {
+context-drawImageBuffer(maskImage.get(), ColorSpaceDeviceRGB, maskRect, CompositeDestinationIn);
+context-endTransparencyLayer();
+}
 }
 
 static inline LayoutUnit resolveWidthForRatio(LayoutUnit height, const FloatSize intrinsicRatio)






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes


[webkit-changes] [107818] trunk/Source/WebCore

2012-02-15 Thread zherczeg
Title: [107818] trunk/Source/WebCore








Revision 107818
Author zherc...@webkit.org
Date 2012-02-15 07:54:31 -0800 (Wed, 15 Feb 2012)


Log Message
Remove clipToImageBuffer from SourceAlpha and feComposite
https://bugs.webkit.org/show_bug.cgi?id=78355

Reviewed by Nikolas Zimmermann.

The implementation of clipToImageBuffer is inefficient on
non-mac platforms, so we would benefit if remove it.

Existing tests cover this feature.

* platform/graphics/filters/FEComposite.cpp:
(WebCore::FEComposite::platformApplySoftware):
* platform/graphics/filters/SourceAlpha.cpp:
(WebCore::SourceAlpha::platformApplySoftware):

Modified Paths

trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/platform/graphics/filters/FEComposite.cpp
trunk/Source/WebCore/platform/graphics/filters/SourceAlpha.cpp




Diff

Modified: trunk/Source/WebCore/ChangeLog (107817 => 107818)

--- trunk/Source/WebCore/ChangeLog	2012-02-15 15:42:30 UTC (rev 107817)
+++ trunk/Source/WebCore/ChangeLog	2012-02-15 15:54:31 UTC (rev 107818)
@@ -1,3 +1,20 @@
+2012-02-15  Zoltan Herczeg  zherc...@webkit.org
+
+Remove clipToImageBuffer from SourceAlpha and feComposite
+https://bugs.webkit.org/show_bug.cgi?id=78355
+
+Reviewed by Nikolas Zimmermann.
+
+The implementation of clipToImageBuffer is inefficient on
+non-mac platforms, so we would benefit if remove it.
+
+Existing tests cover this feature.
+
+* platform/graphics/filters/FEComposite.cpp:
+(WebCore::FEComposite::platformApplySoftware):
+* platform/graphics/filters/SourceAlpha.cpp:
+(WebCore::SourceAlpha::platformApplySoftware):
+
 2012-02-15  Simon Hausmann  simon.hausm...@nokia.com
 
 [Qt] Replace use of QGLWidget/QGLContext with QOpenGLContext and QSurface for Qt 5


Modified: trunk/Source/WebCore/platform/graphics/filters/FEComposite.cpp (107817 => 107818)

--- trunk/Source/WebCore/platform/graphics/filters/FEComposite.cpp	2012-02-15 15:42:30 UTC (rev 107817)
+++ trunk/Source/WebCore/platform/graphics/filters/FEComposite.cpp	2012-02-15 15:54:31 UTC (rev 107818)
@@ -239,9 +239,19 @@
 filterContext-drawImageBuffer(in-asImageBuffer(), ColorSpaceDeviceRGB, drawingRegionOfInputImage(in-absolutePaintRect()));
 break;
 case FECOMPOSITE_OPERATOR_IN: {
-GraphicsContextStateSaver stateSaver(*filterContext);
-filterContext-clipToImageBuffer(in2-asImageBuffer(), drawingRegionOfInputImage(in2-absolutePaintRect()));
-filterContext-drawImageBuffer(in-asImageBuffer(), ColorSpaceDeviceRGB, drawingRegionOfInputImage(in-absolutePaintRect()));
+// Applies only to the intersected region.
+IntRect destinationRect = in-absolutePaintRect();
+destinationRect.intersect(in2-absolutePaintRect());
+destinationRect.intersect(absolutePaintRect());
+if (destinationRect.isEmpty())
+break;
+IntPoint destinationPoint(destinationRect.x() - absolutePaintRect().x(), destinationRect.y() - absolutePaintRect().y());
+IntRect sourceRect(IntPoint(destinationRect.x() - in-absolutePaintRect().x(),
+destinationRect.y() - in-absolutePaintRect().y()), destinationRect.size());
+IntRect source2Rect(IntPoint(destinationRect.x() - in2-absolutePaintRect().x(),
+ destinationRect.y() - in2-absolutePaintRect().y()), destinationRect.size());
+filterContext-drawImageBuffer(in2-asImageBuffer(), ColorSpaceDeviceRGB, destinationPoint, source2Rect);
+filterContext-drawImageBuffer(in-asImageBuffer(), ColorSpaceDeviceRGB, destinationPoint, sourceRect, CompositeSourceIn);
 break;
 }
 case FECOMPOSITE_OPERATOR_OUT:


Modified: trunk/Source/WebCore/platform/graphics/filters/SourceAlpha.cpp (107817 => 107818)

--- trunk/Source/WebCore/platform/graphics/filters/SourceAlpha.cpp	2012-02-15 15:42:30 UTC (rev 107817)
+++ trunk/Source/WebCore/platform/graphics/filters/SourceAlpha.cpp	2012-02-15 15:54:31 UTC (rev 107818)
@@ -63,9 +63,8 @@
 
 FloatRect imageRect(FloatPoint(), absolutePaintRect().size());
 GraphicsContext* filterContext = resultImage-context();
-GraphicsContextStateSaver stateSaver(*filterContext);
-filterContext-clipToImageBuffer(filter-sourceImage(), imageRect);
 filterContext-fillRect(imageRect, Color::black, ColorSpaceDeviceRGB);
+filterContext-drawImageBuffer(filter-sourceImage(), ColorSpaceDeviceRGB, IntPoint(), CompositeDestinationIn);
 }
 
 void SourceAlpha::dump()






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes


[webkit-changes] [107369] trunk

2012-02-09 Thread zherczeg
Title: [107369] trunk








Revision 107369
Author zherc...@webkit.org
Date 2012-02-09 23:49:35 -0800 (Thu, 09 Feb 2012)


Log Message
Heap-buffer-overflow in WebCore::CSSParser::lex
https://bugs.webkit.org/show_bug.cgi?id=77402

Reviewed by Antti Koivisto.

Source/WebCore: 

Comments should check only a single \0 terminator.

Test: fast/css/parsing-css-comment.html

* css/CSSParser.cpp:
(WebCore::CSSParser::lex):

LayoutTests: 

Test various comment types.

* fast/css/parsing-css-comment-expected.txt: Added.
* fast/css/parsing-css-comment.html: Added.

Modified Paths

trunk/LayoutTests/ChangeLog
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/css/CSSParser.cpp


Added Paths

trunk/LayoutTests/fast/css/parsing-css-comment-expected.txt
trunk/LayoutTests/fast/css/parsing-css-comment.html




Diff

Modified: trunk/LayoutTests/ChangeLog (107368 => 107369)

--- trunk/LayoutTests/ChangeLog	2012-02-10 07:38:15 UTC (rev 107368)
+++ trunk/LayoutTests/ChangeLog	2012-02-10 07:49:35 UTC (rev 107369)
@@ -1,3 +1,15 @@
+2012-02-09  Zoltan Herczeg  zherc...@webkit.org
+
+Heap-buffer-overflow in WebCore::CSSParser::lex
+https://bugs.webkit.org/show_bug.cgi?id=77402
+
+Reviewed by Antti Koivisto.
+
+Test various comment types.
+
+* fast/css/parsing-css-comment-expected.txt: Added.
+* fast/css/parsing-css-comment.html: Added.
+
 2012-02-09  Kenichi Ishibashi  ba...@chromium.org
 
 Add WebSocket extension support


Added: trunk/LayoutTests/fast/css/parsing-css-comment-expected.txt (0 => 107369)

--- trunk/LayoutTests/fast/css/parsing-css-comment-expected.txt	(rev 0)
+++ trunk/LayoutTests/fast/css/parsing-css-comment-expected.txt	2012-02-10 07:49:35 UTC (rev 107369)
@@ -0,0 +1,19 @@
+Test parsing of CSS nth-child tokens.
+
+SUCCESS
+
+Rules from the stylesheet:
+
+#a { color: green; }
+#b { color: green; }
+#e { color: green; }
+#g { color: green; }
+#i { color: green; }
+Expected result:
+
+#a { color: green; }
+#b { color: green; }
+#e { color: green; }
+#g { color: green; }
+#i { color: green; }
+


Added: trunk/LayoutTests/fast/css/parsing-css-comment.html (0 => 107369)

--- trunk/LayoutTests/fast/css/parsing-css-comment.html	(rev 0)
+++ trunk/LayoutTests/fast/css/parsing-css-comment.html	2012-02-10 07:49:35 UTC (rev 107369)
@@ -0,0 +1,75 @@
+head
+style
+#a { color: green; }
+/*
+  Normal multiline comment
+*/
+#b { color: green; }
+/*
+#c { color: green; }
+*/
+
+/* Single line comment */
+/* #d { color: green; } */
+#e { color: green; }
+
+/*/ Comment start with slash */
+/*/ #f { color: green; } */
+
+/* #g is valid: */
+/**/ #g { color: green; } */
+#h { color: green; }
+/**/#i/**/{/**/color/**/:/**/green/**/;/**/}
+
+/* Unterminated comment. Should not crash.
+/style
+
+script
+
+function runTest()
+{
+if (window.layoutTestController)
+layoutTestController.dumpAsText();
+
+var rules = document.styleSheets[0].cssRules;
+var text = ;
+for (var i = 0; i  rules.length; i++) {
+text += rules.item(i).cssText;
+text += \n;
+}
+
+document.getElementById(result).appendChild(document.createTextNode(text));
+
+if (document.getElementById(result).firstChild.data ="" document.getElementById(expected).firstChild.data)
+document.getElementById(message).firstChild.data = ""
+else
+document.getElementById(message).firstChild.data = ""
+}
+
+/script
+
+/head
+
+body _onload_=runTest()
+
+pTest parsing of CSS nth-child tokens./p
+
+p id=messageTEST DID NOT COMPLETE/p
+
+pRules from the stylesheet:/p
+
+pre id=result/pre
+
+pExpected result:/p
+
+pre id=expected#a { color: green; }
+#b { color: green; }
+#e { color: green; }
+#g { color: green; }
+#i { color: green; }
+/pre
+
+script
+
+/script
+/body


Modified: trunk/Source/WebCore/ChangeLog (107368 => 107369)

--- trunk/Source/WebCore/ChangeLog	2012-02-10 07:38:15 UTC (rev 107368)
+++ trunk/Source/WebCore/ChangeLog	2012-02-10 07:49:35 UTC (rev 107369)
@@ -1,3 +1,17 @@
+2012-02-09  Zoltan Herczeg  zherc...@webkit.org
+
+Heap-buffer-overflow in WebCore::CSSParser::lex
+https://bugs.webkit.org/show_bug.cgi?id=77402
+
+Reviewed by Antti Koivisto.
+
+Comments should check only a single \0 terminator.
+
+Test: fast/css/parsing-css-comment.html
+
+* css/CSSParser.cpp:
+(WebCore::CSSParser::lex):
+
 2012-02-09  Kentaro Hara  hara...@chromium.org
 
 Replace [CustomArgumentHandling] with [CallWith=ScriptArguments|CallStack]


Modified: trunk/Source/WebCore/css/CSSParser.cpp (107368 => 107369)

--- trunk/Source/WebCore/css/CSSParser.cpp	2012-02-10 07:38:15 UTC (rev 107368)
+++ trunk/Source/WebCore/css/CSSParser.cpp	2012-02-10 07:49:35 UTC (rev 107369)
@@ -8525,9 +8525,9 @@
 if (*m_currentCharacter == '*') {
 ++m_currentCharacter;
 while (m_currentCharacter[0] != '*' || m_currentCharacter[1] != '/') {
-if 

[webkit-changes] [106227] trunk/Source/WebCore

2012-01-30 Thread zherczeg
Title: [106227] trunk/Source/WebCore








Revision 106227
Author zherc...@webkit.org
Date 2012-01-30 03:17:58 -0800 (Mon, 30 Jan 2012)


Log Message
Custom written CSS lexer
https://bugs.webkit.org/show_bug.cgi?id=70107

Rubber Stamped by Csaba Osztrogonác.

Do not advance pointer at the end of input, just
keep returning with END_TOKEN.

* css/CSSParser.cpp:
(WebCore::CSSParser::lex):

Modified Paths

trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/css/CSSParser.cpp




Diff

Modified: trunk/Source/WebCore/ChangeLog (106226 => 106227)

--- trunk/Source/WebCore/ChangeLog	2012-01-30 10:48:38 UTC (rev 106226)
+++ trunk/Source/WebCore/ChangeLog	2012-01-30 11:17:58 UTC (rev 106227)
@@ -1,3 +1,16 @@
+2012-01-30  Zoltan Herczeg  zherc...@webkit.org
+
+Custom written CSS lexer
+https://bugs.webkit.org/show_bug.cgi?id=70107
+
+Rubber Stamped by Csaba Osztrogonác.
+
+Do not advance pointer at the end of input, just
+keep returning with END_TOKEN.
+
+* css/CSSParser.cpp:
+(WebCore::CSSParser::lex):
+
 2012-01-30  Shinya Kawanaka  shin...@google.com
 
 The query selector for HTMLContentElement should follow the shadow dom spec.


Modified: trunk/Source/WebCore/css/CSSParser.cpp (106226 => 106227)

--- trunk/Source/WebCore/css/CSSParser.cpp	2012-01-30 10:48:38 UTC (rev 106226)
+++ trunk/Source/WebCore/css/CSSParser.cpp	2012-01-30 11:17:58 UTC (rev 106227)
@@ -7446,6 +7446,7 @@
 CharacterDash,
 
 CharacterOther,
+CharacterNull,
 CharacterWhiteSpace,
 CharacterEndMediaQuery,
 CharacterEndNthChild,
@@ -7467,7 +7468,7 @@
 
 // 128 ASCII codes
 static const CharacterType typesOfASCIICharacters[128] = {
-/*   0 - Null   */ CharacterOther,
+/*   0 - Null   */ CharacterNull,
 /*   1 - Start of Heading   */ CharacterOther,
 /*   2 - Start of Text  */ CharacterOther,
 /*   3 - End of Text*/ CharacterOther,
@@ -8427,6 +8428,11 @@
 // m_token is simply the current character.
 break;
 
+case CharacterNull:
+// Do not advance pointer at the end of input.
+--m_currentCharacter;
+break;
+
 case CharacterWhiteSpace:
 m_token = WHITESPACE;
 // Might start with a '\n'.






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes


[webkit-changes] [104824] trunk

2012-01-12 Thread zherczeg
Title: [104824] trunk








Revision 104824
Author zherc...@webkit.org
Date 2012-01-12 06:57:26 -0800 (Thu, 12 Jan 2012)


Log Message
Fix turbulence bug when stitch tiles enabled and rendered in parallel
https://bugs.webkit.org/show_bug.cgi?id=76042

Reviewed by Nikolas Zimmermann.

Source/WebCore: 

The width / height / wrap members are used by all threads
in the same time. The patch makes them local for all threads.

Fixes the layout fails in svg/dynamic-updates/SVGFETurbulence*

* platform/graphics/filters/FETurbulence.cpp:
(WebCore::FETurbulence::noise2D):
(WebCore::Noise::if):
(WebCore::FETurbulence::calculateTurbulenceValueForPoint):
(WebCore::FETurbulence::fillRegion):
* platform/graphics/filters/FETurbulence.h:
(WebCore::FETurbulence::PaintingData::PaintingData):
(WebCore::FETurbulence::StitchData::StitchData):

LayoutTests: 

Update mac expected files.

* platform/mac-snowleopard/svg/dynamic-updates/SVGFETurbulenceElement-dom-stitchTiles-attr-expected.png:
* platform/mac-snowleopard/svg/dynamic-updates/SVGFETurbulenceElement-svgdom-stitchTiles-prop-expected.png:
* platform/mac/svg/dynamic-updates/SVGFETurbulenceElement-dom-stitchTiles-attr-expected.png:
* platform/mac/svg/dynamic-updates/SVGFETurbulenceElement-svgdom-stitchTiles-prop-expected.png:

Modified Paths

trunk/LayoutTests/ChangeLog
trunk/LayoutTests/platform/mac/svg/dynamic-updates/SVGFETurbulenceElement-dom-stitchTiles-attr-expected.png
trunk/LayoutTests/platform/mac/svg/dynamic-updates/SVGFETurbulenceElement-svgdom-stitchTiles-prop-expected.png
trunk/LayoutTests/platform/mac-snowleopard/svg/dynamic-updates/SVGFETurbulenceElement-dom-stitchTiles-attr-expected.png
trunk/LayoutTests/platform/mac-snowleopard/svg/dynamic-updates/SVGFETurbulenceElement-svgdom-stitchTiles-prop-expected.png
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/platform/graphics/filters/FETurbulence.cpp
trunk/Source/WebCore/platform/graphics/filters/FETurbulence.h




Diff

Modified: trunk/LayoutTests/ChangeLog (104823 => 104824)

--- trunk/LayoutTests/ChangeLog	2012-01-12 14:52:57 UTC (rev 104823)
+++ trunk/LayoutTests/ChangeLog	2012-01-12 14:57:26 UTC (rev 104824)
@@ -1,3 +1,17 @@
+2012-01-12  Zoltan Herczeg  zherc...@webkit.org
+
+Fix turbulence bug when stitch tiles enabled and rendered in parallel
+https://bugs.webkit.org/show_bug.cgi?id=76042
+
+Reviewed by Nikolas Zimmermann.
+
+Update mac expected files.
+
+* platform/mac-snowleopard/svg/dynamic-updates/SVGFETurbulenceElement-dom-stitchTiles-attr-expected.png:
+* platform/mac-snowleopard/svg/dynamic-updates/SVGFETurbulenceElement-svgdom-stitchTiles-prop-expected.png:
+* platform/mac/svg/dynamic-updates/SVGFETurbulenceElement-dom-stitchTiles-attr-expected.png:
+* platform/mac/svg/dynamic-updates/SVGFETurbulenceElement-svgdom-stitchTiles-prop-expected.png:
+
 2012-01-12  Hans Wennborg  h...@chromium.org
 
 Speech input: Send text to correct element even if focus has changed


Modified: trunk/LayoutTests/platform/mac/svg/dynamic-updates/SVGFETurbulenceElement-dom-stitchTiles-attr-expected.png

(Binary files differ)


Modified: trunk/LayoutTests/platform/mac/svg/dynamic-updates/SVGFETurbulenceElement-svgdom-stitchTiles-prop-expected.png

(Binary files differ)


Modified: trunk/LayoutTests/platform/mac-snowleopard/svg/dynamic-updates/SVGFETurbulenceElement-dom-stitchTiles-attr-expected.png

(Binary files differ)


Modified: trunk/LayoutTests/platform/mac-snowleopard/svg/dynamic-updates/SVGFETurbulenceElement-svgdom-stitchTiles-prop-expected.png

(Binary files differ)


Modified: trunk/Source/WebCore/ChangeLog (104823 => 104824)

--- trunk/Source/WebCore/ChangeLog	2012-01-12 14:52:57 UTC (rev 104823)
+++ trunk/Source/WebCore/ChangeLog	2012-01-12 14:57:26 UTC (rev 104824)
@@ -1,3 +1,24 @@
+2012-01-12  Zoltan Herczeg  zherc...@webkit.org
+
+Fix turbulence bug when stitch tiles enabled and rendered in parallel
+https://bugs.webkit.org/show_bug.cgi?id=76042
+
+Reviewed by Nikolas Zimmermann.
+
+The width / height / wrap members are used by all threads
+in the same time. The patch makes them local for all threads.
+
+Fixes the layout fails in svg/dynamic-updates/SVGFETurbulence*
+
+* platform/graphics/filters/FETurbulence.cpp:
+(WebCore::FETurbulence::noise2D):
+(WebCore::Noise::if):
+(WebCore::FETurbulence::calculateTurbulenceValueForPoint):
+(WebCore::FETurbulence::fillRegion):
+* platform/graphics/filters/FETurbulence.h:
+(WebCore::FETurbulence::PaintingData::PaintingData):
+(WebCore::FETurbulence::StitchData::StitchData):
+
 2012-01-12  Ilya Tikhonovsky  loi...@chromium.org
 
 Web Inspector: performance: restore 'log 300 messages into console' test.


Modified: trunk/Source/WebCore/platform/graphics/filters/FETurbulence.cpp (104823 => 104824)

--- trunk/Source/WebCore/platform/graphics/filters/FETurbulence.cpp	

[webkit-changes] [104688] trunk/Source/WebCore

2012-01-11 Thread zherczeg
Title: [104688] trunk/Source/WebCore








Revision 104688
Author zherc...@webkit.org
Date 2012-01-11 02:11:55 -0800 (Wed, 11 Jan 2012)


Log Message
Fix more style errors in CSSParser.cpp (followup patch)
https://bugs.webkit.org/show_bug.cgi?id=76038

Reviewed by Nikolas Zimmermann.

The previous was: http://trac.webkit.org/changeset/104576
This patch mostly contains C cast to static_cast rewrites.

* css/CSSParser.cpp:
(WebCore::CSSParser::parseValue):
(WebCore::CSSParser::parseBackgroundColor):
(WebCore::ShadowParseContext::commitLength):
(WebCore::BorderImageSliceParseContext::commitNumber):
(WebCore::BorderImageSliceParseContext::commitBorderImageSlice):
(WebCore::BorderImageQuadParseContext::commitNumber):
(WebCore::BorderImageQuadParseContext::commitBorderImageQuad):
(WebCore::parseDeprecatedGradientPoint):

Modified Paths

trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/css/CSSParser.cpp




Diff

Modified: trunk/Source/WebCore/ChangeLog (104687 => 104688)

--- trunk/Source/WebCore/ChangeLog	2012-01-11 10:05:48 UTC (rev 104687)
+++ trunk/Source/WebCore/ChangeLog	2012-01-11 10:11:55 UTC (rev 104688)
@@ -1,3 +1,23 @@
+2012-01-11  Zoltan Herczeg  zherc...@webkit.org
+
+Fix more style errors in CSSParser.cpp (followup patch)
+https://bugs.webkit.org/show_bug.cgi?id=76038
+
+Reviewed by Nikolas Zimmermann.
+
+The previous was: http://trac.webkit.org/changeset/104576
+This patch mostly contains C cast to static_cast rewrites.
+
+* css/CSSParser.cpp:
+(WebCore::CSSParser::parseValue):
+(WebCore::CSSParser::parseBackgroundColor):
+(WebCore::ShadowParseContext::commitLength):
+(WebCore::BorderImageSliceParseContext::commitNumber):
+(WebCore::BorderImageSliceParseContext::commitBorderImageSlice):
+(WebCore::BorderImageQuadParseContext::commitNumber):
+(WebCore::BorderImageQuadParseContext::commitBorderImageQuad):
+(WebCore::parseDeprecatedGradientPoint):
+
 2012-01-11  Nikolas Zimmermann  nzimmerm...@rim.com
 
 Not reviewed. Fix release builds.


Modified: trunk/Source/WebCore/css/CSSParser.cpp (104687 => 104688)

--- trunk/Source/WebCore/css/CSSParser.cpp	2012-01-11 10:05:48 UTC (rev 104687)
+++ trunk/Source/WebCore/css/CSSParser.cpp	2012-01-11 10:11:55 UTC (rev 104688)
@@ -1037,7 +1037,7 @@
 
 case CSSPropertyClear:// none | left | right | both | inherit
 if (id == CSSValueNone || id == CSSValueLeft
-|| id == CSSValueRight|| id == CSSValueBoth)
+|| id == CSSValueRight || id == CSSValueBoth)
 validPrimitive = true;
 break;
 
@@ -2883,7 +2883,7 @@
 int id = m_valueList-current()-id;
 if (id == CSSValueWebkitText || (id = CSSValueAqua  id = CSSValueWindowtext) || id == CSSValueMenu || id == CSSValueCurrentcolor ||
 (id = CSSValueGrey  id  CSSValueWebkitText  !m_strict))
-   return cssValuePool()-createIdentifierValue(id);
+return cssValuePool()-createIdentifierValue(id);
 return parseColor();
 }
 
@@ -5054,7 +5054,7 @@
 
 void commitLength(CSSParserValue* v)
 {
-RefPtrCSSPrimitiveValue val = m_cssValuePool-createValue(v-fValue, (CSSPrimitiveValue::UnitTypes)v-unit);
+RefPtrCSSPrimitiveValue val = m_cssValuePool-createValue(v-fValue, static_castCSSPrimitiveValue::UnitTypes(v-unit));
 
 if (allowX) {
 x = val.release();
@@ -5512,7 +5512,7 @@
 
 void commitNumber(CSSParserValue* v)
 {
-RefPtrCSSPrimitiveValue val = m_cssValuePool-createValue(v-fValue, (CSSPrimitiveValue::UnitTypes)v-unit);
+RefPtrCSSPrimitiveValue val = m_cssValuePool-createValue(v-fValue, static_castCSSPrimitiveValue::UnitTypes(v-unit));
 if (!m_top)
 m_top = val;
 else if (!m_right)
@@ -5539,16 +5539,16 @@
 // We need to clone and repeat values for any omissions.
 ASSERT(m_top);
 if (!m_right) {
-m_right = m_cssValuePool-createValue(m_top-getDoubleValue(), (CSSPrimitiveValue::UnitTypes)m_top-primitiveType());
-m_bottom = m_cssValuePool-createValue(m_top-getDoubleValue(), (CSSPrimitiveValue::UnitTypes)m_top-primitiveType());
-m_left = m_cssValuePool-createValue(m_top-getDoubleValue(), (CSSPrimitiveValue::UnitTypes)m_top-primitiveType());
+m_right = m_cssValuePool-createValue(m_top-getDoubleValue(), static_castCSSPrimitiveValue::UnitTypes(m_top-primitiveType()));
+m_bottom = m_cssValuePool-createValue(m_top-getDoubleValue(), static_castCSSPrimitiveValue::UnitTypes(m_top-primitiveType()));
+m_left = m_cssValuePool-createValue(m_top-getDoubleValue(), static_castCSSPrimitiveValue::UnitTypes(m_top-primitiveType()));
 }
 if (!m_bottom) {
-m_bottom = m_cssValuePool-createValue(m_top-getDoubleValue(), (CSSPrimitiveValue::UnitTypes)m_top-primitiveType());
-m_left = 

[webkit-changes] [104576] trunk/Source/WebCore

2012-01-10 Thread zherczeg
Title: [104576] trunk/Source/WebCore








Revision 104576
Author zherc...@webkit.org
Date 2012-01-10 07:58:31 -0800 (Tue, 10 Jan 2012)


Log Message
Fix style errors in CSSParser.cpp
https://bugs.webkit.org/show_bug.cgi?id=75854

Reviewed by Nikolas Zimmermann.

Minor style fixes.

* css/CSSParser.cpp:
(WebCore::CSSParser::parseValue):
(WebCore::CSSParser::parseShadow):
(WebCore::isBorderImageRepeatKeyword):
(WebCore::BorderImageSliceParseContext::commitBorderImageSlice):
(WebCore::BorderImageQuadParseContext::commitBorderImageQuad):
(WebCore::filterInfoForName):

Modified Paths

trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/css/CSSParser.cpp




Diff

Modified: trunk/Source/WebCore/ChangeLog (104575 => 104576)

--- trunk/Source/WebCore/ChangeLog	2012-01-10 15:48:40 UTC (rev 104575)
+++ trunk/Source/WebCore/ChangeLog	2012-01-10 15:58:31 UTC (rev 104576)
@@ -1,3 +1,20 @@
+2012-01-10  Zoltan Herczeg  zherc...@webkit.org
+
+Fix style errors in CSSParser.cpp
+https://bugs.webkit.org/show_bug.cgi?id=75854
+
+Reviewed by Nikolas Zimmermann.
+
+Minor style fixes.
+
+* css/CSSParser.cpp:
+(WebCore::CSSParser::parseValue):
+(WebCore::CSSParser::parseShadow):
+(WebCore::isBorderImageRepeatKeyword):
+(WebCore::BorderImageSliceParseContext::commitBorderImageSlice):
+(WebCore::BorderImageQuadParseContext::commitBorderImageQuad):
+(WebCore::filterInfoForName):
+
 2012-01-10  Julien Chaffraix  jchaffr...@webkit.org
 
 REGRESSION (r93614): Safari Reader doesn't repaint correctly when scrolling


Modified: trunk/Source/WebCore/css/CSSParser.cpp (104575 => 104576)

--- trunk/Source/WebCore/css/CSSParser.cpp	2012-01-10 15:48:40 UTC (rev 104575)
+++ trunk/Source/WebCore/css/CSSParser.cpp	2012-01-10 15:58:31 UTC (rev 104576)
@@ -906,28 +906,28 @@
 break;
 
 case CSSPropertyPosition: // static | relative | absolute | fixed | inherit
-if (id == CSSValueStatic ||
- id == CSSValueRelative ||
- id == CSSValueAbsolute ||
- id == CSSValueFixed)
+if (id == CSSValueStatic
+|| id == CSSValueRelative
+|| id == CSSValueAbsolute
+|| id == CSSValueFixed)
 validPrimitive = true;
 break;
 
-case CSSPropertyPageBreakAfter: // auto | always | avoid | left | right | inherit
+case CSSPropertyPageBreakAfter:   // auto | always | avoid | left | right | inherit
 case CSSPropertyPageBreakBefore:
 case CSSPropertyWebkitColumnBreakAfter:
 case CSSPropertyWebkitColumnBreakBefore:
 case CSSPropertyWebkitRegionBreakAfter:
 case CSSPropertyWebkitRegionBreakBefore:
-if (id == CSSValueAuto ||
- id == CSSValueAlways ||
- id == CSSValueAvoid ||
- id == CSSValueLeft ||
- id == CSSValueRight)
+if (id == CSSValueAuto
+|| id == CSSValueAlways
+|| id == CSSValueAvoid
+|| id == CSSValueLeft
+|| id == CSSValueRight)
 validPrimitive = true;
 break;
 
-case CSSPropertyPageBreakInside:// avoid | auto | inherit
+case CSSPropertyPageBreakInside: // avoid | auto | inherit
 case CSSPropertyWebkitColumnBreakInside:
 case CSSPropertyWebkitRegionBreakInside:
 if (id == CSSValueAuto || id == CSSValueAvoid)
@@ -935,8 +935,8 @@
 break;
 
 case CSSPropertyEmptyCells:  // show | hide | inherit
-if (id == CSSValueShow ||
- id == CSSValueHide)
+if (id == CSSValueShow
+|| id == CSSValueHide)
 validPrimitive = true;
 break;
 
@@ -945,11 +945,11 @@
 return parseContent(propId, important);
 
 case CSSPropertyWhiteSpace:  // normal | pre | nowrap | inherit
-if (id == CSSValueNormal ||
-id == CSSValuePre ||
-id == CSSValuePreWrap ||
-id == CSSValuePreLine ||
-id == CSSValueNowrap)
+if (id == CSSValueNormal
+|| id == CSSValuePre
+|| id == CSSValuePreWrap
+|| id == CSSValuePreLine
+|| id == CSSValueNowrap)
 validPrimitive = true;
 break;
 
@@ -1036,8 +1036,8 @@
 break;
 
 case CSSPropertyClear:// none | left | right | both | inherit
-if (id == CSSValueNone || id == CSSValueLeft ||
- id == CSSValueRight|| id == CSSValueBoth)
+if (id == CSSValueNone || id == CSSValueLeft
+|| id == CSSValueRight|| id == CSSValueBoth)
 validPrimitive = true;
 break;
 
@@ -1045,7 +1045,7 @@
 // left | right | center | justify | webkit_left | webkit_right | webkit_center | webkit_match_parent |
 // start | end | string | inherit
 if ((id = CSSValueWebkitAuto  id = CSSValueWebkitMatchParent) || id == CSSValueStart || id == CSSValueEnd
- 

[webkit-changes] [104134] trunk/Source/JavaScriptCore

2012-01-05 Thread zherczeg
Title: [104134] trunk/Source/_javascript_Core








Revision 104134
Author zherc...@webkit.org
Date 2012-01-05 03:06:35 -0800 (Thu, 05 Jan 2012)


Log Message
Unreviewed gardening after r75605.

Rubber stamped by NOBODY Csaba Osztrogonác.

* wtf/Assertions.cpp:

Modified Paths

trunk/Source/_javascript_Core/ChangeLog
trunk/Source/_javascript_Core/wtf/Assertions.cpp




Diff

Modified: trunk/Source/_javascript_Core/ChangeLog (104133 => 104134)

--- trunk/Source/_javascript_Core/ChangeLog	2012-01-05 10:57:16 UTC (rev 104133)
+++ trunk/Source/_javascript_Core/ChangeLog	2012-01-05 11:06:35 UTC (rev 104134)
@@ -1,3 +1,11 @@
+2012-01-05  Zoltan Herczeg  zherc...@webkit.org
+
+Unreviewed gardening after r75605.
+
+Rubber stamped by NOBODY Csaba Osztrogonác.
+
+* wtf/Assertions.cpp:
+
 2012-01-05  Benjamin Poulain  benja...@webkit.org
 
 Improve charactersAreAllASCII() to compare multiple characters at a time


Modified: trunk/Source/_javascript_Core/wtf/Assertions.cpp (104133 => 104134)

--- trunk/Source/_javascript_Core/wtf/Assertions.cpp	2012-01-05 10:57:16 UTC (rev 104133)
+++ trunk/Source/_javascript_Core/wtf/Assertions.cpp	2012-01-05 11:06:35 UTC (rev 104134)
@@ -136,7 +136,7 @@
 vfprintf(stderr, format, args);
 }
 
-#if COMPILER(GCC)
+#if COMPILER(GCC)  GCC_VERSION_AT_LEAST(4, 6, 0)
 #pragma GCC diagnostic push
 #pragma GCC diagnostic ignored -Wformat-nonliteral
 #endif
@@ -169,7 +169,7 @@
 vprintf_stderr_common(formatWithNewline.get(), args);
 }
 
-#if COMPILER(GCC)
+#if COMPILER(GCC)  GCC_VERSION_AT_LEAST(4, 6, 0)
 #pragma GCC diagnostic pop
 #endif
 






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes


[webkit-changes] [104135] trunk/Source/JavaScriptCore

2012-01-05 Thread zherczeg
Title: [104135] trunk/Source/_javascript_Core








Revision 104135
Author zherc...@webkit.org
Date 2012-01-05 03:24:23 -0800 (Thu, 05 Jan 2012)


Log Message
Unreviewed gardening after r104134.

* wtf/Assertions.cpp:

Modified Paths

trunk/Source/_javascript_Core/ChangeLog
trunk/Source/_javascript_Core/wtf/Assertions.cpp




Diff

Modified: trunk/Source/_javascript_Core/ChangeLog (104134 => 104135)

--- trunk/Source/_javascript_Core/ChangeLog	2012-01-05 11:06:35 UTC (rev 104134)
+++ trunk/Source/_javascript_Core/ChangeLog	2012-01-05 11:24:23 UTC (rev 104135)
@@ -1,5 +1,11 @@
 2012-01-05  Zoltan Herczeg  zherc...@webkit.org
 
+Unreviewed gardening after r104134.
+
+* wtf/Assertions.cpp:
+
+2012-01-05  Zoltan Herczeg  zherc...@webkit.org
+
 Unreviewed gardening after r75605.
 
 Rubber stamped by NOBODY Csaba Osztrogonác.


Modified: trunk/Source/_javascript_Core/wtf/Assertions.cpp (104134 => 104135)

--- trunk/Source/_javascript_Core/wtf/Assertions.cpp	2012-01-05 11:06:35 UTC (rev 104134)
+++ trunk/Source/_javascript_Core/wtf/Assertions.cpp	2012-01-05 11:24:23 UTC (rev 104135)
@@ -136,7 +136,7 @@
 vfprintf(stderr, format, args);
 }
 
-#if COMPILER(GCC)  GCC_VERSION_AT_LEAST(4, 6, 0)
+#if COMPILER(CLANG) || (COMPILER(GCC)  GCC_VERSION_AT_LEAST(4, 6, 0))
 #pragma GCC diagnostic push
 #pragma GCC diagnostic ignored -Wformat-nonliteral
 #endif
@@ -169,7 +169,7 @@
 vprintf_stderr_common(formatWithNewline.get(), args);
 }
 
-#if COMPILER(GCC)  GCC_VERSION_AT_LEAST(4, 6, 0)
+#if COMPILER(CLANG) || (COMPILER(GCC)  GCC_VERSION_AT_LEAST(4, 6, 0))
 #pragma GCC diagnostic pop
 #endif
 






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes


[webkit-changes] [102904] trunk/LayoutTests

2011-12-15 Thread zherczeg
Title: [102904] trunk/LayoutTests








Revision 102904
Author zherc...@webkit.org
Date 2011-12-15 01:29:56 -0800 (Thu, 15 Dec 2011)


Log Message
Unreviewed, Qt rebaseline after r102149 and r102846.

* platform/qt/editing/pasteboard/paste-text-013-expected.txt:
* platform/qt/editing/pasteboard/paste-text-014-expected.txt:
* platform/qt/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt: Added.

Modified Paths

trunk/LayoutTests/ChangeLog
trunk/LayoutTests/platform/qt/editing/pasteboard/paste-text-013-expected.txt
trunk/LayoutTests/platform/qt/editing/pasteboard/paste-text-014-expected.txt


Added Paths

trunk/LayoutTests/platform/qt/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt




Diff

Modified: trunk/LayoutTests/ChangeLog (102903 => 102904)

--- trunk/LayoutTests/ChangeLog	2011-12-15 09:23:19 UTC (rev 102903)
+++ trunk/LayoutTests/ChangeLog	2011-12-15 09:29:56 UTC (rev 102904)
@@ -1,3 +1,11 @@
+2011-12-15  Zoltan Herczeg  zherc...@webkit.org
+
+Unreviewed, Qt rebaseline after r102149 and r102846.
+
+* platform/qt/editing/pasteboard/paste-text-013-expected.txt:
+* platform/qt/editing/pasteboard/paste-text-014-expected.txt:
+* platform/qt/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt: Added.
+
 2011-12-15  Hajime Morrita  morr...@chromium.org
 
 Unreviewed, removing incorrect expectaions.


Modified: trunk/LayoutTests/platform/qt/editing/pasteboard/paste-text-013-expected.txt (102903 => 102904)

--- trunk/LayoutTests/platform/qt/editing/pasteboard/paste-text-013-expected.txt	2011-12-15 09:23:19 UTC (rev 102903)
+++ trunk/LayoutTests/platform/qt/editing/pasteboard/paste-text-013-expected.txt	2011-12-15 09:29:56 UTC (rev 102904)
@@ -40,10 +40,9 @@
 RenderText {#text} at (14,14) size 12x30
   text run at (14,14) width 12: x
   RenderBlock {DIV} at (0,251) size 784x136 [border: (2px solid #FF)]
-RenderBlock {DIV} at (14,38) size 756x30
-  RenderBlock {BLOCKQUOTE} at (40,0) size 676x30
-RenderText {#text} at (0,0) size 33x30
-  text run at (0,0) width 33: foo
+RenderBlock {BLOCKQUOTE} at (54,38) size 676x30
+  RenderText {#text} at (0,0) size 33x30
+text run at (0,0) width 33: foo
 RenderBlock {DIV} at (14,92) size 756x30
   RenderBR {BR} at (0,0) size 0x30
 caret: position 0 of child 0 {BR} of child 1 {DIV} of child 8 {DIV} of body


Modified: trunk/LayoutTests/platform/qt/editing/pasteboard/paste-text-014-expected.txt (102903 => 102904)

--- trunk/LayoutTests/platform/qt/editing/pasteboard/paste-text-014-expected.txt	2011-12-15 09:23:19 UTC (rev 102903)
+++ trunk/LayoutTests/platform/qt/editing/pasteboard/paste-text-014-expected.txt	2011-12-15 09:29:56 UTC (rev 102904)
@@ -35,10 +35,9 @@
 RenderText {#text} at (0,0) size 33x30
   text run at (0,0) width 33: foo
   RenderBlock {DIV} at (0,172) size 784x106 [border: (2px solid #FF)]
-RenderBlock {DIV} at (14,38) size 756x30
-  RenderBlock {BLOCKQUOTE} at (40,0) size 676x30
-RenderText {#text} at (0,0) size 33x30
-  text run at (0,0) width 33: foo
+RenderBlock {BLOCKQUOTE} at (54,38) size 676x30
+  RenderText {#text} at (0,0) size 33x30
+text run at (0,0) width 33: foo
   RenderBlock {DIV} at (0,278) size 784x58 [border: (2px solid #FF)]
 RenderText {#text} at (14,14) size 12x30
   text run at (14,14) width 12: x


Added: trunk/LayoutTests/platform/qt/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt (0 => 102904)

--- trunk/LayoutTests/platform/qt/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt	(rev 0)
+++ trunk/LayoutTests/platform/qt/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt	2011-12-15 09:29:56 UTC (rev 102904)
@@ -0,0 +1,279 @@
+Computed style of an element whose parent's 'display' value is 'none':
+background-attachment: scroll
+background-clip: border-box
+background-color: rgb(0, 128, 0)
+background-image: none
+background-origin: padding-box
+background-position: 0% 0%
+background-repeat: repeat
+background-size: auto
+border-bottom-color: rgb(0, 0, 0)
+border-bottom-left-radius: 0px
+border-bottom-right-radius: 0px
+border-bottom-style: none
+border-bottom-width: 0px
+border-collapse: separate
+border-image-outset: 0px
+border-image-repeat: stretch
+border-image-slice: 100%
+border-image-source: none
+border-image-width: 1
+border-left-color: rgb(0, 0, 0)
+border-left-style: none
+border-left-width: 0px
+border-right-color: rgb(0, 0, 0)
+border-right-style: none
+border-right-width: 0px
+border-top-color: rgb(0, 0, 0)
+border-top-left-radius: 0px
+border-top-right-radius: 0px
+border-top-style: solid
+   

[webkit-changes] [102560] trunk

2011-12-11 Thread zherczeg
Title: [102560] trunk








Revision 102560
Author zherc...@webkit.org
Date 2011-12-11 23:44:02 -0800 (Sun, 11 Dec 2011)


Log Message
Add new CSS nth-children parsing tests
https://bugs.webkit.org/show_bug.cgi?id=74178

Source/WebCore: 

Reviewed by Darin Adler.

Test: fast/css/parsing-css-nth-child.html

* css/CSSParser.cpp:
(WebCore::isValidNthToken): Add -n to the possible identifiers.

LayoutTests: 

The test covers several valid and invalid nth-child tokens.

Reviewed by Darin Adler.

* fast/css/parsing-css-nth-child-expected.txt: Added.
* fast/css/parsing-css-nth-child.html: Added.

Modified Paths

trunk/LayoutTests/ChangeLog
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/css/CSSParser.cpp


Added Paths

trunk/LayoutTests/fast/css/parsing-css-nth-child-expected.txt
trunk/LayoutTests/fast/css/parsing-css-nth-child.html




Diff

Modified: trunk/LayoutTests/ChangeLog (102559 => 102560)

--- trunk/LayoutTests/ChangeLog	2011-12-12 07:07:56 UTC (rev 102559)
+++ trunk/LayoutTests/ChangeLog	2011-12-12 07:44:02 UTC (rev 102560)
@@ -1,3 +1,15 @@
+2011-12-11  Zoltan Herczeg  zherc...@webkit.org
+
+Add new CSS nth-children parsing tests
+https://bugs.webkit.org/show_bug.cgi?id=74178
+
+The test covers several valid and invalid nth-child tokens.
+
+Reviewed by Darin Adler.
+
+* fast/css/parsing-css-nth-child-expected.txt: Added.
+* fast/css/parsing-css-nth-child.html: Added.
+
 2011-12-11  Andreas Kling  kl...@webkit.org
 
 WK2/NetscapePlugin: Incorrect mouse event coordinates when frameScaleFactor != 1.


Added: trunk/LayoutTests/fast/css/parsing-css-nth-child-expected.txt (0 => 102560)

--- trunk/LayoutTests/fast/css/parsing-css-nth-child-expected.txt	(rev 0)
+++ trunk/LayoutTests/fast/css/parsing-css-nth-child-expected.txt	2011-12-12 07:44:02 UTC (rev 102560)
@@ -0,0 +1,51 @@
+Test parsing of CSS nth-child tokens.
+
+SUCCESS
+
+Rules from the stylesheet:
+
+#a:nth-child(1n+0) { color: green; }
+#b:nth-child(n+0) { color: green; }
+#c:nth-child(n) { color: green; }
+#d:nth-child(-n+0) { color: green; }
+#e:nth-child(-n) { color: green; }
+#f:nth-child(1N+0) { color: green; }
+#g:nth-child(N+0) { color: green; }
+#h:nth-child(N) { color: green; }
+#i:nth-child(-N+0) { color: green; }
+#j:nth-child(-N) { color: green; }
+#l:nth-child(-1N
+  -
+  123) { color: green; }
+#m:nth-child(N- 123) { color: green; }
+#o:nth-child(23n
+
+ +
+
+123) { color: green; }
+#t:nth-child(+n+3) { color: green; }
+#u:nth-child(+n + 7) { color: green; }
+Expected result:
+
+#a:nth-child(1n+0) { color: green; }
+#b:nth-child(n+0) { color: green; }
+#c:nth-child(n) { color: green; }
+#d:nth-child(-n+0) { color: green; }
+#e:nth-child(-n) { color: green; }
+#f:nth-child(1N+0) { color: green; }
+#g:nth-child(N+0) { color: green; }
+#h:nth-child(N) { color: green; }
+#i:nth-child(-N+0) { color: green; }
+#j:nth-child(-N) { color: green; }
+#l:nth-child(-1N
+  -
+  123) { color: green; }
+#m:nth-child(N- 123) { color: green; }
+#o:nth-child(23n
+
+ +
+
+123) { color: green; }
+#t:nth-child(+n+3) { color: green; }
+#u:nth-child(+n + 7) { color: green; }
+


Added: trunk/LayoutTests/fast/css/parsing-css-nth-child.html (0 => 102560)

--- trunk/LayoutTests/fast/css/parsing-css-nth-child.html	(rev 0)
+++ trunk/LayoutTests/fast/css/parsing-css-nth-child.html	2011-12-12 07:44:02 UTC (rev 102560)
@@ -0,0 +1,100 @@
+head
+style
+#a:nth-child(1n+0) { color: green; }
+#b:nth-child(n+0)  { color: green; }
+#c:nth-child(n){ color: green; }
+#d:nth-child(-n+0) { color: green; }
+#e:nth-child(-n)   { color: green; }
+
+#f:nth-child(1N+0) { color: green; }
+#g:nth-child(N+0)  { color: green; }
+#h:nth-child(N){ color: green; }
+#i:nth-child(-N+0) { color: green; }
+#j:nth-child(-N)   { color: green; }
+
+#k:nth-child(+ 1n) { color: green; }
+#l:nth-child(-1N
+  -
+  123  ) { color: green; }
+#m:nth-child(  N- 123) { color: green; }
+#n:nth-child(  n +12 3) { color: green; }
+#o:nth-child( 23n
+
+ +
+
+123  ) { color: green; }
+#p:nth-child(  12 n ) { color: green; }
+#q:nth-child(+12n-0+1) { color: green; }
+#r:nth-child(+12N -- 1) { color: green; }
+#s:nth-child(+12 N ) { color: green; }
+#t:nth-child(+n+3) { color: green; }
+#u:nth-child( +n + 7 ) { color: green; }
+#v:nth-child(+ n + 7) { color: green; }
+/style
+
+script
+
+/** Changes the result text font size. */
+function runTest()
+{
+if (window.layoutTestController)
+layoutTestController.dumpAsText();
+
+var rules = document.styleSheets[0].cssRules;
+var text = ;
+for (var i = 0; i  rules.length; i++) {
+text += rules.item(i).cssText;
+text += \n;
+}
+
+document.getElementById(result).appendChild(document.createTextNode(text));
+
+if (document.getElementById(result).firstChild.data ="" document.getElementById(expected).firstChild.data)
+document.getElementById(message).firstChild.data = ""
+else
+

[webkit-changes] [99828] trunk/LayoutTests

2011-11-10 Thread zherczeg
Title: [99828] trunk/LayoutTests








Revision 99828
Author zherc...@webkit.org
Date 2011-11-10 02:03:56 -0800 (Thu, 10 Nov 2011)


Log Message
Unreviewed gardening.

* fast/harness/results-expected.txt: Update after r99818.

Modified Paths

trunk/LayoutTests/ChangeLog
trunk/LayoutTests/fast/harness/results-expected.txt




Diff

Modified: trunk/LayoutTests/ChangeLog (99827 => 99828)

--- trunk/LayoutTests/ChangeLog	2011-11-10 09:46:16 UTC (rev 99827)
+++ trunk/LayoutTests/ChangeLog	2011-11-10 10:03:56 UTC (rev 99828)
@@ -1,3 +1,9 @@
+2011-11-10  Zoltan Herczeg  zherc...@webkit.org
+
+Unreviewed gardening.
+
+* fast/harness/results-expected.txt: Update after r99818.
+
 2011-11-10  Yuta Kitamura  yu...@chromium.org
 
 [Chromium] Unreviewed, rebaseline SVG test results changed after r99539. (part 4)


Modified: trunk/LayoutTests/fast/harness/results-expected.txt (99827 => 99828)

--- trunk/LayoutTests/fast/harness/results-expected.txt	2011-11-10 09:46:16 UTC (rev 99827)
+++ trunk/LayoutTests/fast/harness/results-expected.txt	2011-11-10 10:03:56 UTC (rev 99828)
@@ -96,70 +96,84 @@
 TEST-16: PASS
 TEST-16: PASS
 TEST-16: PASS
+TEST-16: PASS
+TEST-16: PASS
+TEST-16: PASS
 TEST-17: PASS
-TEST-17: PASS
-TEST-17: PASS
 TEST-18: PASS
 TEST-18: PASS
 TEST-18: PASS
 TEST-18: PASS
+TEST-18: PASS
+TEST-18: PASS
 TEST-19: PASS
-TEST-19: PASS
 TEST-20: PASS
 TEST-20: PASS
 TEST-20: PASS
-TEST-20: PASS
-TEST-20: PASS
 TEST-21: PASS
 TEST-21: PASS
+TEST-21: PASS
 TEST-22: PASS
 TEST-22: PASS
+TEST-22: PASS
+TEST-22: PASS
 TEST-23: PASS
+TEST-23: PASS
 TEST-24: PASS
 TEST-24: PASS
+TEST-24: PASS
+TEST-24: PASS
+TEST-24: PASS
 TEST-25: PASS
+TEST-25: PASS
 TEST-26: PASS
+TEST-26: PASS
 TEST-27: PASS
 TEST-28: PASS
 TEST-28: PASS
-TEST-28: PASS
 TEST-29: PASS
-TEST-29: PASS
 TEST-30: PASS
-TEST-30: PASS
-TEST-30: PASS
-TEST-30: PASS
-TEST-30: PASS
-TEST-30: PASS
-TEST-30: PASS
-TEST-30: PASS
-TEST-30: PASS
-TEST-30: PASS
-TEST-30: PASS
-TEST-30: PASS
-TEST-30: PASS
-TEST-30: PASS
-TEST-30: PASS
-TEST-30: PASS
 TEST-31: PASS
-TEST-31: PASS
-TEST-31: PASS
-TEST-31: PASS
-TEST-31: PASS
-TEST-31: PASS
-TEST-31: PASS
-TEST-31: PASS
-TEST-31: PASS
-TEST-31: PASS
 TEST-32: PASS
 TEST-32: PASS
 TEST-32: PASS
-TEST-32: PASS
 TEST-33: PASS
 TEST-33: PASS
-TEST-33: PASS
-TEST-33: PASS
-TEST-33: PASS
-TEST-33: PASS
-TEST-33: PASS
-TEST-33: PASS
+TEST-34: PASS
+TEST-34: PASS
+TEST-34: PASS
+TEST-34: PASS
+TEST-34: PASS
+TEST-34: PASS
+TEST-34: PASS
+TEST-34: PASS
+TEST-34: PASS
+TEST-34: PASS
+TEST-34: PASS
+TEST-34: PASS
+TEST-34: PASS
+TEST-34: PASS
+TEST-34: PASS
+TEST-34: PASS
+TEST-35: PASS
+TEST-35: PASS
+TEST-35: PASS
+TEST-35: PASS
+TEST-35: PASS
+TEST-35: PASS
+TEST-35: PASS
+TEST-35: PASS
+TEST-35: PASS
+TEST-35: PASS
+TEST-36: PASS
+TEST-36: PASS
+TEST-36: PASS
+TEST-36: PASS
+TEST-37: PASS
+TEST-37: PASS
+TEST-37: PASS
+TEST-37: PASS
+TEST-37: PASS
+TEST-37: PASS
+TEST-37: PASS
+TEST-37: PASS






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes


[webkit-changes] [99183] trunk/Source/WebCore

2011-11-03 Thread zherczeg
Title: [99183] trunk/Source/WebCore








Revision 99183
Author zherc...@webkit.org
Date 2011-11-03 06:48:11 -0700 (Thu, 03 Nov 2011)


Log Message
Optimizing arithmetic composite filter to ARM-neon SIMD instruction set
https://bugs.webkit.org/show_bug.cgi?id=65522

Reviewed by Dirk Schulze.

This patch continues the work of Felician Marton.

The arithmetic operation of feComposite SVG filter can be greatly enhanced
by the ARM-NEON instruction set since we are able to calculate the pixel data
for all four RGBA channels at the same time.

It won't affect the general behavior and existing tests will cover the changes.

* CMakeLists.txt:
* GNUmakefile.list.am:
* WebCore.gypi:
* WebCore.pro:
* WebCore.vcproj/WebCore.vcproj:
* platform/graphics/filters/FEComposite.cpp:
(WebCore::computeArithmeticPixels):
(WebCore::arithmeticSoftware):
(WebCore::FEComposite::platformArithmeticSoftware):
(WebCore::FEComposite::platformApplySoftware):
* platform/graphics/filters/FEComposite.h:
* platform/graphics/filters/arm/FECompositeArithmeticNEON.cpp: Added.
* platform/graphics/filters/arm/FECompositeArithmeticNEON.h: Added.
(WebCore::FEComposite::platformArithmeticNeon):

Modified Paths

trunk/Source/WebCore/CMakeLists.txt
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/GNUmakefile.list.am
trunk/Source/WebCore/WebCore.gypi
trunk/Source/WebCore/WebCore.pro
trunk/Source/WebCore/WebCore.vcproj/WebCore.vcproj
trunk/Source/WebCore/platform/graphics/filters/FEComposite.cpp
trunk/Source/WebCore/platform/graphics/filters/FEComposite.h


Added Paths

trunk/Source/WebCore/platform/graphics/filters/arm/FECompositeArithmeticNEON.cpp
trunk/Source/WebCore/platform/graphics/filters/arm/FECompositeArithmeticNEON.h




Diff

Modified: trunk/Source/WebCore/CMakeLists.txt (99182 => 99183)

--- trunk/Source/WebCore/CMakeLists.txt	2011-11-03 13:40:03 UTC (rev 99182)
+++ trunk/Source/WebCore/CMakeLists.txt	2011-11-03 13:48:11 UTC (rev 99183)
@@ -1135,6 +1135,8 @@
 platform/graphics/filters/SourceAlpha.cpp
 platform/graphics/filters/SourceGraphic.cpp
 
+platform/graphics/filters/arm/FECompositeArithmeticNEON.cpp
+platform/graphics/filters/arm/FECompositeArithmeticNEON.h
 platform/graphics/filters/arm/FEGaussianBlurNEON.cpp
 platform/graphics/filters/arm/FEGaussianBlurNEON.h
 platform/graphics/filters/arm/FELightingNEON.cpp


Modified: trunk/Source/WebCore/ChangeLog (99182 => 99183)

--- trunk/Source/WebCore/ChangeLog	2011-11-03 13:40:03 UTC (rev 99182)
+++ trunk/Source/WebCore/ChangeLog	2011-11-03 13:48:11 UTC (rev 99183)
@@ -1,3 +1,33 @@
+2011-11-03  Zoltan Herczeg  zherc...@webkit.org
+
+Optimizing arithmetic composite filter to ARM-neon SIMD instruction set
+https://bugs.webkit.org/show_bug.cgi?id=65522
+
+Reviewed by Dirk Schulze.
+
+This patch continues the work of Felician Marton.
+
+The arithmetic operation of feComposite SVG filter can be greatly enhanced
+by the ARM-NEON instruction set since we are able to calculate the pixel data
+for all four RGBA channels at the same time.
+
+It won't affect the general behavior and existing tests will cover the changes.
+
+* CMakeLists.txt:
+* GNUmakefile.list.am:
+* WebCore.gypi:
+* WebCore.pro:
+* WebCore.vcproj/WebCore.vcproj:
+* platform/graphics/filters/FEComposite.cpp:
+(WebCore::computeArithmeticPixels):
+(WebCore::arithmeticSoftware):
+(WebCore::FEComposite::platformArithmeticSoftware):
+(WebCore::FEComposite::platformApplySoftware):
+* platform/graphics/filters/FEComposite.h:
+* platform/graphics/filters/arm/FECompositeArithmeticNEON.cpp: Added.
+* platform/graphics/filters/arm/FECompositeArithmeticNEON.h: Added.
+(WebCore::FEComposite::platformArithmeticNeon):
+
 2011-11-03  Simon Hausmann  simon.hausm...@nokia.com
 
 [Qt] Add pageScaleFactor to the serialized history item


Modified: trunk/Source/WebCore/GNUmakefile.list.am (99182 => 99183)

--- trunk/Source/WebCore/GNUmakefile.list.am	2011-11-03 13:40:03 UTC (rev 99182)
+++ trunk/Source/WebCore/GNUmakefile.list.am	2011-11-03 13:48:11 UTC (rev 99183)
@@ -2553,6 +2553,8 @@
 	Source/WebCore/platform/graphics/filters/SourceGraphic.h \
 	Source/WebCore/platform/graphics/filters/SpotLightSource.cpp \
 	Source/WebCore/platform/graphics/filters/SpotLightSource.h \
+	Source/WebCore/platform/graphics/filters/arm/FECompositeArithmeticNEON.cpp \
+	Source/WebCore/platform/graphics/filters/arm/FECompositeArithmeticNEON.h \
 	Source/WebCore/platform/graphics/filters/arm/FEGaussianBlurNEON.cpp \
 	Source/WebCore/platform/graphics/filters/arm/FEGaussianBlurNEON.h \
 	Source/WebCore/platform/graphics/filters/arm/FELightingNEON.cpp \


Modified: trunk/Source/WebCore/WebCore.gypi (99182 => 99183)

--- trunk/Source/WebCore/WebCore.gypi	2011-11-03 13:40:03 UTC (rev 99182)
+++ trunk/Source/WebCore/WebCore.gypi	2011-11-03 13:48:11 UTC (rev 99183)

[webkit-changes] [99086] trunk

2011-11-02 Thread zherczeg
Title: [99086] trunk








Revision 99086
Author zherc...@webkit.org
Date 2011-11-02 11:53:07 -0700 (Wed, 02 Nov 2011)


Log Message
Match allowed CSS string characters to Firefox and Opera browsers
https://bugs.webkit.org/show_bug.cgi?id=71000

Reviewed by Darin Adler.

Source/WebCore:

Firefox and Opera accepts any character inside a string except
newline and starting quote character. This behaviour matches to
the CSS 2.1 grammar: http://www.w3.org/TR/CSS2/grammar.html
WebKit should follow them.

Test: fast/css/parsing-css-allowed-string-characters.html

* css/tokenizer.flex:

LayoutTests:

Add a new test case which contains various uncommon characters.

* fast/css/parsing-css-allowed-string-characters-expected.txt: Added.
* fast/css/parsing-css-allowed-string-characters.html: Added.

Modified Paths

trunk/LayoutTests/ChangeLog
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/css/tokenizer.flex


Added Paths

trunk/LayoutTests/fast/css/parsing-css-allowed-string-characters-expected.txt
trunk/LayoutTests/fast/css/parsing-css-allowed-string-characters.html




Diff

Modified: trunk/LayoutTests/ChangeLog (99085 => 99086)

--- trunk/LayoutTests/ChangeLog	2011-11-02 18:43:29 UTC (rev 99085)
+++ trunk/LayoutTests/ChangeLog	2011-11-02 18:53:07 UTC (rev 99086)
@@ -1,3 +1,15 @@
+2011-11-02  Zoltan Herczeg  zherc...@webkit.org
+
+Match allowed CSS string characters to Firefox and Opera browsers
+https://bugs.webkit.org/show_bug.cgi?id=71000
+
+Reviewed by Darin Adler.
+
+Add a new test case which contains various uncommon characters.
+
+* fast/css/parsing-css-allowed-string-characters-expected.txt: Added.
+* fast/css/parsing-css-allowed-string-characters.html: Added.
+
 2011-11-02  Devdatta Deshpande  pwj...@motorola.com
 
 Assert in MoveSelectionCommand::doApply


Added: trunk/LayoutTests/fast/css/parsing-css-allowed-string-characters-expected.txt (0 => 99086)

--- trunk/LayoutTests/fast/css/parsing-css-allowed-string-characters-expected.txt	(rev 0)
+++ trunk/LayoutTests/fast/css/parsing-css-allowed-string-characters-expected.txt	2011-11-02 18:53:07 UTC (rev 99086)
@@ -0,0 +1,39 @@
+Test parsing of CSS escapes.
+
+SUCCESS
+
+Rules from the stylesheet:
+
+#a::after { content: '65:A 9:\9'; }
+#b::after { content: '128:€ 159:Ÿ'; }
+#c::after { content: '1:\1'; }
+#d::after { content: '8:\8'; }
+#e::after { content: '9:\9'; }
+#f::after { }
+#g::after { content: '11:\b'; }
+#h::after { }
+#i::after { }
+#j::after { content: '14:\e'; }
+#k::after { content: '20:\14'; }
+#l::after { content: '30:\1e'; }
+#m::after { content: '31:\1f'; }
+#n::after { content: '127:\7f'; }
+#o::after { content: '384:ƀ'; }
+Expected result:
+
+#a::after { content: '65:A 9:\9'; }
+#b::after { content: '128:€ 159:Ÿ'; }
+#c::after { content: '1:\1'; }
+#d::after { content: '8:\8'; }
+#e::after { content: '9:\9'; }
+#f::after { }
+#g::after { content: '11:\b'; }
+#h::after { }
+#i::after { }
+#j::after { content: '14:\e'; }
+#k::after { content: '20:\14'; }
+#l::after { content: '30:\1e'; }
+#m::after { content: '31:\1f'; }
+#n::after { content: '127:\7f'; }
+#o::after { content: '384:ƀ'; }
+


Added: trunk/LayoutTests/fast/css/parsing-css-allowed-string-characters.html (0 => 99086)

--- trunk/LayoutTests/fast/css/parsing-css-allowed-string-characters.html	(rev 0)
+++ trunk/LayoutTests/fast/css/parsing-css-allowed-string-characters.html	2011-11-02 18:53:07 UTC (rev 99086)
@@ -0,0 +1,91 @@
+html
+head
+meta http-equiv=Content-Type content=text/html;charset=UTF-8
+style
+/*
+  === Warning ===
+
+  This file intentionally contains scpecial characters which
+  cannot be displayed by some text editors. Please
+  carefully edit the file.
+*/
+
+#a:after { content:65:A 9:	; }
+#b:after { content:128:€ 159:Ÿ; }
+#c:after { content:1:; }
+#d:after { content:8:; }
+#e:after { content:9:	; }
+#f:after { content:10:
+; } /* Newline. Invalid string */
+#g:after { content:11:; }
+#h:after { content:12:; } /* Newline. Invalid string */
+#i:after { content:13:
+; } /* Newline. Invalid string */
+#j:after { content:14:; }
+#k:after { content:20:; }
+#l:after { content:30:; }
+#m:after { content:31:; }
+#n:after { content:127: }
+#o:after { content:384:ƀ }
+/style
+
+script
+
+function runTest()
+{
+if (window.layoutTestController)
+layoutTestController.dumpAsText();
+
+var rules = document.styleSheets[0].cssRules;
+var text = ;
+for (var i = 0; i  rules.length; i++) {
+text += rules.item(i).cssText;
+text += \n;
+}
+
+document.getElementById(result).appendChild(document.createTextNode(text));
+
+if (document.getElementById(result).firstChild.data ="" document.getElementById(expected).firstChild.data)
+document.getElementById(message).firstChild.data = ""
+else
+document.getElementById(message).firstChild.data = ""
+}
+
+/script
+
+/head
+
+body _onload_=runTest()
+

[webkit-changes] [98744] trunk/LayoutTests

2011-10-28 Thread zherczeg
Title: [98744] trunk/LayoutTests








Revision 98744
Author zherc...@webkit.org
Date 2011-10-28 11:40:14 -0700 (Fri, 28 Oct 2011)


Log Message
Testing allowed identifier chars which are greater than 126
https://bugs.webkit.org/show_bug.cgi?id=71097

Reviewed by Darin Adler.

* fast/css/parsing-css-nonascii-expected.txt: Added.
* fast/css/parsing-css-nonascii.html: Added.

Modified Paths

trunk/LayoutTests/ChangeLog


Added Paths

trunk/LayoutTests/fast/css/parsing-css-nonascii-expected.txt
trunk/LayoutTests/fast/css/parsing-css-nonascii.html




Diff

Modified: trunk/LayoutTests/ChangeLog (98743 => 98744)

--- trunk/LayoutTests/ChangeLog	2011-10-28 18:29:35 UTC (rev 98743)
+++ trunk/LayoutTests/ChangeLog	2011-10-28 18:40:14 UTC (rev 98744)
@@ -1,3 +1,13 @@
+2011-10-28  Zoltan Herczeg  zherc...@webkit.org
+
+Testing allowed identifier chars which are greater than 126
+https://bugs.webkit.org/show_bug.cgi?id=71097
+
+Reviewed by Darin Adler.
+
+* fast/css/parsing-css-nonascii-expected.txt: Added.
+* fast/css/parsing-css-nonascii.html: Added.
+
 2011-10-28  Ryosuke Niwa  rn...@webkit.org
 
 Unskip tables/mozilla_expected_failures/bugs/bug14007-2.html


Added: trunk/LayoutTests/fast/css/parsing-css-nonascii-expected.txt (0 => 98744)

--- trunk/LayoutTests/fast/css/parsing-css-nonascii-expected.txt	(rev 0)
+++ trunk/LayoutTests/fast/css/parsing-css-nonascii-expected.txt	2011-10-28 18:40:14 UTC (rev 98744)
@@ -0,0 +1,27 @@
+Test parsing of CSS escapes.
+
+SUCCESS
+
+Rules from the stylesheet:
+
+#a127_\ { color: red; }
+#a128_€ { color: red; }
+#b128_€ { color: green; }
+#a159_Ÿ { color: red; }
+#b159_Ÿ { color: green; }
+#a160_  { color: red; }
+#b160_  { color: green; }
+#a384_ƀ { color: red; }
+#b384_ƀ { color: green; }
+Expected result:
+
+#a127_\ { color: red; }
+#a128_€ { color: red; }
+#b128_€ { color: green; }
+#a159_Ÿ { color: red; }
+#b159_Ÿ { color: green; }
+#a160_  { color: red; }
+#b160_  { color: green; }
+#a384_ƀ { color: red; }
+#b384_ƀ { color: green; }
+


Added: trunk/LayoutTests/fast/css/parsing-css-nonascii.html (0 => 98744)

--- trunk/LayoutTests/fast/css/parsing-css-nonascii.html	(rev 0)
+++ trunk/LayoutTests/fast/css/parsing-css-nonascii.html	2011-10-28 18:40:14 UTC (rev 98744)
@@ -0,0 +1,81 @@
+html
+head
+meta http-equiv=Content-Type content=text/html;charset=UTF-8
+style
+/*
+  === Warning ===
+
+  This file intentionally contains scpecial characters which
+  cannot be displayed by some text editors. Please
+  carefully edit the file.
+*/
+
+#a127_\7F { color:red; }
+#b127_ { color:green; }
+
+#a128_\80 { color:red; }
+#b128_€ { color:green; }
+
+#a159_\9F { color:red; }
+#b159_Ÿ { color:green; }
+
+#a160_\A0 { color:red; }
+#b160_  { color:green; }
+
+#a384_\180 { color:red; }
+#b384_ƀ { color:green; }
+/style
+
+script
+
+function runTest()
+{
+if (window.layoutTestController)
+layoutTestController.dumpAsText();
+
+var rules = document.styleSheets[0].cssRules;
+var text = ;
+for (var i = 0; i  rules.length; i++) {
+text += rules.item(i).cssText;
+text += \n;
+}
+
+document.getElementById(result).appendChild(document.createTextNode(text));
+
+if (document.getElementById(result).firstChild.data ="" document.getElementById(expected).firstChild.data)
+document.getElementById(message).firstChild.data = ""
+else
+document.getElementById(message).firstChild.data = ""
+}
+
+/script
+
+/head
+
+body _onload_=runTest()
+
+pTest parsing of CSS escapes./p
+
+p id=messageTEST DID NOT COMPLETE/p
+
+pRules from the stylesheet:/p
+
+pre id=result/pre
+
+pExpected result:/p
+
+pre id=expected#a127_\ { color: red; }
+#a128_€ { color: red; }
+#b128_€ { color: green; }
+#a159_Ÿ { color: red; }
+#b159_Ÿ { color: green; }
+#a160_  { color: red; }
+#b160_  { color: green; }
+#a384_ƀ { color: red; }
+#b384_ƀ { color: green; }
+/pre
+
+script
+
+/script
+/body






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes


[webkit-changes] [98504] trunk/LayoutTests

2011-10-26 Thread zherczeg
Title: [98504] trunk/LayoutTests








Revision 98504
Author zherc...@webkit.org
Date 2011-10-26 11:29:55 -0700 (Wed, 26 Oct 2011)


Log Message
Add new CSS escape sequence parsing tests
https://bugs.webkit.org/show_bug.cgi?id=70909

Reviewed by Darin Adler.

Test escape various sequences terminated by \n \f \r and \r\n.

* fast/css/parsing-css-escapes-expected.txt: Added.
* fast/css/parsing-css-escapes.html: Added.

Modified Paths

trunk/LayoutTests/ChangeLog


Added Paths

trunk/LayoutTests/fast/css/parsing-css-escapes-expected.txt
trunk/LayoutTests/fast/css/parsing-css-escapes.html




Diff

Modified: trunk/LayoutTests/ChangeLog (98503 => 98504)

--- trunk/LayoutTests/ChangeLog	2011-10-26 18:11:51 UTC (rev 98503)
+++ trunk/LayoutTests/ChangeLog	2011-10-26 18:29:55 UTC (rev 98504)
@@ -1,3 +1,15 @@
+2011-10-26  Zoltan Herczeg  zherc...@webkit.org
+
+Add new CSS escape sequence parsing tests
+https://bugs.webkit.org/show_bug.cgi?id=70909
+
+Reviewed by Darin Adler.
+
+Test escape various sequences terminated by \n \f \r and \r\n.
+
+* fast/css/parsing-css-escapes-expected.txt: Added.
+* fast/css/parsing-css-escapes.html: Added.
+
 2011-10-26  Ryosuke Niwa  rn...@webkit.org
 
 Chromium Mac editing test rebaseline.


Added: trunk/LayoutTests/fast/css/parsing-css-escapes-expected.txt (0 => 98504)

--- trunk/LayoutTests/fast/css/parsing-css-escapes-expected.txt	(rev 0)
+++ trunk/LayoutTests/fast/css/parsing-css-escapes-expected.txt	2011-10-26 18:29:55 UTC (rev 98504)
@@ -0,0 +1,19 @@
+Test parsing of CSS escapes.
+
+SUCCESS
+
+Rules from the stylesheet:
+
+#a { background-color: green; }
+#b { background-color: green; }
+#c { background-color: green; }
+#d { }
+#e { }
+Expected result:
+
+#a { background-color: green; }
+#b { background-color: green; }
+#c { background-color: green; }
+#d { }
+#e { }
+


Added: trunk/LayoutTests/fast/css/parsing-css-escapes.html (0 => 98504)

--- trunk/LayoutTests/fast/css/parsing-css-escapes.html	(rev 0)
+++ trunk/LayoutTests/fast/css/parsing-css-escapes.html	2011-10-26 18:29:55 UTC (rev 98504)
@@ -0,0 +1,69 @@
+head
+style
+/*
+  === Warning ===
+
+  This file intentionally contains special newline characters: \n \f \r \r\n
+  Keep them when you move the file between different OSes / file-systems
+*/
+
+/* Base form of the rule: */
+#a { background-color: green; }
+#b { b\61 C\06bG\0052o\00075n\64-col\6Fr: green; }
+#c { b\61	c\6b
+g\72o\75
+n\64
+-color:green }
+#d { background-colo\072: green; /* Invalid: too many hex characters. */}
+#e { b\61ckground-color: green; /* Invalid: \0b is not a newline character. */}
+/style
+
+script
+
+function runTest()
+{
+if (window.layoutTestController)
+layoutTestController.dumpAsText();
+
+var rules = document.styleSheets[0].cssRules;
+var text = ;
+for (var i = 0; i  rules.length; i++) {
+text += rules.item(i).cssText;
+text += \n;
+}
+
+document.getElementById(result).appendChild(document.createTextNode(text));
+
+if (document.getElementById(result).firstChild.data ="" document.getElementById(expected).firstChild.data)
+document.getElementById(message).firstChild.data = ""
+else
+document.getElementById(message).firstChild.data = ""
+}
+
+/script
+
+/head
+
+body _onload_=runTest()
+
+pTest parsing of CSS escapes./p
+
+p id=messageTEST DID NOT COMPLETE/p
+
+pRules from the stylesheet:/p
+
+pre id=result/pre
+
+pExpected result:/p
+
+pre id=expected#a { background-color: green; }
+#b { background-color: green; }
+#c { background-color: green; }
+#d { }
+#e { }
+/pre
+
+script
+
+/script
+/body






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes


[webkit-changes] [98366] trunk/LayoutTests

2011-10-25 Thread zherczeg
Title: [98366] trunk/LayoutTests








Revision 98366
Author zherc...@webkit.org
Date 2011-10-25 11:24:04 -0700 (Tue, 25 Oct 2011)


Log Message
Add new uri tokenizer tests
https://bugs.webkit.org/show_bug.cgi?id=70807

Reviewed by Darin Adler.

Test more cases!

* fast/css/uri-token-parsing-expected.txt:
* fast/css/uri-token-parsing.html:

Modified Paths

trunk/LayoutTests/ChangeLog
trunk/LayoutTests/fast/css/uri-token-parsing-expected.txt
trunk/LayoutTests/fast/css/uri-token-parsing.html




Diff

Modified: trunk/LayoutTests/ChangeLog (98365 => 98366)

--- trunk/LayoutTests/ChangeLog	2011-10-25 18:17:37 UTC (rev 98365)
+++ trunk/LayoutTests/ChangeLog	2011-10-25 18:24:04 UTC (rev 98366)
@@ -1,3 +1,15 @@
+2011-10-25  Zoltan Herczeg  zherc...@webkit.org
+
+Add new uri tokenizer tests
+https://bugs.webkit.org/show_bug.cgi?id=70807
+
+Reviewed by Darin Adler.
+
+Test more cases!
+
+* fast/css/uri-token-parsing-expected.txt:
+* fast/css/uri-token-parsing.html:
+
 2011-10-25  Julien Chaffraix  jchaffr...@webkit.org
 
 Unreviewed gardening.


Modified: trunk/LayoutTests/fast/css/uri-token-parsing-expected.txt (98365 => 98366)

--- trunk/LayoutTests/fast/css/uri-token-parsing-expected.txt	2011-10-25 18:17:37 UTC (rev 98365)
+++ trunk/LayoutTests/fast/css/uri-token-parsing-expected.txt	2011-10-25 18:24:04 UTC (rev 98366)
@@ -24,6 +24,9 @@
 #t { -webkit-mask-image: url('url(t)'); }
 #u { -webkit-border-image: url('url(u)') 1 2 3 4 fill stretch round; }
 #v { -webkit-mask-box-image: url('url(v)') 1 2 3 4 fill stretch round; }
+#w { content: url(ww); }
+#x { content: url(x%20xx); }
+#y { content: url(y%20yy); }
 Expected result:
 
 #a { content: url(c); }
@@ -46,4 +49,7 @@
 #t { -webkit-mask-image: url('url(t)'); }
 #u { -webkit-border-image: url('url(u)') 1 2 3 4 fill stretch round; }
 #v { -webkit-mask-box-image: url('url(v)') 1 2 3 4 fill stretch round; }
+#w { content: url(ww); }
+#x { content: url(x%20xx); }
+#y { content: url(y%20yy); }
 


Modified: trunk/LayoutTests/fast/css/uri-token-parsing.html (98365 => 98366)

--- trunk/LayoutTests/fast/css/uri-token-parsing.html	2011-10-25 18:17:37 UTC (rev 98365)
+++ trunk/LayoutTests/fast/css/uri-token-parsing.html	2011-10-25 18:24:04 UTC (rev 98366)
@@ -20,6 +20,10 @@
 #t { -webkit-mask-image: url(url(t)); }
 #u { -webkit-border-image: url(url(u)) 1 2 3 4 fill stretch round; }
 #v { -webkit-mask-box-image: url(url(v)) 1 2 3 4 fill stretch round; }
+#w { content: url(w\0d\
+w); /* URL string contains twice newline */ }
+#x { content: url( x x	x ); } /* URL string contains space and tab */
+#y { content: url( y\20y\9y ); } /* URL string contains space and tab */
 /style
 
 script
@@ -94,6 +98,9 @@
 #t { -webkit-mask-image: url('url(t)'); }
 #u { -webkit-border-image: url('url(u)') 1 2 3 4 fill stretch round; }
 #v { -webkit-mask-box-image: url('url(v)') 1 2 3 4 fill stretch round; }
+#w { content: url(ww); }
+#x { content: url(x%20xx); }
+#y { content: url(y%20yy); }
 /pre
 
 script






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes


[webkit-changes] [97970] trunk/Source/WebCore

2011-10-20 Thread zherczeg
Title: [97970] trunk/Source/WebCore








Revision 97970
Author zherc...@webkit.org
Date 2011-10-20 06:05:21 -0700 (Thu, 20 Oct 2011)


Log Message
Improve NEON based GaussianBlur
https://bugs.webkit.org/show_bug.cgi?id=70493

Reviewed by Csaba Osztrogonác.

vmov instruction is less complex than vtbl.

* platform/graphics/filters/arm/FEGaussianBlurNEON.cpp:

Modified Paths

trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/platform/graphics/filters/arm/FEGaussianBlurNEON.cpp




Diff

Modified: trunk/Source/WebCore/ChangeLog (97969 => 97970)

--- trunk/Source/WebCore/ChangeLog	2011-10-20 12:57:47 UTC (rev 97969)
+++ trunk/Source/WebCore/ChangeLog	2011-10-20 13:05:21 UTC (rev 97970)
@@ -1,3 +1,14 @@
+2011-10-20  Zoltan Herczeg  zherc...@webkit.org
+
+Improve NEON based GaussianBlur
+https://bugs.webkit.org/show_bug.cgi?id=70493
+
+Reviewed by Csaba Osztrogonác.
+
+vmov instruction is less complex than vtbl.
+
+* platform/graphics/filters/arm/FEGaussianBlurNEON.cpp:
+
 2011-10-20  Pavel Feldman  pfeld...@google.com
 
 Not reviewed: follow up to 97961 - dispatching of hide on detach was missing.


Modified: trunk/Source/WebCore/platform/graphics/filters/arm/FEGaussianBlurNEON.cpp (97969 => 97970)

--- trunk/Source/WebCore/platform/graphics/filters/arm/FEGaussianBlurNEON.cpp	2011-10-20 12:57:47 UTC (rev 97969)
+++ trunk/Source/WebCore/platform/graphics/filters/arm/FEGaussianBlurNEON.cpp	2011-10-20 13:05:21 UTC (rev 97970)
@@ -34,8 +34,6 @@
 namespace WebCore {
 
 static WTF_ALIGNED(unsigned char, s_FEGaussianBlurConstantsForNeon[], 16) = {
-// Mapping from ARM to NEON registers.
-0, 16, 16, 16, 1,  16, 16, 16, 2,  16, 16, 16, 3,  16, 16, 16,
 // Mapping from NEON to ARM registers.
 0, 4,  8,  12, 16, 16, 16, 16
 };
@@ -93,10 +91,7 @@
 #define PIXEL_D11   d5[1]
 #define REMAINING_STRIDES_S0s12
 
-#define READ_RANGE  d16-d18
-#define REMAP_ARM_NEON1_Q   d16
-#define REMAP_ARM_NEON2_Q   d17
-#define REMAP_NEON_ARM_Qd18
+#define REMAP_NEON_ARM_Qd16
 
 asm ( // NOLINT
 .globl  TOSTRING(neonDrawAllChannelGaussianBlur) NL
@@ -119,7 +114,7 @@
 movcs  MAX_KERNEL_SIZE_R ,  STRIDE_WIDTH_R NL
 add  SOURCE_LINE_END_R ,  SOURCE_LINE_END_R ,  SOURCE_R NL
 vdup.f32  INVERTED_KERNEL_SIZE_Q ,  INIT_INVERTED_KERNEL_SIZE_R NL
-vld1.f32 {  READ_RANGE  }, [ INIT_PAINTING_CONSTANTS_R ]! NL
+vld1.f32 {  REMAP_NEON_ARM_Q  }, [ INIT_PAINTING_CONSTANTS_R ]! NL
 
 .allChannelMainLoop: NL
 
@@ -131,8 +126,8 @@
 bcs .allChannelInitSumDone NL
 .allChannelInitSum: NL
 vld1.u32  PIXEL_D00 , [ INIT_SUM_R ],  STRIDE_R NL
-vtbl.8  PIXEL_D1 , { PIXEL_D0 },  REMAP_ARM_NEON2_Q NL
-vtbl.8  PIXEL_D0 , { PIXEL_D0 },  REMAP_ARM_NEON1_Q NL
+vmovl.u8  PIXEL_Q ,  PIXEL_D0 NL
+vmovl.u16  PIXEL_Q ,  PIXEL_D0 NL
 vadd.u32  SUM_Q ,  SUM_Q ,  PIXEL_Q NL
 cmp  INIT_SUM_R ,  SOURCE_END_R NL
 bcc .allChannelInitSum NL
@@ -154,16 +149,16 @@
 cmp  LEFT_R ,  SOURCE_R NL
 bcc .allChannelSkipLeft NL
 vld1.u32  PIXEL_D00 , [ LEFT_R ] NL
-vtbl.8  PIXEL_D1 , { PIXEL_D0 },  REMAP_ARM_NEON2_Q NL
-vtbl.8  PIXEL_D0 , { PIXEL_D0 },  REMAP_ARM_NEON1_Q NL
+vmovl.u8  PIXEL_Q ,  PIXEL_D0 NL
+vmovl.u16  PIXEL_Q ,  PIXEL_D0 NL
 vsub.u32  SUM_Q ,  SUM_Q ,  PIXEL_Q NL
 .allChannelSkipLeft:  NL
 
 cmp  RIGHT_R ,  SOURCE_END_R NL
 bcs .allChannelSkipRight NL
 vld1.u32  PIXEL_D00 , [ RIGHT_R ] NL
-vtbl.8  PIXEL_D1 , { PIXEL_D0 },  REMAP_ARM_NEON2_Q NL
-vtbl.8  PIXEL_D0 , { PIXEL_D0 },  REMAP_ARM_NEON1_Q NL
+vmovl.u8  PIXEL_Q ,  PIXEL_D0 NL
+vmovl.u16  PIXEL_Q ,  PIXEL_D0 NL
 vadd.u32  SUM_Q ,  SUM_Q ,  PIXEL_Q NL
 .allChannelSkipRight:  NL
 






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes


[webkit-changes] [96795] trunk/LayoutTests

2011-10-06 Thread zherczeg
Title: [96795] trunk/LayoutTests








Revision 96795
Author zherc...@webkit.org
Date 2011-10-06 02:30:24 -0700 (Thu, 06 Oct 2011)


Log Message
Unreviewed, Qt rebaseline after r96789.

* platform/qt/http/tests/navigation/postredirect-basic-expected.txt:
* platform/qt/http/tests/navigation/postredirect-frames-expected.txt: Added.
* platform/qt/http/tests/navigation/postredirect-goback1-expected.txt:
* platform/qt/http/tests/navigation/postredirect-goback2-expected.txt: Added.

Modified Paths

trunk/LayoutTests/ChangeLog
trunk/LayoutTests/platform/qt/http/tests/navigation/postredirect-basic-expected.txt
trunk/LayoutTests/platform/qt/http/tests/navigation/postredirect-goback1-expected.txt


Added Paths

trunk/LayoutTests/platform/qt/http/tests/navigation/postredirect-frames-expected.txt
trunk/LayoutTests/platform/qt/http/tests/navigation/postredirect-goback2-expected.txt




Diff

Modified: trunk/LayoutTests/ChangeLog (96794 => 96795)

--- trunk/LayoutTests/ChangeLog	2011-10-06 09:08:25 UTC (rev 96794)
+++ trunk/LayoutTests/ChangeLog	2011-10-06 09:30:24 UTC (rev 96795)
@@ -1,3 +1,12 @@
+2011-10-06  Zoltan Herczeg  zherc...@webkit.org
+
+Unreviewed, Qt rebaseline after r96789.
+
+* platform/qt/http/tests/navigation/postredirect-basic-expected.txt:
+* platform/qt/http/tests/navigation/postredirect-frames-expected.txt: Added.
+* platform/qt/http/tests/navigation/postredirect-goback1-expected.txt:
+* platform/qt/http/tests/navigation/postredirect-goback2-expected.txt: Added.
+
 2011-10-03  Hans Wennborg  h...@chromium.org
 
 IndexedDB: Remove SQLite-LevelDB migration code


Modified: trunk/LayoutTests/platform/qt/http/tests/navigation/postredirect-basic-expected.txt (96794 => 96795)

--- trunk/LayoutTests/platform/qt/http/tests/navigation/postredirect-basic-expected.txt	2011-10-06 09:08:25 UTC (rev 96794)
+++ trunk/LayoutTests/platform/qt/http/tests/navigation/postredirect-basic-expected.txt	2011-10-06 09:30:24 UTC (rev 96795)
@@ -24,5 +24,5 @@
 == Back Forward List ==
 http://127.0.0.1:8000/navigation/postredirect-basic.html  **nav target**
 http://127.0.0.1:8000/navigation/resources/success200.html  **nav target**
-curr-  http://127.0.0.1:8000/navigation/resources/postresult.pl?submitwithpostredirect=Submit%20with%20POST%20followed%20by%20a%20redirecttextfield1=New%20form%20text%20from%20usertextfield2=radiogroup1=femalecheckbox2=onselectgroup1=fiattextarea1=More%20new%20form%20text%20from%20user%2C%20which%20should%20be%20restored%20when%20we%20return%20to%20this%20page.redirectHappened=true  **nav target**
+curr-  http://127.0.0.1:8000/navigation/resources/postresult.pl?submitwithpostredirect=Submit with POST followed by a redirecttextfield1=New form text from usertextfield2=radiogroup1=femalecheckbox2=onselectgroup1=fiattextarea1=More new form text from user, which should be restored when we return to this page.redirectHappened=true  **nav target**
 ===


Added: trunk/LayoutTests/platform/qt/http/tests/navigation/postredirect-frames-expected.txt (0 => 96795)

--- trunk/LayoutTests/platform/qt/http/tests/navigation/postredirect-frames-expected.txt	(rev 0)
+++ trunk/LayoutTests/platform/qt/http/tests/navigation/postredirect-frames-expected.txt	2011-10-06 09:30:24 UTC (rev 96795)
@@ -0,0 +1,26 @@
+
+
+
+Frame: 'main'
+
+This is just a minimal page that we navigate in response to an HTTP POST. 
+
+If the next line is empty after the colon, it probably means that we made a mistake and requested this page with a GET with no query instead of a POST. 
+
+The first text field contained: New form text from user 
+This page was requested with an HTTP GET
+
+
+Frame: 'footer'
+
+This is just a minimal page that we navigate to as part of testing back/forward.
+
+== Back Forward List ==
+http://127.0.0.1:8000/navigation/postredirect-frames.html  **nav target**
+http://127.0.0.1:8000/navigation/resources/frameset.pl?frameURL=success200.html  **nav target**
+http://127.0.0.1:8000/navigation/resources/otherpage.html (in frame footer)
+http://127.0.0.1:8000/navigation/resources/success200.html (in frame main)
+curr-  http://127.0.0.1:8000/navigation/resources/frameset.pl?frameURL=success200.html
+http://127.0.0.1:8000/navigation/resources/otherpage.html (in frame footer)
+http://127.0.0.1:8000/navigation/resources/postresult.pl?submitwithpostredirect=Submit with POST followed by a redirecttextfield1=New form text from usertextfield2=radiogroup1=femalecheckbox2=onselectgroup1=fiattextarea1=More new form text from user, which should be restored when we return to this page.redirectHappened=true (in frame main)  **nav target**
+===


Modified: 

[webkit-changes] [95178] trunk/LayoutTests

2011-09-15 Thread zherczeg
Title: [95178] trunk/LayoutTests








Revision 95178
Author zherc...@webkit.org
Date 2011-09-15 01:09:56 -0700 (Thu, 15 Sep 2011)


Log Message
Renaming PeerConnection to webkitPeerConnection affects all ports,
not just chromium. Updating layout test after r95151.

Unreviewed gardening.

* fast/dom/call-a-constructor-as-a-function-expected.txt:

Modified Paths

trunk/LayoutTests/ChangeLog
trunk/LayoutTests/fast/dom/call-a-constructor-as-a-function-expected.txt




Diff

Modified: trunk/LayoutTests/ChangeLog (95177 => 95178)

--- trunk/LayoutTests/ChangeLog	2011-09-15 07:56:20 UTC (rev 95177)
+++ trunk/LayoutTests/ChangeLog	2011-09-15 08:09:56 UTC (rev 95178)
@@ -1,3 +1,12 @@
+2011-09-15  Zoltan Herczeg  zherc...@webkit.org
+
+Renaming PeerConnection to webkitPeerConnection affects all ports,
+not just chromium. Updating layout test after r95151.
+
+Unreviewed gardening.
+
+* fast/dom/call-a-constructor-as-a-function-expected.txt:
+
 2011-09-15  Keishi Hattori  kei...@webkit.org
 
 [chromium] rebaseline expectations due to r95121


Modified: trunk/LayoutTests/fast/dom/call-a-constructor-as-a-function-expected.txt (95177 => 95178)

--- trunk/LayoutTests/fast/dom/call-a-constructor-as-a-function-expected.txt	2011-09-15 07:56:20 UTC (rev 95177)
+++ trunk/LayoutTests/fast/dom/call-a-constructor-as-a-function-expected.txt	2011-09-15 08:09:56 UTC (rev 95178)
@@ -15,7 +15,7 @@
 PASS Int32Array() threw exception TypeError: '[object Int32ArrayConstructor]' is not a function (evaluating 'Int32Array()').
 PASS Int8Array() threw exception TypeError: '[object Int8ArrayConstructor]' is not a function (evaluating 'Int8Array()').
 PASS MessageChannel() threw exception TypeError: '[object MessageChannelConstructor]' is not a function (evaluating 'MessageChannel()').
-SKIP PeerConnection is not implemented.
+SKIP webkitPeerConnection is not implemented.
 PASS SharedWorker() threw exception TypeError: '[object SharedWorkerConstructor]' is not a function (evaluating 'SharedWorker()').
 PASS Uint16Array() threw exception TypeError: '[object Uint16ArrayConstructor]' is not a function (evaluating 'Uint16Array()').
 PASS Uint32Array() threw exception TypeError: '[object Uint32ArrayConstructor]' is not a function (evaluating 'Uint32Array()').






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes


[webkit-changes] [95179] trunk/LayoutTests

2011-09-15 Thread zherczeg
Title: [95179] trunk/LayoutTests








Revision 95179
Author zherc...@webkit.org
Date 2011-09-15 01:37:49 -0700 (Thu, 15 Sep 2011)


Log Message
[Qt] Unreviewed gardening. Adding new test expectations.

* platform/qt/fast/borders/border-image-massive-scale-expected.txt: Added.
* platform/qt/fast/borders/border-image-scaled-gradient-expected.txt: Added.
* platform/qt/fast/borders/border-image-scrambled-expected.txt: Added.
* platform/qt/fast/parser/innerhtml-with-prefixed-elements-expected.txt: Added.

Modified Paths

trunk/LayoutTests/ChangeLog


Added Paths

trunk/LayoutTests/platform/qt/fast/borders/border-image-massive-scale-expected.txt
trunk/LayoutTests/platform/qt/fast/borders/border-image-scaled-gradient-expected.txt
trunk/LayoutTests/platform/qt/fast/borders/border-image-scrambled-expected.txt
trunk/LayoutTests/platform/qt/fast/parser/innerhtml-with-prefixed-elements-expected.txt




Diff

Modified: trunk/LayoutTests/ChangeLog (95178 => 95179)

--- trunk/LayoutTests/ChangeLog	2011-09-15 08:09:56 UTC (rev 95178)
+++ trunk/LayoutTests/ChangeLog	2011-09-15 08:37:49 UTC (rev 95179)
@@ -1,5 +1,14 @@
 2011-09-15  Zoltan Herczeg  zherc...@webkit.org
 
+[Qt] Unreviewed gardening. Adding new test expectations.
+
+* platform/qt/fast/borders/border-image-massive-scale-expected.txt: Added.
+* platform/qt/fast/borders/border-image-scaled-gradient-expected.txt: Added.
+* platform/qt/fast/borders/border-image-scrambled-expected.txt: Added.
+* platform/qt/fast/parser/innerhtml-with-prefixed-elements-expected.txt: Added.
+
+2011-09-15  Zoltan Herczeg  zherc...@webkit.org
+
 Renaming PeerConnection to webkitPeerConnection affects all ports,
 not just chromium. Updating layout test after r95151.
 


Added: trunk/LayoutTests/platform/qt/fast/borders/border-image-massive-scale-expected.txt (0 => 95179)

--- trunk/LayoutTests/platform/qt/fast/borders/border-image-massive-scale-expected.txt	(rev 0)
+++ trunk/LayoutTests/platform/qt/fast/borders/border-image-massive-scale-expected.txt	2011-09-15 08:37:49 UTC (rev 95179)
@@ -0,0 +1,17 @@
+layer at (0,0) size 784x728
+  RenderView at (0,0) size 784x600
+layer at (0,0) size 784x728
+  RenderBlock {HTML} at (0,0) size 784x728
+RenderBody {BODY} at (8,8) size 768x712
+  RenderBlock {DIV} at (10,10) size 330x330 [border: (105px none #00) (150px none #00) (105px none #00)]
+  RenderText {#text} at (350,334) size 4x22
+text run at (350,334) width 4:  
+  RenderBlock {DIV} at (364,10) size 330x330 [border: (105px none #00) (150px none #00) (105px none #00)]
+  RenderText {#text} at (704,334) size 4x22
+text run at (704,334) width 4:  
+  RenderBR {BR} at (0,0) size 0x0
+  RenderBlock {DIV} at (10,366) size 330x330 [border: (105px none #00) (150px none #00) (105px none #00)]
+  RenderText {#text} at (350,690) size 4x22
+text run at (350,690) width 4:  
+  RenderBlock {DIV} at (364,366) size 330x330 [border: (105px none #00) (150px none #00) (105px none #00)]
+  RenderText {#text} at (0,0) size 0x0


Added: trunk/LayoutTests/platform/qt/fast/borders/border-image-scaled-gradient-expected.txt (0 => 95179)

--- trunk/LayoutTests/platform/qt/fast/borders/border-image-scaled-gradient-expected.txt	(rev 0)
+++ trunk/LayoutTests/platform/qt/fast/borders/border-image-scaled-gradient-expected.txt	2011-09-15 08:37:49 UTC (rev 95179)
@@ -0,0 +1,6 @@
+layer at (0,0) size 800x600
+  RenderView at (0,0) size 800x600
+layer at (0,0) size 800x320
+  RenderBlock {HTML} at (0,0) size 800x320
+RenderBody {BODY} at (8,8) size 784x304
+  RenderBlock {DIV} at (0,0) size 304x304 [border: (2px solid #FF)]


Added: trunk/LayoutTests/platform/qt/fast/borders/border-image-scrambled-expected.txt (0 => 95179)

--- trunk/LayoutTests/platform/qt/fast/borders/border-image-scrambled-expected.txt	(rev 0)
+++ trunk/LayoutTests/platform/qt/fast/borders/border-image-scrambled-expected.txt	2011-09-15 08:37:49 UTC (rev 95179)
@@ -0,0 +1,17 @@
+layer at (0,0) size 800x600
+  RenderView at (0,0) size 800x600
+layer at (0,0) size 800x600
+  RenderBlock {HTML} at (0,0) size 800x600
+RenderBody {BODY} at (8,8) size 784x584
+  RenderBlock {DIV} at (10,10) size 126x126 [border: (21px none #00) (30px none #00) (21px none #00)]
+  RenderText {#text} at (146,130) size 4x22
+text run at (146,130) width 4:  
+  RenderBlock {DIV} at (160,10) size 126x126 [border: (21px none #00) (30px none #00) (21px none #00)]
+  RenderText {#text} at (296,130) size 4x22
+text run at (296,130) width 4:  
+  RenderBR {BR} at (0,0) size 0x0
+  RenderBlock {DIV} at (10,162) size 126x126 [border: (21px none #00) (30px none #00) (21px none #00)]
+  RenderText {#text} at (146,282) size 4x22
+text run at (146,282) 

[webkit-changes] [94283] trunk/LayoutTests

2011-09-01 Thread zherczeg
Title: [94283] trunk/LayoutTests








Revision 94283
Author zherc...@webkit.org
Date 2011-09-01 03:22:47 -0700 (Thu, 01 Sep 2011)


Log Message
Unreviewed, skipping failing tests after r94259.

r94259 fails on Qt and Gtk
https://bugs.webkit.org/show_bug.cgi?id=67383

* platform/gtk/Skipped:
* platform/qt/Skipped:

Modified Paths

trunk/LayoutTests/ChangeLog
trunk/LayoutTests/platform/gtk/Skipped
trunk/LayoutTests/platform/qt/Skipped




Diff

Modified: trunk/LayoutTests/ChangeLog (94282 => 94283)

--- trunk/LayoutTests/ChangeLog	2011-09-01 09:58:52 UTC (rev 94282)
+++ trunk/LayoutTests/ChangeLog	2011-09-01 10:22:47 UTC (rev 94283)
@@ -1,3 +1,13 @@
+2011-09-01  Zoltan Herczeg  zherc...@webkit.org
+
+Unreviewed, skipping failing tests after r94259.
+
+r94259 fails on Qt and Gtk
+https://bugs.webkit.org/show_bug.cgi?id=67383
+
+* platform/gtk/Skipped:
+* platform/qt/Skipped:
+
 2011-09-01  Yuta Kitamura  yu...@chromium.org
 
 WebSocket: Fix bufferedAmount after WebSocket is closed


Modified: trunk/LayoutTests/platform/gtk/Skipped (94282 => 94283)

--- trunk/LayoutTests/platform/gtk/Skipped	2011-09-01 09:58:52 UTC (rev 94282)
+++ trunk/LayoutTests/platform/gtk/Skipped	2011-09-01 10:22:47 UTC (rev 94283)
@@ -1623,3 +1623,7 @@
 
 # https://bugs.webkit.org/show_bug.cgi?id=67277
 fast/loader/location-port.html
+
+# https://bugs.webkit.org/show_bug.cgi?id=67383
+fast/events/document-elementFromPoint.html
+plugins/mouse-events-fixedpos.html


Modified: trunk/LayoutTests/platform/qt/Skipped (94282 => 94283)

--- trunk/LayoutTests/platform/qt/Skipped	2011-09-01 09:58:52 UTC (rev 94282)
+++ trunk/LayoutTests/platform/qt/Skipped	2011-09-01 10:22:47 UTC (rev 94283)
@@ -2411,3 +2411,7 @@
 fast/multicol/positive-leading.html
 fast/multicol/overflow-unsplittable.html
 fast/multicol/overflow-across-columns.html
+
+# [Qt] Incorrect absolute mouse pointer results after r94259
+fast/events/document-elementFromPoint.html
+plugins/mouse-events-fixedpos.html






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes


[webkit-changes] [93955] trunk/Source/WebCore

2011-08-29 Thread zherczeg
Title: [93955] trunk/Source/WebCore








Revision 93955
Author zherc...@webkit.org
Date 2011-08-29 00:07:27 -0700 (Mon, 29 Aug 2011)


Log Message
[Qt] Unreviewed minimal-build buildfix after r93937.

* platform/qt/RenderThemeQt.cpp:
(WebCore::RenderThemeQt::paintInnerSpinButton):
* platform/qt/RenderThemeQt.h:

Patch by Zoltan Herczeg zherc...@webkit.org on 2011-08-29

Modified Paths

trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/platform/qt/RenderThemeQt.cpp
trunk/Source/WebCore/platform/qt/RenderThemeQt.h




Diff

Modified: trunk/Source/WebCore/ChangeLog (93954 => 93955)

--- trunk/Source/WebCore/ChangeLog	2011-08-29 06:09:42 UTC (rev 93954)
+++ trunk/Source/WebCore/ChangeLog	2011-08-29 07:07:27 UTC (rev 93955)
@@ -1,3 +1,11 @@
+2011-08-29  Zoltan Herczeg  zherc...@webkit.org
+
+[Qt] Unreviewed minimal-build buildfix after r93937.
+
+* platform/qt/RenderThemeQt.cpp:
+(WebCore::RenderThemeQt::paintInnerSpinButton):
+* platform/qt/RenderThemeQt.h:
+
 2011-08-28  Keishi Hattori  kei...@webkit.org
 
 Chromium Win: Setting square-button appearance reaches NOTREACHED


Modified: trunk/Source/WebCore/platform/qt/RenderThemeQt.cpp (93954 => 93955)

--- trunk/Source/WebCore/platform/qt/RenderThemeQt.cpp	2011-08-29 06:09:42 UTC (rev 93954)
+++ trunk/Source/WebCore/platform/qt/RenderThemeQt.cpp	2011-08-29 07:07:27 UTC (rev 93955)
@@ -1068,6 +1068,7 @@
 return RenderTheme::paintSearchFieldResultsDecoration(o, pi, r);
 }
 
+#ifndef QT_NO_SPINBOX
 void RenderThemeQt::adjustInnerSpinButtonStyle(CSSStyleSelector* selector, RenderStyle* style,
Element* e) const
 {
@@ -1119,6 +1120,7 @@
 p.drawComplexControl(QStyle::CC_SpinBox, option);
 return false;
 }
+#endif
 
 bool RenderThemeQt::supportsFocus(ControlPart appearance) const
 {


Modified: trunk/Source/WebCore/platform/qt/RenderThemeQt.h (93954 => 93955)

--- trunk/Source/WebCore/platform/qt/RenderThemeQt.h	2011-08-29 06:09:42 UTC (rev 93954)
+++ trunk/Source/WebCore/platform/qt/RenderThemeQt.h	2011-08-29 07:07:27 UTC (rev 93955)
@@ -138,8 +138,10 @@
 virtual void adjustSearchFieldResultsDecorationStyle(CSSStyleSelector*, RenderStyle*, Element*) const;
 virtual bool paintSearchFieldResultsDecoration(RenderObject*, const PaintInfo, const IntRect);
 
+#ifndef QT_NO_SPINBOX
 virtual void adjustInnerSpinButtonStyle(CSSStyleSelector*, RenderStyle*, Element*) const;
 virtual bool paintInnerSpinButton(RenderObject*, const PaintInfo, const IntRect);
+#endif
 
 #if ENABLE(PROGRESS_TAG)
 // Returns the repeat interval of the animation for the progress bar.






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes


[webkit-changes] [93763] trunk/LayoutTests

2011-08-25 Thread zherczeg
Title: [93763] trunk/LayoutTests








Revision 93763
Author zherc...@webkit.org
Date 2011-08-25 01:04:45 -0700 (Thu, 25 Aug 2011)


Log Message
[Qt] Add Qt specific expected after r93713 since ENABLE_MEDIA_STREAM is disabled.

Unreviewed, test expectations update.

* platform/qt/fast/files/create-blob-url-crash-expected.txt: Added.

Patch by Zoltan Herczeg zherc...@webkit.org on 2011-08-25

Modified Paths

trunk/LayoutTests/ChangeLog


Added Paths

trunk/LayoutTests/platform/qt/fast/files/
trunk/LayoutTests/platform/qt/fast/files/create-blob-url-crash-expected.txt




Diff

Modified: trunk/LayoutTests/ChangeLog (93762 => 93763)

--- trunk/LayoutTests/ChangeLog	2011-08-25 06:58:33 UTC (rev 93762)
+++ trunk/LayoutTests/ChangeLog	2011-08-25 08:04:45 UTC (rev 93763)
@@ -1,3 +1,11 @@
+2011-08-25  Zoltan Herczeg  zherc...@webkit.org
+
+[Qt] Add Qt specific expected after r93713 since ENABLE_MEDIA_STREAM is disabled.
+
+Unreviewed, test expectations update.
+
+* platform/qt/fast/files/create-blob-url-crash-expected.txt: Added.
+
 2011-08-24  Kentaro Hara  hara...@google.com
 
 Implement an Event constructor.


Added: trunk/LayoutTests/platform/qt/fast/files/create-blob-url-crash-expected.txt (0 => 93763)

--- trunk/LayoutTests/platform/qt/fast/files/create-blob-url-crash-expected.txt	(rev 0)
+++ trunk/LayoutTests/platform/qt/fast/files/create-blob-url-crash-expected.txt	2011-08-25 08:04:45 UTC (rev 93763)
@@ -0,0 +1,4 @@
+Test that createObjectURL with no argument should throw an exception.
+PASS: Not enough arguments
+DONE
+






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes


[webkit-changes] [90624] trunk/Tools

2011-07-08 Thread zherczeg
Title: [90624] trunk/Tools








Revision 90624
Author zherc...@webkit.org
Date 2011-07-08 06:19:25 -0700 (Fri, 08 Jul 2011)


Log Message
Reviewed by Andreas Kling.

Adding myself as a reviewer.

* Scripts/webkitpy/common/config/committers.py:

Modified Paths

trunk/Tools/ChangeLog
trunk/Tools/Scripts/webkitpy/common/config/committers.py




Diff

Modified: trunk/Tools/ChangeLog (90623 => 90624)

--- trunk/Tools/ChangeLog	2011-07-08 12:55:27 UTC (rev 90623)
+++ trunk/Tools/ChangeLog	2011-07-08 13:19:25 UTC (rev 90624)
@@ -1,3 +1,11 @@
+2011-07-08  Herczeg Zoltan  zherc...@webkit.org
+
+Reviewed by Andreas Kling.
+
+Adding myself as a reviewer.
+
+* Scripts/webkitpy/common/config/committers.py:
+
 2011-07-07  Kevin Ollivier  kev...@theolliviers.com
 
 [wx] Unreviewed build fix, remove old files from the tree and update paths to new ones.


Modified: trunk/Tools/Scripts/webkitpy/common/config/committers.py (90623 => 90624)

--- trunk/Tools/Scripts/webkitpy/common/config/committers.py	2011-07-08 12:55:27 UTC (rev 90623)
+++ trunk/Tools/Scripts/webkitpy/common/config/committers.py	2011-07-08 13:19:25 UTC (rev 90624)
@@ -264,7 +264,6 @@
 Committer(Yuta Kitamura, yu...@chromium.org, yutak),
 Committer(Yuzo Fujishima, y...@google.com, yuzo),
 Committer(Zhenyao Mo, z...@google.com, zhenyao),
-Committer(Zoltan Herczeg, zherc...@webkit.org, zherczeg),
 Committer(Zoltan Horvath, [zol...@webkit.org, hzol...@inf.u-szeged.hu, horvath.zolta...@stud.u-szeged.hu], zoltan),
 Committer(Naoki Takano, hon...@chromium.org, honten),
 ]
@@ -369,6 +368,7 @@
 Reviewer(Xan Lopez, [xan.lo...@gmail.com, x...@gnome.org, x...@webkit.org], xan),
 Reviewer(Yury Semikhatsky, yu...@chromium.org, yurys),
 Reviewer(Zack Rusin, z...@kde.org, zackr),
+Reviewer(Zoltan Herczeg, zherc...@webkit.org, zherczeg),
 ]
 
 






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes


[webkit-changes] [89556] trunk

2011-06-23 Thread zherczeg
Title: [89556] trunk








Revision 89556
Author zherc...@webkit.org
Date 2011-06-23 05:02:58 -0700 (Thu, 23 Jun 2011)


Log Message
Inspector may close at the start of the next inspector test in DRT
https://bugs.webkit.org/show_bug.cgi?id=60881

Patch by Zoltan Herczeg zherc...@inf.u-szeged.hu on 2011-06-23
Reviewed by Pavel Feldman.

Source/WebCore: 

Add a new function to the inspector, which tests whether the
dispatch queue is empty.

* inspector/front-end/inspector.js:
(WebInspector.dispatchQueueIsEmpty):

LayoutTests: 

The didEvaluateForTestInFrontend function call is postponed until
all messages in the dispatch queue is processed.

* http/tests/inspector/inspector-test.js:
(initialize_InspectorTest.InspectorTest.completeTest):

Modified Paths

trunk/LayoutTests/ChangeLog
trunk/LayoutTests/http/tests/inspector/inspector-test.js
trunk/Source/WebCore/ChangeLog
trunk/Source/WebCore/inspector/front-end/inspector.js




Diff

Modified: trunk/LayoutTests/ChangeLog (89555 => 89556)

--- trunk/LayoutTests/ChangeLog	2011-06-23 11:52:15 UTC (rev 89555)
+++ trunk/LayoutTests/ChangeLog	2011-06-23 12:02:58 UTC (rev 89556)
@@ -1,3 +1,16 @@
+2011-06-23  Zoltan Herczeg  zherc...@inf.u-szeged.hu
+
+Reviewed by Pavel Feldman.
+
+Inspector may close at the start of the next inspector test in DRT
+https://bugs.webkit.org/show_bug.cgi?id=60881
+
+The didEvaluateForTestInFrontend function call is postponed until
+all messages in the dispatch queue is processed.
+
+* http/tests/inspector/inspector-test.js:
+(initialize_InspectorTest.InspectorTest.completeTest):
+
 2011-06-23  Sergio Villar Senin  svil...@igalia.com
 
 Unreviewed, final round of baselines for the new CSS2.1 tests added in r88913.


Modified: trunk/LayoutTests/http/tests/inspector/inspector-test.js (89555 => 89556)

--- trunk/LayoutTests/http/tests/inspector/inspector-test.js	2011-06-23 11:52:15 UTC (rev 89555)
+++ trunk/LayoutTests/http/tests/inspector/inspector-test.js	2011-06-23 12:02:58 UTC (rev 89556)
@@ -14,7 +14,15 @@
 
 InspectorTest.completeTest = function()
 {
-RuntimeAgent.evaluate(didEvaluateForTestInFrontend( + InspectorTest.completeTestCallId + , \\), test);
+function testDispatchQueueIsEmpty() {
+if (!WebInspector.dispatchQueueIsEmpty()) {
+// Wait for unprocessed messages.
+setTimeout(testDispatchQueueIsEmpty, 10);
+return;
+}
+RuntimeAgent.evaluate(didEvaluateForTestInFrontend( + InspectorTest.completeTestCallId + , \\), test);
+}
+testDispatchQueueIsEmpty();
 }
 
 InspectorTest.evaluateInConsole = function(code, callback)


Modified: trunk/Source/WebCore/ChangeLog (89555 => 89556)

--- trunk/Source/WebCore/ChangeLog	2011-06-23 11:52:15 UTC (rev 89555)
+++ trunk/Source/WebCore/ChangeLog	2011-06-23 12:02:58 UTC (rev 89556)
@@ -1,3 +1,16 @@
+2011-06-23  Zoltan Herczeg  zherc...@inf.u-szeged.hu
+
+Reviewed by Pavel Feldman.
+
+Inspector may close at the start of the next inspector test in DRT
+https://bugs.webkit.org/show_bug.cgi?id=60881
+
+Add a new function to the inspector, which tests whether the
+dispatch queue is empty.
+
+* inspector/front-end/inspector.js:
+(WebInspector.dispatchQueueIsEmpty):
+
 2011-06-23  Jeffrey Pfau  jp...@apple.com
 
 Reviewed by Nikolas Zimmermann.


Modified: trunk/Source/WebCore/inspector/front-end/inspector.js (89555 => 89556)

--- trunk/Source/WebCore/inspector/front-end/inspector.js	2011-06-23 11:52:15 UTC (rev 89555)
+++ trunk/Source/WebCore/inspector/front-end/inspector.js	2011-06-23 12:02:58 UTC (rev 89556)
@@ -537,6 +537,10 @@
 
 var messagesToDispatch = [];
 
+WebInspector.dispatchQueueIsEmpty = function() {
+return messagesToDispatch.length == 0;
+}
+
 WebInspector.dispatch = function(message) {
 messagesToDispatch.push(message);
 setTimeout(function() {






___
webkit-changes mailing list
webkit-changes@lists.webkit.org
http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes