resolve conflict

Makefile

@@ -4,6 +4,7 @@ LIBSUFFIX=a
 CP=cp
 ROOTDIR=$(PWD)
 
+
 ifeq (,$(wildcard ./gtest))
 HAVE_GTEST=No
 else
@@ -13,20 +14,22 @@ endif
 # Configurations
 ifeq ($(BUILDTYPE), Release)
 CFLAGS += -O3
-ifneq ($(ENABLE64BIT), Yes)
 USE_ASM = Yes
-endif
 else
 CFLAGS = -g
 USE_ASM = No
 endif
+
 ifeq ($(ENABLE64BIT), Yes)
 CFLAGS += -m64
 LDFLAGS += -m64
+ASMFLAGS += -DUNIX64
 else
 CFLAGS += -m32
 LDFLAGS += -m32
+ASMFLAGS += -DX86_32
 endif
+
 include build/platform-$(UNAME).mk
 
 ifeq ($(USE_ASM),Yes)
@@ -40,7 +43,8 @@ ASMFLAGS += -DNO_DYNAMIC_VP
 
 #### No user-serviceable parts below this line
 INCLUDES = -Icodec/api/svc  -Icodec/common -Igtest/include
-ASM_INCLUDES = -Iprocessing/src/asm/
+#ASM_INCLUDES = -Iprocessing/src/asm/
+ASM_INCLUDES = -Icodec/common/
 
 COMMON_INCLUDES = \
     -Icodec/decoder/core/inc
@@ -83,7 +87,7 @@ test:
 include codec/common/targets.mk
 include codec/decoder/targets.mk
 include codec/encoder/targets.mk
-include processing/targets.mk
+include codec/processing/targets.mk
 include codec/console/dec/targets.mk
 include codec/console/enc/targets.mk
 

build/mktargets.sh

@@ -2,7 +2,7 @@
 (cd codec/decoder; python ../../build/mktargets.py --directory codec/decoder --library decoder --exclude StdAfx.cpp)
 (cd codec/encoder; python ../../build/mktargets.py --directory codec/encoder --library encoder --exclude DllEntry.cpp)
 (cd codec/common; python ../../build/mktargets.py --directory codec/common --library common)
-(cd processing; python ../build/mktargets.py --directory processing --library processing --exclude wels_process.cpp --exclude WelsVideoProcessor.cpp)
+(cd codec/processing; python ../../build/mktargets.py --directory codec/processing --library processing --exclude wels_process.cpp --exclude WelsVideoProcessor.cpp)
 
 (cd codec/console/dec; python ../../../build/mktargets.py --directory codec/console/dec --binary h264dec --exclude dec_console.h --exclude load_bundle_functions.cpp)
 (cd codec/console/enc; python ../../../build/mktargets.py --directory codec/console/enc --binary h264enc --exclude enc_console.h --exclude bundlewelsenc.cpp)

build/platform-darwin.mk

@@ -1,5 +1,11 @@
-USE_ASM = No  # We don't have ASM working on Mac yet
+
 ASM = nasm
 CFLAGS += -Werror -fPIC
 LDFLAGS += -lpthread
-ASMFLAGS += -f macho --prefix _ -DNOPREFIX
+ASMFLAGS += --prefix _ -DNOPREFIX
+ifeq ($(ENABLE64BIT), Yes)
+ASMFLAGS += -f macho64
+else
+ASMFLAGS += -f macho
+endif
+

build/platform-linux.mk

@@ -1,5 +1,10 @@
 ASM = nasm
 CFLAGS += -Werror -fPIC -DLINUX -D__NO_CTYPE
 LDFLAGS += -lpthread
-ASMFLAGS += -f elf -DNOPREFIX
+ASMFLAGS += -DNOPREFIX
+ifeq ($(ENABLE64BIT), Yes)
+ASMFLAGS += -f elf64
+else 
+ASMFLAGS += -f elf32
+endif
 

codec/build/win32/dec/WelsDecCore.vcproj

@@ -348,44 +348,6 @@
 				Name="asm"
 				Filter="*.asm;*.inc"
 				>
-				<File
-					RelativePath="..\..\..\decoder\core\asm\asm_inc.asm"
-					>
-					<FileConfiguration
-						Name="Release|Win32"
-						>
-						<Tool
-							Name="VCCustomBuildTool"
-							CommandLine="nasm  -I$(InputDir) -f win32 -O3 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
-							Outputs="$(IntDir)\$(InputName).obj"
-						/>
-					</FileConfiguration>
-					<FileConfiguration
-						Name="Release|x64"
-						ExcludedFromBuild="true"
-						>
-						<Tool
-							Name="VCCustomBuildTool"
-							CommandLine="nasm  -I$(InputDir) -f win32 -O3 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
-							Outputs="$(IntDir)\$(InputName).obj"
-						/>
-					</FileConfiguration>
-					<FileConfiguration
-						Name="Debug|Win32"
-						>
-						<Tool
-							Name="VCCustomBuildTool"
-						/>
-					</FileConfiguration>
-					<FileConfiguration
-						Name="Debug|x64"
-						ExcludedFromBuild="true"
-						>
-						<Tool
-							Name="VCCustomBuildTool"
-						/>
-					</FileConfiguration>
-				</File>
 				<File
 					RelativePath="..\..\..\decoder\core\asm\block_add.asm"
 					>
@@ -394,17 +356,16 @@
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm -I$(InputDir) -f win32 -O3 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win32 -DPREFIX -DX86_32 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
 					<FileConfiguration
 						Name="Release|x64"
-						ExcludedFromBuild="true"
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm -I$(InputDir) -f win32 -O3 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
@@ -413,40 +374,38 @@
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win32 -DPREFIX -DX86_32 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
 					<FileConfiguration
 						Name="Debug|x64"
-						ExcludedFromBuild="true"
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
 				</File>
 				<File
-					RelativePath="..\..\..\decoder\core\asm\cpuid.asm"
+					RelativePath="..\..\..\common\cpuid.asm"
 					>
 					<FileConfiguration
 						Name="Release|Win32"
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm  -I$(InputDir) -f win32 -O3 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win32 -DPREFIX -DX86_32 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
 					<FileConfiguration
 						Name="Release|x64"
-						ExcludedFromBuild="true"
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm  -I$(InputDir) -f win32 -O3 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
@@ -455,17 +414,16 @@
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win32 -DPREFIX -DX86_32 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
 					<FileConfiguration
 						Name="Debug|x64"
-						ExcludedFromBuild="true"
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
@@ -478,17 +436,16 @@
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm -I$(InputDir) -f win32 -O3 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win32 -DPREFIX -DX86_32 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
 					<FileConfiguration
 						Name="Release|x64"
-						ExcludedFromBuild="true"
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm -I$(InputDir) -f win32 -O3 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
@@ -497,40 +454,38 @@
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm  -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win32 -DPREFIX -DX86_32 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
 					<FileConfiguration
 						Name="Debug|x64"
-						ExcludedFromBuild="true"
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm  -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
 				</File>
 				<File
-					RelativePath="..\..\..\decoder\core\asm\deblock.asm"
+					RelativePath="..\..\..\common\deblock.asm"
 					>
 					<FileConfiguration
 						Name="Release|Win32"
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm -I$(InputDir) -f win32 -O3 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win32 -DPREFIX -DX86_32 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
 					<FileConfiguration
 						Name="Release|x64"
-						ExcludedFromBuild="true"
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm -I$(InputDir) -f win32 -O3 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
@@ -539,40 +494,38 @@
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win32 -DPREFIX -DX86_32 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
 					<FileConfiguration
 						Name="Debug|x64"
-						ExcludedFromBuild="true"
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
 				</File>
 				<File
-					RelativePath="..\..\..\decoder\core\asm\expand_picture.asm"
+					RelativePath="..\..\..\common\expand_picture.asm"
 					>
 					<FileConfiguration
 						Name="Release|Win32"
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm -I$(InputDir) -f win32 -O3 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win32 -DPREFIX -DX86_32 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
 					<FileConfiguration
 						Name="Release|x64"
-						ExcludedFromBuild="true"
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm -I$(InputDir) -f win32 -O3 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
@@ -581,17 +534,16 @@
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win32 -DPREFIX -DX86_32 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
 					<FileConfiguration
 						Name="Debug|x64"
-						ExcludedFromBuild="true"
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
@@ -604,17 +556,16 @@
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm -I$(InputDir) -f win32 -O3 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win32 -DPREFIX -DX86_32 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
 					<FileConfiguration
 						Name="Release|x64"
-						ExcludedFromBuild="true"
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm -I$(InputDir) -f win32 -O3 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
@@ -623,40 +574,38 @@
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win32 -DPREFIX -DX86_32 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
 					<FileConfiguration
 						Name="Debug|x64"
-						ExcludedFromBuild="true"
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
 				</File>
 				<File
-					RelativePath="..\..\..\decoder\core\asm\mb_copy.asm"
+					RelativePath="..\..\..\common\mb_copy.asm"
 					>
 					<FileConfiguration
 						Name="Release|Win32"
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm -I$(InputDir) -f win32 -O3 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win32 -DPREFIX -DX86_32 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
 					<FileConfiguration
 						Name="Release|x64"
-						ExcludedFromBuild="true"
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm -I$(InputDir) -f win32 -O3 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
@@ -665,40 +614,38 @@
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm  -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win32 -DPREFIX -DX86_32 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
 					<FileConfiguration
 						Name="Debug|x64"
-						ExcludedFromBuild="true"
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm  -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
 				</File>
 				<File
-					RelativePath="..\..\..\decoder\core\asm\mc_chroma.asm"
+					RelativePath="..\..\..\common\mc_chroma.asm"
 					>
 					<FileConfiguration
 						Name="Release|Win32"
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm -I$(InputDir) -f win32 -O3 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win32 -DPREFIX -DX86_32 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
 					<FileConfiguration
 						Name="Release|x64"
-						ExcludedFromBuild="true"
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm -I$(InputDir) -f win32 -O3 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
@@ -707,40 +654,38 @@
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm  -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win32 -DPREFIX -DX86_32 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
 					<FileConfiguration
 						Name="Debug|x64"
-						ExcludedFromBuild="true"
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm  -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
 				</File>
 				<File
-					RelativePath="..\..\..\decoder\core\asm\mc_luma.asm"
+					RelativePath="..\..\..\common\mc_luma.asm"
 					>
 					<FileConfiguration
 						Name="Release|Win32"
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm  -I$(InputDir) -f win32 -O3 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win32 -DPREFIX -DX86_32 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
 					<FileConfiguration
 						Name="Release|x64"
-						ExcludedFromBuild="true"
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm  -I$(InputDir) -f win32 -O3 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
@@ -749,59 +694,16 @@
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win32 -DPREFIX -DX86_32 -o $(IntDir)\$(InputName).obj $(InputPath)"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>
 					<FileConfiguration
 						Name="Debug|x64"
-						ExcludedFromBuild="true"
 						>
 						<Tool
 							Name="VCCustomBuildTool"
-							CommandLine="nasm -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+							CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)"
-							Outputs="$(IntDir)\$(InputName).obj"
-						/>
-					</FileConfiguration>
-				</File>
-				<File
-					RelativePath="..\..\..\decoder\core\asm\memzero.asm"
-					>
-					<FileConfiguration
-						Name="Release|Win32"
-						>
-						<Tool
-							Name="VCCustomBuildTool"
-							CommandLine="nasm  -I$(InputDir) -f win32 -O3 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
-							Outputs="$(IntDir)\$(InputName).obj"
-						/>
-					</FileConfiguration>
-					<FileConfiguration
-						Name="Release|x64"
-						ExcludedFromBuild="true"
-						>
-						<Tool
-							Name="VCCustomBuildTool"
-							CommandLine="nasm  -I$(InputDir) -f win32 -O3 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
-							Outputs="$(IntDir)\$(InputName).obj"
-						/>
-					</FileConfiguration>
-					<FileConfiguration
-						Name="Debug|Win32"
-						>
-						<Tool
-							Name="VCCustomBuildTool"
-							CommandLine="nasm -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
-							Outputs="$(IntDir)\$(InputName).obj"
-						/>
-					</FileConfiguration>
-					<FileConfiguration
-						Name="Debug|x64"
-						ExcludedFromBuild="true"
-						>
-						<Tool
-							Name="VCCustomBuildTool"
-							CommandLine="nasm -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
 							Outputs="$(IntDir)\$(InputName).obj"
 						/>
 					</FileConfiguration>

codec/build/win32/dec/WelsDecCore_2010.vcxproj

@@ -94,8 +94,8 @@
     <ClCompile>
       <Optimization>MaxSpeed</Optimization>
       <InlineFunctionExpansion>OnlyExplicitInline</InlineFunctionExpansion>
-      <AdditionalIncludeDirectories>..\..\..\decoder\core\inc;..\..\..\common\inc;..\..\..\api\svc;..\..\..\hwDecoder\core\inc;..\..\..\hwDecoder\dxva\inc;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories>
+      <AdditionalIncludeDirectories>..\..\..\decoder\core\inc;..\..\..\common;..\..\..\api\svc;..\..\..\hwDecoder\core\inc;..\..\..\hwDecoder\dxva\inc;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories>
-      <PreprocessorDefinitions>WIN32;NDEBUG;_LIB;HAVE_CACHE_LINE_ALIGN;%(PreprocessorDefinitions)</PreprocessorDefinitions>
+      <PreprocessorDefinitions>WIN32;NDEBUG;X86_ASM;_LIB;HAVE_CACHE_LINE_ALIGN;%(PreprocessorDefinitions)</PreprocessorDefinitions>
       <StringPooling>true</StringPooling>
       <RuntimeLibrary>MultiThreadedDLL</RuntimeLibrary>
       <FunctionLevelLinking>true</FunctionLevelLinking>
@@ -125,8 +125,8 @@
     <ClCompile>
       <Optimization>MaxSpeed</Optimization>
       <InlineFunctionExpansion>OnlyExplicitInline</InlineFunctionExpansion>
-      <AdditionalIncludeDirectories>..\..\..\decoder\core\inc;..\..\..\common\inc;..\..\..\api\svc;..\..\..\hwDecoder\core\inc;..\..\..\hwDecoder\dxva\inc;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories>
+      <AdditionalIncludeDirectories>..\..\..\decoder\core\inc;..\..\..\common;..\..\..\api\svc;..\..\..\hwDecoder\core\inc;..\..\..\hwDecoder\dxva\inc;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories>
-      <PreprocessorDefinitions>WIN64;NDEBUG;_LIB;HAVE_CACHE_LINE_ALIGN;%(PreprocessorDefinitions)</PreprocessorDefinitions>
+      <PreprocessorDefinitions>WIN64;NDEBUG;X86_ASM;_LIB;HAVE_CACHE_LINE_ALIGN;%(PreprocessorDefinitions)</PreprocessorDefinitions>
       <StringPooling>true</StringPooling>
       <RuntimeLibrary>MultiThreadedDLL</RuntimeLibrary>
       <FunctionLevelLinking>true</FunctionLevelLinking>
@@ -151,11 +151,15 @@
       <SuppressStartupBanner>true</SuppressStartupBanner>
       <OutputFile>$(OutDir)\WelsDecCore.bsc</OutputFile>
     </Bscmake>
+    <CustomBuild>
+      <Outputs>$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command>nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win64  -DWIN64 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+    </CustomBuild>
   </ItemDefinitionGroup>
   <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">
     <ClCompile>
       <Optimization>Disabled</Optimization>
-      <AdditionalIncludeDirectories>..\..\..\decoder\core\inc;..\..\..\common\inc;..\..\..\api\svc;..\..\..\hwDecoder\core\inc;..\..\..\hwDecoder\dxva\inc;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories>
+      <AdditionalIncludeDirectories>..\..\..\decoder\core\inc;..\..\..\common;..\..\..\api\svc;..\..\..\hwDecoder\core\inc;..\..\..\hwDecoder\dxva\inc;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories>
       <PreprocessorDefinitions>WIN32;_DEBUG;_LIB;X86_ASM;HAVE_CACHE_LINE_ALIGN;%(PreprocessorDefinitions)</PreprocessorDefinitions>
       <MinimalRebuild>true</MinimalRebuild>
       <BasicRuntimeChecks>EnableFastChecks</BasicRuntimeChecks>
@@ -184,7 +188,7 @@
   <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">
     <ClCompile>
       <Optimization>Disabled</Optimization>
-      <AdditionalIncludeDirectories>..\..\..\decoder\core\inc;..\..\..\common\inc;..\..\..\api\svc;..\..\..\hwDecoder\core\inc;..\..\..\hwDecoder\dxva\inc;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories>
+      <AdditionalIncludeDirectories>..\..\..\decoder\core\inc;..\..\..\common;..\..\..\api\svc;..\..\..\hwDecoder\core\inc;..\..\..\hwDecoder\dxva\inc;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories>
       <PreprocessorDefinitions>WIN64;_DEBUG;_LIB;HAVE_CACHE_LINE_ALIGN;%(PreprocessorDefinitions)</PreprocessorDefinitions>
       <BasicRuntimeChecks>EnableFastChecks</BasicRuntimeChecks>
       <RuntimeLibrary>MultiThreadedDebugDLL</RuntimeLibrary>
@@ -208,176 +212,45 @@
       <SuppressStartupBanner>true</SuppressStartupBanner>
       <OutputFile>$(OutDir)\WelsDecCore.bsc</OutputFile>
     </Bscmake>
+    <CustomBuild>
+      <Command>nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win64  -DWIN64 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs>$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+    </CustomBuild>
   </ItemDefinitionGroup>
   <ItemGroup>
-    <CustomBuild Include="..\..\..\decoder\core\asm\asm_inc.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
-    </CustomBuild>
     <CustomBuild Include="..\..\..\decoder\core\asm\block_add.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32  -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win64  -DWIN64 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win32  -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win64  -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
-    </CustomBuild>
-    <CustomBuild Include="..\..\..\decoder\core\asm\cpuid.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
     </CustomBuild>
     <CustomBuild Include="..\..\..\decoder\core\asm\dct.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32  -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win64  -DWIN64 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win32  -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win64  -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
-    </CustomBuild>
-    <CustomBuild Include="..\..\..\decoder\core\asm\deblock.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
-    </CustomBuild>
-    <CustomBuild Include="..\..\..\decoder\core\asm\expand_picture.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
     </CustomBuild>
     <CustomBuild Include="..\..\..\decoder\core\asm\intra_pred.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32  -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win64  -DWIN64 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win32  -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win64  -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
-    </CustomBuild>
-    <CustomBuild Include="..\..\..\decoder\core\asm\mb_copy.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
-    </CustomBuild>
-    <CustomBuild Include="..\..\..\decoder\core\asm\mc_chroma.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
-    </CustomBuild>
-    <CustomBuild Include="..\..\..\decoder\core\asm\mc_luma.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
-    </CustomBuild>
-    <CustomBuild Include="..\..\..\decoder\core\asm\memzero.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
     </CustomBuild>
   </ItemGroup>
   <ItemGroup>
+    <ClInclude Include="..\..\..\common\logging.h" />
     <ClInclude Include="..\..\..\decoder\core\inc\as264_common.h" />
     <ClInclude Include="..\..\..\decoder\core\inc\au_parser.h" />
     <ClInclude Include="..\..\..\decoder\core\inc\bit_stream.h" />
@@ -419,6 +292,7 @@
     <ClInclude Include="..\..\..\decoder\core\inc\wels_const.h" />
   </ItemGroup>
   <ItemGroup>
+    <ClCompile Include="..\..\..\common\logging.cpp" />
     <ClCompile Include="..\..\..\decoder\core\src\au_parser.cpp" />
     <ClCompile Include="..\..\..\decoder\core\src\bit_stream.cpp" />
     <ClCompile Include="..\..\..\decoder\core\src\cpu.cpp" />
@@ -441,6 +315,68 @@
     <ClCompile Include="..\..\..\decoder\core\src\decoder_core.cpp" />
     <ClCompile Include="..\..\..\decoder\core\src\utils.cpp" />
   </ItemGroup>
+  <ItemGroup>
+    <CustomBuild Include="..\..\..\common\cpuid.asm">
+      <FileType>Document</FileType>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32  -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win32  -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win64  -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win64  -DWIN64 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+    </CustomBuild>
+    <CustomBuild Include="..\..\..\common\deblock.asm">
+      <FileType>Document</FileType>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32  -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win32  -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win64  -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win64  -DWIN64 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+    </CustomBuild>
+    <CustomBuild Include="..\..\..\common\expand_picture.asm">
+      <FileType>Document</FileType>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32  -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win32  -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win64  -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win64  -DWIN64 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+    </CustomBuild>
+    <CustomBuild Include="..\..\..\common\mb_copy.asm">
+      <FileType>Document</FileType>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32  -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win32  -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win64  -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win64  -DWIN64 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+    </CustomBuild>
+    <CustomBuild Include="..\..\..\common\mc_chroma.asm">
+      <FileType>Document</FileType>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32  -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win32  -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win64  -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win64  -DWIN64 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+    </CustomBuild>
+    <CustomBuild Include="..\..\..\common\mc_luma.asm">
+      <FileType>Document</FileType>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32  -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win32  -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win64  -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm  -I ..\..\..\common\ -I%(RootDir)%(Directory) -f win64  -DWIN64 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+    </CustomBuild>
+  </ItemGroup>
   <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
   <ImportGroup Label="ExtensionTargets">
     <Import Project="$(VCTargetsPath)\BuildCustomizations\masm.targets" />

codec/build/win32/dec/WelsDecCore_2010.vcxproj.filters

@@ -64,6 +64,9 @@
     <ClCompile Include="..\..\..\decoder\core\src\utils.cpp">
       <Filter>sources</Filter>
     </ClCompile>
+    <ClCompile Include="..\..\..\common\logging.cpp">
+      <Filter>sources</Filter>
+    </ClCompile>
   </ItemGroup>
   <ItemGroup>
     <ClInclude Include="..\..\..\decoder\core\inc\as264_common.h">
@@ -183,39 +186,36 @@
     <ClInclude Include="..\..\..\decoder\core\inc\wels_common_basis.h">
       <Filter>headers</Filter>
     </ClInclude>
+    <ClInclude Include="..\..\..\common\logging.h">
+      <Filter>headers</Filter>
+    </ClInclude>
   </ItemGroup>
   <ItemGroup>
-    <CustomBuild Include="..\..\..\decoder\core\asm\asm_inc.asm">
-      <Filter>ASM</Filter>
-    </CustomBuild>
     <CustomBuild Include="..\..\..\decoder\core\asm\block_add.asm">
       <Filter>ASM</Filter>
     </CustomBuild>
-    <CustomBuild Include="..\..\..\decoder\core\asm\cpuid.asm">
-      <Filter>ASM</Filter>
-    </CustomBuild>
     <CustomBuild Include="..\..\..\decoder\core\asm\dct.asm">
       <Filter>ASM</Filter>
     </CustomBuild>
-    <CustomBuild Include="..\..\..\decoder\core\asm\deblock.asm">
-      <Filter>ASM</Filter>
-    </CustomBuild>
-    <CustomBuild Include="..\..\..\decoder\core\asm\expand_picture.asm">
-      <Filter>ASM</Filter>
-    </CustomBuild>
     <CustomBuild Include="..\..\..\decoder\core\asm\intra_pred.asm">
       <Filter>ASM</Filter>
     </CustomBuild>
-    <CustomBuild Include="..\..\..\decoder\core\asm\mb_copy.asm">
+    <CustomBuild Include="..\..\..\common\mc_luma.asm">
       <Filter>ASM</Filter>
     </CustomBuild>
-    <CustomBuild Include="..\..\..\decoder\core\asm\mc_chroma.asm">
+    <CustomBuild Include="..\..\..\common\mc_chroma.asm">
       <Filter>ASM</Filter>
     </CustomBuild>
-    <CustomBuild Include="..\..\..\decoder\core\asm\mc_luma.asm">
+    <CustomBuild Include="..\..\..\common\mb_copy.asm">
       <Filter>ASM</Filter>
     </CustomBuild>
-    <CustomBuild Include="..\..\..\decoder\core\asm\memzero.asm">
+    <CustomBuild Include="..\..\..\common\expand_picture.asm">
+      <Filter>ASM</Filter>
+    </CustomBuild>
+    <CustomBuild Include="..\..\..\common\deblock.asm">
+      <Filter>ASM</Filter>
+    </CustomBuild>
+    <CustomBuild Include="..\..\..\common\cpuid.asm">
       <Filter>ASM</Filter>
     </CustomBuild>
   </ItemGroup>

codec/build/win32/dec/WelsDecPlus_2010.vcxproj

@@ -107,7 +107,7 @@
     <ClCompile>
       <Optimization>MaxSpeed</Optimization>
       <InlineFunctionExpansion>OnlyExplicitInline</InlineFunctionExpansion>
-      <AdditionalIncludeDirectories>..\..\..\decoder\plus\inc;..\..\..\decoder\core\inc;..\..\..\api\svc;..\..\..\common;..\..\..\hwDecoder\plus\inc;..\..\..\hwDecoder\core\inc;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories>
+      <AdditionalIncludeDirectories>..\..\..\common;..\..\..\decoder\plus\inc;..\..\..\decoder\core\inc;..\..\..\api\svc;..\..\..\common;..\..\..\hwDecoder\plus\inc;..\..\..\hwDecoder\core\inc;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories>
       <PreprocessorDefinitions>WIN32;NDEBUG;_WINDOWS;_USRDLL;WELSDECPLUS_EXPORTS;HAVE_CACHE_LINE_ALIGN;%(PreprocessorDefinitions)</PreprocessorDefinitions>
       <StringPooling>true</StringPooling>
       <RuntimeLibrary>MultiThreadedDLL</RuntimeLibrary>
@@ -156,7 +156,7 @@
     <ClCompile>
       <Optimization>MaxSpeed</Optimization>
       <InlineFunctionExpansion>OnlyExplicitInline</InlineFunctionExpansion>
-      <AdditionalIncludeDirectories>..\..\..\decoder\plus\inc;..\..\..\decoder\core\inc;..\..\..\api\svc;..\..\..\common;..\..\..\hwDecoder\plus\inc;..\..\..\hwDecoder\core\inc;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories>
+      <AdditionalIncludeDirectories>..\..\..\common;..\..\..\decoder\plus\inc;..\..\..\decoder\core\inc;..\..\..\api\svc;..\..\..\common;..\..\..\hwDecoder\plus\inc;..\..\..\hwDecoder\core\inc;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories>
       <PreprocessorDefinitions>WIN32;NDEBUG;_WINDOWS;_USRDLL;WELSDECPLUS_EXPORTS;HAVE_CACHE_LINE_ALIGN;%(PreprocessorDefinitions)</PreprocessorDefinitions>
       <StringPooling>true</StringPooling>
       <RuntimeLibrary>MultiThreadedDLL</RuntimeLibrary>
@@ -204,7 +204,7 @@
     </Midl>
     <ClCompile>
       <Optimization>Disabled</Optimization>
-      <AdditionalIncludeDirectories>..\..\..\decoder\plus\inc;..\..\..\decoder\core\inc;..\..\..\api\svc;..\..\..\common;..\..\..\hwDecoder\plus\inc;..\..\..\hwDecoder\core\inc;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories>
+      <AdditionalIncludeDirectories>..\..\..\common;..\..\..\decoder\plus\inc;..\..\..\decoder\core\inc;..\..\..\api\svc;..\..\..\common;..\..\..\hwDecoder\plus\inc;..\..\..\hwDecoder\core\inc;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories>
       <PreprocessorDefinitions>WIN32;_DEBUG;_WINDOWS;_USRDLL;WELSDECPLUS_EXPORTS;HAVE_CACHE_LINE_ALIGN;%(PreprocessorDefinitions)</PreprocessorDefinitions>
       <MinimalRebuild>true</MinimalRebuild>
       <BasicRuntimeChecks>EnableFastChecks</BasicRuntimeChecks>

codec/build/win32/dec/decConsole_2010.vcxproj

@@ -102,7 +102,7 @@
     <ClCompile>
       <Optimization>MaxSpeed</Optimization>
       <InlineFunctionExpansion>OnlyExplicitInline</InlineFunctionExpansion>
-      <AdditionalIncludeDirectories>..\..\..\console\dec\inc;..\..\..\api\svc;..\..\..\common;..\..\..\encoder\core\inc;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories>
+      <AdditionalIncludeDirectories>..\..\..\common;..\..\..\console\dec\inc;..\..\..\api\svc;..\..\..\common;..\..\..\encoder\core\inc;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories>
       <PreprocessorDefinitions>WIN32;NDEBUG;_CONSOLE;%(PreprocessorDefinitions)</PreprocessorDefinitions>
       <StringPooling>true</StringPooling>
       <RuntimeLibrary>MultiThreadedDLL</RuntimeLibrary>
@@ -144,7 +144,7 @@
     <ClCompile>
       <Optimization>MaxSpeed</Optimization>
       <InlineFunctionExpansion>OnlyExplicitInline</InlineFunctionExpansion>
-      <AdditionalIncludeDirectories>..\..\..\console\dec\inc;..\..\..\api\svc;..\..\..\common;..\..\..\encoder\core\inc;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories>
+      <AdditionalIncludeDirectories>..\..\..\common;..\..\..\console\dec\inc;..\..\..\api\svc;..\..\..\common;..\..\..\encoder\core\inc;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories>
       <PreprocessorDefinitions>WIN32;NDEBUG;_CONSOLE;%(PreprocessorDefinitions)</PreprocessorDefinitions>
       <StringPooling>true</StringPooling>
       <RuntimeLibrary>MultiThreadedDLL</RuntimeLibrary>
@@ -227,7 +227,7 @@
     </Midl>
     <ClCompile>
       <Optimization>Disabled</Optimization>
-      <AdditionalIncludeDirectories>..\..\..\console\dec\inc;..\..\..\api\svc;..\..\..\common;..\..\..\encoder\core\inc;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories>
+      <AdditionalIncludeDirectories>..\..\..\common;..\..\..\console\dec\inc;..\..\..\api\svc;..\..\..\common;..\..\..\encoder\core\inc;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories>
       <PreprocessorDefinitions>WIN64;_DEBUG;_CONSOLE;%(PreprocessorDefinitions)</PreprocessorDefinitions>
       <BasicRuntimeChecks>EnableFastChecks</BasicRuntimeChecks>
       <RuntimeLibrary>MultiThreadedDebugDLL</RuntimeLibrary>

codec/build/win32/enc/WelsEncCore_2010.vcxproj

@@ -127,7 +127,7 @@
     <ClCompile>
       <Optimization>Disabled</Optimization>
       <AdditionalIncludeDirectories>..\..\..\encoder\core\inc;..\..\..\api\svc;..\..\..\WelsThreadLib\api;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories>
-      <PreprocessorDefinitions>WIN64;_DEBUG;_LIB;WELS_SVC;ENCODER_CORE;HAVE_CACHE_LINE_ALIGN;MT_ENABLED;%(PreprocessorDefinitions)</PreprocessorDefinitions>
+      <PreprocessorDefinitions>WIN64;_DEBUG;X86_ASM;_LIB;WELS_SVC;ENCODER_CORE;HAVE_CACHE_LINE_ALIGN;MT_ENABLED;%(PreprocessorDefinitions)</PreprocessorDefinitions>
       <BasicRuntimeChecks>EnableFastChecks</BasicRuntimeChecks>
       <RuntimeLibrary>MultiThreadedDebugDLL</RuntimeLibrary>
       <PrecompiledHeaderOutputFile>.\..\..\..\obj\encoder\core\Debug/WelsEncCore.pch</PrecompiledHeaderOutputFile>
@@ -197,7 +197,7 @@
       <FavorSizeOrSpeed>Speed</FavorSizeOrSpeed>
       <WholeProgramOptimization>true</WholeProgramOptimization>
       <AdditionalIncludeDirectories>..\..\..\encoder\core\inc;..\..\..\api\svc;..\..\..\WelsThreadLib\api;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories>
-      <PreprocessorDefinitions>WIN64;NDEBUG;_LIB;WELS_SVC;ENCODER_CORE;HAVE_CACHE_LINE_ALIGN;MT_ENABLED;%(PreprocessorDefinitions)</PreprocessorDefinitions>
+      <PreprocessorDefinitions>WIN64;NDEBUG;X86_ASM;_LIB;WELS_SVC;ENCODER_CORE;HAVE_CACHE_LINE_ALIGN;MT_ENABLED;%(PreprocessorDefinitions)</PreprocessorDefinitions>
       <StringPooling>true</StringPooling>
       <RuntimeLibrary>MultiThreadedDLL</RuntimeLibrary>
       <FunctionLevelLinking>true</FunctionLevelLinking>
@@ -565,255 +565,154 @@
     <ClInclude Include="..\..\..\encoder\core\inc\wels_preprocess.h" />
   </ItemGroup>
   <ItemGroup>
-    <CustomBuild Include="..\..\..\encoder\core\asm\asm_inc.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
-    </CustomBuild>
     <CustomBuild Include="..\..\..\encoder\core\asm\coeff.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
-    </CustomBuild>
-    <CustomBuild Include="..\..\..\encoder\core\asm\cpuid.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
     </CustomBuild>
     <CustomBuild Include="..\..\..\encoder\core\asm\dct.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
-    </CustomBuild>
-    <CustomBuild Include="..\..\..\encoder\core\asm\deblock.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
-    </CustomBuild>
-    <CustomBuild Include="..\..\..\encoder\core\asm\expand_picture.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
     </CustomBuild>
     <CustomBuild Include="..\..\..\encoder\core\asm\intra_pred.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
-    </CustomBuild>
-    <CustomBuild Include="..\..\..\encoder\core\asm\intra_pred_util.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
-    </CustomBuild>
-    <CustomBuild Include="..\..\..\encoder\core\asm\mb_copy.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
-    </CustomBuild>
-    <CustomBuild Include="..\..\..\encoder\core\asm\mc_chroma.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
-    </CustomBuild>
-    <CustomBuild Include="..\..\..\encoder\core\asm\mc_luma.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
     </CustomBuild>
     <CustomBuild Include="..\..\..\encoder\core\asm\memzero.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
     </CustomBuild>
     <CustomBuild Include="..\..\..\encoder\core\asm\quant.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
     </CustomBuild>
     <CustomBuild Include="..\..\..\encoder\core\asm\satd_sad.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
     </CustomBuild>
     <CustomBuild Include="..\..\..\encoder\core\asm\score.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
     </CustomBuild>
-    <CustomBuild Include="..\..\..\encoder\core\asm\vaa.asm">
+  </ItemGroup>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+  <ItemGroup>
-</Command>
+    <CustomBuild Include="..\..\..\common\cpuid.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <FileType>Document</FileType>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -O3 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+    </CustomBuild>
+    <CustomBuild Include="..\..\..\common\deblock.asm">
+      <FileType>Document</FileType>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+    </CustomBuild>
+    <CustomBuild Include="..\..\..\common\expand_picture.asm">
+      <FileType>Document</FileType>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+    </CustomBuild>
+    <CustomBuild Include="..\..\..\common\mb_copy.asm">
+      <FileType>Document</FileType>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+    </CustomBuild>
+    <CustomBuild Include="..\..\..\common\mc_chroma.asm">
+      <FileType>Document</FileType>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+    </CustomBuild>
+    <CustomBuild Include="..\..\..\common\mc_luma.asm">
+      <FileType>Document</FileType>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+    </CustomBuild>
+    <CustomBuild Include="..\..\..\common\vaa.asm">
+      <FileType>Document</FileType>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
     </CustomBuild>
   </ItemGroup>
   <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />

codec/build/win32/enc/WelsEncCore_2010.vcxproj.filters

@@ -278,39 +278,15 @@
     </ClInclude>
   </ItemGroup>
   <ItemGroup>
-    <CustomBuild Include="..\..\..\encoder\core\asm\asm_inc.asm">
-      <Filter>ASM</Filter>
-    </CustomBuild>
     <CustomBuild Include="..\..\..\encoder\core\asm\coeff.asm">
       <Filter>ASM</Filter>
     </CustomBuild>
-    <CustomBuild Include="..\..\..\encoder\core\asm\cpuid.asm">
-      <Filter>ASM</Filter>
-    </CustomBuild>
     <CustomBuild Include="..\..\..\encoder\core\asm\dct.asm">
       <Filter>ASM</Filter>
     </CustomBuild>
-    <CustomBuild Include="..\..\..\encoder\core\asm\deblock.asm">
-      <Filter>ASM</Filter>
-    </CustomBuild>
-    <CustomBuild Include="..\..\..\encoder\core\asm\expand_picture.asm">
-      <Filter>ASM</Filter>
-    </CustomBuild>
     <CustomBuild Include="..\..\..\encoder\core\asm\intra_pred.asm">
       <Filter>ASM</Filter>
     </CustomBuild>
-    <CustomBuild Include="..\..\..\encoder\core\asm\intra_pred_util.asm">
-      <Filter>ASM</Filter>
-    </CustomBuild>
-    <CustomBuild Include="..\..\..\encoder\core\asm\mb_copy.asm">
-      <Filter>ASM</Filter>
-    </CustomBuild>
-    <CustomBuild Include="..\..\..\encoder\core\asm\mc_chroma.asm">
-      <Filter>ASM</Filter>
-    </CustomBuild>
-    <CustomBuild Include="..\..\..\encoder\core\asm\mc_luma.asm">
-      <Filter>ASM</Filter>
-    </CustomBuild>
     <CustomBuild Include="..\..\..\encoder\core\asm\memzero.asm">
       <Filter>ASM</Filter>
     </CustomBuild>
@@ -323,7 +299,25 @@
     <CustomBuild Include="..\..\..\encoder\core\asm\score.asm">
       <Filter>ASM</Filter>
     </CustomBuild>
-    <CustomBuild Include="..\..\..\encoder\core\asm\vaa.asm">
+    <CustomBuild Include="..\..\..\common\mc_luma.asm">
+      <Filter>ASM</Filter>
+    </CustomBuild>
+    <CustomBuild Include="..\..\..\common\mc_chroma.asm">
+      <Filter>ASM</Filter>
+    </CustomBuild>
+    <CustomBuild Include="..\..\..\common\mb_copy.asm">
+      <Filter>ASM</Filter>
+    </CustomBuild>
+    <CustomBuild Include="..\..\..\common\expand_picture.asm">
+      <Filter>ASM</Filter>
+    </CustomBuild>
+    <CustomBuild Include="..\..\..\common\deblock.asm">
+      <Filter>ASM</Filter>
+    </CustomBuild>
+    <CustomBuild Include="..\..\..\common\cpuid.asm">
+      <Filter>ASM</Filter>
+    </CustomBuild>
+    <CustomBuild Include="..\..\..\common\vaa.asm">
       <Filter>ASM</Filter>
     </CustomBuild>
   </ItemGroup>

codec/build/win32/enc/WelsEncoder_2008.sln

@@ -17,7 +17,7 @@ Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "encConsole", "encConsole.vc
 		{E8DFAFA1-8DAC-4127-8D27-FBD5819EE562} = {E8DFAFA1-8DAC-4127-8D27-FBD5819EE562}
 	EndProjectSection
 EndProject
-Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "WelsVP", "..\..\..\..\processing\build\win32\WelsVP_2008.vcproj", "{E8DFAFA1-8DAC-4127-8D27-FBD5819EE562}"
+Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "WelsVP", "..\..\..\processing\build\win32\WelsVP_2008.vcproj", "{E8DFAFA1-8DAC-4127-8D27-FBD5819EE562}"
 EndProject
 Global
 	GlobalSection(SolutionConfigurationPlatforms) = preSolution

codec/build/win32/enc/WelsEncoder_2010.sln

@@ -10,7 +10,7 @@ Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "WelsEncPlus_2010", "WelsEnc
 EndProject
 Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "encConsole_2010", "encConsole_2010.vcxproj", "{8509E2A8-2CBD-49E2-B564-3EFF1E927459}"
 EndProject
-Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "WelsVP_2010", "..\..\..\..\processing\build\win32\WelsVP_2010.vcxproj", "{E8DFAFA1-8DAC-4127-8D27-FBD5819EE562}"
+Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "WelsVP_2010", "..\..\..\processing\build\win32\WelsVP_2010.vcxproj", "{E8DFAFA1-8DAC-4127-8D27-FBD5819EE562}"
 EndProject
 Global
 	GlobalSection(SolutionConfigurationPlatforms) = preSolution

codec/common/asm_inc.asm

@@ -55,12 +55,286 @@
 	%define WELSEMMS
 %endif
 
-BITS 32
 
 ;***********************************************************************
 ; Macros
 ;***********************************************************************
 
+DEFAULT REL
+
+%ifdef WIN64 ; Windows x64 ;************************************
+
+BITS 64
+
+%define arg1 rcx
+%define arg2 rdx
+%define arg3 r8
+%define arg4 r9
+%define arg5 [rsp + push_num*8 + 40]
+%define arg6 [rsp + push_num*8 + 48]
+%define arg7 [rsp + push_num*8 + 56]
+%define arg8 [rsp + push_num*8 + 64]
+%define arg9 [rsp + push_num*8 + 72]
+%define arg10 [rsp + push_num*8 + 80]
+
+%define r0 rcx
+%define r1 rdx
+%define r2 r8
+%define r3 r9
+%define r4 rax
+%define r5 r10
+%define r6 r11
+%define r7 rsp
+
+%define r0d ecx
+%define r1d edx
+%define r2d r8d
+%define r3d r9d
+%define r4d eax
+%define r5d r10d
+%define r6d r11d
+
+%define r0w  cx
+%define r1w  dx
+%define r2w  r8w
+%define r3w  r9w
+
+%define r0b  cl
+%define r1b  dl
+%define r2b  r8l
+%define r3b  r9l
+
+%define  PUSHRFLAGS     pushfq
+%define  POPRFLAGS      popfq
+%define  retrq          rax
+%define  retrd          eax
+
+%elifdef UNIX64 ; Unix x64 ;************************************
+
+BITS 64
+
+%define arg1 rdi
+%define arg2 rsi
+%define arg3 rdx
+%define arg4 rcx
+%define arg5 r8
+%define arg6 r9
+%define arg7 [rsp + push_num*8 + 8]
+%define arg8 [rsp + push_num*8 + 16]
+%define arg9 [rsp + push_num*8 + 24]
+%define arg10 [rsp + push_num*8 + 32]
+
+%define r0 rdi
+%define r1 rsi
+%define r2 rdx
+%define r3 rcx
+%define r4 r8
+%define r5 r9
+%define r6 r10
+%define r7 rsp
+
+%define r0d edi
+%define r1d esi
+%define r2d edx
+%define r3d ecx
+%define r4d r8d
+%define r5d r9d
+%define r6d r10d
+
+%define r0w  di
+%define r1w  si
+%define r2w  dx
+%define r3w  cx
+
+%define r0b  dil
+%define r1b  sil
+%define r2b  dl
+%define r3b  cl
+
+%define  PUSHRFLAGS     pushfq
+%define  POPRFLAGS      popfq
+%define  retrq          rax
+%define  retrd          eax 
+
+%elifdef X86_32 ; X86_32 ;************************************
+
+BITS 32
+
+%define arg1 [esp + push_num*4 + 4]
+%define arg2 [esp + push_num*4 + 8]
+%define arg3 [esp + push_num*4 + 12]
+%define arg4 [esp + push_num*4 + 16]
+%define arg5 [esp + push_num*4 + 20]
+%define arg6 [esp + push_num*4 + 24]
+%define arg7 [esp + push_num*4 + 28]
+%define arg8 [esp + push_num*4 + 32]
+%define arg9 [esp + push_num*4 + 36]
+%define arg10 [esp + push_num*4 + 40]
+
+%define r0 eax
+%define r1 ecx
+%define r2 edx
+%define r3 ebx
+%define r4 esi
+%define r5 edi
+%define r6 ebp
+%define r7 esp
+
+%define r0d eax
+%define r1d ecx
+%define r2d edx
+%define r3d ebx
+%define r4d esi
+%define r5d edi
+%define r6d ebp
+
+%define r0w ax
+%define r1w cx
+%define r2w dx
+%define r3w bx
+
+%define r0b al
+%define r1b cl
+%define r2b dl
+%define r3b bl
+
+%define  PUSHRFLAGS     pushfd
+%define  POPRFLAGS      popfd
+%define  retrq          eax      ; 32 bit mode do not support 64 bits regesters
+%define  retrd          eax
+
+%endif
+
+%macro LOAD_PARA 2
+    mov %1, %2
+%endmacro
+
+%macro LOAD_1_PARA 0
+    %ifdef X86_32
+	mov r0, [esp + push_num*4 + 4]
+    %endif
+%endmacro
+
+%macro LOAD_2_PARA 0
+    %ifdef X86_32
+        mov r0, [esp + push_num*4 + 4]
+        mov r1, [esp + push_num*4 + 8]
+    %endif
+%endmacro
+
+%macro LOAD_3_PARA 0
+    %ifdef X86_32
+        mov r0, [esp + push_num*4 + 4]
+	mov r1, [esp + push_num*4 + 8]
+	mov r2, [esp + push_num*4 + 12]
+    %endif
+%endmacro
+
+%macro LOAD_4_PARA 0
+    %ifdef X86_32
+        push r3
+        %assign  push_num push_num+1	
+        mov r0, [esp + push_num*4 + 4]
+        mov r1, [esp + push_num*4 + 8]
+        mov r2, [esp + push_num*4 + 12]
+        mov r3, [esp + push_num*4 + 16]
+    %endif
+%endmacro
+
+%macro LOAD_5_PARA 0
+    %ifdef X86_32
+        push r3
+        push r4
+        %assign  push_num push_num+2	
+        mov r0, [esp + push_num*4 + 4]
+        mov r1, [esp + push_num*4 + 8]
+        mov r2, [esp + push_num*4 + 12]
+        mov r3, [esp + push_num*4 + 16]
+        mov r4, [esp + push_num*4 + 20]
+    %elifdef WIN64
+        mov r4, [rsp + push_num*8 + 40]
+    %endif
+%endmacro
+
+%macro LOAD_6_PARA 0
+    %ifdef X86_32
+	push r3
+        push r4
+        push r5
+        %assign  push_num push_num+3	
+        mov r0, [esp + push_num*4 + 4]
+        mov r1, [esp + push_num*4 + 8]
+        mov r2, [esp + push_num*4 + 12]
+        mov r3, [esp + push_num*4 + 16]
+        mov r4, [esp + push_num*4 + 20]
+        mov r5, [esp + push_num*4 + 24]
+    %elifdef WIN64
+        mov r4, [rsp + push_num*8 + 40]
+        mov r5, [rsp + push_num*8 + 48]
+    %endif
+%endmacro
+
+%macro LOAD_7_PARA 0
+    %ifdef X86_32
+        push r3
+        push r4
+        push r5
+        push r6
+        %assign  push_num push_num+4	
+        mov r0, [esp + push_num*4 + 4]
+        mov r1, [esp + push_num*4 + 8]
+        mov r2, [esp + push_num*4 + 12]
+        mov r3, [esp + push_num*4 + 16]
+        mov r4, [esp + push_num*4 + 20]
+        mov r5, [esp + push_num*4 + 24]
+        mov r6, [esp + push_num*4 + 28]
+    %elifdef WIN64
+        mov r4, [rsp + push_num*8 + 40]
+        mov r5, [rsp + push_num*8 + 48]
+        mov r6, [rsp + push_num*8 + 56]
+    %elifdef UNIX64
+        mov r6, [rsp + push_num*8 + 8]
+    %endif
+%endmacro
+
+
+
+%macro LOAD_4_PARA_POP 0
+    %ifdef X86_32
+	pop r3
+    %endif
+%endmacro
+
+%macro LOAD_5_PARA_POP 0
+    %ifdef X86_32
+        pop r4
+	pop r3
+    %endif
+%endmacro
+
+%macro LOAD_6_PARA_POP 0
+    %ifdef X86_32
+        pop r5
+  	pop r4
+ 	pop r3
+    %endif
+%endmacro
+
+%macro LOAD_7_PARA_POP 0
+    %ifdef X86_32
+        pop r6
+        pop r5
+        pop r4
+        pop r3
+    %endif
+%endmacro
+
+%macro SIGN_EXTENTION 2
+    %ifndef X86_32
+            movsx %1, %2
+    %endif
+%endmacro
+ 
 %macro WELS_EXTERN 1
     %ifdef PREFIX
         global _%1

codec/common/cpuid.asm

@@ -39,20 +39,12 @@
 ;*
 ;*************************************************************************/
 
-bits 32
+%include "asm_inc.asm"
 
 ;******************************************************************************************
 ; Macros
 ;******************************************************************************************
 
-%macro WELS_EXTERN 1
-	%ifdef PREFIX
-		global _%1
-		%define %1 _%1
-	%else
-		global %1
-	%endif
-%endmacro
 
 ;******************************************************************************************
 ; Code
@@ -69,13 +61,16 @@ ALIGN 16
 ;   int32_t WelsCPUIdVerify()
 ;******************************************************************************************
 WelsCPUIdVerify:
-    pushfd					; decrease the SP by 4 and load EFLAGS register onto stack, pushfd 32 bit and pushf for 16 bit
+    push    r1
-	pushfd					; need push 2 EFLAGS, one for processing and the another one for storing purpose
+    PUSHRFLAGS
-    pop     ecx				; get EFLAGS to bit manipulation
+    PUSHRFLAGS
-    mov     eax, ecx		; store into ecx followed
+
-    xor     eax, 00200000h	; get ID flag (bit 21) of EFLAGS to directly indicate cpuid support or not
+    pop      r1
-	xor		eax, ecx		; get the ID flag bitwise, eax - 0: not support; otherwise: support
+    mov      eax, r1d
-    popfd					; store back EFLAGS and keep unchanged for system
+    xor      eax, 00200000h
+    xor      eax, r1d
+    POPRFLAGS
+    pop      r1
     ret
 
 WELS_EXTERN WelsCPUId
@@ -83,6 +78,44 @@ ALIGN 16
 ;****************************************************************************************************
 ;   void WelsCPUId( int32_t uiIndex, int32_t *pFeatureA, int32_t *pFeatureB, int32_t *pFeatureC, int32_t *pFeatureD )
 ;****************************************************************************************************
+%ifdef       WIN64
+
+WelsCPUId:
+    push     rbx        
+    push     rdx    
+ 
+    mov      eax,     ecx
+    cpuid  
+    mov      [r9],    ecx
+    mov      [r8],    ebx
+    mov      rcx,    [rsp + 2*8 + 40]        
+    mov      [rcx],   edx
+    pop      rdx 
+    mov      [rdx],   eax
+
+    pop      rbx
+    ret
+
+%elifdef     UNIX64
+WelsCPUId:
+    push     rbx
+    push     rcx
+    push     rdx
+
+    mov      eax,     edi    
+    cpuid
+    mov      [r8],    edx
+    pop      rdx    
+    pop      r8
+    mov      [r8],   ecx
+    mov      [rdx],   ebx
+    mov      [rsi],   eax
+
+    pop      rbx
+    ret
+
+%elifdef     X86_32
+
 WelsCPUId:
     push	ebx
     push	edi
@@ -104,6 +137,8 @@ WelsCPUId:
     pop     ebx
     ret
 
+%endif
+
 WELS_EXTERN WelsCPUSupportAVX
 ; need call after cpuid=1 and eax, ecx flag got then
 ALIGN 16
@@ -111,8 +146,16 @@ ALIGN 16
 ;   int32_t WelsCPUSupportAVX( uint32_t eax, uint32_t ecx )
 ;****************************************************************************************************
 WelsCPUSupportAVX:
+%ifdef     WIN64
+        mov   eax,    ecx
+        mov   ecx,    edx
+%elifdef   UNIX64
+        mov eax, edi
+        mov ecx, esi
+%else 
         mov eax, [esp+4]
         mov ecx, [esp+8]  
+%endif
 
         ; refer to detection of AVX addressed in INTEL AVX manual document
         and ecx, 018000000H
@@ -130,6 +173,7 @@ avx_not_supported:
         mov eax, 0
         ret
 
+
 WELS_EXTERN  WelsCPUSupportFMA
 ; need call after cpuid=1 and eax, ecx flag got then
 ALIGN 16
@@ -137,9 +181,16 @@ ALIGN 16
 ;   int32_t WelsCPUSupportFMA( uint32_t eax, uint32_t ecx )
 ;****************************************************************************************************
 WelsCPUSupportFMA:
+%ifdef     WIN64
+        mov   eax,   ecx
+        mov   ecx,   edx
+%elifdef   UNIX64
+        mov   eax,   edi
+        mov   ecx,   esi
+%else
 	mov eax, [esp+4]
 	mov ecx, [esp+8]
-
+%endif
 	; refer to detection of FMA addressed in INTEL AVX manual document
 	and ecx, 018001000H
 	cmp ecx, 018001000H		; check OSXSAVE, AVX, FMA feature flags

codec/common/expand_picture.asm

@@ -0,0 +1,740 @@
+;*!
+;* \copy
+;*     Copyright (c)  2009-2013, Cisco Systems
+;*     All rights reserved.
+;*
+;*     Redistribution and use in source and binary forms, with or without
+;*     modification, are permitted provided that the following conditions
+;*     are met:
+;*
+;*        * Redistributions of source code must retain the above copyright
+;*          notice, this list of conditions and the following disclaimer.
+;*
+;*        * Redistributions in binary form must reproduce the above copyright
+;*          notice, this list of conditions and the following disclaimer in
+;*          the documentation and/or other materials provided with the
+;*          distribution.
+;*
+;*     THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+;*     "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+;*     LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
+;*     FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
+;*     COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
+;*     INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
+;*     BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+;*     LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+;*     CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+;*     LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
+;*     ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+;*     POSSIBILITY OF SUCH DAMAGE.
+;*
+;*
+;*  expand_picture.asm
+;*
+;*  Abstract
+;*      mmxext/sse for expand_frame
+;*
+;*  History
+;*      09/25/2009 Created
+;*
+;*
+;*************************************************************************/
+
+%include "asm_inc.asm"
+
+
+
+;***********************************************************************
+; Macros and other preprocessor constants
+;***********************************************************************
+
+;***********************************************************************
+; Local Data (Read Only)
+;***********************************************************************
+
+;SECTION .rodata pData align=16
+
+;***********************************************************************
+; Various memory constants (trigonometric values or rounding values)
+;***********************************************************************
+;%define PADDING_SIZE_ASM 	32 	; PADDING_LENGTH
+
+;***********************************************************************
+; Code
+;***********************************************************************
+
+
+
+SECTION .text
+
+WELS_EXTERN ExpandPictureLuma_sse2
+WELS_EXTERN ExpandPictureChromaAlign_sse2	; for chroma alignment
+WELS_EXTERN ExpandPictureChromaUnalign_sse2	; for chroma unalignment
+
+;;;;;;;expanding result;;;;;;;
+
+;aaaa|attttttttttttttttb|bbbb
+;aaaa|attttttttttttttttb|bbbb
+;aaaa|attttttttttttttttb|bbbb
+;aaaa|attttttttttttttttb|bbbb
+;----------------------------
+;aaaa|attttttttttttttttb|bbbb
+;llll|l                r|rrrr
+;llll|l                r|rrrr
+;llll|l                r|rrrr
+;llll|l                r|rrrr
+;llll|l                r|rrrr
+;cccc|ceeeeeeeeeeeeeeeed|dddd
+;----------------------------
+;cccc|ceeeeeeeeeeeeeeeed|dddd
+;cccc|ceeeeeeeeeeeeeeeed|dddd
+;cccc|ceeeeeeeeeeeeeeeed|dddd
+;cccc|ceeeeeeeeeeeeeeeed|dddd
+
+%macro mov_line_8x4_mmx		3	; dst, stride, mm?
+	movq [%1], %3
+	movq [%1+%2], %3
+	lea %1, [%1+2*%2]
+	movq [%1], %3
+	movq [%1+%2], %3
+	lea %1, [%1+2*%2]
+%endmacro
+
+%macro mov_line_end8x4_mmx		3	; dst, stride, mm?
+	movq [%1], %3
+	movq [%1+%2], %3
+	lea %1, [%1+2*%2]
+	movq [%1], %3
+	movq [%1+%2], %3
+	lea %1, [%1+%2]
+%endmacro
+
+%macro mov_line_16x4_sse2	4	; dst, stride, xmm?, u/a
+	movdq%4 [%1], %3 		; top(bottom)_0
+	movdq%4 [%1+%2], %3		; top(bottom)_1
+	lea %1, [%1+2*%2]
+	movdq%4 [%1], %3 		; top(bottom)_2
+	movdq%4 [%1+%2], %3		; top(bottom)_3
+	lea %1, [%1+2*%2]
+%endmacro
+
+%macro mov_line_end16x4_sse2	4	; dst, stride, xmm?, u/a
+	movdq%4 [%1], %3 		; top(bottom)_0
+	movdq%4 [%1+%2], %3		; top(bottom)_1
+	lea %1, [%1+2*%2]
+	movdq%4 [%1], %3 		; top(bottom)_2
+	movdq%4 [%1+%2], %3		; top(bottom)_3
+	lea %1, [%1+%2]
+%endmacro
+
+%macro mov_line_32x4_sse2	3	; dst, stride, xmm?
+	movdqa [%1], %3 		; top(bottom)_0
+	movdqa [%1+16], %3 		; top(bottom)_0
+	movdqa [%1+%2], %3		; top(bottom)_1
+	movdqa [%1+%2+16], %3		; top(bottom)_1
+	lea %1, [%1+2*%2]
+	movdqa [%1], %3 		; top(bottom)_2
+	movdqa [%1+16], %3 		; top(bottom)_2
+	movdqa [%1+%2], %3		; top(bottom)_3
+	movdqa [%1+%2+16], %3		; top(bottom)_3
+	lea %1, [%1+2*%2]
+%endmacro
+
+%macro mov_line_end32x4_sse2	3	; dst, stride, xmm?
+	movdqa [%1], %3 		; top(bottom)_0
+	movdqa [%1+16], %3 		; top(bottom)_0
+	movdqa [%1+%2], %3		; top(bottom)_1
+	movdqa [%1+%2+16], %3		; top(bottom)_1
+	lea %1, [%1+2*%2]
+	movdqa [%1], %3 		; top(bottom)_2
+	movdqa [%1+16], %3 		; top(bottom)_2
+	movdqa [%1+%2], %3		; top(bottom)_3
+	movdqa [%1+%2+16], %3		; top(bottom)_3
+	lea %1, [%1+%2]
+%endmacro
+
+%macro exp_top_bottom_sse2	1	; iPaddingSize [luma(32)/chroma(16)]
+    ;r2 [width/16(8)]
+    ;r0 [pSrc +0], r5 [pSrc -width] r1[-stride], 32(16) ;top
+    ;r3 [pSrc +(h-1)*stride], r4 [pSrc + (h+31)*stride],32(16); bottom
+
+%if %1 == 32		; for luma
+	sar r2, 04h 	; width / 16(8) pixels
+.top_bottom_loops:
+	; top
+	movdqa xmm0, [r0]		; first line of picture pData
+	mov_line_16x4_sse2 r5, r1, xmm0, a	; dst, stride, xmm?
+	mov_line_16x4_sse2 r5, r1, xmm0, a
+	mov_line_16x4_sse2 r5, r1, xmm0, a
+	mov_line_16x4_sse2 r5, r1, xmm0, a
+	mov_line_16x4_sse2 r5, r1, xmm0, a	; dst, stride, xmm?
+	mov_line_16x4_sse2 r5, r1, xmm0, a
+	mov_line_16x4_sse2 r5, r1, xmm0, a
+	mov_line_end16x4_sse2 r5, r1, xmm0, a
+
+	; bottom
+	movdqa xmm1, [r3] 		; last line of picture pData
+	mov_line_16x4_sse2 r4, r1, xmm1, a	; dst, stride, xmm?
+	mov_line_16x4_sse2 r4, r1, xmm1, a
+	mov_line_16x4_sse2 r4, r1, xmm1, a
+	mov_line_16x4_sse2 r4, r1, xmm1, a
+	mov_line_16x4_sse2 r4, r1, xmm1, a	; dst, stride, xmm?
+	mov_line_16x4_sse2 r4, r1, xmm1, a
+	mov_line_16x4_sse2 r4, r1, xmm1, a
+	mov_line_end16x4_sse2 r4, r1, xmm1, a
+
+	lea r0, [r0+16]		; top pSrc
+	lea r5, [r5+16]		; top dst
+	lea r3, [r3+16]		; bottom pSrc
+	lea r4, [r4+16]		; bottom dst
+	neg r1 			; positive/negative stride need for next loop?
+
+	dec r2
+	jnz near .top_bottom_loops
+%elif %1 == 16	; for chroma ??
+	mov r6, r2
+	sar r2, 04h 	; (width / 16) pixels
+.top_bottom_loops:
+	; top
+	movdqa xmm0, [r0]		; first line of picture pData
+	mov_line_16x4_sse2 r5, r1, xmm0, a	; dst, stride, xmm?
+	mov_line_16x4_sse2 r5, r1, xmm0, a
+	mov_line_16x4_sse2 r5, r1, xmm0, a
+	mov_line_end16x4_sse2 r5, r1, xmm0, a
+
+	; bottom
+	movdqa xmm1, [r3] 		; last line of picture pData
+	mov_line_16x4_sse2 r4, r1, xmm1, a	; dst, stride, xmm?
+	mov_line_16x4_sse2 r4, r1, xmm1, a
+	mov_line_16x4_sse2 r4, r1, xmm1, a
+	mov_line_end16x4_sse2 r4, r1, xmm1, a
+
+	lea r0, [r0+16]		; top pSrc
+	lea r5, [r5+16]		; top dst
+	lea r3, [r3+16]		; bottom pSrc
+	lea r4, [r4+16]		; bottom dst
+	neg r1 			; positive/negative stride need for next loop?
+
+	dec r2
+	jnz near .top_bottom_loops
+
+	; for remaining 8 bytes
+	and r6, 0fh		; any 8 bytes left?
+	test r6, r6
+	jz near .to_be_continued	; no left to exit here
+
+	; top
+	movq mm0, [r0]		; remained 8 byte
+	mov_line_8x4_mmx r5, r1, mm0	; dst, stride, mm?
+	mov_line_8x4_mmx r5, r1, mm0	; dst, stride, mm?
+	mov_line_8x4_mmx r5, r1, mm0	; dst, stride, mm?
+	mov_line_end8x4_mmx r5, r1, mm0	; dst, stride, mm?
+	; bottom
+	movq mm1, [r3]
+	mov_line_8x4_mmx r4, r1, mm1	; dst, stride, mm?
+	mov_line_8x4_mmx r4, r1, mm1	; dst, stride, mm?
+	mov_line_8x4_mmx r4, r1, mm1	; dst, stride, mm?
+	mov_line_end8x4_mmx r4, r1, mm1	; dst, stride, mm?
+	WELSEMMS
+
+.to_be_continued:
+%endif
+%endmacro
+
+%macro exp_left_right_sse2	2	; iPaddingSize [luma(32)/chroma(16)], u/a
+    ;r6 [height]
+    ;r0 [pSrc+0]  r5[pSrc-32] r1[stride]
+    ;r3 [pSrc+(w-1)] r4[pSrc+w] 
+
+%if %1 == 32		; for luma
+.left_right_loops:
+	; left
+	movzx r2d, byte [r0]		; pixel pData for left border
+	SSE2_Copy16Times	xmm0, r2d				; dst, tmp, pSrc [generic register name: a/b/c/d]
+	movdqa [r5], xmm0
+	movdqa [r5+16], xmm0
+
+	; right
+	movzx r2d, byte [r3]
+	SSE2_Copy16Times	xmm1, r2d				; dst, tmp, pSrc [generic register name: a/b/c/d]
+	movdqa [r4], xmm1
+	movdqa [r4+16], xmm1
+
+	lea r0, [r0+r1]		; left pSrc
+	lea r5, [r5+r1]		; left dst
+	lea r3, [r3+r1]		; right pSrc
+	lea r4, [r4+r1]		; right dst
+
+	dec r6
+	jnz near .left_right_loops
+%elif %1 == 16	; for chroma ??
+.left_right_loops:
+	; left
+	movzx r2d, byte [r0]		; pixel pData for left border
+	SSE2_Copy16Times	xmm0, r2d				; dst, tmp, pSrc [generic register name: a/b/c/d]
+	movdqa [r5], xmm0
+
+	; right
+	movzx r2d, byte [r3]
+	SSE2_Copy16Times	xmm1, r2d				; dst, tmp, pSrc [generic register name: a/b/c/d]
+	movdq%2 [r4], xmm1								; might not be aligned 16 bytes in case chroma planes
+
+	lea r0, [r0+r1]		; left pSrc
+	lea r5, [r5+r1]		; left dst
+	lea r3, [r3+r1]		; right pSrc
+	lea r4, [r4+r1]		; right dst
+
+	dec r6
+	jnz near .left_right_loops
+%endif
+%endmacro
+
+%macro exp_cross_sse2	2	; iPaddingSize [luma(32)/chroma(16)], u/a
+	; top-left: (x)mm3, top-right: (x)mm4, bottom-left: (x)mm5, bottom-right: (x)mm6
+	; edi: TL, ebp: TR, eax: BL, ebx: BR, ecx, -stride
+    ;r3:TL ,r4:TR,r5:BL,r6:BR r1:-stride
+%if %1 == 32		; luma
+	; TL
+	mov_line_32x4_sse2	r3, r1, xmm3	; dst, stride, xmm?
+	mov_line_32x4_sse2	r3, r1, xmm3	; dst, stride, xmm?
+	mov_line_32x4_sse2	r3, r1, xmm3	; dst, stride, xmm?
+	mov_line_32x4_sse2	r3, r1, xmm3	; dst, stride, xmm?
+	mov_line_32x4_sse2	r3, r1, xmm3	; dst, stride, xmm?
+	mov_line_32x4_sse2	r3, r1, xmm3	; dst, stride, xmm?
+	mov_line_32x4_sse2	r3, r1, xmm3	; dst, stride, xmm?
+	mov_line_end32x4_sse2	r3, r1, xmm3	; dst, stride, xmm?
+
+	; TR
+	mov_line_32x4_sse2	r4, r1, xmm4	; dst, stride, xmm?
+	mov_line_32x4_sse2	r4, r1, xmm4	; dst, stride, xmm?
+	mov_line_32x4_sse2	r4, r1, xmm4	; dst, stride, xmm?
+	mov_line_32x4_sse2	r4, r1, xmm4	; dst, stride, xmm?
+	mov_line_32x4_sse2	r4, r1, xmm4	; dst, stride, xmm?
+	mov_line_32x4_sse2	r4, r1, xmm4	; dst, stride, xmm?
+	mov_line_32x4_sse2	r4, r1, xmm4	; dst, stride, xmm?
+	mov_line_end32x4_sse2	r4, r1, xmm4	; dst, stride, xmm?
+
+	; BL
+	mov_line_32x4_sse2	r5, r1, xmm5	; dst, stride, xmm?
+	mov_line_32x4_sse2	r5, r1, xmm5	; dst, stride, xmm?
+	mov_line_32x4_sse2	r5, r1, xmm5	; dst, stride, xmm?
+	mov_line_32x4_sse2	r5, r1, xmm5	; dst, stride, xmm?
+	mov_line_32x4_sse2	r5, r1, xmm5	; dst, stride, xmm?
+	mov_line_32x4_sse2	r5, r1, xmm5	; dst, stride, xmm?
+	mov_line_32x4_sse2	r5, r1, xmm5	; dst, stride, xmm?
+	mov_line_end32x4_sse2	r5, r1, xmm5	; dst, stride, xmm?
+
+	; BR
+	mov_line_32x4_sse2	r6, r1, xmm6	; dst, stride, xmm?
+	mov_line_32x4_sse2	r6, r1, xmm6	; dst, stride, xmm?
+	mov_line_32x4_sse2	r6, r1, xmm6	; dst, stride, xmm?
+	mov_line_32x4_sse2	r6, r1, xmm6	; dst, stride, xmm?
+	mov_line_32x4_sse2	r6, r1, xmm6	; dst, stride, xmm?
+	mov_line_32x4_sse2	r6, r1, xmm6	; dst, stride, xmm?
+	mov_line_32x4_sse2	r6, r1, xmm6	; dst, stride, xmm?
+	mov_line_end32x4_sse2	r6, r1, xmm6	; dst, stride, xmm?
+%elif %1 == 16	; chroma
+	; TL
+	mov_line_16x4_sse2	r3, r1, xmm3, a	; dst, stride, xmm?
+	mov_line_16x4_sse2	r3, r1, xmm3, a	; dst, stride, xmm?
+	mov_line_16x4_sse2	r3, r1, xmm3, a	; dst, stride, xmm?
+	mov_line_end16x4_sse2	r3, r1, xmm3, a	; dst, stride, xmm?
+
+	; TR
+	mov_line_16x4_sse2	r4, r1, xmm4, %2	; dst, stride, xmm?
+	mov_line_16x4_sse2	r4, r1, xmm4, %2	; dst, stride, xmm?
+	mov_line_16x4_sse2	r4, r1, xmm4, %2	; dst, stride, xmm?
+	mov_line_end16x4_sse2 r4, r1, xmm4, %2	; dst, stride, xmm?
+
+	; BL
+	mov_line_16x4_sse2	r5, r1, xmm5, a	; dst, stride, xmm?
+	mov_line_16x4_sse2	r5, r1, xmm5, a	; dst, stride, xmm?
+	mov_line_16x4_sse2	r5, r1, xmm5, a	; dst, stride, xmm?
+	mov_line_end16x4_sse2	r5, r1, xmm5, a	; dst, stride, xmm?
+
+	; BR
+	mov_line_16x4_sse2	r6, r1, xmm6, %2	; dst, stride, xmm?
+	mov_line_16x4_sse2	r6, r1, xmm6, %2	; dst, stride, xmm?
+	mov_line_16x4_sse2	r6, r1, xmm6, %2	; dst, stride, xmm?
+	mov_line_end16x4_sse2	r6, r1, xmm6, %2	; dst, stride, xmm?
+%endif
+%endmacro
+
+ALIGN 16
+;***********************************************************************----------------
+; void ExpandPictureLuma_sse2(	uint8_t *pDst,
+;									const int32_t iStride,
+;									const int32_t iWidth,
+;									const int32_t iHeight	);
+;***********************************************************************----------------
+ExpandPictureLuma_sse2:
+
+    push r4
+    push r5
+    push r6
+
+    %assign push_num 3
+    LOAD_4_PARA
+    
+    SIGN_EXTENTION r1, r1d
+    SIGN_EXTENTION r2, r2d
+    SIGN_EXTENTION r3, r3d
+
+    ;also prepare for cross border pData top-left:xmm3
+    
+    movzx r6d,byte[r0]
+    SSE2_Copy16Times xmm3,r6d         ;xmm3: pSrc[0]
+
+    neg r1
+    lea r5,[r0+r1]              ;last line of top border r5= dst top  pSrc[-stride]
+    neg r1
+
+    push r3
+
+
+    dec r3                      ;h-1
+    imul r3,r1                  ;(h-1)*stride
+    lea  r3,[r0+r3]             ;pSrc[(h-1)*stride]  r3 = src bottom
+    
+    mov r6,r1                    ;r6 = stride
+    sal r6,05h                   ;r6 = 32*stride
+    lea r4,[r3+r6]               ;r4 = dst bottom
+    
+    ;also prepare for cross border data: bottom-left with xmm5,bottom-right xmm6
+    
+    movzx r6d,byte [r3]             ;bottom-left
+    SSE2_Copy16Times xmm5,r6d
+    
+    lea r6,[r3+r2-1]
+    movzx r6d,byte [r6]
+    SSE2_Copy16Times xmm6,r6d ;bottom-right
+    
+    neg r1  ;r1 = -stride
+    
+    push r0
+    push r1
+    push r2
+
+    exp_top_bottom_sse2 32
+
+	; for both left and right border
+	;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
+    
+    pop r2
+    pop r1
+    pop r0
+
+    lea r5,[r0-32]                          ;left border dst  luma =32 chroma = -16
+    
+    lea r3,[r0+r2-1]                        ;right border src
+    lea r4,[r3+1]                           ;right border dst
+
+    ;prepare for cross border data: top-rigth with xmm4
+     movzx r6d,byte [r3]                         ;top -rigth
+     SSE2_Copy16Times xmm4,r6d
+    
+    neg r1   ;r1 = stride
+
+
+    pop r6  ;  r6 = height
+
+
+
+    push r0
+    push r1
+    push r2
+    push r6
+    
+    exp_left_right_sse2  32,a
+
+    pop r6
+    pop r2
+    pop r1
+    pop r0
+
+	; for cross border [top-left, top-right, bottom-left, bottom-right]
+	;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
+    ; have done xmm3,..,xmm6 cross pData initialization above, perform pading as below, To be continued..
+    
+    neg r1  ;r1 = -stride
+    lea r3,[r0-32]
+    lea r3,[r3+r1]    ;last line of top-left border
+    
+    lea r4,[r0+r2]    ;psrc +width
+    lea r4,[r4+r1]    ;psrc +width -stride
+    
+    
+    neg r1  ;r1 = stride
+    add r6,32         ;height +32(16) ,luma = 32, chroma = 16
+    imul r6,r1
+    
+    lea r5,[r3+r6]    ;last line of bottom-left border
+    lea r6,[r4+r6]    ;last line of botoom-right border
+    
+    neg r1 ; r1 = -stride
+
+    ; for left & right border expanding
+    exp_cross_sse2 32,a
+
+    LOAD_4_PARA_POP
+    
+    pop r6
+    pop r5
+    pop r4
+    
+    %assign push_num 0
+
+
+	ret
+
+ALIGN 16
+;***********************************************************************----------------
+; void ExpandPictureChromaAlign_sse2(	uint8_t *pDst,
+;										const int32_t iStride,
+;										const int32_t iWidth,
+;										const int32_t iHeight	);
+;***********************************************************************----------------
+ExpandPictureChromaAlign_sse2:
+	
+    push r4
+    push r5
+    push r6
+
+    %assign push_num 3
+    LOAD_4_PARA
+
+    SIGN_EXTENTION r1,r1d
+    SIGN_EXTENTION r2,r2d
+    SIGN_EXTENTION r3,r3d
+
+    ;also prepare for cross border pData top-left:xmm3
+    
+    movzx r6d,byte [r0]
+    SSE2_Copy16Times xmm3,r6d         ;xmm3: pSrc[0]
+
+    neg r1
+    lea r5,[r0+r1]              ;last line of top border r5= dst top  pSrc[-stride]
+    neg r1
+
+    push r3
+
+
+    dec r3                      ;h-1
+    imul r3,r1                  ;(h-1)*stride
+    lea  r3,[r0+r3]             ;pSrc[(h-1)*stride]  r3 = src bottom
+    
+    mov r6,r1                    ;r6 = stride
+    sal r6,04h                   ;r6 = 32*stride
+    lea r4,[r3+r6]               ;r4 = dst bottom 
+    
+    ;also prepare for cross border data: bottom-left with xmm5,bottom-right xmm6
+    
+    movzx r6d,byte [r3]             ;bottom-left
+    SSE2_Copy16Times xmm5,r6d
+    
+    lea r6,[r3+r2-1]
+    movzx r6d,byte [r6]
+    SSE2_Copy16Times xmm6,r6d ;bottom-right
+    
+    neg r1  ;r1 = -stride
+    
+    push r0
+    push r1 
+    push r2
+
+    exp_top_bottom_sse2 16
+
+	; for both left and right border
+	;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
+    
+    pop r2
+    pop r1
+    pop r0
+
+    lea r5,[r0-16]                          ;left border dst  luma =32 chroma = -16
+    
+    lea r3,[r0+r2-1]                        ;right border src 
+    lea r4,[r3+1]                           ;right border dst
+
+    ;prepare for cross border data: top-rigth with xmm4
+    movzx r6d,byte [r3]                         ;top -rigth
+    SSE2_Copy16Times xmm4,r6d
+    
+    neg r1   ;r1 = stride
+
+
+    pop r6  ;  r6 = height
+
+
+
+    push r0
+    push r1 
+    push r2
+	push r6
+    exp_left_right_sse2 16,a
+
+    pop r6
+    pop r2
+    pop r1
+    pop r0
+
+	; for cross border [top-left, top-right, bottom-left, bottom-right]
+	;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
+    ; have done xmm3,..,xmm6 cross pData initialization above, perform pading as below, To be continued..
+    
+    neg r1  ;r1 = -stride
+    lea r3,[r0-16]
+    lea r3,[r3+r1]    ;last line of top-left border
+    
+    lea r4,[r0+r2]    ;psrc +width
+    lea r4,[r4+r1]    ;psrc +width -stride  
+    
+    
+    neg r1  ;r1 = stride
+    add r6,16         ;height +32(16) ,luma = 32, chroma = 16
+    imul r6,r1
+    
+    lea r5,[r3+r6]    ;last line of bottom-left border
+    lea r6,[r4+r6]    ;last line of botoom-right border
+    
+    neg r1 ; r1 = -stride
+
+    ; for left & right border expanding
+    exp_cross_sse2 16,a
+
+    LOAD_4_PARA_POP
+    
+    pop r6
+    pop r5
+    pop r4
+    
+    %assign push_num 0
+
+
+	ret
+
+ALIGN 16
+;***********************************************************************----------------
+; void ExpandPictureChromaUnalign_sse2(	uint8_t *pDst,
+;										const int32_t iStride,
+;										const int32_t iWidth,
+;										const int32_t iHeight	);
+;***********************************************************************----------------
+ExpandPictureChromaUnalign_sse2:
+	push r4
+    push r5
+    push r6
+
+    %assign push_num 3
+    LOAD_4_PARA
+
+    SIGN_EXTENTION r1,r1d
+    SIGN_EXTENTION r2,r2d
+    SIGN_EXTENTION r3,r3d
+
+    ;also prepare for cross border pData top-left:xmm3
+    
+    movzx r6d,byte [r0]
+    SSE2_Copy16Times xmm3,r6d         ;xmm3: pSrc[0]
+
+    neg r1
+    lea r5,[r0+r1]              ;last line of top border r5= dst top  pSrc[-stride]
+    neg r1
+
+    push r3
+
+
+    dec r3                      ;h-1
+    imul r3,r1                  ;(h-1)*stride
+    lea  r3,[r0+r3]             ;pSrc[(h-1)*stride]  r3 = src bottom
+    
+    mov r6,r1                    ;r6 = stride
+    sal r6,04h                   ;r6 = 32*stride
+    lea r4,[r3+r6]               ;r4 = dst bottom 
+    
+    ;also prepare for cross border data: bottom-left with xmm5,bottom-right xmm6
+    
+    movzx r6d,byte [r3]             ;bottom-left
+    SSE2_Copy16Times xmm5,r6d
+    
+    lea r6,[r3+r2-1]
+    movzx r6d,byte [r6]
+    SSE2_Copy16Times xmm6,r6d ;bottom-right
+    
+    neg r1  ;r1 = -stride
+    
+    push r0
+    push r1 
+    push r2
+
+    exp_top_bottom_sse2 16
+
+	; for both left and right border
+	;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
+    
+    pop r2
+    pop r1
+    pop r0
+
+    lea r5,[r0-16]                          ;left border dst  luma =32 chroma = -16
+    
+    lea r3,[r0+r2-1]                        ;right border src 
+    lea r4,[r3+1]                           ;right border dst
+
+    ;prepare for cross border data: top-rigth with xmm4
+    movzx r6d,byte [r3]                         ;top -rigth
+    SSE2_Copy16Times xmm4,r6d
+    
+    neg r1   ;r1 = stride
+
+
+    pop r6  ;  r6 = height
+
+
+
+    push r0
+    push r1 
+    push r2
+	push r6
+    exp_left_right_sse2 16,u
+
+    pop r6
+    pop r2
+    pop r1
+    pop r0
+
+	; for cross border [top-left, top-right, bottom-left, bottom-right]
+	;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
+    ; have done xmm3,..,xmm6 cross pData initialization above, perform pading as below, To be continued..
+    
+    neg r1  ;r1 = -stride
+    lea r3,[r0-16]
+    lea r3,[r3+r1]    ;last line of top-left border
+    
+    lea r4,[r0+r2]    ;psrc +width
+    lea r4,[r4+r1]    ;psrc +width -stride  
+    
+    
+    neg r1  ;r1 = stride
+    add r6,16         ;height +32(16) ,luma = 32, chroma = 16
+    imul r6,r1
+    
+    lea r5,[r3+r6]    ;last line of bottom-left border
+    lea r6,[r4+r6]    ;last line of botoom-right border
+    
+    neg r1 ; r1 = -stride
+
+    ; for left & right border expanding
+    exp_cross_sse2 16,u
+
+    LOAD_4_PARA_POP
+    
+    pop r6
+    pop r5
+    pop r4
+    
+    %assign push_num 0
+
+
+	ret
+    

codec/common/mb_copy.asm

@@ -0,0 +1,701 @@
+;*!
+;* \copy
+;*     Copyright (c)  2009-2013, Cisco Systems
+;*     All rights reserved.
+;*
+;*     Redistribution and use in source and binary forms, with or without
+;*     modification, are permitted provided that the following conditions
+;*     are met:
+;*
+;*        * Redistributions of source code must retain the above copyright
+;*          notice, this list of conditions and the following disclaimer.
+;*
+;*        * Redistributions in binary form must reproduce the above copyright
+;*          notice, this list of conditions and the following disclaimer in
+;*          the documentation and/or other materials provided with the
+;*          distribution.
+;*
+;*     THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+;*     "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+;*     LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
+;*     FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
+;*     COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
+;*     INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
+;*     BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+;*     LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+;*     CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+;*     LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
+;*     ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+;*     POSSIBILITY OF SUCH DAMAGE.
+;*
+;*
+;*  mb_copy.asm
+;*
+;*  Abstract
+;*      mb_copy and mb_copy1
+;*
+;*  History
+;*      15/09/2009 Created
+;*		12/28/2009 Modified with larger throughput
+;*		12/29/2011 Tuned WelsCopy16x16NotAligned_sse2, added UpdateMbMv_sse2 WelsCopy16x8NotAligned_sse2,
+;*				   WelsCopy16x8_mmx, WelsCopy8x16_mmx etc;
+;*
+;*
+;*********************************************************************************************/
+%include "asm_inc.asm"
+
+;***********************************************************************
+; Macros and other preprocessor constants
+;***********************************************************************
+
+;***********************************************************************
+; Code
+;***********************************************************************
+
+SECTION .text
+
+WELS_EXTERN WelsCopy16x16_sse2
+WELS_EXTERN WelsCopy16x16NotAligned_sse2
+WELS_EXTERN WelsCopy8x8_mmx
+WELS_EXTERN WelsCopy16x8NotAligned_sse2	;
+WELS_EXTERN WelsCopy8x16_mmx		;
+WELS_EXTERN UpdateMbMv_sse2		;
+
+;***********************************************************************
+; void WelsCopy16x16_sse2(	uint8_t* Dst,
+;							int32_t  iStrideD,
+;							uint8_t* Src,
+;							int32_t  iStrideS )
+;***********************************************************************
+ALIGN 16
+WelsCopy16x16_sse2:
+
+	push r4
+	push r5
+	%assign  push_num 2
+    LOAD_4_PARA
+
+	lea r4, [r1+2*r1]	;ebx, [eax+2*eax]	; x3
+	lea r5, [r3+2*r3]	;edx, [ecx+2*ecx]	; x3
+
+	movdqa xmm0, [r2]
+	movdqa xmm1, [r2+r3]
+	movdqa xmm2, [r2+2*r3]
+	movdqa xmm3, [r2+r5]
+	lea r2, [r2+4*r3]
+	movdqa xmm4, [r2]
+	movdqa xmm5, [r2+r3]
+	movdqa xmm6, [r2+2*r3]
+	movdqa xmm7, [r2+r5]
+	lea r2, [r2+4*r3]
+
+	movdqa [r0], xmm0
+	movdqa [r0+r1], xmm1
+	movdqa [r0+2*r1], xmm2
+	movdqa [r0+r4], xmm3
+	lea r0, [r0+4*r1]
+	movdqa [r0], xmm4
+	movdqa [r0+r1], xmm5
+	movdqa [r0+2*r1], xmm6
+	movdqa [r0+r4], xmm7
+	lea r0, [r0+4*r1]
+
+	movdqa xmm0, [r2]
+	movdqa xmm1, [r2+r3]
+	movdqa xmm2, [r2+2*r3]
+	movdqa xmm3, [r2+r5]
+	lea r2, [r2+4*r3]
+	movdqa xmm4, [r2]
+	movdqa xmm5, [r2+r3]
+	movdqa xmm6, [r2+2*r3]
+	movdqa xmm7, [r2+r5]
+
+	movdqa [r0], xmm0
+	movdqa [r0+r1], xmm1
+	movdqa [r0+2*r1], xmm2
+	movdqa [r0+r4], xmm3
+	lea r0, [r0+4*r1]
+	movdqa [r0], xmm4
+	movdqa [r0+r1], xmm5
+	movdqa [r0+2*r1], xmm6
+	movdqa [r0+r4], xmm7
+	LOAD_4_PARA_POP
+	pop r5
+	pop r4	
+	ret
+
+;***********************************************************************
+; void WelsCopy16x16NotAligned_sse2(	uint8_t* Dst,
+;							int32_t  iStrideD,
+;							uint8_t* Src,
+;							int32_t  iStrideS )
+;***********************************************************************
+ALIGN 16
+; dst can be align with 16 bytes, but not sure about pSrc, 12/29/2011
+WelsCopy16x16NotAligned_sse2:
+	;push esi
+	;push edi
+	;push ebx
+
+	;mov edi, [esp+16]	; Dst
+	;mov eax, [esp+20]	; iStrideD
+	;mov esi, [esp+24]	; Src
+	;mov ecx, [esp+28]	; iStrideS
+	
+	push r4
+	push r5
+	%assign  push_num 2
+    LOAD_4_PARA
+
+	lea r4, [r1+2*r1]	;ebx, [eax+2*eax]	; x3
+	lea r5, [r3+2*r3]	;edx, [ecx+2*ecx]	; x3
+
+	movdqu xmm0, [r2]
+	movdqu xmm1, [r2+r3]
+	movdqu xmm2, [r2+2*r3]
+	movdqu xmm3, [r2+r5]
+	lea r2, [r2+4*r3]
+	movdqu xmm4, [r2]
+	movdqu xmm5, [r2+r3]
+	movdqu xmm6, [r2+2*r3]
+	movdqu xmm7, [r2+r5]
+	lea r2, [r2+4*r3]
+
+	movdqa [r0], xmm0
+	movdqa [r0+r1], xmm1
+	movdqa [r0+2*r1], xmm2
+	movdqa [r0+r4], xmm3
+	lea r0, [r0+4*r1]
+	movdqa [r0], xmm4
+	movdqa [r0+r1], xmm5
+	movdqa [r0+2*r1], xmm6
+	movdqa [r0+r4], xmm7
+	lea r0, [r0+4*r1]
+
+	movdqu xmm0, [r2]
+	movdqu xmm1, [r2+r3]
+	movdqu xmm2, [r2+2*r3]
+	movdqu xmm3, [r2+r5]
+	lea r2, [r2+4*r3]
+	movdqu xmm4, [r2]
+	movdqu xmm5, [r2+r3]
+	movdqu xmm6, [r2+2*r3]
+	movdqu xmm7, [r2+r5]
+
+	movdqa [r0], xmm0
+	movdqa [r0+r1], xmm1
+	movdqa [r0+2*r1], xmm2
+	movdqa [r0+r4], xmm3
+	lea r0, [r0+4*r1]
+	movdqa [r0], xmm4
+	movdqa [r0+r1], xmm5
+	movdqa [r0+2*r1], xmm6
+	movdqa [r0+r4], xmm7
+	LOAD_4_PARA_POP
+	pop r5
+	pop r4
+	ret
+
+; , 12/29/2011
+;***********************************************************************
+; void WelsCopy16x8NotAligned_sse2(uint8_t* Dst,
+;							int32_t  iStrideD,
+;							uint8_t* Src,
+;							int32_t  iStrideS )
+;***********************************************************************
+ALIGN 16
+WelsCopy16x8NotAligned_sse2:
+	;push esi
+	;push edi
+	;push ebx
+
+	;mov edi, [esp+16]	; Dst
+	;mov eax, [esp+20]	; iStrideD
+	;mov esi, [esp+24]	; Src
+	;mov ecx, [esp+28]	; iStrideS
+	
+	push r4
+	push r5
+	%assign  push_num 2
+    LOAD_4_PARA
+
+	lea r4, [r1+2*r1]	;ebx, [eax+2*eax]	; x3
+	lea r5, [r3+2*r3]	;edx, [ecx+2*ecx]	; x3
+
+	movdqu xmm0, [r2]
+	movdqu xmm1, [r2+r3]
+	movdqu xmm2, [r2+2*r3]
+	movdqu xmm3, [r2+r5]
+	lea r2, [r2+4*r3]
+	movdqu xmm4, [r2]
+	movdqu xmm5, [r2+r3]
+	movdqu xmm6, [r2+2*r3]
+	movdqu xmm7, [r2+r5]
+
+	movdqa [r0], xmm0
+	movdqa [r0+r1], xmm1
+	movdqa [r0+2*r1], xmm2
+	movdqa [r0+r4], xmm3
+	lea r0, [r0+4*r1]
+	movdqa [r0], xmm4
+	movdqa [r0+r1], xmm5
+	movdqa [r0+2*r1], xmm6
+	movdqa [r0+r4], xmm7
+	LOAD_4_PARA_POP
+	pop r5
+	pop r4	
+	ret
+
+
+;***********************************************************************
+; void WelsCopy8x16_mmx(uint8_t* Dst,
+;                       int32_t  iStrideD,
+;                       uint8_t* Src,
+;                       int32_t  iStrideS )
+;***********************************************************************
+ALIGN 16
+WelsCopy8x16_mmx:
+	;push ebx
+
+	;mov eax, [esp + 8 ]           ;Dst
+	;mov ecx, [esp + 12]           ;iStrideD
+	;mov ebx, [esp + 16]           ;Src
+	;mov edx, [esp + 20]           ;iStrideS
+	
+	%assign  push_num 0
+    LOAD_4_PARA
+
+	movq mm0, [r2]
+	movq mm1, [r2+r3]
+	lea r2, [r2+2*r3]
+	movq mm2, [r2]
+	movq mm3, [r2+r3]
+	lea r2, [r2+2*r3]
+	movq mm4, [r2]
+	movq mm5, [r2+r3]
+	lea r2, [r2+2*r3]
+	movq mm6, [r2]
+	movq mm7, [r2+r3]
+	lea r2, [r2+2*r3]
+	
+	movq [r0], mm0
+	movq [r0+r1], mm1
+	lea r0, [r0+2*r1]
+	movq [r0], mm2
+	movq [r0+r1], mm3
+	lea r0, [r0+2*r1]
+	movq [r0], mm4
+	movq [r0+r1], mm5
+	lea r0, [r0+2*r1]
+	movq [r0], mm6
+	movq [r0+r1], mm7
+	lea r0, [r0+2*r1]
+
+	movq mm0, [r2]
+	movq mm1, [r2+r3]
+	lea r2, [r2+2*r3]
+	movq mm2, [r2]
+	movq mm3, [r2+r3]
+	lea r2, [r2+2*r3]
+	movq mm4, [r2]
+	movq mm5, [r2+r3]
+	lea r2, [r2+2*r3]
+	movq mm6, [r2]
+	movq mm7, [r2+r3]
+	
+	movq [r0], mm0
+	movq [r0+r1], mm1
+	lea r0, [r0+2*r1]
+	movq [r0], mm2
+	movq [r0+r1], mm3
+	lea r0, [r0+2*r1]
+	movq [r0], mm4
+	movq [r0+r1], mm5
+	lea r0, [r0+2*r1]
+	movq [r0], mm6
+	movq [r0+r1], mm7
+
+	WELSEMMS
+	LOAD_4_PARA_POP
+	ret
+
+;***********************************************************************
+; void WelsCopy8x8_mmx(  uint8_t* Dst,
+;                        int32_t  iStrideD,
+;                        uint8_t* Src,
+;                        int32_t  iStrideS )
+;***********************************************************************
+ALIGN 16
+WelsCopy8x8_mmx:
+	;push ebx
+	;push esi
+	;mov eax, [esp + 12]           ;Dst
+	;mov ecx, [esp + 16]           ;iStrideD
+	;mov esi, [esp + 20]           ;Src
+	;mov ebx, [esp + 24]           ;iStrideS
+	
+	push r4
+	%assign  push_num 1
+    LOAD_4_PARA
+	lea r4, [r3+2*r3]	;edx, [ebx+2*ebx]
+
+	; to prefetch next loop
+	prefetchnta [r2+2*r3]
+	prefetchnta [r2+r4]
+	movq mm0, [r2]
+	movq mm1, [r2+r3]
+	lea r2, [r2+2*r3]
+	; to prefetch next loop
+	prefetchnta [r2+2*r3]
+	prefetchnta [r2+r4]
+	movq mm2, [r2]
+	movq mm3, [r2+r3]
+	lea r2, [r2+2*r3]
+	; to prefetch next loop
+	prefetchnta [r2+2*r3]
+	prefetchnta [r2+r4]
+	movq mm4, [r2]
+	movq mm5, [r2+r3]
+	lea r2, [r2+2*r3]
+	movq mm6, [r2]
+	movq mm7, [r2+r3]
+
+	movq [r0], mm0
+	movq [r0+r1], mm1
+	lea r0, [r0+2*r1]
+	movq [r0], mm2
+	movq [r0+r1], mm3
+	lea r0, [r0+2*r1]
+	movq [r0], mm4
+	movq [r0+r1], mm5
+	lea r0, [r0+2*r1]
+	movq [r0], mm6
+	movq [r0+r1], mm7
+
+	WELSEMMS
+	;pop esi
+	;pop ebx	
+	LOAD_4_PARA_POP
+	pop r4
+	ret
+
+; (dunhuang@cisco), 12/21/2011
+;***********************************************************************
+; void UpdateMbMv_sse2( SMVUnitXY *pMvBuffer, const SMVUnitXY sMv )
+;***********************************************************************
+ALIGN 16
+UpdateMbMv_sse2:
+
+    %assign  push_num 0
+    LOAD_2_PARA
+    
+	;mov eax, [esp+4]	; mv_buffer
+	;movd xmm0, [esp+8]	; _mv
+	movd xmm0, r1d	; _mv
+	pshufd xmm1, xmm0, $0
+	movdqa [r0     ], xmm1
+	movdqa [r0+0x10], xmm1
+	movdqa [r0+0x20], xmm1
+	movdqa [r0+0x30], xmm1
+	ret
+
+;*******************************************************************************
+; Macros and other preprocessor constants
+;*******************************************************************************
+
+;*******************************************************************************
+; Local Data (Read Only)
+;*******************************************************************************
+
+;SECTION .rodata data align=16
+
+;*******************************************************************************
+; Various memory constants (trigonometric values or rounding values)
+;*******************************************************************************
+
+ALIGN 16
+
+;*******************************************************************************
+; Code
+;*******************************************************************************
+
+SECTION .text
+
+WELS_EXTERN PixelAvgWidthEq4_mmx
+WELS_EXTERN PixelAvgWidthEq8_mmx
+WELS_EXTERN PixelAvgWidthEq16_sse2
+
+WELS_EXTERN McCopyWidthEq4_mmx
+WELS_EXTERN McCopyWidthEq8_mmx
+WELS_EXTERN McCopyWidthEq16_sse2
+
+
+ALIGN 16
+;*******************************************************************************
+; void_t PixelAvgWidthEq4_mmx( uint8_t *pDst,  int iDstStride,
+;                           uint8_t *pSrcA, int iSrcAStride,
+;                           uint8_t *pSrcB, int iSrcBStride,
+;                           int iHeight );
+;*******************************************************************************
+PixelAvgWidthEq4_mmx:
+ 
+    %assign  push_num 0
+    LOAD_7_PARA
+
+%ifndef X86_32
+	movsx	r1, r1d
+	movsx	r3, r3d
+	movsx	r5, r5d
+	movsx	r6, r6d
+%endif
+
+ALIGN 4
+.height_loop:
+	movd        mm0, [r4]
+    pavgb       mm0, [r2]
+    movd        [r0], mm0
+
+    dec         r6
+    lea         r0, [r0+r1]
+    lea         r2, [r2+r3]
+    lea         r4, [r4+r5]
+    jne         .height_loop
+
+	WELSEMMS
+	LOAD_7_PARA_POP
+    ret
+
+
+ALIGN 16
+;*******************************************************************************
+; void_t PixelAvgWidthEq8_mmx( uint8_t *pDst,  int iDstStride,
+;                           uint8_t *pSrcA, int iSrcAStride,
+;                           uint8_t *pSrcB, int iSrcBStride,
+;                           int iHeight );
+;*******************************************************************************
+PixelAvgWidthEq8_mmx:
+
+    ;push        esi
+    ;push        edi
+    ;push        ebp
+    ;push        ebx
+
+    ;mov         edi, [esp+20]       ; pDst
+    ;mov         eax, [esp+24]       ; iDstStride
+    ;mov         esi, [esp+28]       ; pSrcA
+    ;mov         ecx, [esp+32]       ; iSrcAStride
+    ;mov         ebp, [esp+36]       ; pSrcB
+    ;mov         edx, [esp+40]       ; iSrcBStride
+    ;mov         ebx, [esp+44]       ; iHeight
+    
+    %assign  push_num 0
+    LOAD_7_PARA
+
+%ifndef X86_32
+	movsx	r1, r1d
+	movsx	r3, r3d
+	movsx	r5, r5d
+	movsx	r6, r6d
+%endif
+    
+ALIGN 4
+.height_loop:
+	movq        mm0, [r2]
+    pavgb       mm0, [r4]
+    movq        [r0], mm0
+    movq        mm0, [r2+r3]
+    pavgb       mm0, [r4+r5]
+    movq		[r0+r1], mm0
+
+    lea			r2,  [r2+2*r3]
+    lea			r4,  [r4+2*r5]
+    lea			r0,  [r0+2*r1]
+
+    sub         r6, 2
+    jnz         .height_loop
+
+	WELSEMMS
+	LOAD_7_PARA_POP
+    ret
+
+
+
+ALIGN 16
+;*******************************************************************************
+; void_t PixelAvgWidthEq16_sse2( uint8_t *pDst,  int iDstStride,
+;                          uint8_t *pSrcA, int iSrcAStride,
+;                          uint8_t *pSrcB, int iSrcBStride,
+;                          int iHeight );
+;*******************************************************************************
+PixelAvgWidthEq16_sse2:
+        
+    %assign  push_num 0
+    LOAD_7_PARA
+%ifndef X86_32
+	movsx	r1, r1d
+	movsx	r3, r3d
+	movsx	r5, r5d
+	movsx	r6, r6d
+%endif
+ALIGN 4
+.height_loop:
+	movdqu      xmm0, [r2]
+	movdqu	    xmm1, [r4]
+	pavgb	    xmm0, xmm1
+	;pavgb       xmm0, [r4]
+    movdqu      [r0], xmm0
+
+	movdqu      xmm0, [r2+r3]
+	movdqu      xmm1, [r4+r5]
+	pavgb	    xmm0, xmm1
+    movdqu      [r0+r1], xmm0
+
+	movdqu      xmm0, [r2+2*r3]
+	movdqu       xmm1, [r4+2*r5]
+	pavgb	    xmm0, xmm1
+    movdqu      [r0+2*r1], xmm0
+
+    lea         r2, [r2+2*r3]
+    lea			r4, [r4+2*r5]
+    lea			r0, [r0+2*r1]
+
+	movdqu      xmm0, [r2+r3]
+	movdqu      xmm1, [r4+r5]
+	pavgb	    xmm0, xmm1
+    movdqu      [r0+r1], xmm0
+
+    lea         r2, [r2+2*r3]
+    lea			r4, [r4+2*r5]
+    lea			r0, [r0+2*r1]
+    
+    sub         r6, 4
+    jne         .height_loop
+
+	WELSEMMS
+	LOAD_7_PARA_POP
+    ret
+
+ALIGN 16
+;*******************************************************************************
+;  void_t McCopyWidthEq4_mmx( uint8_t *pSrc, int iSrcStride,
+;                          uint8_t *pDst, int iDstStride, int iHeight )
+;*******************************************************************************
+McCopyWidthEq4_mmx:
+    ;push    esi
+    ;push    edi
+    ;push    ebx
+
+
+    ;mov esi,  [esp+16]
+    ;mov eax, [esp+20]
+    ;mov edi,  [esp+24]
+    ;mov ecx,  [esp+28]
+    ;mov edx,  [esp+32]
+    
+    push	r5
+    %assign  push_num 1
+    LOAD_5_PARA
+   
+%ifndef X86_32
+	movsx	r1, r1d
+	movsx	r3, r3d
+	movsx	r4, r4d
+%endif
+    
+ALIGN 4
+.height_loop:
+	mov r5d, [r0]
+	mov [r2], r5d
+
+	add r0, r1
+	add r2, r3
+	dec r4
+	jnz .height_loop
+	WELSEMMS
+    LOAD_5_PARA_POP
+    pop	   r5
+    ret
+
+ALIGN 16
+;*******************************************************************************
+;   void_t McCopyWidthEq8_mmx( uint8_t *pSrc, int iSrcStride,
+;                           uint8_t *pDst, int iDstStride, int iHeight )
+;*******************************************************************************
+McCopyWidthEq8_mmx:
+    ;push  esi
+    ;push  edi
+	;mov  esi, [esp+12]
+	;mov eax, [esp+16]
+	;mov edi, [esp+20]
+	;mov ecx, [esp+24]
+	;mov edx, [esp+28]
+	
+    %assign  push_num 0
+    LOAD_5_PARA
+
+%ifndef X86_32
+	movsx	r1, r1d
+	movsx	r3, r3d
+	movsx	r4, r4d
+%endif
+
+ALIGN 4
+.height_loop:
+	movq mm0, [r0]
+	movq [r2], mm0
+	add r0, r1
+	add r2, r3
+	dec r4
+	jnz .height_loop
+
+	WELSEMMS
+	LOAD_5_PARA_POP
+    ret
+
+
+ALIGN 16
+;*******************************************************************************
+;   void_t McCopyWidthEq16_sse2( uint8_t *pSrc, int iSrcStride, uint8_t *pDst, int iDstStride, int iHeight )
+;*******************************************************************************
+;read unaligned memory
+%macro SSE_READ_UNA 2
+	movq	%1, [%2]
+	movhps	%1,	[%2+8]
+%endmacro
+
+;write unaligned memory
+%macro SSE_WRITE_UNA 2
+	movq	[%1],	%2
+	movhps	[%1+8], %2
+%endmacro
+McCopyWidthEq16_sse2:
+    ;push    esi
+    ;push    edi
+
+    ;mov     esi, [esp+12]       ; pSrc
+    ;mov     eax, [esp+16]       ; iSrcStride
+    ;mov     edi, [esp+20]       ; pDst
+    ;mov     edx, [esp+24]       ; iDstStride
+    ;mov     ecx, [esp+28]       ; iHeight
+
+    %assign  push_num 0
+    LOAD_5_PARA
+%ifndef X86_32
+	movsx	r1, r1d
+	movsx	r3, r3d
+	movsx	r4, r4d
+%endif
+ALIGN 4
+.height_loop:
+    SSE_READ_UNA	xmm0, r0
+    SSE_READ_UNA	xmm1, r0+r1
+    SSE_WRITE_UNA	r2, xmm0
+    SSE_WRITE_UNA	r2+r3, xmm1
+
+	sub		r4,	2
+    lea     r0, [r0+r1*2]
+    lea     r2, [r2+r3*2]
+    jnz     .height_loop
+
+	LOAD_5_PARA_POP
+    ret

codec/common/mc_chroma.asm

@@ -41,8 +41,6 @@
 ;*************************************************************************/
 %include "asm_inc.asm"
 
-BITS 32
-
 ;***********************************************************************
 ; Local Data (Read Only)
 ;***********************************************************************
@@ -78,12 +76,21 @@ ALIGN 16
 ;*******************************************************************************
 WELS_EXTERN McChromaWidthEq4_mmx
 McChromaWidthEq4_mmx:
-	push esi
+	;push esi
-	push edi
+	;push edi
-	push ebx
+	;push ebx
 	
-	mov eax, [esp +12 + 20]
+	%assign  push_num 0
-	movd mm3, [eax]
+	LOAD_6_PARA 
+%ifndef X86_32
+	movsx	r1, r1d
+	movsx	r3, r3d
+	movsx	r5, r5d
+%endif
+	
+	;mov eax, [esp +12 + 20]
+	
+	movd mm3, [r4];	[eax]
 	WELS_Zero mm7
 	punpcklbw mm3, mm3
 	movq      mm4, mm3
@@ -98,15 +105,15 @@ McChromaWidthEq4_mmx:
 	punpcklbw mm4, mm7
 	punpckhbw mm6, mm7
 
-	mov esi, [esp +12+ 4]
+	;mov esi, [esp +12+ 4]
-	mov eax, [esp + 12 + 8]
+	;mov eax, [esp + 12 + 8]
-	mov edi, [esp + 12 + 12]
+	;mov edi, [esp + 12 + 12]
-	mov edx, [esp + 12 + 16]
+	;mov edx, [esp + 12 + 16]
-    mov ecx, [esp + 12 + 24]
+    ;mov ecx, [esp + 12 + 24]
 
-	lea ebx, [esi + eax]
+	lea r4, [r0 + r1] ;lea ebx, [esi + eax]
-	movd mm0, [esi]
+	movd mm0, [r0]
-	movd mm1, [esi+1]
+	movd mm1, [r0+1]
 	punpcklbw mm0, mm7
 	punpcklbw mm1, mm7
 .xloop:
@@ -115,13 +122,13 @@ McChromaWidthEq4_mmx:
 	pmullw mm1, mm5
 	paddw  mm0, mm1
 
-	movd  mm1, [ebx]
+	movd  mm1, [r4]
 	punpcklbw mm1, mm7
 	movq mm2, mm1
 	pmullw mm1, mm4
 	paddw mm0, mm1
 
-	movd mm1, [ebx+1]
+	movd mm1, [r4+1]
 	punpcklbw mm1, mm7
 	movq mm7, mm1
 	pmullw mm1,mm6
@@ -133,19 +140,20 @@ McChromaWidthEq4_mmx:
 
 	WELS_Zero mm7
 	packuswb mm0, mm7
-	movd [edi], mm0
+	movd [r2], mm0
 
 	movq mm0, mm2
 
-	lea edi, [edi +edx  ]
+	lea r2, [r2 + r3]
-	lea ebx, [ebx + eax]
+	lea r4, [r4 + r1]
 
-	dec ecx
+	dec r5
 	jnz near .xloop
 	WELSEMMS
-	pop ebx
+	LOAD_6_PARA_POP
-	pop edi
+	;pop ebx
-	pop esi
+	;pop edi
+	;pop esi
 	ret
 
 
@@ -160,12 +168,20 @@ ALIGN 16
 ;*******************************************************************************
 WELS_EXTERN McChromaWidthEq8_sse2
 McChromaWidthEq8_sse2:
-	push esi
+	;push esi
-	push edi
+	;push edi
-	push ebx
+	;push ebx
 
-	mov eax, [esp +12 + 20]
+	%assign  push_num 0
-	movd xmm3, [eax]
+	LOAD_6_PARA 	
+%ifndef X86_32
+	movsx	r1, r1d
+	movsx	r3, r3d
+	movsx	r5, r5d
+%endif
+
+	;mov eax, [esp +12 + 20]
+	movd xmm3, [r4]
 	WELS_Zero xmm7
 	punpcklbw  xmm3, xmm3
 	punpcklwd  xmm3, xmm3
@@ -181,15 +197,15 @@ McChromaWidthEq8_sse2:
 	punpcklbw  xmm4, xmm7
 	punpckhbw  xmm6, xmm7
 
-	mov esi, [esp +12+ 4]
+	;mov esi, [esp +12+ 4]
-	mov eax, [esp + 12 + 8]
+	;mov eax, [esp + 12 + 8]
-	mov edi, [esp + 12 + 12]
+	;mov edi, [esp + 12 + 12]
-	mov edx, [esp + 12 + 16]
+	;mov edx, [esp + 12 + 16]
-    mov ecx, [esp + 12 + 24]
+    ;mov ecx, [esp + 12 + 24]
 
-	lea ebx, [esi + eax]
+	lea r4, [r0 + r1] ;lea ebx, [esi + eax]
-	movq xmm0, [esi]
+	movq xmm0, [r0]
-	movq xmm1, [esi+1]
+	movq xmm1, [r0+1]
 	punpcklbw xmm0, xmm7
 	punpcklbw xmm1, xmm7
 .xloop:
@@ -198,13 +214,13 @@ McChromaWidthEq8_sse2:
 	pmullw xmm1, xmm5
 	paddw  xmm0, xmm1
 
-	movq  xmm1, [ebx]
+	movq  xmm1, [r4]
 	punpcklbw xmm1, xmm7
 	movdqa xmm2, xmm1
 	pmullw xmm1, xmm4
 	paddw xmm0, xmm1
 
-	movq xmm1, [ebx+1]
+	movq xmm1, [r4+1]
 	punpcklbw xmm1, xmm7
 	movdqa xmm7, xmm1
 	pmullw xmm1, xmm6
@@ -216,19 +232,21 @@ McChromaWidthEq8_sse2:
 
 	WELS_Zero xmm7
 	packuswb xmm0, xmm7
-	movq [edi], xmm0
+	movq [r2], xmm0
 
 	movdqa xmm0, xmm2
 
-	lea edi, [edi +edx  ]
+	lea r2, [r2 + r3]
-	lea ebx, [ebx + eax]
+	lea r4, [r4 + r1]
 
-	dec ecx
+	dec r5
 	jnz near .xloop
 	
-	pop ebx
+	LOAD_6_PARA_POP
-	pop edi
+
-	pop esi
+	;pop ebx
+	;pop edi
+	;pop esi
 	ret
 
 
@@ -245,39 +263,46 @@ ALIGN 16
 ;***********************************************************************
 WELS_EXTERN McChromaWidthEq8_ssse3
 McChromaWidthEq8_ssse3:
-	push ebx
+	;push ebx
-	push esi
+	;push esi
-	push edi
+	;push edi
+	%assign  push_num 0
+	LOAD_6_PARA
+%ifndef X86_32
+	movsx	r1, r1d
+	movsx	r3, r3d
+	movsx	r5, r5d
+%endif
 	
-	mov eax, [esp + 12 + 20]
+	;mov eax, [esp + 12 + 20]
 
     pxor      xmm7, xmm7
-    movd   xmm5, [eax]
+    movd   xmm5, [r4]
     punpcklwd xmm5, xmm5
     punpckldq xmm5, xmm5
     movdqa    xmm6, xmm5
     punpcklqdq xmm5, xmm5
     punpckhqdq xmm6, xmm6
 
-	mov eax, [esp + 12 + 4]
+	;mov eax, [esp + 12 + 4]
-	mov edx, [esp + 12 + 8]
+	;mov edx, [esp + 12 + 8]
-	mov esi, [esp + 12 + 12]
+	;mov esi, [esp + 12 + 12]
-	mov edi, [esp + 12 + 16]
+	;mov edi, [esp + 12 + 16]
-    mov ecx, [esp + 12 + 24]
+    ;mov ecx, [esp + 12 + 24]
 
-    sub esi, edi
+    sub r2, r3 ;sub esi, edi
-    sub esi, edi
+    sub r2, r3
 	movdqa xmm7, [h264_d0x20_sse2]
 
-	movdqu xmm0, [eax]
+	movdqu xmm0, [r0]
 	movdqa xmm1, xmm0
 	psrldq xmm1, 1
 	punpcklbw xmm0, xmm1
 
 .hloop_chroma:
-	lea	esi, [esi+2*edi]
+	lea	r2, [r2+2*r3]
 
-	movdqu xmm2, [eax+edx]
+	movdqu xmm2, [r0+r1]
 	movdqa xmm3, xmm2
 	psrldq xmm3, 1
 	punpcklbw xmm2, xmm3
@@ -289,10 +314,10 @@ McChromaWidthEq8_ssse3:
     paddw      xmm0, xmm7
 	psrlw      xmm0, 6
     packuswb   xmm0, xmm0
-    movq       [esi],xmm0
+    movq       [r2],xmm0
 
-    lea eax, [eax+2*edx]
+    lea r0, [r0+2*r1]
-    movdqu xmm2, [eax]
+    movdqu xmm2, [r0]
     movdqa xmm3, xmm2
     psrldq xmm3, 1
     punpcklbw xmm2, xmm3
@@ -304,13 +329,16 @@ McChromaWidthEq8_ssse3:
     paddw      xmm4, xmm7
 	psrlw      xmm4, 6
     packuswb   xmm4, xmm4
-    movq       [esi+edi],xmm4
+    movq       [r2+r3],xmm4
 
-	sub ecx, 2
+	sub r5, 2
 	jnz .hloop_chroma
-	pop edi
+	
-	pop esi
+	LOAD_6_PARA_POP
-	pop ebx
+	
+	;pop edi
+	;pop esi
+	;pop ebx
 
 	ret
 

codec/common/targets.mk

@@ -6,6 +6,14 @@ COMMON_CPP_SRCS=\
 COMMON_OBJS += $(COMMON_CPP_SRCS:.cpp=.o)
 ifeq ($(USE_ASM), Yes)
 COMMON_ASM_SRCS=\
+	$(COMMON_SRCDIR)/./asm_inc.asm\
+	$(COMMON_SRCDIR)/./cpuid.asm\
+	$(COMMON_SRCDIR)/./deblock.asm\
+	$(COMMON_SRCDIR)/./expand_picture.asm\
+	$(COMMON_SRCDIR)/./mb_copy.asm\
+	$(COMMON_SRCDIR)/./mc_chroma.asm\
+	$(COMMON_SRCDIR)/./mc_luma.asm\
+	$(COMMON_SRCDIR)/./vaa.asm\
 
 COMMON_OBJS += $(COMMON_ASM_SRCS:.asm=.o)
 endif
@@ -14,6 +22,30 @@ OBJS += $(COMMON_OBJS)
 $(COMMON_SRCDIR)/./logging.o: $(COMMON_SRCDIR)/./logging.cpp
 	$(CXX) $(CFLAGS) $(CXXFLAGS) $(INCLUDES) $(COMMON_CFLAGS) $(COMMON_INCLUDES) -c -o $(COMMON_SRCDIR)/./logging.o $(COMMON_SRCDIR)/./logging.cpp
 
+$(COMMON_SRCDIR)/./asm_inc.o: $(COMMON_SRCDIR)/./asm_inc.asm
+	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(COMMON_ASMFLAGS) $(COMMON_ASM_INCLUDES) -o $(COMMON_SRCDIR)/./asm_inc.o $(COMMON_SRCDIR)/./asm_inc.asm
+
+$(COMMON_SRCDIR)/./cpuid.o: $(COMMON_SRCDIR)/./cpuid.asm
+	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(COMMON_ASMFLAGS) $(COMMON_ASM_INCLUDES) -o $(COMMON_SRCDIR)/./cpuid.o $(COMMON_SRCDIR)/./cpuid.asm
+
+$(COMMON_SRCDIR)/./deblock.o: $(COMMON_SRCDIR)/./deblock.asm
+	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(COMMON_ASMFLAGS) $(COMMON_ASM_INCLUDES) -o $(COMMON_SRCDIR)/./deblock.o $(COMMON_SRCDIR)/./deblock.asm
+
+$(COMMON_SRCDIR)/./expand_picture.o: $(COMMON_SRCDIR)/./expand_picture.asm
+	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(COMMON_ASMFLAGS) $(COMMON_ASM_INCLUDES) -o $(COMMON_SRCDIR)/./expand_picture.o $(COMMON_SRCDIR)/./expand_picture.asm
+
+$(COMMON_SRCDIR)/./mb_copy.o: $(COMMON_SRCDIR)/./mb_copy.asm
+	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(COMMON_ASMFLAGS) $(COMMON_ASM_INCLUDES) -o $(COMMON_SRCDIR)/./mb_copy.o $(COMMON_SRCDIR)/./mb_copy.asm
+
+$(COMMON_SRCDIR)/./mc_chroma.o: $(COMMON_SRCDIR)/./mc_chroma.asm
+	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(COMMON_ASMFLAGS) $(COMMON_ASM_INCLUDES) -o $(COMMON_SRCDIR)/./mc_chroma.o $(COMMON_SRCDIR)/./mc_chroma.asm
+
+$(COMMON_SRCDIR)/./mc_luma.o: $(COMMON_SRCDIR)/./mc_luma.asm
+	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(COMMON_ASMFLAGS) $(COMMON_ASM_INCLUDES) -o $(COMMON_SRCDIR)/./mc_luma.o $(COMMON_SRCDIR)/./mc_luma.asm
+
+$(COMMON_SRCDIR)/./vaa.o: $(COMMON_SRCDIR)/./vaa.asm
+	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(COMMON_ASMFLAGS) $(COMMON_ASM_INCLUDES) -o $(COMMON_SRCDIR)/./vaa.o $(COMMON_SRCDIR)/./vaa.asm
+
 $(LIBPREFIX)common.$(LIBSUFFIX): $(COMMON_OBJS)
 	rm -f $(LIBPREFIX)common.$(LIBSUFFIX)
 	$(AR) cr $@ $(COMMON_OBJS)

codec/common/vaa.asm

@@ -42,7 +42,7 @@
 ;*
 ;*************************************************************************/
 %include "asm_inc.asm"
-BITS 32
+
 
 ;***********************************************************************
 ; Macros and other preprocessor constants
@@ -62,8 +62,8 @@ BITS 32
 
 
 %macro VAA_AVG_BLOCK_SSE2 6 ; dst, t0, t1, t2, t3, t4
-	movdqa %1, [esi    ]	; line 0
+	movdqa %1, [r0    ]	; line 0
-	movdqa %2, [esi+ecx]	; line 1
+	movdqa %2, [r0+r1]	; line 1
 	movdqa %3, %1
 	punpcklbw %1, xmm7
 	punpckhbw %3, xmm7
@@ -72,8 +72,8 @@ BITS 32
 	punpckhbw %2, xmm7
 	paddw %1, %4
 	paddw %2, %3
-	movdqa %3, [esi+ebx]	; line 2
+	movdqa %3, [r0+r2]	; line 2
-	movdqa %4, [esi+edx]	; line 3
+	movdqa %4, [r0+r3]	; line 3
 	movdqa %5, %3
 	punpcklbw %3, xmm7
 	punpckhbw %5, xmm7
@@ -105,8 +105,8 @@ BITS 32
 %endmacro
 
 %macro VAA_AVG_BLOCK_SSSE3 6 ; dst, t0, t1, t2, t3, t4
-	movdqa %1, [esi    ]	; line 0
+	movdqa %1, [r0    ]	; line 0
-	movdqa %2, [esi+ecx]	; line 1
+	movdqa %2, [r0+r1]	; line 1
 	movdqa %3, %1
 	punpcklbw %1, xmm7
 	punpckhbw %3, xmm7
@@ -115,8 +115,8 @@ BITS 32
 	punpckhbw %2, xmm7
 	paddw %1, %4
 	paddw %2, %3
-	movdqa %3, [esi+ebx]	; line 2
+	movdqa %3, [r0+r2]	; line 2
-	movdqa %4, [esi+edx]	; line 3
+	movdqa %4, [r0+r3]	; line 3
 	movdqa %5, %3
 	punpcklbw %3, xmm7
 	punpckhbw %5, xmm7
@@ -158,48 +158,53 @@ WELS_EXTERN AnalysisVaaInfoIntra_sse2
 ;***********************************************************************
 ALIGN 16
 AnalysisVaaInfoIntra_sse2:
-	push ebx
-	push edx
-	push esi
-	push edi
-	push ebp
 
-	mov ebp, esp
+    %assign push_num 0
-	and ebp, 0fh
+    LOAD_2_PARA 
-	sub esp, ebp
+    SIGN_EXTENTION r1,r1d
-	sub esp, 32
-	%define PUSH_SIZE	52	; 20 + 32
 
-	mov esi, [esp+ebp+PUSH_SIZE+4]	; data_y
+%ifdef X86_32
-	mov ecx, [esp+ebp+PUSH_SIZE+8]	; iLineSize
+    push r3
+    push r4
+    push r5
+    push r6
+    %assign push_num push_num+4
+%endif
 
-	mov ebx, ecx
+    mov  r5,r7
-	sal ebx, $1			; iLineSize x 2 [ebx]
+    and  r5,0fh
-	mov edx, ebx
+    sub  r7,r5
-	add edx, ecx		; iLineSize x 3 [edx]
+    sub  r7,32
-	mov eax, ebx
+    
-	sal eax, $1			; iLineSize x 4 [eax]
+    
+    mov r2,r1    
+    sal r2,$1   ;r2 = 2*iLineSize
+    mov r3,r2
+    add r3,r1   ;r3 = 3*iLineSize
+    
+    mov r4,r2
+    sal r4,$1   ;r4 = 4*iLineSize
     
 	pxor xmm7, xmm7
 
 	; loops
 	VAA_AVG_BLOCK_SSE2 xmm0, xmm1, xmm2, xmm3, xmm4, xmm5
-	movq [esp], xmm0
+	movq [r7], xmm0
 
-	lea esi, [esi+eax]
+	lea r0, [r0+r4]
 	VAA_AVG_BLOCK_SSE2 xmm0, xmm1, xmm2, xmm3, xmm4, xmm5
-	movq [esp+8], xmm0
+	movq [r7+8], xmm0
 
-	lea esi, [esi+eax]
+	lea r0, [r0+r4]
 	VAA_AVG_BLOCK_SSE2 xmm0, xmm1, xmm2, xmm3, xmm4, xmm5
-	movq [esp+16], xmm0
+	movq [r7+16], xmm0
 
-	lea esi, [esi+eax]
+	lea r0, [r0+r4]
 	VAA_AVG_BLOCK_SSE2 xmm0, xmm1, xmm2, xmm3, xmm4, xmm5
-	movq [esp+24], xmm0
+	movq [r7+24], xmm0
 
-	movdqa xmm0, [esp]		; block 0~7
+	movdqa xmm0, [r7]		; block 0~7
-	movdqa xmm1, [esp+16]	; block 8~15
+	movdqa xmm1, [r7+16]	; block 8~15
 	movdqa xmm2, xmm0
 	paddw xmm0, xmm1
 	SUM_WORD_8x2_SSE2 xmm0, xmm3
@@ -220,22 +225,26 @@ AnalysisVaaInfoIntra_sse2:
 	pshufd xmm2, xmm1, 0B1h
 	paddd xmm1, xmm2
 
-	movd ebx, xmm0
-	and ebx, 0ffffh		; effective low word truncated
-	mov ecx, ebx
-	imul ebx, ecx
-	sar ebx, $4
-	movd eax, xmm1
-	sub eax, ebx
 	
-	%undef PUSH_SIZE
+	
-	add esp, 32
+	movd r2d, xmm0
-	add esp, ebp
+	and r2, 0ffffh		; effective low work truncated
-	pop ebp
+	mov r3, r2
-	pop edi
+	imul r2, r3
-	pop esi
+	sar r2, $4
-	pop edx
+	movd retrd, xmm1
-	pop ebx
+	sub retrd, r2d
+	
+	add r7,32
+	add r7,r5
+
+%ifdef X86_32
+	pop r6
+	pop r5
+	pop r4
+	pop r3
+%endif
+	
 	ret
 
 WELS_EXTERN AnalysisVaaInfoIntra_ssse3
@@ -244,48 +253,55 @@ WELS_EXTERN AnalysisVaaInfoIntra_ssse3
 ;***********************************************************************
 ALIGN 16
 AnalysisVaaInfoIntra_ssse3:
-	push ebx
-	push edx
-	push esi
-	push edi
-	push ebp
 
-	mov ebp, esp
+    %assign push_num 0
-	and ebp, 0fh
+    LOAD_2_PARA 
-	sub esp, ebp
+    SIGN_EXTENTION r1,r1d
-	sub esp, 32
-	%define PUSH_SIZE	52	; 20 + 32
 
-	mov esi, [esp+ebp+PUSH_SIZE+4]	; data_y
+%ifdef X86_32
-	mov ecx, [esp+ebp+PUSH_SIZE+8]	; iLineSize
+    push r3
+    push r4
+    push r5
+    push r6
+    %assign push_num push_num+4
+%endif
    
-	mov ebx, ecx
+    mov  r5,r7
-	sal ebx, $1			; iLineSize x 2 [ebx]
+    and  r5,0fh
-	mov edx, ebx
+    sub  r7,r5
-	add edx, ecx		; iLineSize x 3 [edx]
+    sub  r7,32
-	mov eax, ebx
+    
-	sal eax, $1			; iLineSize x 4 [eax]
+
+    mov r2,r1    
+    sal r2,$1   ;r2 = 2*iLineSize
+    mov r3,r2
+    add r3,r1   ;r3 = 3*iLineSize
+    
+    mov r4,r2
+    sal r4,$1   ;r4 = 4*iLineSize
      
 	pxor xmm7, xmm7
 
 	; loops
 	VAA_AVG_BLOCK_SSSE3 xmm0, xmm1, xmm2, xmm3, xmm4, xmm5
-	movq [esp], xmm0
+    movq [r7],xmm0
     
-	lea esi, [esi+eax]
+	lea r0,[r0+r4]
 	VAA_AVG_BLOCK_SSSE3 xmm1, xmm2, xmm3, xmm4, xmm5, xmm6
-	movq [esp+8], xmm1
+    movq [r7+8],xmm1
     
-	lea esi, [esi+eax]
+    
+	lea r0,[r0+r4]
 	VAA_AVG_BLOCK_SSSE3 xmm0, xmm1, xmm2, xmm3, xmm4, xmm5
-	movq [esp+16], xmm0
+    movq [r7+16],xmm0
     
-	lea esi, [esi+eax]
+	lea r0,[r0+r4]
 	VAA_AVG_BLOCK_SSSE3 xmm1, xmm2, xmm3, xmm4, xmm5, xmm6
-	movq [esp+24], xmm1
+    movq [r7+24],xmm1
     
-	movdqa xmm0, [esp]		; block 0~7
+    
-	movdqa xmm1, [esp+16]	; block 8~15
+	movdqa xmm0,[r7]
+	movdqa xmm1,[r7+16]
 	movdqa xmm2, xmm0
 	paddw xmm0, xmm1
 	SUM_WORD_8x2_SSE2 xmm0, xmm3	; better performance than that of phaddw sets
@@ -306,22 +322,24 @@ AnalysisVaaInfoIntra_ssse3:
 	pshufd xmm2, xmm1, 0B1h
 	paddd xmm1, xmm2
 
-	movd ebx, xmm0
-	and ebx, 0ffffh		; effective low work truncated
-	mov ecx, ebx
-	imul ebx, ecx
-	sar ebx, $4
-	movd eax, xmm1
-	sub eax, ebx
     
-	%undef PUSH_SIZE
+    movd r2d, xmm0
-	add esp, 32
+    and r2, 0ffffh          ; effective low work truncated
-	add esp, ebp
+    mov r3, r2
-	pop ebp
+    imul r2, r3
-	pop edi
+    sar r2, $4
-	pop esi
+    movd retrd, xmm1
-	pop edx
+	sub retrd, r2d
-	pop ebx
+
+	add r7,32
+	add r7,r5
+%ifdef X86_32
+	pop r6
+	pop r5
+	pop r4
+	pop r3
+%endif
+	
 	ret
 
 WELS_EXTERN MdInterAnalysisVaaInfo_sse41
@@ -330,8 +348,9 @@ WELS_EXTERN MdInterAnalysisVaaInfo_sse41
 ;***********************************************************************
 ALIGN 16
 MdInterAnalysisVaaInfo_sse41:
-	mov eax, [esp+4]
+	%assign push_num 0
-	movdqa xmm0, [eax]	; load 4 sad_8x8
+	LOAD_1_PARA
+	movdqa xmm0,[r0]
 	pshufd xmm1, xmm0, 01Bh
 	paddd xmm1, xmm0
 	pshufd xmm2, xmm1, 0B1h
@@ -347,15 +366,16 @@ MdInterAnalysisVaaInfo_sse41:
 	paddd xmm4, xmm3
 	pshufd xmm3, xmm4, 0B1h
 	paddd xmm3, xmm4
-	movd eax, xmm3
+	movd r0d, xmm3
-	cmp eax, 20	; INTER_VARIANCE_SAD_THRESHOLD
+	cmp r0d, 20	; INTER_VARIANCE_SAD_THRESHOLD
+	
 	jb near .threshold_exit
-	pshufd xmm0, xmm0, 0B1h
+	pshufd xmm0, xmm0, 01Bh
 	pcmpgtd xmm0, xmm1	; iSadBlock > iAverageSad
-	movmskps eax, xmm0
+	movmskps retrd, xmm0
 	ret
 .threshold_exit:
-	mov eax, 15
+	mov retrd, 15
 	ret
 
 WELS_EXTERN MdInterAnalysisVaaInfo_sse2
@@ -364,8 +384,9 @@ WELS_EXTERN MdInterAnalysisVaaInfo_sse2
 ;***********************************************************************
 ALIGN 16
 MdInterAnalysisVaaInfo_sse2:
-	mov eax, [esp+4]
+	%assign push_num 0
-	movdqa xmm0, [eax]	; load 4 sad_8x8
+	LOAD_1_PARA
+	movdqa xmm0, [r0]
 	pshufd xmm1, xmm0, 01Bh
 	paddd xmm1, xmm0
 	pshufd xmm2, xmm1, 0B1h
@@ -391,13 +412,14 @@ MdInterAnalysisVaaInfo_sse2:
 	paddd xmm4, xmm5
 	pshufd xmm5, xmm4, 0B1h
 	paddd xmm5, xmm4
-	movd eax, xmm5
+	
-	cmp eax, 20	; INTER_VARIANCE_SAD_THRESHOLD
+	movd r0d, xmm5
+	cmp r0d, 20	; INTER_VARIANCE_SAD_THRESHOLD
 	jb near .threshold_exit
-	pshufd xmm0, xmm0, 0B1h
+	pshufd xmm0, xmm0, 01Bh
 	pcmpgtd xmm0, xmm1	; iSadBlock > iAverageSad
-	movmskps eax, xmm0
+	movmskps retrd, xmm0
 	ret
 .threshold_exit:
-	mov eax, 15
+	mov retrd, 15
 	ret

codec/decoder/core/asm/block_add.asm

@@ -42,262 +42,6 @@
 
 %include  "asm_inc.asm"
 
-BITS 32
-
-;*******************************************************************************
-; Macros and other preprocessor constants
-;*******************************************************************************
-
-%macro   BLOCK_ADD_16_SSE2   4
-	movdqa    xmm0,       [%2]
-	movdqa    xmm1,       [%3]
-    movdqa    xmm2,       [%3+10h]
-	movdqa    xmm6,       xmm0
-
-	punpcklbw    xmm0,    xmm7
-	punpckhbw    xmm6,    xmm7
-
-	paddw        xmm0,    xmm1
-	paddw        xmm6,    xmm2
-
-	packuswb     xmm0,    xmm6
-	movdqa       [%1],    xmm0
-
-	lea          %2,      [%2+%4]
-	lea          %3,      [%3+%4*2]
-	lea          %1,      [%1+%4]
-%endmacro
-
-%macro    BLOCK_ADD_8_MMXEXT   4
-    movq       mm0,       [%2]
-	movq       mm1,       [%3]
-	movq       mm2,       [%3+08h]
-	movq       mm6,       mm0
-
-	punpcklbw    mm0,     mm7
-	punpckhbw    mm6,     mm7
-
-	paddw        mm0,     mm1
-	paddw        mm6,     mm2
-
-	packuswb     mm0,     mm6
-	movq         [%1],    mm0
-
-	lea          %2,      [%2+%4]
-	lea          %3,      [%3+%4*2]
-	lea          %1,      [%1+%4]
-%endmacro
-
-
-%macro    BLOCK_ADD_16_STRIDE_SSE2  5
-    movdqa    xmm0,       [%2]
-	movdqa    xmm1,       [%3]
-    movdqa    xmm2,       [%3+10h]
-	movdqa    xmm6,       xmm0
-
-	punpcklbw    xmm0,    xmm7
-	punpckhbw    xmm6,    xmm7
-
-	paddw        xmm0,    xmm1
-	paddw        xmm6,    xmm2
-
-	packuswb     xmm0,    xmm6
-	movdqa       [%1],    xmm0
-
-	lea          %2,      [%2+%4]
-	lea          %3,      [%3+%5*2]
-	lea          %1,      [%1+%4]
-%endmacro
-
-
-%macro    BLOCK_ADD_8_STRIDE_MMXEXT   5
-    movq       mm0,       [%2]
-	movq       mm1,       [%3]
-	movq       mm2,       [%3+08h]
-	movq       mm6,       mm0
-
-	punpcklbw    mm0,     mm7
-	punpckhbw    mm6,     mm7
-
-	paddw        mm0,     mm1
-	paddw        mm6,     mm2
-
-	packuswb     mm0,     mm6
-	movq         [%1],    mm0
-
-	lea          %2,      [%2+%4]
-	lea          %3,      [%3+%5*2]
-	lea          %1,      [%1+%4]
-%endmacro
-
-%macro    BLOCK_ADD_8_STRIDE_2_LINES_SSE2   5
-	movdqa xmm1, [%3]
-	movq xmm0, [%2]
-	punpcklbw xmm0, xmm7
-	paddw xmm0, xmm1
-	packuswb xmm0, xmm7
-	movq [%1], xmm0
-
-	movdqa xmm3, [%3+%5*2]
-	movq xmm2, [%2+%4]
-	punpcklbw xmm2, xmm7
-	paddw xmm2, xmm3
-	packuswb xmm2, xmm7
-	movq [%1+%4], xmm2
-
-	lea %1, [%1+%4*2]
-	lea %2, [%2+%4*2]
-	lea %3, [%3+%5*4]
-%endmacro
-
-%macro   CHECK_DATA_16_ZERO_SSE4     3
-    mov        eax,      0h
-	movdqa     xmm0,     [%1]
-	movdqa     xmm1,     [%1+10h]
-	mov        ebx,       [ecx]
-
-	por		   xmm0,	 xmm1
-	ptest      xmm7,     xmm0
-	cmovae     eax,      %3
-
-	add        %1,       20h
-	add        ecx,      04h
-	mov        byte [%2+ebx],  al
-%endmacro
-
-%macro  CHECK_RS_4x4_BLOCK_2_ZERO_SSE4   5
-    movdqa     xmm0,      [%1]
-    movdqa     xmm1,      [%1+%3]
-    movdqa     xmm2,      [%1+%3*2]
-    movdqa     xmm3,      [%1+%4]
-
-    mov        eax,       0h
-    mov        ebx,       0h
-    movdqa     xmm4,      xmm0
-    movdqa     xmm5,      xmm2
-
-    punpcklqdq  xmm0,     xmm1
-    punpckhqdq  xmm4,     xmm1
-    punpcklqdq  xmm2,     xmm3
-    punpckhqdq  xmm5,     xmm3
-
-	por			xmm0,	  xmm2
-	por			xmm4,	  xmm5
-
-    ptest       xmm7,     xmm0
-    cmovae      eax,      %5
-    ptest       xmm7,     xmm4
-    cmovae      ebx,      %5
-
-    mov     byte [%2],    al
-    mov     byte [%2+1],  bl
-%endmacro
-
-%macro   DATA_COPY_16x2_SSE2      3
-    movdqa     xmm0,    [%1]
-	movdqa     xmm1,    [%1+10h]
-	movdqa     xmm2,    [%1+%3]
-	movdqa     xmm3,    [%1+%3+10h]
-
-	movdqa     [%2],    xmm0
-	movdqa     [%2+10h],  xmm1
-	movdqa     [%2+20h],  xmm2
-	movdqa     [%2+30h],  xmm3
-
-	lea        %1,      [%1+%3*2]
-	lea        %2,      [%2+40h]
-%endmacro
-
-
-%macro   DATA_COPY_8x4_SSE2      4
-    movdqa     xmm0,         [%1]
-	movdqa     xmm1,         [%1+%3]
-	movdqa     xmm2,         [%1+%3*2]
-	movdqa     xmm3,         [%1+%4]
-
-	movdqa     [%2],         xmm0
-	movdqa     [%2+10h],     xmm1
-	movdqa     [%2+20h],     xmm2
-	movdqa     [%2+30h],     xmm3
-
-	lea        %1,           [%1+%3*4]
-	lea        %2,           [%2+40h]
-%endmacro
-
-
-%macro   CHECK_DATA_16_ZERO_SSE2   3
-    mov        eax,       0h
-    movdqa     xmm0,      [%1]
-    movdqa     xmm1,      [%1+10h]
-    mov        ebx,       [ecx]
-
-    pcmpeqw    xmm0,      xmm7
-    pcmpeqw    xmm1,      xmm7
-    packsswb   xmm0,      xmm1
-    pmovmskb   edx,       xmm0
-    sub        edx,       0ffffh
-
-    cmovb      eax,       ebp
-    add        ecx,       4
-    add        %1,        20h
-    mov      byte [%2+ebx],    al
-%endmacro
-
-
-
-%macro   CHECK_RS_4x4_BLOCK_2_ZERO_SSE2    5
-    movdqa    xmm0,      [%1]
-    movdqa    xmm1,      [%1 + %3]
-    movdqa    xmm2,      [%1 + %3*2]
-    movdqa    xmm3,      [%1 + %4]
-
-    movdqa    xmm4,       xmm0
-    movdqa    xmm5,       xmm2
-
-    punpcklqdq   xmm0,    xmm1
-    punpckhqdq   xmm4,    xmm1
-    punpcklqdq   xmm2,    xmm3
-    punpckhqdq   xmm5,    xmm3
-
-    pcmpeqw      xmm0,    xmm7
-    pcmpeqw      xmm2,    xmm7
-    pcmpeqw      xmm4,    xmm7
-    pcmpeqw      xmm5,    xmm7
-
-    packsswb     xmm0,    xmm2
-    packsswb     xmm4,    xmm5
-    pmovmskb     eax,     xmm0
-    pmovmskb     ebx,     xmm4
-
-    sub          eax,     0ffffh
-    mov          eax,     0
-    cmovb        eax,     %5
-    sub          ebx,     0ffffh
-    mov          ebx,     0
-    cmovb        ebx,     %5
-    mov       byte [%2],    al
-    mov       byte [%2+1],  bl
-%endmacro
-
-;*******************************************************************************
-; Data
-;*******************************************************************************
-
-%ifdef FORMAT_COFF
-SECTION .rodata data
-%else
-SECTION .rodata align=16
-%endif
-
-ALIGN  16
-SubMbScanIdx:
-     dd    0x0,  0x1,  0x4,  0x5,
-	 dd    0x2,  0x3,  0x6,  0x7,
-	 dd    0x8,  0x9,  0xc,  0xd,
-	 dd    0xa,  0xb,  0xe,  0xf,
-	 dd    0x10, 0x11, 0x14, 0x15,
-	 dd    0x12, 0x13, 0x16, 0x17,
-
 ;*******************************************************************************
 ; Code
 ;*******************************************************************************
@@ -312,71 +56,77 @@ ALIGN    16
 ;  void_t WelsResBlockZero16x16_sse2(int16_t* pBlock,int32_t iStride)
 ;*******************************************************************************
 WelsResBlockZero16x16_sse2:
-    push     esi
+        ;push     r0
-
+        %assign push_num 0
-	mov      esi,        [esp+08h]
+        LOAD_2_PARA
-	mov      ecx,        [esp+0ch]
+		%ifndef X86_32
-	lea      ecx,        [ecx*2]
+		movsx r1, r1d
-	lea      eax,        [ecx*3]
+		%endif
+	;mov      r0,        [esp+08h]
+	;mov      r1,        [esp+0ch]
+	;lea      r1,        [r1*2]
+        lea 	r1, 	[r1*2]
+	;lea      r2,        [r1*3]
+        lea 	r2,	[r1*3]
 
 	pxor     xmm7,       xmm7
 
     ; four  lines
-	movdqa   [esi],      xmm7
+	movdqa   [r0],      xmm7
-	movdqa   [esi+10h],  xmm7
+	movdqa   [r0+10h],  xmm7
 
-	movdqa   [esi+ecx],  xmm7
+	movdqa   [r0+r1],  xmm7
-	movdqa   [esi+ecx+10h],     xmm7
+	movdqa   [r0+r1+10h],     xmm7
 
-    movdqa   [esi+ecx*2],   xmm7
+    movdqa   [r0+r1*2],   xmm7
-	movdqa   [esi+ecx*2+10h],   xmm7
+	movdqa   [r0+r1*2+10h],   xmm7
 
-	movdqa   [esi+eax],     xmm7
+	movdqa   [r0+r2],     xmm7
-	movdqa   [esi+eax+10h],     xmm7
+	movdqa   [r0+r2+10h],     xmm7
 
     ;  four lines
-	lea      esi,       [esi+ecx*4]
+	lea      r0,       [r0+r1*4]
-	movdqa   [esi],      xmm7
+	movdqa   [r0],      xmm7
-	movdqa   [esi+10h],  xmm7
+	movdqa   [r0+10h],  xmm7
 
-	movdqa   [esi+ecx],  xmm7
+	movdqa   [r0+r1],  xmm7
-	movdqa   [esi+ecx+10h],     xmm7
+	movdqa   [r0+r1+10h],     xmm7
 
-    movdqa   [esi+ecx*2],   xmm7
+    movdqa   [r0+r1*2],   xmm7
-	movdqa   [esi+ecx*2+10h],   xmm7
+	movdqa   [r0+r1*2+10h],   xmm7
 
-	movdqa   [esi+eax],     xmm7
+	movdqa   [r0+r2],     xmm7
-	movdqa   [esi+eax+10h],     xmm7
+	movdqa   [r0+r2+10h],     xmm7
 
 	;  four lines
-	lea      esi,       [esi+ecx*4]
+	lea      r0,       [r0+r1*4]
-	movdqa   [esi],      xmm7
+	movdqa   [r0],      xmm7
-	movdqa   [esi+10h],  xmm7
+	movdqa   [r0+10h],  xmm7
 
-	movdqa   [esi+ecx],  xmm7
+	movdqa   [r0+r1],  xmm7
-	movdqa   [esi+ecx+10h],     xmm7
+	movdqa   [r0+r1+10h],     xmm7
 
-    movdqa   [esi+ecx*2],   xmm7
+    movdqa   [r0+r1*2],   xmm7
-	movdqa   [esi+ecx*2+10h],   xmm7
+	movdqa   [r0+r1*2+10h],   xmm7
 
-	movdqa   [esi+eax],     xmm7
+	movdqa   [r0+r2],     xmm7
-	movdqa   [esi+eax+10h],     xmm7
+	movdqa   [r0+r2+10h],     xmm7
 
 	;  four lines
-	lea      esi,       [esi+ecx*4]
+	lea      r0,       [r0+r1*4]
-	movdqa   [esi],      xmm7
+	movdqa   [r0],      xmm7
-	movdqa   [esi+10h],  xmm7
+	movdqa   [r0+10h],  xmm7
 
-	movdqa   [esi+ecx],  xmm7
+	movdqa   [r0+r1],  xmm7
-	movdqa   [esi+ecx+10h],     xmm7
+	movdqa   [r0+r1+10h],     xmm7
 
-    movdqa   [esi+ecx*2],   xmm7
+    movdqa   [r0+r1*2],   xmm7
-	movdqa   [esi+ecx*2+10h],   xmm7
+	movdqa   [r0+r1*2+10h],   xmm7
 
-	movdqa   [esi+eax],     xmm7
+	movdqa   [r0+r2],     xmm7
-	movdqa   [esi+eax+10h],     xmm7
+	movdqa   [r0+r2+10h],     xmm7
 
-    pop      esi
+    ;pop      r0
 	ret
 
 
@@ -387,27 +137,31 @@ ALIGN    16
 ;  void_t WelsResBlockZero8x8_sse2(int16_t * pBlock, int32_t iStride)
 ;*******************************************************************************
 WelsResBlockZero8x8_sse2:
-	  push      esi
+	  ;push      r0
-
+	  %assign push_num 0
-      mov       esi,     [esp+08h]
+          LOAD_2_PARA
-	  mov       ecx,     [esp+0ch]
+		  %ifndef X86_32
-	  lea       ecx,     [ecx*2]
+		  movsx r1, r1d
-	  lea       eax,     [ecx*3]
+		  %endif
+      	  ;mov       r0,     [esp+08h]
+	  ;mov       r1,     [esp+0ch]
+	  lea       r1,     [r1*2]
+	  lea       r2,     [r1*3]
 
 	  pxor      xmm7,          xmm7
 
-	  movdqa    [esi],         xmm7
+	  movdqa    [r0],         xmm7
-	  movdqa    [esi+ecx],     xmm7
+	  movdqa    [r0+r1],     xmm7
-	  movdqa    [esi+ecx*2],   xmm7
+	  movdqa    [r0+r1*2],   xmm7
-	  movdqa    [esi+eax],     xmm7
+	  movdqa    [r0+r2],     xmm7
 
-	  lea       esi,     [esi+ecx*4]
+	  lea       r0,     [r0+r1*4]
-	  movdqa    [esi],         xmm7
+	  movdqa    [r0],         xmm7
-	  movdqa    [esi+ecx],     xmm7
+	  movdqa    [r0+r1],     xmm7
-	  movdqa    [esi+ecx*2],   xmm7
+	  movdqa    [r0+r1*2],   xmm7
-	  movdqa    [esi+eax],     xmm7
+	  movdqa    [r0+r2],     xmm7
 
 
-	  pop       esi
+	  ;pop       r0
 	  ret
 

codec/decoder/core/asm/cpuid.asm

@@ -1,169 +0,0 @@
-;*!
-;* \copy
-;*     Copyright (c)  2009-2013, Cisco Systems
-;*     All rights reserved.
-;*
-;*     Redistribution and use in source and binary forms, with or without
-;*     modification, are permitted provided that the following conditions
-;*     are met:
-;*
-;*        * Redistributions of source code must retain the above copyright
-;*          notice, this list of conditions and the following disclaimer.
-;*
-;*        * Redistributions in binary form must reproduce the above copyright
-;*          notice, this list of conditions and the following disclaimer in
-;*          the documentation and/or other materials provided with the
-;*          distribution.
-;*
-;*     THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
-;*     "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
-;*     LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
-;*     FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
-;*     COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
-;*     INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
-;*     BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
-;*     LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
-;*     CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
-;*     LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
-;*     ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
-;*     POSSIBILITY OF SUCH DAMAGE.
-;*
-;*
-;*	cpu_mmx.asm
-;*
-;*  Abstract
-;*		verify cpuid feature support and cpuid detection
-;*
-;*  History
-;*      04/29/2009	Created
-;*
-;*************************************************************************/
-
-bits 32
-
-;******************************************************************************************
-; Macros
-;******************************************************************************************
-
-%macro WELS_EXTERN 1
-	%ifdef PREFIX
-		global _%1
-		%define %1 _%1
-	%else
-		global %1
-	%endif
-%endmacro
-
-;******************************************************************************************
-; Code
-;******************************************************************************************
-
-SECTION .text
-
-; refer to "The IA-32 Intel(R) Architecture Software Developers Manual, Volume 2A A-M"
-; section CPUID - CPU Identification
-
-WELS_EXTERN WelsCPUIdVerify
-ALIGN 16
-;******************************************************************************************
-;   int32_t WelsCPUIdVerify()
-;******************************************************************************************
-WelsCPUIdVerify:
-    pushfd					; decrease the SP by 4 and load EFLAGS register onto stack, pushfd 32 bit and pushf for 16 bit
-	pushfd					; need push 2 EFLAGS, one for processing and the another one for storing purpose
-    pop     ecx				; get EFLAGS to bit manipulation
-    mov     eax, ecx		; store into ecx followed
-    xor     eax, 00200000h	; get ID flag (bit 21) of EFLAGS to directly indicate cpuid support or not
-	xor		eax, ecx		; get the ID flag bitwise, eax - 0: not support; otherwise: support
-    popfd					; store back EFLAGS and keep unchanged for system
-    ret
-
-WELS_EXTERN WelsCPUId
-ALIGN 16
-;****************************************************************************************************
-;   void WelsCPUId( int32_t index, int32_t *uiFeatureA, int32_t *uiFeatureB, int32_t *uiFeatureC, int32_t *uiFeatureD )
-;****************************************************************************************************
-WelsCPUId:
-	push	ebx
-	push	edi
-
-	mov     eax, [esp+12]	; operating index
-    cpuid					; cpuid
-
-	; processing various information return
-	mov     edi, [esp+16]
-    mov     [edi], eax
-    mov     edi, [esp+20]
-    mov     [edi], ebx
-    mov     edi, [esp+24]
-    mov     [edi], ecx
-    mov     edi, [esp+28]
-    mov     [edi], edx
-
-	pop		edi
-    pop     ebx
-	ret
-
-WELS_EXTERN WelsCPUSupportAVX
-; need call after cpuid=1 and eax, ecx flag got then
-ALIGN 16
-;****************************************************************************************************
-;   int32_t WelsCPUSupportAVX( uint32_t eax, uint32_t ecx )
-;****************************************************************************************************
-WelsCPUSupportAVX:
-	mov eax, [esp+4]
-	mov ecx, [esp+8]
-
-	; refer to detection of AVX addressed in INTEL AVX manual document
-	and ecx, 018000000H
-	cmp ecx, 018000000H		; check both OSXSAVE and AVX feature flags
-	jne avx_not_supported
-	; processor supports AVX instructions and XGETBV is enabled by OS
-	mov ecx, 0				; specify 0 for XFEATURE_ENABLED_MASK register
-	XGETBV					; result in EDX:EAX
-	and eax, 06H
-	cmp eax, 06H			; check OS has enabled both XMM and YMM state support
-	jne avx_not_supported
-	mov eax, 1
-	ret
-avx_not_supported:
-	mov eax, 0
-	ret
-
-WELS_EXTERN WelsCPUSupportFMA
-; need call after cpuid=1 and eax, ecx flag got then
-ALIGN 16
-;****************************************************************************************************
-;   int32_t WelsCPUSupportFMA( uint32_t eax, uint32_t ecx )
-;****************************************************************************************************
-WelsCPUSupportFMA:
-	mov eax, [esp+4]
-	mov ecx, [esp+8]
-
-	; refer to detection of FMA addressed in INTEL AVX manual document
-	and ecx, 018001000H
-	cmp ecx, 018001000H		; check OSXSAVE, AVX, FMA feature flags
-	jne fma_not_supported
-	; processor supports AVX,FMA instructions and XGETBV is enabled by OS
-	mov ecx, 0				; specify 0 for XFEATURE_ENABLED_MASK register
-	XGETBV					; result in EDX:EAX
-	and eax, 06H
-	cmp eax, 06H			; check OS has enabled both XMM and YMM state support
-	jne fma_not_supported
-	mov eax, 1
-	ret
-fma_not_supported:
-	mov eax, 0
-	ret
-
-WELS_EXTERN WelsEmms
-ALIGN 16
-;******************************************************************************************
-;   void WelsEmms()
-;******************************************************************************************
-WelsEmms:
-	emms	; empty mmx technology states
-	ret
-
-
-

codec/decoder/core/asm/dct.asm

@@ -42,8 +42,6 @@
 
 %include "asm_inc.asm"
 
-BITS 32
-
 ;*******************************************************************************
 ; Macros and other preprocessor constants
 ;*******************************************************************************
@@ -93,19 +91,15 @@ ALIGN 16
 ;*******************************************************************************
 
 IdctResAddPred_mmx:
-
+    %assign push_num 0
-%define	pushsize	0
+    LOAD_3_PARA
-%define pPred       esp+pushsize+4
+	%ifndef X86_32
-%define kiStride     esp+pushsize+8
+	movsx r1, r1d
-%define pRs         esp+pushsize+12
+	%endif
-
+    movq    mm0, [r2+ 0]
-	mov     eax, [pRs   ]
+    movq    mm1, [r2+ 8]
-    mov     edx, [pPred ]
+    movq    mm2, [r2+16]
-    mov     ecx, [kiStride]
+    movq    mm3, [r2+24]
-    movq    mm0, [eax+ 0]
-    movq    mm1, [eax+ 8]
-    movq    mm2, [eax+16]
-    movq    mm3, [eax+24]
 
 	MMX_Trans4x4W        mm0, mm1, mm2, mm3, mm4
 	MMX_IDCT			mm1, mm2, mm3, mm4, mm0, mm6
@@ -115,15 +109,12 @@ IdctResAddPred_mmx:
     WELS_Zero			mm7
     WELS_DW32			mm6
 
-    MMX_StoreDiff4P    mm3, mm0, mm6, mm7, [edx]
+    MMX_StoreDiff4P    mm3, mm0, mm6, mm7, [r0]
-    MMX_StoreDiff4P    mm4, mm0, mm6, mm7, [edx+ecx]
+    MMX_StoreDiff4P    mm4, mm0, mm6, mm7, [r0+r1]
-    lea     edx, [edx+2*ecx]
+    lea     r0, [r0+2*r1]
-    MMX_StoreDiff4P    mm1, mm0, mm6, mm7, [edx]
+    MMX_StoreDiff4P    mm1, mm0, mm6, mm7, [r0]
-    MMX_StoreDiff4P    mm2, mm0, mm6, mm7, [edx+ecx]
+    MMX_StoreDiff4P    mm2, mm0, mm6, mm7, [r0+r1]
+
 
-%undef	pushsize
-%undef  pPred
-%undef  kiStride
-%undef  pRs
 	emms
     ret

codec/decoder/core/asm/expand_picture.asm

@@ -1,655 +0,0 @@
-;*!
-;* \copy
-;*     Copyright (c)  2009-2013, Cisco Systems
-;*     All rights reserved.
-;*
-;*     Redistribution and use in source and binary forms, with or without
-;*     modification, are permitted provided that the following conditions
-;*     are met:
-;*
-;*        * Redistributions of source code must retain the above copyright
-;*          notice, this list of conditions and the following disclaimer.
-;*
-;*        * Redistributions in binary form must reproduce the above copyright
-;*          notice, this list of conditions and the following disclaimer in
-;*          the documentation and/or other materials provided with the
-;*          distribution.
-;*
-;*     THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
-;*     "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
-;*     LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
-;*     FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
-;*     COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
-;*     INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
-;*     BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
-;*     LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
-;*     CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
-;*     LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
-;*     ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
-;*     POSSIBILITY OF SUCH DAMAGE.
-;*
-;*
-;*  expand_picture.asm
-;*
-;*  Abstract
-;*      mmxext/sse for expand_frame
-;*
-;*  History
-;*      09/25/2009 Created
-;*
-;*
-;*************************************************************************/
-
-%include "asm_inc.asm"
-
-BITS 32
-
-;***********************************************************************
-; Macros and other preprocessor constants
-;***********************************************************************
-
-;***********************************************************************
-; Local Data (Read Only)
-;***********************************************************************
-
-;SECTION .rodata pData align=16
-
-;***********************************************************************
-; Various memory constants (trigonometric values or rounding values)
-;***********************************************************************
-;%define PADDING_SIZE_ASM 	32 	; PADDING_LENGTH
-
-;***********************************************************************
-; Code
-;***********************************************************************
-
-
-
-SECTION .text
-
-;WELS_EXTERN expand_picture_luma_mmx
-;WELS_EXTERN expand_picture_chroma_mmx
-WELS_EXTERN ExpandPictureLuma_sse2
-WELS_EXTERN ExpandPictureChromaAlign_sse2	; for chroma alignment
-WELS_EXTERN ExpandPictureChromaUnalign_sse2	; for chroma unalignment
-
-;;;;;;;expanding result;;;;;;;
-
-;aaaa|attttttttttttttttb|bbbb
-;aaaa|attttttttttttttttb|bbbb
-;aaaa|attttttttttttttttb|bbbb
-;aaaa|attttttttttttttttb|bbbb
-;----------------------------
-;aaaa|attttttttttttttttb|bbbb
-;llll|l                r|rrrr
-;llll|l                r|rrrr
-;llll|l                r|rrrr
-;llll|l                r|rrrr
-;llll|l                r|rrrr
-;cccc|ceeeeeeeeeeeeeeeed|dddd
-;----------------------------
-;cccc|ceeeeeeeeeeeeeeeed|dddd
-;cccc|ceeeeeeeeeeeeeeeed|dddd
-;cccc|ceeeeeeeeeeeeeeeed|dddd
-;cccc|ceeeeeeeeeeeeeeeed|dddd
-
-%macro mov_line_8x4_mmx		3	; dst, stride, mm?
-	movq [%1], %3
-	movq [%1+%2], %3
-	lea %1, [%1+2*%2]
-	movq [%1], %3
-	movq [%1+%2], %3
-	lea %1, [%1+2*%2]
-%endmacro
-
-%macro mov_line_end8x4_mmx		3	; dst, stride, mm?
-	movq [%1], %3
-	movq [%1+%2], %3
-	lea %1, [%1+2*%2]
-	movq [%1], %3
-	movq [%1+%2], %3
-	lea %1, [%1+%2]
-%endmacro
-
-%macro mov_line_16x4_sse2	4	; dst, stride, xmm?, u/a
-	movdq%4 [%1], %3 		; top(bottom)_0
-	movdq%4 [%1+%2], %3		; top(bottom)_1
-	lea %1, [%1+2*%2]
-	movdq%4 [%1], %3 		; top(bottom)_2
-	movdq%4 [%1+%2], %3		; top(bottom)_3
-	lea %1, [%1+2*%2]
-%endmacro
-
-%macro mov_line_end16x4_sse2	4	; dst, stride, xmm?, u/a
-	movdq%4 [%1], %3 		; top(bottom)_0
-	movdq%4 [%1+%2], %3		; top(bottom)_1
-	lea %1, [%1+2*%2]
-	movdq%4 [%1], %3 		; top(bottom)_2
-	movdq%4 [%1+%2], %3		; top(bottom)_3
-	lea %1, [%1+%2]
-%endmacro
-
-%macro mov_line_32x4_sse2	3	; dst, stride, xmm?
-	movdqa [%1], %3 		; top(bottom)_0
-	movdqa [%1+16], %3 		; top(bottom)_0
-	movdqa [%1+%2], %3		; top(bottom)_1
-	movdqa [%1+%2+16], %3		; top(bottom)_1
-	lea %1, [%1+2*%2]
-	movdqa [%1], %3 		; top(bottom)_2
-	movdqa [%1+16], %3 		; top(bottom)_2
-	movdqa [%1+%2], %3		; top(bottom)_3
-	movdqa [%1+%2+16], %3		; top(bottom)_3
-	lea %1, [%1+2*%2]
-%endmacro
-
-%macro mov_line_end32x4_sse2	3	; dst, stride, xmm?
-	movdqa [%1], %3 		; top(bottom)_0
-	movdqa [%1+16], %3 		; top(bottom)_0
-	movdqa [%1+%2], %3		; top(bottom)_1
-	movdqa [%1+%2+16], %3		; top(bottom)_1
-	lea %1, [%1+2*%2]
-	movdqa [%1], %3 		; top(bottom)_2
-	movdqa [%1+16], %3 		; top(bottom)_2
-	movdqa [%1+%2], %3		; top(bottom)_3
-	movdqa [%1+%2+16], %3		; top(bottom)_3
-	lea %1, [%1+%2]
-%endmacro
-
-%macro exp_top_bottom_sse2	1	; iPaddingSize [luma(32)/chroma(16)]
-	; ebx [width/16(8)]
-	; esi [pSrc+0], edi [pSrc-1], ecx [-stride], 32(16)		; top
-	; eax [pSrc+(h-1)*stride], ebp [pSrc+(h+31)*stride], 32(16)	; bottom
-
-%if %1 == 32		; for luma
-	sar ebx, 04h 	; width / 16(8) pixels
-.top_bottom_loops:
-	; top
-	movdqa xmm0, [esi]		; first line of picture pData
-	mov_line_16x4_sse2 edi, ecx, xmm0, a	; dst, stride, xmm?
-	mov_line_16x4_sse2 edi, ecx, xmm0, a
-	mov_line_16x4_sse2 edi, ecx, xmm0, a
-	mov_line_16x4_sse2 edi, ecx, xmm0, a
-	mov_line_16x4_sse2 edi, ecx, xmm0, a	; dst, stride, xmm?
-	mov_line_16x4_sse2 edi, ecx, xmm0, a
-	mov_line_16x4_sse2 edi, ecx, xmm0, a
-	mov_line_end16x4_sse2 edi, ecx, xmm0, a
-
-	; bottom
-	movdqa xmm1, [eax] 		; last line of picture pData
-	mov_line_16x4_sse2 ebp, ecx, xmm1, a	; dst, stride, xmm?
-	mov_line_16x4_sse2 ebp, ecx, xmm1, a
-	mov_line_16x4_sse2 ebp, ecx, xmm1, a
-	mov_line_16x4_sse2 ebp, ecx, xmm1, a
-	mov_line_16x4_sse2 ebp, ecx, xmm1, a	; dst, stride, xmm?
-	mov_line_16x4_sse2 ebp, ecx, xmm1, a
-	mov_line_16x4_sse2 ebp, ecx, xmm1, a
-	mov_line_end16x4_sse2 ebp, ecx, xmm1, a
-
-	lea esi, [esi+16]		; top pSrc
-	lea edi, [edi+16]		; top dst
-	lea eax, [eax+16]		; bottom pSrc
-	lea ebp, [ebp+16]		; bottom dst
-	neg ecx 			; positive/negative stride need for next loop?
-
-	dec ebx
-	jnz near .top_bottom_loops
-%elif %1 == 16	; for chroma ??
-	mov edx, ebx
-	sar ebx, 04h 	; (width / 16) pixels
-.top_bottom_loops:
-	; top
-	movdqa xmm0, [esi]		; first line of picture pData
-	mov_line_16x4_sse2 edi, ecx, xmm0, a	; dst, stride, xmm?
-	mov_line_16x4_sse2 edi, ecx, xmm0, a
-	mov_line_16x4_sse2 edi, ecx, xmm0, a
-	mov_line_end16x4_sse2 edi, ecx, xmm0, a
-
-	; bottom
-	movdqa xmm1, [eax] 		; last line of picture pData
-	mov_line_16x4_sse2 ebp, ecx, xmm1, a	; dst, stride, xmm?
-	mov_line_16x4_sse2 ebp, ecx, xmm1, a
-	mov_line_16x4_sse2 ebp, ecx, xmm1, a
-	mov_line_end16x4_sse2 ebp, ecx, xmm1, a
-
-	lea esi, [esi+16]		; top pSrc
-	lea edi, [edi+16]		; top dst
-	lea eax, [eax+16]		; bottom pSrc
-	lea ebp, [ebp+16]		; bottom dst
-	neg ecx 			; positive/negative stride need for next loop?
-
-	dec ebx
-	jnz near .top_bottom_loops
-
-	; for remaining 8 bytes
-	and edx, 0fh		; any 8 bytes left?
-	test edx, edx
-	jz near .to_be_continued	; no left to exit here
-
-	; top
-	movq mm0, [esi]		; remained 8 byte
-	mov_line_8x4_mmx edi, ecx, mm0	; dst, stride, mm?
-	mov_line_8x4_mmx edi, ecx, mm0	; dst, stride, mm?
-	mov_line_8x4_mmx edi, ecx, mm0	; dst, stride, mm?
-	mov_line_end8x4_mmx edi, ecx, mm0	; dst, stride, mm?
-	; bottom
-	movq mm1, [eax]
-	mov_line_8x4_mmx ebp, ecx, mm1	; dst, stride, mm?
-	mov_line_8x4_mmx ebp, ecx, mm1	; dst, stride, mm?
-	mov_line_8x4_mmx ebp, ecx, mm1	; dst, stride, mm?
-	mov_line_end8x4_mmx ebp, ecx, mm1	; dst, stride, mm?
-	WELSEMMS
-
-.to_be_continued:
-%endif
-%endmacro
-
-%macro exp_left_right_sse2	2	; iPaddingSize [luma(32)/chroma(16)], u/a
-	; ecx [height]
-	; esi [pSrc+0], 	   edi [pSrc-32], edx [stride], 32(16)	; left
-	; ebx [pSrc+(w-1)], ebp [pSrc+w], 32(16)			; right
-;	xor eax, eax 	; for pixel pData (uint8_t)		; make sure eax=0 at least high 24 bits of eax = 0
-
-%if %1 == 32		; for luma
-.left_right_loops:
-	; left
-	mov al, byte [esi]		; pixel pData for left border
-	butterfly_1to16_sse	xmm0, xmm1, a				; dst, tmp, pSrc [generic register name: a/b/c/d]
-	movdqa [edi], xmm0
-	movdqa [edi+16], xmm0
-
-	; right
-	mov al, byte [ebx]
-	butterfly_1to16_sse	xmm1, xmm2, a				; dst, tmp, pSrc [generic register name: a/b/c/d]
-	movdqa [ebp], xmm1
-	movdqa [ebp+16], xmm1
-
-	lea esi, [esi+edx]		; left pSrc
-	lea edi, [edi+edx]		; left dst
-	lea ebx, [ebx+edx]		; right pSrc
-	lea ebp, [ebp+edx]		; right dst
-
-	dec ecx
-	jnz near .left_right_loops
-%elif %1 == 16	; for chroma ??
-.left_right_loops:
-	; left
-	mov al, byte [esi]		; pixel pData for left border
-	butterfly_1to16_sse	xmm0, xmm1, a				; dst, tmp, pSrc [generic register name: a/b/c/d]
-	movdqa [edi], xmm0
-
-	; right
-	mov al, byte [ebx]
-	butterfly_1to16_sse	xmm1, xmm2, a				; dst, tmp, pSrc [generic register name: a/b/c/d]
-	movdq%2 [ebp], xmm1								; might not be aligned 16 bytes in case chroma planes
-
-	lea esi, [esi+edx]		; left pSrc
-	lea edi, [edi+edx]		; left dst
-	lea ebx, [ebx+edx]		; right pSrc
-	lea ebp, [ebp+edx]		; right dst
-
-	dec ecx
-	jnz near .left_right_loops
-%endif
-%endmacro
-
-%macro exp_cross_sse2	2	; iPaddingSize [luma(32)/chroma(16)], u/a
-	; top-left: (x)mm3, top-right: (x)mm4, bottom-left: (x)mm5, bottom-right: (x)mm6
-	; edi: TL, ebp: TR, eax: BL, ebx: BR, ecx, -stride
-%if %1 == 32		; luma
-	; TL
-	mov_line_32x4_sse2	edi, ecx, xmm3	; dst, stride, xmm?
-	mov_line_32x4_sse2	edi, ecx, xmm3	; dst, stride, xmm?
-	mov_line_32x4_sse2	edi, ecx, xmm3	; dst, stride, xmm?
-	mov_line_32x4_sse2	edi, ecx, xmm3	; dst, stride, xmm?
-	mov_line_32x4_sse2	edi, ecx, xmm3	; dst, stride, xmm?
-	mov_line_32x4_sse2	edi, ecx, xmm3	; dst, stride, xmm?
-	mov_line_32x4_sse2	edi, ecx, xmm3	; dst, stride, xmm?
-	mov_line_end32x4_sse2	edi, ecx, xmm3	; dst, stride, xmm?
-
-	; TR
-	mov_line_32x4_sse2	ebp, ecx, xmm4	; dst, stride, xmm?
-	mov_line_32x4_sse2	ebp, ecx, xmm4	; dst, stride, xmm?
-	mov_line_32x4_sse2	ebp, ecx, xmm4	; dst, stride, xmm?
-	mov_line_32x4_sse2	ebp, ecx, xmm4	; dst, stride, xmm?
-	mov_line_32x4_sse2	ebp, ecx, xmm4	; dst, stride, xmm?
-	mov_line_32x4_sse2	ebp, ecx, xmm4	; dst, stride, xmm?
-	mov_line_32x4_sse2	ebp, ecx, xmm4	; dst, stride, xmm?
-	mov_line_end32x4_sse2	ebp, ecx, xmm4	; dst, stride, xmm?
-
-	; BL
-	mov_line_32x4_sse2	eax, ecx, xmm5	; dst, stride, xmm?
-	mov_line_32x4_sse2	eax, ecx, xmm5	; dst, stride, xmm?
-	mov_line_32x4_sse2	eax, ecx, xmm5	; dst, stride, xmm?
-	mov_line_32x4_sse2	eax, ecx, xmm5	; dst, stride, xmm?
-	mov_line_32x4_sse2	eax, ecx, xmm5	; dst, stride, xmm?
-	mov_line_32x4_sse2	eax, ecx, xmm5	; dst, stride, xmm?
-	mov_line_32x4_sse2	eax, ecx, xmm5	; dst, stride, xmm?
-	mov_line_end32x4_sse2	eax, ecx, xmm5	; dst, stride, xmm?
-
-	; BR
-	mov_line_32x4_sse2	ebx, ecx, xmm6	; dst, stride, xmm?
-	mov_line_32x4_sse2	ebx, ecx, xmm6	; dst, stride, xmm?
-	mov_line_32x4_sse2	ebx, ecx, xmm6	; dst, stride, xmm?
-	mov_line_32x4_sse2	ebx, ecx, xmm6	; dst, stride, xmm?
-	mov_line_32x4_sse2	ebx, ecx, xmm6	; dst, stride, xmm?
-	mov_line_32x4_sse2	ebx, ecx, xmm6	; dst, stride, xmm?
-	mov_line_32x4_sse2	ebx, ecx, xmm6	; dst, stride, xmm?
-	mov_line_end32x4_sse2	ebx, ecx, xmm6	; dst, stride, xmm?
-%elif %1 == 16	; chroma
-	; TL
-	mov_line_16x4_sse2	edi, ecx, xmm3, a	; dst, stride, xmm?
-	mov_line_16x4_sse2	edi, ecx, xmm3, a	; dst, stride, xmm?
-	mov_line_16x4_sse2	edi, ecx, xmm3, a	; dst, stride, xmm?
-	mov_line_end16x4_sse2	edi, ecx, xmm3, a	; dst, stride, xmm?
-
-	; TR
-	mov_line_16x4_sse2	ebp, ecx, xmm4, %2	; dst, stride, xmm?
-	mov_line_16x4_sse2	ebp, ecx, xmm4, %2	; dst, stride, xmm?
-	mov_line_16x4_sse2	ebp, ecx, xmm4, %2	; dst, stride, xmm?
-	mov_line_end16x4_sse2 ebp, ecx, xmm4, %2	; dst, stride, xmm?
-
-	; BL
-	mov_line_16x4_sse2	eax, ecx, xmm5, a	; dst, stride, xmm?
-	mov_line_16x4_sse2	eax, ecx, xmm5, a	; dst, stride, xmm?
-	mov_line_16x4_sse2	eax, ecx, xmm5, a	; dst, stride, xmm?
-	mov_line_end16x4_sse2	eax, ecx, xmm5, a	; dst, stride, xmm?
-
-	; BR
-	mov_line_16x4_sse2	ebx, ecx, xmm6, %2	; dst, stride, xmm?
-	mov_line_16x4_sse2	ebx, ecx, xmm6, %2	; dst, stride, xmm?
-	mov_line_16x4_sse2	ebx, ecx, xmm6, %2	; dst, stride, xmm?
-	mov_line_end16x4_sse2	ebx, ecx, xmm6, %2	; dst, stride, xmm?
-%endif
-%endmacro
-
-ALIGN 16
-;***********************************************************************----------------
-; void ExpandPictureLuma_sse2(	uint8_t *pDst,
-;									const int32_t kiStride,
-;									const int32_t kiWidth,
-;									const int32_t kiHeight	);
-;***********************************************************************----------------
-ExpandPictureLuma_sse2:
-	push ebx
-	push edx
-	push esi
-	push edi
-	push ebp
-
-	; for both top and bottom border
-	;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
-	mov esi, [esp+24]						; pDst
-	mov edx, [esp+28]						; kiStride
-	mov ebx, [esp+32]						; kiWidth
-	mov eax, [esp+36]						; kiHeight
-	; also prepare for cross border pData top-left: xmm3
-;	xor ecx, ecx
-	mov cl, byte [esi]
-	butterfly_1to16_sse xmm3, xmm4, c		; pDst, tmp, pSrc [generic register name: a/b/c/d]
-	; load top border
-	mov ecx, edx							; kiStride
-	neg ecx 								; -kiStride
-	lea edi, [esi+ecx]						; last line of top border
-	; load bottom border
-	dec eax									; h-1
-	imul eax, edx 							; (h-1)*kiStride
-	lea eax, [esi+eax]						; last line of picture pData
-	sal edx, 05h							; 32*kiStride
-	lea ebp, [eax+edx]						; last line of bottom border, (h-1)*stride + 32 * stride
-	; also prepare for cross border pData: bottom-left with xmm5, bottom-right xmm6
-	dec ebx									; kiWidth-1
-	lea ebx, [eax+ebx]						; dst[w-1][h-1]
-;	xor edx, edx
-	mov dl, byte [eax]						; bottom-left
-	butterfly_1to16_sse xmm5, xmm6, d		; dst, tmp, pSrc [generic register name: a/b/c/d]
-	mov dl, byte [ebx]						; bottom-right
-	butterfly_1to16_sse xmm6, xmm4, d		; dst, tmp, pSrc [generic register name: a/b/c/d]
-	; for top & bottom expanding
-	mov ebx, [esp+32]						; kiWidth
-	exp_top_bottom_sse2	32
-
-	; for both left and right border
-	;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
-	mov esi, [esp+24]						; p_dst: left border pSrc
-	mov edx, [esp+28]						; kiStride
-	mov ebx, [esp+32]						; kiWidth
-	mov ecx, [esp+36]						; kiHeight
-	; load left border
-	mov eax, -32 							; luma=-32, chroma=-16
-	lea edi, [esi+eax]						; left border dst
-	dec ebx
-	lea ebx, [esi+ebx]						; right border pSrc, (p_dst + width - 1)
-	lea ebp, [ebx+1]						; right border dst
-	; prepare for cross border pData: top-right with xmm4
-;	xor eax, eax
-	mov al, byte [ebx]						; top-right
-	butterfly_1to16_sse xmm4, xmm0, a		; pDst, tmp, pSrc [generic register name: a/b/c/d]
-	; for left & right border expanding
-	exp_left_right_sse2	32, a
-
-	; for cross border [top-left, top-right, bottom-left, bottom-right]
-	;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
-	mov esi, [esp+24]						; pDst
-	mov ecx, [esp+28]						; kiStride
-	mov ebx, [esp+32]						; kiWidth
-	mov edx, [esp+36]						; kiHeight
-	; have done xmm3,..,xmm6 cross pData initialization above, perform pading as below, To be continued..
-	mov eax, -32							; luma=-32, chroma=-16
-	neg ecx										; -stride
-	lea edi, [esi+eax]
-	lea edi, [edi+ecx]				; last line of top-left border
-	lea ebp, [esi+ebx]
-	lea ebp, [ebp+ecx]				; last line of top-right border
-	add edx, 32								; height+32(16), luma=32, chroma=16
-	mov ecx, [esp+28]					; kiStride
-	imul edx, ecx							; (height+32(16)) * stride
-	lea eax, [edi+edx]						; last line of bottom-left border
-	lea ebx, [ebp+edx]						; last line of bottom-right border
-	neg ecx										; -kiStride
-	; for left & right border expanding
-	exp_cross_sse2		32, a
-
-;	sfence									; commit cache write back memory
-
-	pop ebp
-	pop edi
-	pop esi
-	pop edx
-	pop ebx
-
-	ret
-
-ALIGN 16
-;***********************************************************************----------------
-; void ExpandPictureChromaAlign_sse2(	uint8_t *pDst,
-;										const int32_t kiStride,
-;										const int32_t kiWidth,
-;										const int32_t kiHeight	);
-;***********************************************************************----------------
-ExpandPictureChromaAlign_sse2:
-	push ebx
-	push edx
-	push esi
-	push edi
-	push ebp
-
-	; for both top and bottom border
-	;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
-	mov esi, [esp+24]						; pDst
-	mov edx, [esp+28]						; kiStride
-	mov ebx, [esp+32]						; kiWidth
-	mov eax, [esp+36]						; kiHeight
-	; also prepare for cross border pData top-left: xmm3
-;	xor ecx, ecx
-	mov cl, byte [esi]
-	butterfly_1to16_sse xmm3, xmm4, c		; pDst, tmp, pSrc [generic register name: a/b/c/d]
-	; load top border
-	mov ecx, edx							; kiStride
-	neg ecx 								; -kiStride
-	lea edi, [esi+ecx]						; last line of top border
-	; load bottom border
-	dec eax									; h-1
-	imul eax, edx 							; (h-1)*kiStride
-	lea eax, [esi+eax]						; last line of picture pData
-	sal edx, 04h							; 16*kiStride
-	lea ebp, [eax+edx]						; last line of bottom border, (h-1)*kiStride + 16 * kiStride
-	; also prepare for cross border pData: bottom-left with xmm5, bottom-right xmm6
-	dec ebx									; kiWidth-1
-	lea ebx, [eax+ebx]						; pDst[w-1][h-1]
-;	xor edx, edx
-	mov dl, byte [eax]						; bottom-left
-	butterfly_1to16_sse xmm5, xmm6, d		; dst, tmp, pSrc [generic register name: a/b/c/d]
-	mov dl, byte [ebx]						; bottom-right
-	butterfly_1to16_sse xmm6, xmm4, d		; dst, tmp, pSrc [generic register name: a/b/c/d]
-	; for top & bottom expanding
-	mov ebx, [esp+32]						; kiWidth
-	exp_top_bottom_sse2	16
-
-	; for both left and right border
-	;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
-	mov esi, [esp+24]						; pDst: left border pSrc
-	mov edx, [esp+28]						; kiStride
-	mov ebx, [esp+32]						; kiWidth
-	mov ecx, [esp+36]						; kiHeight
-	; load left border
-	mov eax, -16 							; luma=-32, chroma=-16
-	lea edi, [esi+eax]						; left border dst
-	dec ebx
-	lea ebx, [esi+ebx]						; right border pSrc, (p_dst + width - 1)
-	lea ebp, [ebx+1]						; right border dst
-	; prepare for cross border pData: top-right with xmm4
-;	xor eax, eax
-	mov al, byte [ebx]						; top-right
-	butterfly_1to16_sse xmm4, xmm0, a		; pDst, tmp, pSrc [generic register name: a/b/c/d]
-	; for left & right border expanding
-	exp_left_right_sse2	16, a
-
-	; for cross border [top-left, top-right, bottom-left, bottom-right]
-	;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
-	mov esi, [esp+24]						; pDst
-	mov ecx, [esp+28]						; kiStride
-	mov ebx, [esp+32]						; kiWidth
-	mov edx, [esp+36]						; kiHeight
-	; have done xmm3,..,xmm6 cross pData initialization above, perform pading as below, To be continued..
-	mov eax, -16							; chroma=-16
-	neg ecx										; -stride
-	lea edi, [esi+eax]
-	lea edi, [edi+ecx]				; last line of top-left border
-	lea ebp, [esi+ebx]
-	lea ebp, [ebp+ecx]				; last line of top-right border
-	mov ecx, [esp+28]						; kiStride
-	add edx, 16							; height+16, luma=32, chroma=16
-	imul edx, ecx							; (kiHeight+16) * kiStride
-	lea eax, [edi+edx]						; last line of bottom-left border
-	lea ebx, [ebp+edx]						; last line of bottom-right border
-	neg ecx										; -kiStride
-	; for left & right border expanding
-	exp_cross_sse2		16, a
-
-;	sfence									; commit cache write back memory
-
-	pop ebp
-	pop edi
-	pop esi
-	pop edx
-	pop ebx
-
-	ret
-
-ALIGN 16
-;***********************************************************************----------------
-; void ExpandPictureChromaUnalign_sse2(	uint8_t *pDst,
-;										const int32_t kiStride,
-;										const int32_t kiWidth,
-;										const int32_t kiHeight	);
-;***********************************************************************----------------
-ExpandPictureChromaUnalign_sse2:
-	push ebx
-	push edx
-	push esi
-	push edi
-	push ebp
-
-	; for both top and bottom border
-	;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
-	mov esi, [esp+24]						; pDst
-	mov edx, [esp+28]						; kiStride
-	mov ebx, [esp+32]						; kiWidth
-	mov eax, [esp+36]						; kiHeight
-	; also prepare for cross border pData top-left: xmm3
-;	xor ecx, ecx
-	mov cl, byte [esi]
-	butterfly_1to16_sse xmm3, xmm4, c		; pDst, tmp, pSrc [generic register name: a/b/c/d]
-	; load top border
-	mov ecx, edx							; kiStride
-	neg ecx 								; -kiStride
-	lea edi, [esi+ecx]						; last line of top border
-	; load bottom border
-	dec eax									; h-1
-	imul eax, edx 							; (h-1)*kiStride
-	lea eax, [esi+eax]						; last line of picture pData
-	sal edx, 04h							; 16*kiStride
-	lea ebp, [eax+edx]						; last line of bottom border, (h-1)*kiStride + 16 * kiStride
-	; also prepare for cross border pData: bottom-left with xmm5, bottom-right xmm6
-	dec ebx									; kiWidth-1
-	lea ebx, [eax+ebx]						; dst[w-1][h-1]
-;	xor edx, edx
-	mov dl, byte [eax]						; bottom-left
-	butterfly_1to16_sse xmm5, xmm6, d		; dst, tmp, pSrc [generic register name: a/b/c/d]
-	mov dl, byte [ebx]						; bottom-right
-	butterfly_1to16_sse xmm6, xmm4, d		; dst, tmp, pSrc [generic register name: a/b/c/d]
-	; for top & bottom expanding
-	mov ebx, [esp+32]						; kiWidth
-	exp_top_bottom_sse2	16
-
-	; for both left and right border
-	;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
-	mov esi, [esp+24]						; p_dst: left border pSrc
-	mov edx, [esp+28]						; kiStride
-	mov ebx, [esp+32]						; kiWidth
-	mov ecx, [esp+36]						; kiHeight
-	; load left border
-	mov eax, -16 							; luma=-32, chroma=-16
-	lea edi, [esi+eax]						; left border dst
-	dec ebx
-	lea ebx, [esi+ebx]						; right border pSrc, (p_dst + width - 1)
-	lea ebp, [ebx+1]						; right border dst
-	; prepare for cross border pData: top-right with xmm4
-;	xor eax, eax
-	mov al, byte [ebx]						; top-right
-	butterfly_1to16_sse xmm4, xmm0, a		; dst, tmp, pSrc [generic register name: a/b/c/d]
-	; for left & right border expanding
-	exp_left_right_sse2	16, u
-
-	; for cross border [top-left, top-right, bottom-left, bottom-right]
-	;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
-	mov esi, [esp+24]						; p_dst
-	mov ecx, [esp+28]						; kiStride
-	mov ebx, [esp+32]						; kiWidth
-	mov edx, [esp+36]						; kiHeight
-	; have done xmm3,..,xmm6 cross pData initialization above, perform pading as below, To be continued..
-	neg ecx									; -kiStride
-	mov eax, -16							; chroma=-16
-	lea edi, [esi+eax]
-	lea edi, [edi+ecx]				; last line of top-left border
-	lea ebp, [esi+ebx]
-	lea ebp, [ebp+ecx]				; last line of top-right border
-	mov ecx, [esp+28]						; kiStride
-	add edx, 16							; kiHeight+16, luma=32, chroma=16
-	imul edx, ecx							; (kiHeight+16) * kiStride
-	lea eax, [edi+edx]						; last line of bottom-left border
-	lea ebx, [ebp+edx]						; last line of bottom-right border
-	neg ecx									; -kiStride
-	; for left & right border expanding
-	exp_cross_sse2		16, u
-
-;	sfence									; commit cache write back memory
-
-	pop ebp
-	pop edi
-	pop esi
-	pop edx
-	pop ebx
-
-	ret
-

codec/decoder/core/asm/mb_copy.asm

@@ -1,330 +0,0 @@
-;*!
-;* \copy
-;*     Copyright (c)  2009-2013, Cisco Systems
-;*     All rights reserved.
-;*
-;*     Redistribution and use in source and binary forms, with or without
-;*     modification, are permitted provided that the following conditions
-;*     are met:
-;*
-;*        * Redistributions of source code must retain the above copyright
-;*          notice, this list of conditions and the following disclaimer.
-;*
-;*        * Redistributions in binary form must reproduce the above copyright
-;*          notice, this list of conditions and the following disclaimer in
-;*          the documentation and/or other materials provided with the
-;*          distribution.
-;*
-;*     THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
-;*     "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
-;*     LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
-;*     FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
-;*     COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
-;*     INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
-;*     BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
-;*     LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
-;*     CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
-;*     LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
-;*     ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
-;*     POSSIBILITY OF SUCH DAMAGE.
-;*
-;*
-;*  mb_copy.asm
-;*
-;*  Abstract
-;*      mb_copy and mb_copy1
-;*
-;*  History
-;*      15/09/2009 Created
-;*		12/28/2009 Modified with larger throughput
-;*		12/29/2011 Tuned WelsCopy16x16NotAligned_sse2, added UpdateMbMv_sse2 WelsCopy16x8NotAligned_sse2,
-;*				   WelsCopy16x8_mmx, WelsCopy8x16_mmx etc;
-;*
-;*
-;*********************************************************************************************/
-%include "asm_inc.asm"
-BITS 32
-
-;*******************************************************************************
-; Macros and other preprocessor constants
-;*******************************************************************************
-
-;*******************************************************************************
-; Local Data (Read Only)
-;*******************************************************************************
-
-;SECTION .rodata data align=16
-
-;*******************************************************************************
-; Various memory constants (trigonometric values or rounding values)
-;*******************************************************************************
-
-ALIGN 16
-
-;*******************************************************************************
-; Code
-;*******************************************************************************
-
-SECTION .text
-
-WELS_EXTERN PixelAvgWidthEq4_mmx
-WELS_EXTERN PixelAvgWidthEq8_mmx
-WELS_EXTERN PixelAvgWidthEq16_sse2
-
-WELS_EXTERN McCopyWidthEq4_mmx
-WELS_EXTERN McCopyWidthEq8_mmx
-WELS_EXTERN McCopyWidthEq16_sse2
-
-
-ALIGN 16
-;*******************************************************************************
-; void_t PixelAvgWidthEq4_mmx( uint8_t *pDst,  int iDstStride,
-;                           uint8_t *pSrcA, int iSrcAStride,
-;                           uint8_t *pSrcB, int iSrcBStride,
-;                           int iHeight );
-;*******************************************************************************
-PixelAvgWidthEq4_mmx:
-
-    push        esi
-    push        edi
-    push        ebp
-    push        ebx
-
-    mov         edi, [esp+20]       ; pDst
-    mov         eax, [esp+24]       ; iDstStride
-    mov         esi, [esp+28]       ; pSrcA
-    mov         ecx, [esp+32]       ; iSrcAStride
-    mov         ebp, [esp+36]       ; pSrcB
-    mov         edx, [esp+40]       ; iSrcBStride
-    mov         ebx, [esp+44]       ; iHeight
-ALIGN 4
-.height_loop:
-	movd        mm0, [ebp]
-    pavgb       mm0, [esi]
-    movd        [edi], mm0
-
-    dec         ebx
-    lea         edi, [edi+eax]
-    lea         esi, [esi+ecx]
-    lea         ebp, [ebp+edx]
-    jne         .height_loop
-
-	WELSEMMS
-    pop         ebx
-    pop         ebp
-    pop         edi
-    pop         esi
-    ret
-
-ALIGN 16
-;*******************************************************************************
-; void_t PixelAvgWidthEq8_mmx( uint8_t *pDst,  int iDstStride,
-;                           uint8_t *pSrcA, int iSrcAStride,
-;                           uint8_t *pSrcB, int iSrcBStride,
-;                           int iHeight );
-;*******************************************************************************
-PixelAvgWidthEq8_mmx:
-
-    push        esi
-    push        edi
-    push        ebp
-    push        ebx
-
-    mov         edi, [esp+20]       ; pDst
-    mov         eax, [esp+24]       ; iDstStride
-    mov         esi, [esp+28]       ; pSrcA
-    mov         ecx, [esp+32]       ; iSrcAStride
-    mov         ebp, [esp+36]       ; pSrcB
-    mov         edx, [esp+40]       ; iSrcBStride
-    mov         ebx, [esp+44]       ; iHeight
-ALIGN 4
-.height_loop:
-	movq        mm0, [esi]
-    pavgb       mm0, [ebp]
-    movq        [edi], mm0
-    movq        mm0, [esi+ecx]
-    pavgb       mm0, [ebp+edx]
-    movq		[edi+eax], mm0
-
-    lea			esi,  [esi+2*ecx]
-    lea			ebp, [ebp+2*edx]
-    lea			edi,  [edi+2*eax]
-
-    sub           ebx, 2
-    jnz         .height_loop
-
-	WELSEMMS
-    pop         ebx
-    pop         ebp
-    pop         edi
-    pop         esi
-    ret
-
-
-
-ALIGN 16
-;*******************************************************************************
-; void_t PixelAvgWidthEq16_sse2( uint8_t *pDst,  int iDstStride,
-;                          uint8_t *pSrcA, int iSrcAStride,
-;                          uint8_t *pSrcB, int iSrcBStride,
-;                          int iHeight );
-;*******************************************************************************
-PixelAvgWidthEq16_sse2:
-    push        esi
-    push        edi
-    push        ebp
-    push        ebx
-
-
-    mov         edi, [esp+20]       ; pDst
-    mov         eax, [esp+24]       ; iDstStride
-    mov         esi, [esp+28]       ; pSrcA
-    mov         ecx, [esp+32]       ; iSrcAStride
-    mov         ebp, [esp+36]       ; pSrcB
-    mov         edx, [esp+40]       ; iSrcBStride
-    mov         ebx, [esp+44]       ; iHeight
-ALIGN 4
-.height_loop:
-	movdqu      xmm0, [esi]
-	pavgb         xmm0, [ebp]
-    movdqu      [edi], xmm0
-
-	movdqu      xmm0, [esi+ecx]
-	pavgb         xmm0, [ebp+edx]
-    movdqu      [edi+eax], xmm0
-
-	movdqu      xmm0, [esi+2*ecx]
-	pavgb         xmm0, [ebp+2*edx]
-    movdqu      [edi+2*eax], xmm0
-
-    lea              esi,  [esi+2*ecx]
-    lea			   ebp, [ebp+2*edx]
-    lea			   edi,  [edi+2*eax]
-
-	movdqu      xmm0, [esi+ecx]
-	pavgb         xmm0, [ebp+edx]
-    movdqu      [edi+eax], xmm0
-
-    lea              esi,  [esi+2*ecx]
-    lea			   ebp, [ebp+2*edx]
-    lea			   edi,  [edi+2*eax]
-
-
-    sub         ebx, 4
-    jne         .height_loop
-
-	WELSEMMS
-	pop         ebx
-    pop         ebp
-    pop         edi
-    pop         esi
-
-    ret
-
-
-ALIGN 16
-;*******************************************************************************
-;  void_t McCopyWidthEq4_mmx( uint8_t *pSrc, int iSrcStride,
-;                          uint8_t *pDst, int iDstStride, int iHeight )
-;*******************************************************************************
-McCopyWidthEq4_mmx:
-    push    esi
-    push    edi
-    push    ebx
-
-
-    mov esi,  [esp+16]
-    mov eax, [esp+20]
-    mov edi,  [esp+24]
-    mov ecx,  [esp+28]
-    mov edx,  [esp+32]
-ALIGN 4
-.height_loop:
-	mov ebx, [esi]
-	mov [edi], ebx
-
-	add esi, eax
-	add edi, ecx
-	dec edx
-	jnz .height_loop
-	WELSEMMS
-	pop	   ebx
-    pop     edi
-    pop     esi
-    ret
-
-ALIGN 16
-;*******************************************************************************
-;   void_t McCopyWidthEq8_mmx( uint8_t *pSrc, int iSrcStride,
-;                           uint8_t *pDst, int iDstStride, int iHeight )
-;*******************************************************************************
-McCopyWidthEq8_mmx:
-    push  esi
-    push  edi
-	mov  esi, [esp+12]
-	mov eax, [esp+16]
-	mov edi, [esp+20]
-	mov ecx, [esp+24]
-	mov edx, [esp+28]
-
-ALIGN 4
-.height_loop:
-	movq mm0, [esi]
-	movq [edi], mm0
-	add esi, eax
-	add edi, ecx
-	dec edx
-	jnz .height_loop
-
-	WELSEMMS
-    pop     edi
-    pop     esi
-    ret
-
-
-
-
-
-
-
-
-ALIGN 16
-;*******************************************************************************
-;   void_t McCopyWidthEq16_sse2( uint8_t *pSrc, int iSrcStride, uint8_t *pDst, int iDstStride, int iHeight )
-;*******************************************************************************
-;read unaligned memory
-%macro SSE_READ_UNA 2
-	movq	%1, [%2]
-	movhps	%1,	[%2+8]
-%endmacro
-
-;write unaligned memory
-%macro SSE_WRITE_UNA 2
-	movq	[%1],	%2
-	movhps	[%1+8], %2
-%endmacro
-McCopyWidthEq16_sse2:
-    push    esi
-    push    edi
-
-    mov     esi, [esp+12]       ; pSrc
-    mov     eax, [esp+16]       ; iSrcStride
-    mov     edi, [esp+20]       ; pDst
-    mov     edx, [esp+24]       ; iDstStride
-    mov     ecx, [esp+28]       ; iHeight
-
-ALIGN 4
-.height_loop:
-    SSE_READ_UNA	xmm0, esi
-    SSE_READ_UNA	xmm1, esi+eax
-    SSE_WRITE_UNA	edi, xmm0
-    SSE_WRITE_UNA	edi+edx, xmm1
-
-	sub		ecx,	2
-    lea     esi, [esi+eax*2]
-    lea     edi, [edi+edx*2]
-    jnz     .height_loop
-
-    pop     edi
-    pop     esi
-    ret

codec/decoder/core/asm/mc_chroma.asm

@@ -1,317 +0,0 @@
-;*!
-;* \copy
-;*     Copyright (c)  2004-2013, Cisco Systems
-;*     All rights reserved.
-;*
-;*     Redistribution and use in source and binary forms, with or without
-;*     modification, are permitted provided that the following conditions
-;*     are met:
-;*
-;*        * Redistributions of source code must retain the above copyright
-;*          notice, this list of conditions and the following disclaimer.
-;*
-;*        * Redistributions in binary form must reproduce the above copyright
-;*          notice, this list of conditions and the following disclaimer in
-;*          the documentation and/or other materials provided with the
-;*          distribution.
-;*
-;*     THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
-;*     "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
-;*     LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
-;*     FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
-;*     COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
-;*     INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
-;*     BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
-;*     LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
-;*     CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
-;*     LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
-;*     ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
-;*     POSSIBILITY OF SUCH DAMAGE.
-;*
-;*
-;*  mc_chroma.asm
-;*
-;*  Abstract
-;*      mmx motion compensation for chroma
-;*
-;*  History
-;*      10/13/2004 Created
-;*
-;*
-;*************************************************************************/
-%include "asm_inc.asm"
-
-BITS 32
-
-;***********************************************************************
-; Local Data (Read Only)
-;***********************************************************************
-
-SECTION .rodata align=16
-
-;***********************************************************************
-; Various memory constants (trigonometric values or rounding values)
-;***********************************************************************
-
-ALIGN 16
-h264_d0x20_sse2:
-	dw 32,32,32,32,32,32,32,32
-ALIGN 16
-h264_d0x20_mmx:
-	dw 32,32,32,32
-
-
-;=============================================================================
-; Code
-;=============================================================================
-
-SECTION .text
-
-ALIGN 16
-;*******************************************************************************
-; void McChromaWidthEq4_mmx( uint8_t *src,
-;							int32_t iSrcStride,
-;							uint8_t *pDst,
-;							int32_t iDstStride,
-;							uint8_t *pABCD,
-;							int32_t iHeigh );
-;*******************************************************************************
-WELS_EXTERN McChromaWidthEq4_mmx
-McChromaWidthEq4_mmx:
-	push esi
-	push edi
-	push ebx
-
-	mov eax, [esp +12 + 20]
-	movd mm3, [eax]
-	WELS_Zero mm7
-	punpcklbw mm3, mm3
-	movq      mm4, mm3
-	punpcklwd mm3, mm3
-	punpckhwd mm4, mm4
-
-	movq	  mm5, mm3
-	punpcklbw mm3, mm7
-	punpckhbw mm5, mm7
-
-	movq	  mm6, mm4
-	punpcklbw mm4, mm7
-	punpckhbw mm6, mm7
-
-	mov esi, [esp +12+ 4]
-	mov eax, [esp + 12 + 8]
-	mov edi, [esp + 12 + 12]
-	mov edx, [esp + 12 + 16]
-    mov ecx, [esp + 12 + 24]
-
-	lea ebx, [esi + eax]
-	movd mm0, [esi]
-	movd mm1, [esi+1]
-	punpcklbw mm0, mm7
-	punpcklbw mm1, mm7
-.xloop:
-
-	pmullw mm0, mm3
-	pmullw mm1, mm5
-	paddw  mm0, mm1
-
-	movd  mm1, [ebx]
-	punpcklbw mm1, mm7
-	movq mm2, mm1
-	pmullw mm1, mm4
-	paddw mm0, mm1
-
-	movd mm1, [ebx+1]
-	punpcklbw mm1, mm7
-	movq mm7, mm1
-	pmullw mm1,mm6
-	paddw mm0, mm1
-	movq mm1,mm7
-
-	paddw mm0, [h264_d0x20_mmx]
-	psrlw mm0, 6
-
-	WELS_Zero mm7
-	packuswb mm0, mm7
-	movd [edi], mm0
-
-	movq mm0, mm2
-
-	lea edi, [edi +edx  ]
-	lea ebx, [ebx + eax]
-
-	dec ecx
-	jnz near .xloop
-	WELSEMMS
-	pop ebx
-	pop edi
-	pop esi
-	ret
-
-
-ALIGN 16
-;*******************************************************************************
-; void McChromaWidthEq8_sse2( uint8_t *pSrc,
-;						int32_t iSrcStride,
-;						uint8_t *pDst,
-;						int32_t iDstStride,
-;						uint8_t *pABCD,
-;						int32_t iheigh );
-;*******************************************************************************
-WELS_EXTERN McChromaWidthEq8_sse2
-McChromaWidthEq8_sse2:
-	push esi
-	push edi
-	push ebx
-
-	mov eax, [esp +12 + 20]
-	movd xmm3, [eax]
-	WELS_Zero xmm7
-	punpcklbw  xmm3, xmm3
-	punpcklwd  xmm3, xmm3
-
-	movdqa	   xmm4, xmm3
-	punpckldq  xmm3, xmm3
-	punpckhdq  xmm4, xmm4
-	movdqa     xmm5, xmm3
-	movdqa	   xmm6, xmm4
-
-	punpcklbw  xmm3, xmm7
-	punpckhbw  xmm5, xmm7
-	punpcklbw  xmm4, xmm7
-	punpckhbw  xmm6, xmm7
-
-	mov esi, [esp +12+ 4]
-	mov eax, [esp + 12 + 8]
-	mov edi, [esp + 12 + 12]
-	mov edx, [esp + 12 + 16]
-    mov ecx, [esp + 12 + 24]
-
-	lea ebx, [esi + eax]
-	movq xmm0, [esi]
-	movq xmm1, [esi+1]
-	punpcklbw xmm0, xmm7
-	punpcklbw xmm1, xmm7
-.xloop:
-
-	pmullw xmm0, xmm3
-	pmullw xmm1, xmm5
-	paddw  xmm0, xmm1
-
-	movq  xmm1, [ebx]
-	punpcklbw xmm1, xmm7
-	movdqa xmm2, xmm1
-	pmullw xmm1, xmm4
-	paddw xmm0, xmm1
-
-	movq xmm1, [ebx+1]
-	punpcklbw xmm1, xmm7
-	movdqa xmm7, xmm1
-	pmullw xmm1, xmm6
-	paddw xmm0, xmm1
-	movdqa xmm1,xmm7
-
-	paddw xmm0, [h264_d0x20_sse2]
-	psrlw xmm0, 6
-
-	WELS_Zero xmm7
-	packuswb xmm0, xmm7
-	movq [edi], xmm0
-
-	movdqa xmm0, xmm2
-
-	lea edi, [edi +edx  ]
-	lea ebx, [ebx + eax]
-
-	dec ecx
-	jnz near .xloop
-
-	pop ebx
-	pop edi
-	pop esi
-	ret
-
-
-
-
-ALIGN 16
-;***********************************************************************
-; void McChromaWidthEq8_ssse3( uint8_t *pSrc,
-;						 int32_t iSrcStride,
-;                        uint8_t *pDst,
-;                        int32_t iDstStride,
-;                        uint8_t *pABCD,
-;					     int32_t iHeigh);
-;***********************************************************************
-WELS_EXTERN McChromaWidthEq8_ssse3
-McChromaWidthEq8_ssse3:
-	push ebx
-	push esi
-	push edi
-
-	mov eax, [esp + 12 + 20]
-
-    pxor      xmm7, xmm7
-    movd   xmm5, [eax]
-    punpcklwd xmm5, xmm5
-    punpckldq xmm5, xmm5
-    movdqa    xmm6, xmm5
-    punpcklqdq xmm5, xmm5
-    punpckhqdq xmm6, xmm6
-
-	mov eax, [esp + 12 + 4]
-	mov edx, [esp + 12 + 8]
-	mov esi, [esp + 12 + 12]
-	mov edi, [esp + 12 + 16]
-    mov ecx, [esp + 12 + 24]
-
-    sub esi, edi
-    sub esi, edi
-	movdqa xmm7, [h264_d0x20_sse2]
-
-	movdqu xmm0, [eax]
-	movdqa xmm1, xmm0
-	psrldq xmm1, 1
-	punpcklbw xmm0, xmm1
-
-.hloop_chroma:
-	lea	esi, [esi+2*edi]
-
-	movdqu xmm2, [eax+edx]
-	movdqa xmm3, xmm2
-	psrldq xmm3, 1
-	punpcklbw xmm2, xmm3
-	movdqa      xmm4, xmm2
-
-    pmaddubsw  xmm0, xmm5
-    pmaddubsw  xmm2, xmm6
-    paddw      xmm0, xmm2
-    paddw      xmm0, xmm7
-	psrlw      xmm0, 6
-    packuswb   xmm0, xmm0
-    movq       [esi],xmm0
-
-    lea eax, [eax+2*edx]
-    movdqu xmm2, [eax]
-    movdqa xmm3, xmm2
-    psrldq xmm3, 1
-    punpcklbw xmm2, xmm3
-    movdqa      xmm0, xmm2
-
-    pmaddubsw  xmm4, xmm5
-    pmaddubsw  xmm2, xmm6
-    paddw      xmm4, xmm2
-    paddw      xmm4, xmm7
-	psrlw      xmm4, 6
-    packuswb   xmm4, xmm4
-    movq       [esi+edi],xmm4
-
-	sub ecx, 2
-	jnz .hloop_chroma
-	pop edi
-	pop esi
-	pop ebx
-
-	ret
-
-

codec/decoder/core/asm/mc_luma.asm

@@ -1,615 +0,0 @@
-;*!
-;* \copy
-;*     Copyright (c)  2009-2013, Cisco Systems
-;*     All rights reserved.
-;*
-;*     Redistribution and use in source and binary forms, with or without
-;*     modification, are permitted provided that the following conditions
-;*     are met:
-;*
-;*        * Redistributions of source code must retain the above copyright
-;*          notice, this list of conditions and the following disclaimer.
-;*
-;*        * Redistributions in binary form must reproduce the above copyright
-;*          notice, this list of conditions and the following disclaimer in
-;*          the documentation and/or other materials provided with the
-;*          distribution.
-;*
-;*     THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
-;*     "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
-;*     LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
-;*     FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
-;*     COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
-;*     INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
-;*     BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
-;*     LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
-;*     CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
-;*     LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
-;*     ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
-;*     POSSIBILITY OF SUCH DAMAGE.
-;*
-;*
-;*  mc_luma.asm
-;*
-;*  Abstract
-;*      sse2 motion compensation
-;*
-;*  History
-;*      17/08/2009 Created
-;*
-;*
-;*************************************************************************/
-%include "asm_inc.asm"
-
-BITS 32
-
-;*******************************************************************************
-; Local Data (Read Only)
-;*******************************************************************************
-
-SECTION .rodata align=16
-
-;*******************************************************************************
-; Various memory constants (trigonometric values or rounding values)
-;*******************************************************************************
-
-ALIGN 16
-h264_w0x10:
-	dw 16, 16, 16, 16
-
-
-;*******************************************************************************
-; Code
-;*******************************************************************************
-
-SECTION .text
-
-WELS_EXTERN McHorVer20WidthEq4_mmx
-
-
-ALIGN 16
-;*******************************************************************************
-; void_t McHorVer20WidthEq4_mmx( uint8_t *pSrc,
-;                       int iSrcStride,
-;						uint8_t *pDst,
-;						int iDstStride,
-;						int iHeight)
-;*******************************************************************************
-McHorVer20WidthEq4_mmx:
-	push esi
-	push edi
-
-	mov  esi, [esp+12]
-	mov eax, [esp+16]
-	mov edi, [esp+20]
-	mov ecx, [esp+24]
-	mov edx, [esp+28]
-	sub esi, 2
-	WELS_Zero mm7
-	movq mm6, [h264_w0x10]
-.height_loop:
-	movd mm0, [esi]
-	punpcklbw mm0, mm7
-	movd mm1, [esi+5]
-	punpcklbw mm1, mm7
-	movd mm2, [esi+1]
-	punpcklbw mm2, mm7
-	movd mm3, [esi+4]
-	punpcklbw mm3, mm7
-	movd mm4, [esi+2]
-	punpcklbw mm4, mm7
-	movd mm5, [esi+3]
-	punpcklbw mm5, mm7
-
-	paddw mm2, mm3
-	paddw mm4, mm5
-	psllw mm4, 2
-	psubw mm4, mm2
-	paddw mm0, mm1
-	paddw mm0, mm4
-	psllw mm4, 2
-	paddw mm0, mm4
-	paddw mm0, mm6
-	psraw mm0, 5
-	packuswb mm0, mm7
-	movd [edi], mm0
-
-	add esi, eax
-	add edi, ecx
-	dec edx
-	jnz .height_loop
-
-	WELSEMMS
-	pop edi
-	pop esi
-	ret
-
-;*******************************************************************************
-; Macros and other preprocessor constants
-;*******************************************************************************
-
-
-%macro SSE_LOAD_8P 3
-	movq %1, %3
-	punpcklbw %1, %2
-%endmacro
-
-%macro FILTER_HV_W8 9
-	paddw	%1, %6
-	movdqa	%8, %3
-	movdqa	%7, %2
-	paddw	%1, [h264_w0x10_1]
-	paddw	%8, %4
-	paddw	%7, %5
-	psllw	%8, 2
-	psubw	%8, %7
-	paddw	%1, %8
-	psllw	%8, 2
-	paddw	%1, %8
-	psraw   %1, 5
-	WELS_Zero %8
-	packuswb %1, %8
-	movq    %9, %1
-%endmacro
-
-;*******************************************************************************
-; Local Data (Read Only)
-;*******************************************************************************
-
-SECTION .rodata align=16
-
-;*******************************************************************************
-; Various memory constants (trigonometric values or rounding values)
-;*******************************************************************************
-
-ALIGN 16
-h264_w0x10_1:
-	dw 16, 16, 16, 16, 16, 16, 16, 16
-ALIGN 16
-h264_mc_hc_32:
-dw 32, 32, 32, 32, 32, 32, 32, 32
-;*******************************************************************************
-; Code
-;*******************************************************************************
-
-SECTION .text
-WELS_EXTERN McHorVer22Width8HorFirst_sse2
-WELS_EXTERN McHorVer22VerLast_sse2
-WELS_EXTERN McHorVer02WidthEq8_sse2
-WELS_EXTERN McHorVer20WidthEq8_sse2
-WELS_EXTERN McHorVer20WidthEq16_sse2
-
-ALIGN 16
-;***********************************************************************
-; void_t McHorVer22Width8HorFirst_sse2(int16_t *pSrc,
-;                       int16_t iSrcStride,
-;						uint8_t *pDst,
-;						int32_t iDstStride
-;						int32_t iHeight
-;                       )
-;***********************************************************************
-McHorVer22Width8HorFirst_sse2:
-	push esi
-	push edi
-	push ebx
-	mov esi, [esp+16]     ;pSrc
-	mov eax, [esp+20]	;iSrcStride
-	mov edi, [esp+24]		;pDst
-	mov edx, [esp+28]	;iDstStride
-	mov ebx, [esp+32]	;iHeight
-	pxor xmm7, xmm7
-
-	sub esi, eax				;;;;;;;;need more 5 lines.
-	sub esi, eax
-
-.yloop_width_8:
-	movq xmm0, [esi]
-	punpcklbw xmm0, xmm7
-	movq xmm1, [esi+5]
-	punpcklbw xmm1, xmm7
-	movq xmm2, [esi+1]
-	punpcklbw xmm2, xmm7
-	movq xmm3, [esi+4]
-	punpcklbw xmm3, xmm7
-	movq xmm4, [esi+2]
-	punpcklbw xmm4, xmm7
-	movq xmm5, [esi+3]
-	punpcklbw xmm5, xmm7
-
-	paddw xmm2, xmm3
-	paddw xmm4, xmm5
-	psllw xmm4, 2
-	psubw xmm4, xmm2
-	paddw xmm0, xmm1
-	paddw xmm0, xmm4
-	psllw xmm4, 2
-	paddw xmm0, xmm4
-	movdqa [edi], xmm0
-
-	add esi, eax
-	add edi, edx
-	dec ebx
-	jnz .yloop_width_8
-	pop ebx
-	pop edi
-	pop esi
-	ret
-
-ALIGN 16
-;***********************************************************************
-;void_t McHorVer22VerLast_sse2(
-;											uint8_t *pSrc,
-;											int32_t pSrcStride,
-;											uint8_t * pDst,
-;											int32_t iDstStride,
-;											int32_t iWidth,
-;											int32_t iHeight);
-;***********************************************************************
-
-%macro FILTER_VER 9
-	paddw  %1, %6
-	movdqa %7, %2
-	movdqa %8, %3
-
-
-	paddw %7, %5
-	paddw %8, %4
-
-	psubw  %1, %7
-	psraw   %1, 2
-	paddw  %1, %8
-	psubw  %1, %7
-	psraw   %1, 2
-	paddw  %8, %1
-	paddw  %8, [h264_mc_hc_32]
-	psraw   %8, 6
-	packuswb %8, %8
-	movq %9, %8
-%endmacro
-
-McHorVer22VerLast_sse2:
-	push esi
-	push edi
-	push ebx
-	push ebp
-
-	mov esi, [esp+20]
-	mov eax, [esp+24]
-	mov edi, [esp+28]
-	mov edx, [esp+32]
-	mov ebx, [esp+36]
-	mov ecx, [esp+40]
-	shr ebx, 3
-
-.width_loop:
-	movdqa xmm0, [esi]
-	movdqa xmm1, [esi+eax]
-	lea esi, [esi+2*eax]
-	movdqa xmm2, [esi]
-	movdqa xmm3, [esi+eax]
-	lea esi, [esi+2*eax]
-	movdqa xmm4, [esi]
-	movdqa xmm5, [esi+eax]
-
-	FILTER_VER xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, [edi]
-	dec ecx
-	lea esi, [esi+2*eax]
-	movdqa xmm6, [esi]
-
-	movdqa xmm0, xmm1
-	movdqa xmm1, xmm2
-	movdqa xmm2, xmm3
-	movdqa xmm3, xmm4
-	movdqa xmm4, xmm5
-	movdqa xmm5, xmm6
-
-	add edi, edx
-	sub esi, eax
-
-.start:
-	FILTER_VER xmm0,xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, [edi]
-	dec ecx
-	jz near .x_loop_dec
-
-	lea esi, [esi+2*eax]
-	movdqa xmm6, [esi]
-	FILTER_VER xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm0,[edi+edx]
-	dec ecx
-	jz near .x_loop_dec
-
-	lea edi, [edi+2*edx]
-	movdqa xmm7, [esi+eax]
-	FILTER_VER  xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm0, xmm1, [edi]
-	dec ecx
-	jz near .x_loop_dec
-
-	lea esi, [esi+2*eax]
-	movdqa xmm0, [esi]
-	FILTER_VER  xmm3, xmm4, xmm5, xmm6, xmm7, xmm0, xmm1, xmm2,[edi+edx]
-	dec ecx
-	jz near .x_loop_dec
-
-	lea edi, [edi+2*edx]
-	movdqa xmm1, [esi+eax]
-	FILTER_VER  xmm4, xmm5, xmm6, xmm7, xmm0, xmm1, xmm2, xmm3,[edi]
-	dec ecx
-	jz near .x_loop_dec
-
-	lea esi, [esi+2*eax]
-	movdqa xmm2, [esi]
-	FILTER_VER  xmm5, xmm6, xmm7, xmm0, xmm1, xmm2, xmm3,xmm4,[edi+edx]
-	dec ecx
-	jz near .x_loop_dec
-
-	lea edi, [edi+2*edx]
-	movdqa xmm3, [esi+eax]
-	FILTER_VER  xmm6, xmm7, xmm0, xmm1, xmm2, xmm3,xmm4,xmm5,[edi]
-	dec ecx
-	jz near .x_loop_dec
-
-	lea esi, [esi+2*eax]
-	movdqa xmm4, [esi]
-	FILTER_VER  xmm7, xmm0, xmm1, xmm2, xmm3,xmm4,xmm5,xmm6, [edi+edx]
-	dec ecx
-	jz near .x_loop_dec
-
-	lea edi, [edi+2*edx]
-	movdqa xmm5, [esi+eax]
-	jmp near .start
-
-.x_loop_dec:
-	dec ebx
-	jz near .exit
-	mov esi, [esp+20]
-	mov edi, [esp+28]
-	mov ecx, [esp+40]
-	add esi, 16
-	add edi, 8
-	jmp .width_loop
-
-
-
-.exit:
-	pop ebp
-	pop ebx
-	pop edi
-	pop esi
-	ret
-
-
-ALIGN 16
-;*******************************************************************************
-; void_t McHorVer20WidthEq8_sse2(  uint8_t *pSrc,
-;                       int iSrcStride,
-;												uint8_t *pDst,
-;												int iDstStride,
-;												int iHeight,
-;                      );
-;*******************************************************************************
-McHorVer20WidthEq8_sse2:
-	push	esi
-	push	edi
-
-	mov esi, [esp + 12]         ;pSrc
-	mov eax, [esp + 16]         ;iSrcStride
-	mov edi, [esp + 20]         ;pDst
-	mov ecx, [esp + 28]         ;iHeight
-	mov edx, [esp + 24]			;iDstStride
-
-	lea esi, [esi-2]            ;pSrc -= 2;
-
-	pxor xmm7, xmm7
-	movdqa xmm6, [h264_w0x10_1]
-.y_loop:
-	movq xmm0, [esi]
-	punpcklbw xmm0, xmm7
-	movq xmm1, [esi+5]
-	punpcklbw xmm1, xmm7
-	movq xmm2, [esi+1]
-	punpcklbw xmm2, xmm7
-	movq xmm3, [esi+4]
-	punpcklbw xmm3, xmm7
-	movq xmm4, [esi+2]
-	punpcklbw xmm4, xmm7
-	movq xmm5, [esi+3]
-	punpcklbw xmm5, xmm7
-
-	paddw xmm2, xmm3
-	paddw xmm4, xmm5
-	psllw xmm4, 2
-	psubw xmm4, xmm2
-	paddw xmm0, xmm1
-	paddw xmm0, xmm4
-	psllw xmm4, 2
-	paddw xmm0, xmm4
-	paddw xmm0, xmm6
-	psraw xmm0, 5
-
-	packuswb xmm0, xmm7
-	movq [edi], xmm0
-
-	lea edi, [edi+edx]
-	lea esi, [esi+eax]
-	dec ecx
-	jnz near .y_loop
-
-	pop edi
-	pop esi
-	ret
-
-ALIGN 16
-;*******************************************************************************
-; void_t McHorVer20WidthEq16_sse2(  uint8_t *pSrc,
-;                       int iSrcStride,
-;												uint8_t *pDst,
-;												int iDstStride,
-;												int iHeight,
-;                      );
-;*******************************************************************************
-McHorVer20WidthEq16_sse2:
-	push	esi
-	push	edi
-
-
-	mov esi, [esp + 12]         ;pSrc
-	mov eax, [esp + 16]         ;iSrcStride
-	mov edi, [esp + 20]         ;pDst
-	mov ecx, [esp + 28]         ;iHeight
-	mov edx, [esp + 24]			;iDstStride
-
-	lea esi, [esi-2]            ;pSrc -= 2;
-
-	pxor xmm7, xmm7
-	movdqa xmm6, [h264_w0x10_1]
-.y_loop:
-
-	movq xmm0, [esi]
-	punpcklbw xmm0, xmm7
-	movq xmm1, [esi+5]
-	punpcklbw xmm1, xmm7
-	movq xmm2, [esi+1]
-	punpcklbw xmm2, xmm7
-	movq xmm3, [esi+4]
-	punpcklbw xmm3, xmm7
-	movq xmm4, [esi+2]
-	punpcklbw xmm4, xmm7
-	movq xmm5, [esi+3]
-	punpcklbw xmm5, xmm7
-
-	paddw xmm2, xmm3
-	paddw xmm4, xmm5
-	psllw xmm4, 2
-	psubw xmm4, xmm2
-	paddw xmm0, xmm1
-	paddw xmm0, xmm4
-	psllw xmm4, 2
-	paddw xmm0, xmm4
-	paddw xmm0, xmm6
-	psraw xmm0, 5
-	packuswb xmm0, xmm7
-	movq [edi], xmm0
-
-	movq xmm0, [esi+8]
-	punpcklbw xmm0, xmm7
-	movq xmm1, [esi+5+8]
-	punpcklbw xmm1, xmm7
-	movq xmm2, [esi+1+8]
-	punpcklbw xmm2, xmm7
-	movq xmm3, [esi+4+8]
-	punpcklbw xmm3, xmm7
-	movq xmm4, [esi+2+8]
-	punpcklbw xmm4, xmm7
-	movq xmm5, [esi+3+8]
-	punpcklbw xmm5, xmm7
-
-	paddw xmm2, xmm3
-	paddw xmm4, xmm5
-	psllw xmm4, 2
-	psubw xmm4, xmm2
-	paddw xmm0, xmm1
-	paddw xmm0, xmm4
-	psllw xmm4, 2
-	paddw xmm0, xmm4
-	paddw xmm0, xmm6
-	psraw xmm0, 5
-	packuswb xmm0, xmm7
-	movq [edi+8], xmm0
-
-	lea edi, [edi+edx]
-	lea esi, [esi+eax]
-	dec ecx
-	jnz near .y_loop
-	pop edi
-	pop esi
-	ret
-
-
-;*******************************************************************************
-; void_t McHorVer02WidthEq8_sse2( uint8_t *pSrc,
-;                       int iSrcStride,
-;                       uint8_t *pDst,
-;                       int iDstStride,
-;                       int iHeight )
-;*******************************************************************************
-ALIGN 16
-McHorVer02WidthEq8_sse2:
-	push esi
-	push edi
-
-	mov esi, [esp + 12]           ;pSrc
-	mov edx, [esp + 16]	          ;iSrcStride
-	mov edi, [esp + 20]           ;pDst
-	mov eax, [esp + 24]           ;iDstStride
-	mov ecx, [esp + 28]           ;iHeight
-
-	sub esi, edx
-	sub esi, edx
-
-	WELS_Zero xmm7
-
-	SSE_LOAD_8P xmm0, xmm7, [esi]
-	SSE_LOAD_8P xmm1, xmm7, [esi+edx]
-	lea esi, [esi+2*edx]
-	SSE_LOAD_8P xmm2, xmm7, [esi]
-	SSE_LOAD_8P xmm3, xmm7, [esi+edx]
-	lea esi, [esi+2*edx]
-	SSE_LOAD_8P xmm4, xmm7, [esi]
-	SSE_LOAD_8P xmm5, xmm7, [esi+edx]
-
-.start:
-	FILTER_HV_W8 xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, [edi]
-	dec ecx
-	jz near .xx_exit
-
-	lea esi, [esi+2*edx]
-	SSE_LOAD_8P xmm6, xmm7, [esi]
-	FILTER_HV_W8 xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm0, [edi+eax]
-	dec ecx
-	jz near .xx_exit
-
-	lea edi, [edi+2*eax]
-	SSE_LOAD_8P xmm7, xmm0, [esi+edx]
-	FILTER_HV_W8 xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm0, xmm1, [edi]
-	dec ecx
-	jz near .xx_exit
-
-	lea esi, [esi+2*edx]
-	SSE_LOAD_8P xmm0, xmm1, [esi]
-	FILTER_HV_W8 xmm3, xmm4, xmm5, xmm6, xmm7, xmm0, xmm1, xmm2, [edi+eax]
-	dec ecx
-	jz near .xx_exit
-
-	lea edi, [edi+2*eax]
-	SSE_LOAD_8P xmm1, xmm2, [esi+edx]
-	FILTER_HV_W8 xmm4, xmm5, xmm6, xmm7, xmm0, xmm1, xmm2, xmm3, [edi]
-	dec ecx
-	jz near .xx_exit
-
-	lea esi, [esi+2*edx]
-	SSE_LOAD_8P xmm2, xmm3, [esi]
-	FILTER_HV_W8 xmm5, xmm6, xmm7, xmm0, xmm1, xmm2, xmm3, xmm4, [edi+eax]
-	dec ecx
-	jz near .xx_exit
-
-	lea edi, [edi+2*eax]
-	SSE_LOAD_8P xmm3, xmm4, [esi+edx]
-	FILTER_HV_W8 xmm6, xmm7, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, [edi]
-	dec ecx
-	jz near .xx_exit
-
-	lea esi, [esi+2*edx]
-	SSE_LOAD_8P xmm4, xmm5, [esi]
-	FILTER_HV_W8 xmm7, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, [edi+eax]
-	dec ecx
-	jz near .xx_exit
-
-	lea edi, [edi+2*eax]
-	SSE_LOAD_8P xmm5, xmm6, [esi+edx]
-	jmp near .start
-
-.xx_exit:
-	pop edi
-	pop esi
-	ret
-
-

codec/decoder/core/asm/memzero.asm

@@ -1,135 +0,0 @@
-;*!
-;* \copy
-;*     Copyright (c)  2009-2013, Cisco Systems
-;*     All rights reserved.
-;*
-;*     Redistribution and use in source and binary forms, with or without
-;*     modification, are permitted provided that the following conditions
-;*     are met:
-;*
-;*        * Redistributions of source code must retain the above copyright
-;*          notice, this list of conditions and the following disclaimer.
-;*
-;*        * Redistributions in binary form must reproduce the above copyright
-;*          notice, this list of conditions and the following disclaimer in
-;*          the documentation and/or other materials provided with the
-;*          distribution.
-;*
-;*     THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
-;*     "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
-;*     LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
-;*     FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
-;*     COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
-;*     INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
-;*     BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
-;*     LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
-;*     CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
-;*     LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
-;*     ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
-;*     POSSIBILITY OF SUCH DAMAGE.
-;*
-;*
-;*  memzero.asm
-;*
-;*  Abstract
-;*
-;*
-;*  History
-;*      9/16/2009 Created
-;*
-;*
-;*************************************************************************/
-
-BITS 32
-
-%include "asm_inc.asm"
-;***********************************************************************
-; Code
-;***********************************************************************
-
-SECTION .text
-
-ALIGN 16
-;***********************************************************************
-;_inline void __cdecl WelsPrefetchZero_mmx(int8_t const*_A);
-;***********************************************************************
-WELS_EXTERN WelsPrefetchZero_mmx
-WelsPrefetchZero_mmx:
-	mov  eax,[esp+4]
-	prefetchnta [eax]
-	ret
-
-
-ALIGN 16
-;***********************************************************************
-;   void WelsSetMemZeroAligned64_sse2(void *dst, int32_t size)
-;***********************************************************************
-WELS_EXTERN WelsSetMemZeroAligned64_sse2
-WelsSetMemZeroAligned64_sse2:
-		mov		eax,	[esp + 4]          ; dst
-		mov		ecx,	[esp + 8]
-		neg		ecx
-
-		pxor	xmm0,		xmm0
-.memzeroa64_sse2_loops:
-		movdqa	[eax],		xmm0
-		movdqa	[eax+16],	xmm0
-		movdqa	[eax+32],	xmm0
-		movdqa	[eax+48],	xmm0
-		add		eax, 0x40
-
-		add ecx, 0x40
-		jnz near .memzeroa64_sse2_loops
-
-		ret
-
-ALIGN 16
-;***********************************************************************
-;   void WelsSetMemZeroSize64_mmx(void *dst, int32_t size)
-;***********************************************************************
-WELS_EXTERN WelsSetMemZeroSize64_mmx
-WelsSetMemZeroSize64_mmx:
-		mov		eax,	[esp + 4]          ; dst
-		mov		ecx,	[esp + 8]
-		neg		ecx
-
-		pxor	mm0,		mm0
-.memzero64_mmx_loops:
-		movq	[eax],		mm0
-		movq	[eax+8],	mm0
-		movq	[eax+16],	mm0
-		movq	[eax+24],	mm0
-		movq	[eax+32],	mm0
-		movq	[eax+40],	mm0
-		movq	[eax+48],	mm0
-		movq	[eax+56],	mm0
-		add		eax,		0x40
-
-		add ecx, 0x40
-		jnz near .memzero64_mmx_loops
-
-		WELSEMMS
-		ret
-
-ALIGN 16
-;***********************************************************************
-;   void WelsSetMemZeroSize8_mmx(void *dst, int32_t size)
-;***********************************************************************
-WELS_EXTERN WelsSetMemZeroSize8_mmx
-WelsSetMemZeroSize8_mmx:
-		mov		eax,	[esp + 4]		; dst
-		mov		ecx,	[esp + 8]		; size
-		neg		ecx
-		pxor	mm0,		mm0
-
-.memzero8_mmx_loops:
-		movq	[eax],		mm0
-		add		eax,		0x08
-
-		add		ecx,		0x08
-		jnz near .memzero8_mmx_loops
-
-		WELSEMMS
-		ret
-
-

codec/decoder/core/inc/mc.h

@@ -78,7 +78,7 @@ extern void_t McHorVer02WidthEq8_sse2 (uint8_t* pSrc, int32_t iSrcStride, uint8_
                                        int32_t iHeight);
 extern void_t McHorVer22Width8HorFirst_sse2 (uint8_t* pSrc, int32_t iSrcStride, uint8_t* pDst, int32_t iDstStride,
     int32_t iHeight);
-extern void_t McHorVer22VerLast_sse2 (uint8_t* pTap, int32_t iTapStride, uint8_t* pDst, int32_t iDstStride,
+extern void_t McHorVer22Width8VerLastAlign_sse2 (uint8_t* pTap, int32_t iTapStride, uint8_t* pDst, int32_t iDstStride,
                                       int32_t iWidth, int32_t iHeight);
 extern void_t PixelAvgWidthEq16_sse2 (uint8_t* pDst, int32_t iDstStride, uint8_t* pSrcA, int32_t iSrcAStride,
                                       uint8_t* pSrcB, int32_t iSrcBStride, int32_t iHeight);

codec/decoder/core/src/mc.cpp

@@ -362,7 +362,7 @@ static inline void_t McHorVer22WidthEq8_sse2 (uint8_t* pSrc, int32_t iSrcStride,
     int32_t iHeight) {
   ENFORCE_STACK_ALIGN_2D (int16_t, iTap, 21, 8, 16)
   McHorVer22Width8HorFirst_sse2 (pSrc - 2, iSrcStride, (uint8_t*)iTap, 16, iHeight + 5);
-  McHorVer22VerLast_sse2 ((uint8_t*)iTap, 16, pDst, iDstStride, 8, iHeight);
+  McHorVer22Width8VerLastAlign_sse2 ((uint8_t*)iTap, 16, pDst, iDstStride, 8, iHeight);
 }
 
 static inline void_t McHorVer02WidthEq16_sse2 (uint8_t* pSrc, int32_t iSrcStride, uint8_t* pDst, int32_t iDstStride,

codec/decoder/targets.mk

@@ -28,17 +28,9 @@ DECODER_CPP_SRCS=\
 DECODER_OBJS += $(DECODER_CPP_SRCS:.cpp=.o)
 ifeq ($(USE_ASM), Yes)
 DECODER_ASM_SRCS=\
-	$(DECODER_SRCDIR)/./core/asm/asm_inc.asm\
 	$(DECODER_SRCDIR)/./core/asm/block_add.asm\
-	$(DECODER_SRCDIR)/./core/asm/cpuid.asm\
 	$(DECODER_SRCDIR)/./core/asm/dct.asm\
-	$(DECODER_SRCDIR)/./core/asm/deblock.asm\
-	$(DECODER_SRCDIR)/./core/asm/expand_picture.asm\
 	$(DECODER_SRCDIR)/./core/asm/intra_pred.asm\
-	$(DECODER_SRCDIR)/./core/asm/mb_copy.asm\
-	$(DECODER_SRCDIR)/./core/asm/mc_chroma.asm\
-	$(DECODER_SRCDIR)/./core/asm/mc_luma.asm\
-	$(DECODER_SRCDIR)/./core/asm/memzero.asm\
 
 DECODER_OBJS += $(DECODER_ASM_SRCS:.asm=.o)
 endif
@@ -113,39 +105,15 @@ $(DECODER_SRCDIR)/./plus/src/welsCodecTrace.o: $(DECODER_SRCDIR)/./plus/src/wels
 $(DECODER_SRCDIR)/./plus/src/welsDecoderExt.o: $(DECODER_SRCDIR)/./plus/src/welsDecoderExt.cpp
 	$(CXX) $(CFLAGS) $(CXXFLAGS) $(INCLUDES) $(DECODER_CFLAGS) $(DECODER_INCLUDES) -c -o $(DECODER_SRCDIR)/./plus/src/welsDecoderExt.o $(DECODER_SRCDIR)/./plus/src/welsDecoderExt.cpp
 
-$(DECODER_SRCDIR)/./core/asm/asm_inc.o: $(DECODER_SRCDIR)/./core/asm/asm_inc.asm
-	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(DECODER_ASMFLAGS) $(DECODER_ASM_INCLUDES) -o $(DECODER_SRCDIR)/./core/asm/asm_inc.o $(DECODER_SRCDIR)/./core/asm/asm_inc.asm
-
 $(DECODER_SRCDIR)/./core/asm/block_add.o: $(DECODER_SRCDIR)/./core/asm/block_add.asm
 	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(DECODER_ASMFLAGS) $(DECODER_ASM_INCLUDES) -o $(DECODER_SRCDIR)/./core/asm/block_add.o $(DECODER_SRCDIR)/./core/asm/block_add.asm
 
-$(DECODER_SRCDIR)/./core/asm/cpuid.o: $(DECODER_SRCDIR)/./core/asm/cpuid.asm
-	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(DECODER_ASMFLAGS) $(DECODER_ASM_INCLUDES) -o $(DECODER_SRCDIR)/./core/asm/cpuid.o $(DECODER_SRCDIR)/./core/asm/cpuid.asm
-
 $(DECODER_SRCDIR)/./core/asm/dct.o: $(DECODER_SRCDIR)/./core/asm/dct.asm
 	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(DECODER_ASMFLAGS) $(DECODER_ASM_INCLUDES) -o $(DECODER_SRCDIR)/./core/asm/dct.o $(DECODER_SRCDIR)/./core/asm/dct.asm
 
-$(DECODER_SRCDIR)/./core/asm/deblock.o: $(DECODER_SRCDIR)/./core/asm/deblock.asm
-	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(DECODER_ASMFLAGS) $(DECODER_ASM_INCLUDES) -o $(DECODER_SRCDIR)/./core/asm/deblock.o $(DECODER_SRCDIR)/./core/asm/deblock.asm
-
-$(DECODER_SRCDIR)/./core/asm/expand_picture.o: $(DECODER_SRCDIR)/./core/asm/expand_picture.asm
-	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(DECODER_ASMFLAGS) $(DECODER_ASM_INCLUDES) -o $(DECODER_SRCDIR)/./core/asm/expand_picture.o $(DECODER_SRCDIR)/./core/asm/expand_picture.asm
-
 $(DECODER_SRCDIR)/./core/asm/intra_pred.o: $(DECODER_SRCDIR)/./core/asm/intra_pred.asm
 	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(DECODER_ASMFLAGS) $(DECODER_ASM_INCLUDES) -o $(DECODER_SRCDIR)/./core/asm/intra_pred.o $(DECODER_SRCDIR)/./core/asm/intra_pred.asm
 
-$(DECODER_SRCDIR)/./core/asm/mb_copy.o: $(DECODER_SRCDIR)/./core/asm/mb_copy.asm
-	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(DECODER_ASMFLAGS) $(DECODER_ASM_INCLUDES) -o $(DECODER_SRCDIR)/./core/asm/mb_copy.o $(DECODER_SRCDIR)/./core/asm/mb_copy.asm
-
-$(DECODER_SRCDIR)/./core/asm/mc_chroma.o: $(DECODER_SRCDIR)/./core/asm/mc_chroma.asm
-	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(DECODER_ASMFLAGS) $(DECODER_ASM_INCLUDES) -o $(DECODER_SRCDIR)/./core/asm/mc_chroma.o $(DECODER_SRCDIR)/./core/asm/mc_chroma.asm
-
-$(DECODER_SRCDIR)/./core/asm/mc_luma.o: $(DECODER_SRCDIR)/./core/asm/mc_luma.asm
-	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(DECODER_ASMFLAGS) $(DECODER_ASM_INCLUDES) -o $(DECODER_SRCDIR)/./core/asm/mc_luma.o $(DECODER_SRCDIR)/./core/asm/mc_luma.asm
-
-$(DECODER_SRCDIR)/./core/asm/memzero.o: $(DECODER_SRCDIR)/./core/asm/memzero.asm
-	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(DECODER_ASMFLAGS) $(DECODER_ASM_INCLUDES) -o $(DECODER_SRCDIR)/./core/asm/memzero.o $(DECODER_SRCDIR)/./core/asm/memzero.asm
-
 $(LIBPREFIX)decoder.$(LIBSUFFIX): $(DECODER_OBJS)
 	rm -f $(LIBPREFIX)decoder.$(LIBSUFFIX)
 	$(AR) cr $@ $(DECODER_OBJS)

codec/encoder/core/asm/asm_inc.asm

@@ -1,235 +0,0 @@
-;*!
-;* \copy
-;*     Copyright (c)  2009-2013, Cisco Systems
-;*     All rights reserved.
-;*
-;*     Redistribution and use in source and binary forms, with or without
-;*     modification, are permitted provided that the following conditions
-;*     are met:
-;*
-;*        * Redistributions of source code must retain the above copyright
-;*          notice, this list of conditions and the following disclaimer.
-;*
-;*        * Redistributions in binary form must reproduce the above copyright
-;*          notice, this list of conditions and the following disclaimer in
-;*          the documentation and/or other materials provided with the
-;*          distribution.
-;*
-;*     THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
-;*     "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
-;*     LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
-;*     FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
-;*     COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
-;*     INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
-;*     BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
-;*     LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
-;*     CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
-;*     LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
-;*     ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
-;*     POSSIBILITY OF SUCH DAMAGE.
-;*
-;*
-;*  sse2inc.asm
-;*
-;*  Abstract
-;*      macro and constant
-;*
-;*  History
-;*      8/5/2009 Created
-;*
-;*
-;*************************************************************************/
-;***********************************************************************
-; Options, for DEBUG
-;***********************************************************************
-
-%if 1
-	%define MOVDQ movdqa
-%else
-	%define MOVDQ movdqu
-%endif
-
-%if 1
-	%define WELSEMMS	emms
-%else
-	%define WELSEMMS
-%endif
-
-BITS 32
-
-;***********************************************************************
-; Macros
-;***********************************************************************
-
-%macro WELS_EXTERN 1
-	%ifdef PREFIX
-		global _%1
-		%define %1 _%1
-	%else
-		global %1
-	%endif
-%endmacro
-
-%macro WELS_AbsW 2
-	pxor        %2, %2
-    psubw       %2, %1
-    pmaxsw      %1, %2
-%endmacro
-
-%macro MMX_XSwap  4
-    movq		%4, %2
-    punpckh%1   %4, %3
-    punpckl%1   %2, %3
-%endmacro
-
-; pOut mm1, mm4, mm5, mm3
-%macro MMX_Trans4x4W 5
-    MMX_XSwap wd, %1, %2, %5
-    MMX_XSwap wd, %3, %4, %2
-    MMX_XSwap dq, %1, %3, %4
-    MMX_XSwap dq, %5, %2, %3
-%endmacro
-
-;for TRANSPOSE
-%macro SSE2_XSawp 4
-    movdqa      %4, %2
-    punpckl%1   %2, %3
-    punpckh%1   %4, %3
-%endmacro
-
-; in: xmm1, xmm2, xmm3, xmm4  pOut:  xmm1, xmm4, xmm5, mm3
-%macro SSE2_Trans4x4D 5
-    SSE2_XSawp dq,  %1, %2, %5
-    SSE2_XSawp dq,  %3, %4, %2
-    SSE2_XSawp qdq, %1, %3, %4
-    SSE2_XSawp qdq, %5, %2, %3
-%endmacro
-
-;in: xmm0, xmm1, xmm2, xmm3  pOut:  xmm0, xmm1, xmm3, xmm4
-%macro SSE2_TransTwo4x4W 5
-    SSE2_XSawp wd,  %1, %2, %5
-    SSE2_XSawp wd,  %3, %4, %2
-    SSE2_XSawp dq,  %1, %3, %4
-    SSE2_XSawp dq,  %5, %2, %3
-    SSE2_XSawp qdq, %1, %5, %2
-    SSE2_XSawp qdq, %4, %3, %5
-%endmacro
-
-;in:  m1, m2, m3, m4, m5, m6, m7, m8
-;pOut: m5, m3, m4, m8, m6, m2, m7, m1
-%macro SSE2_TransTwo8x8B 9
-	movdqa	%9,	%8
-	SSE2_XSawp bw,  %1, %2, %8
-	SSE2_XSawp bw,  %3, %4, %2
-	SSE2_XSawp bw,  %5, %6, %4
-	movdqa	%6, %9
-	movdqa	%9, %4
-	SSE2_XSawp bw,  %7, %6, %4
-
-	SSE2_XSawp wd,  %1, %3, %6
-	SSE2_XSawp wd,  %8, %2, %3
-	SSE2_XSawp wd,  %5, %7, %2
-	movdqa	%7, %9
-	movdqa	%9, %3
-	SSE2_XSawp wd,  %7, %4, %3
-
-	SSE2_XSawp dq,  %1, %5, %4
-	SSE2_XSawp dq,  %6, %2, %5
-	SSE2_XSawp dq,  %8, %7, %2
-	movdqa	%7, %9
-	movdqa	%9, %5
-	SSE2_XSawp dq,  %7, %3, %5
-
-	SSE2_XSawp qdq,  %1, %8, %3
-	SSE2_XSawp qdq,  %4, %2, %8
-	SSE2_XSawp qdq,  %6, %7, %2
-	movdqa	%7, %9
-	movdqa	%9, %1
-	SSE2_XSawp qdq,  %7, %5, %1
-	movdqa	%5, %9
-%endmacro
-
-;xmm0, xmm6, xmm7, [eax], [ecx]
-;xmm7 = 0, eax = pix1, ecx = pix2, xmm0 save the result
-%macro SSE2_LoadDiff8P 5
-    movq         %1, %4
-    punpcklbw    %1, %3
-    movq         %2, %5
-    punpcklbw    %2, %3
-    psubw        %1, %2
-%endmacro
-
-; m2 = m1 + m2, m1 = m1 - m2
-%macro SSE2_SumSub 3
-	movdqa  %3, %2
-    paddw   %2, %1
-    psubw   %1, %3
-%endmacro
-
-
-%macro butterfly_1to16_sse	3	; xmm? for dst, xmm? for tmp, one byte for pSrc [generic register name: a/b/c/d]
-	mov %3h, %3l
-	movd %1, e%3x		; i.e, 1% = eax (=b0)
-	pshuflw %2, %1, 00h	; ..., b0 b0 b0 b0 b0 b0 b0 b0
-	pshufd %1, %2, 00h	; b0 b0 b0 b0, b0 b0 b0 b0, b0 b0 b0 b0, b0 b0 b0 b0
-%endmacro
-
-;copy a dw into a xmm for 8 times
-%macro  SSE2_Copy8Times 2
-		movd	%1, %2
-		punpcklwd %1, %1
-		pshufd	%1,	%1,	0
-%endmacro
-
-;copy a db into a xmm for 16 times
-%macro  SSE2_Copy16Times 2
-		movd		%1, %2
-		pshuflw		%1, %1, 0
-		punpcklqdq	%1, %1
-		packuswb	%1,	%1
-%endmacro
-
-
-
-;***********************************************************************
-;preprocessor constants
-;***********************************************************************
-;dw 32,32,32,32,32,32,32,32 for xmm
-;dw 32,32,32,32 for mm
-%macro WELS_DW32 1
-	pcmpeqw %1,%1
-	psrlw %1,15
-	psllw %1,5
-%endmacro
-
-;dw 1, 1, 1, 1, 1, 1, 1, 1 for xmm
-;dw 1, 1, 1, 1 for mm
-%macro WELS_DW1 1
-	pcmpeqw %1,%1
-	psrlw %1,15
-%endmacro
-
-;all 0 for xmm and mm
-%macro	WELS_Zero 1
-	pxor %1, %1
-%endmacro
-
-;dd 1, 1, 1, 1 for xmm
-;dd 1, 1 for mm
-%macro WELS_DD1 1
-	pcmpeqw %1,%1
-	psrld %1,31
-%endmacro
-
-;dB 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1
-%macro WELS_DB1 1
-	pcmpeqw %1,%1
-	psrlw %1,15
-	packuswb %1,%1
-%endmacro
-
-
-
-
-
-

codec/encoder/core/asm/coeff.asm

@@ -44,7 +44,7 @@
 
 
 
-
+%ifdef X86_32
 SECTION .rodata align=16
 
 align 16
@@ -457,3 +457,4 @@ CavlcParamCal_sse2:
 	pop edi
 	pop ebx
 	ret
+%endif

codec/encoder/core/asm/dct.asm

@@ -42,8 +42,6 @@
 
 %include "asm_inc.asm"
 
-BITS 32
-
 SECTION .rodata align=16
 
 ;***********************************************************************
@@ -131,22 +129,27 @@ SSE2_DeQuant8 dw  10, 13, 10, 13, 13, 16, 13, 16,
     packuswb   %1, %2
     movd       %5, %1
 %endmacro
-
+SECTION .text
 ALIGN 16
 ;***********************************************************************
 ;   void __cdecl WelsDctT4_mmx( int16_t *pDct[4], uint8_t *pix1, int32_t i_pix1, uint8_t *pix2, int32_t i_pix2 )
 ;***********************************************************************
 WELS_EXTERN WelsDctT4_mmx
 WelsDctT4_mmx:
-    push    ebx
+    ;push    ebx
-    mov     eax, [esp+12]   ; pix1
+    ;mov     eax, [esp+12]   ; pix1
-    mov     ebx, [esp+16]   ; i_pix1
+    ;mov     ebx, [esp+16]   ; i_pix1
-    mov     ecx, [esp+20]   ; pix2
+    ;mov     ecx, [esp+20]   ; pix2
-    mov     edx, [esp+24]   ; i_pix2
+    ;mov     edx, [esp+24]   ; i_pix2
-
+    %assign push_num 0
+    LOAD_5_PARA
+	%ifndef X86_32
+	movsx r2, r2d
+	movsx r4, r4d
+	%endif
     WELS_Zero    mm7
 
-    MMX_LoadDiff4x4P mm1, mm2, mm3, mm4, eax, ebx, ecx, edx, mm0, mm7
+    MMX_LoadDiff4x4P mm1, mm2, mm3, mm4, r1, r2, r3, r4, mm0, mm7
 
     MMX_DCT			mm1, mm2, mm3 ,mm4, mm5, mm6
     MMX_Trans4x4W	mm3, mm1, mm4, mm5, mm2
@@ -154,14 +157,14 @@ WelsDctT4_mmx:
     MMX_DCT			mm3, mm5, mm2 ,mm4, mm1, mm6
     MMX_Trans4x4W	mm2, mm3, mm4, mm1, mm5
 
-    mov     eax, [esp+ 8]   ; pDct
+    ;mov     eax, [esp+ 8]   ; pDct
-    movq    [eax+ 0],   mm2
+    movq    [r0+ 0],   mm2
-    movq    [eax+ 8],   mm1
+    movq    [r0+ 8],   mm1
-    movq    [eax+16],   mm5
+    movq    [r0+16],   mm5
-    movq    [eax+24],   mm4
+    movq    [r0+24],   mm4
-
     WELSEMMS
-    pop     ebx
+    LOAD_5_PARA_POP
+    ;pop     ebx
     ret
 
 
@@ -170,23 +173,28 @@ WelsDctT4_mmx:
 ;***********************************************************************
 WELS_EXTERN WelsIDctT4Rec_mmx
 WelsIDctT4Rec_mmx:
-	push   ebx
+	;push   ebx
-%define	pushsize	4
+;%define	pushsize	4
-%define     p_dst       esp+pushsize+4
+;%define     p_dst       esp+pushsize+4
-%define     i_dst       esp+pushsize+8
+;%define     i_dst       esp+pushsize+8
-%define     p_pred      esp+pushsize+12
+;%define     p_pred      esp+pushsize+12
-%define     i_pred      esp+pushsize+16
+;%define     i_pred      esp+pushsize+16
-%define     pDct        esp+pushsize+20
+;%define     pDct        esp+pushsize+20
-
+    %assign push_num 0
-	mov     eax, [pDct   ]
+    LOAD_5_PARA
-    movq    mm0, [eax+ 0]
+	%ifndef X86_32
-    movq    mm1, [eax+ 8]
+	movsx r1, r1d
-    movq    mm2, [eax+16]
+	movsx r3, r3d
-    movq    mm3, [eax+24]
+	%endif
-    mov     edx, [p_dst ]
+;	mov     eax, [pDct   ] 
-    mov     ecx, [i_dst ]
+    movq    mm0, [r4+ 0]
-    mov     eax, [p_pred]
+    movq    mm1, [r4+ 8]
-    mov     ebx, [i_pred]
+    movq    mm2, [r4+16]
+    movq    mm3, [r4+24]
+    ;mov     edx, [p_dst ] ; r0
+    ;mov     ecx, [i_dst ] ; r1
+    ;mov     eax, [p_pred] ; r2
+    ;mov     ebx, [i_pred] ; r3
 
 	MMX_Trans4x4W		mm0, mm1, mm2, mm3, mm4
 	MMX_IDCT			mm1, mm2, mm3, mm4, mm0, mm6
@@ -196,21 +204,22 @@ WelsIDctT4Rec_mmx:
     WELS_Zero			mm7
     WELS_DW32			mm6
 
-    MMX_StoreDiff4P		mm3, mm0, mm6, mm7, [edx], [eax]
+    MMX_StoreDiff4P		mm3, mm0, mm6, mm7, [r0], [r2]
-    MMX_StoreDiff4P		mm4, mm0, mm6, mm7, [edx+ecx], [eax+ebx]
+    MMX_StoreDiff4P		mm4, mm0, mm6, mm7, [r0+r1], [r2+r3]
-    lea     edx, [edx+2*ecx]
+    lea     r0, [r0+2*r1]
-    lea     eax, [eax+2*ebx]
+    lea     r2, [r2+2*r3]
-    MMX_StoreDiff4P		mm1, mm0, mm6, mm7, [edx], [eax]
+    MMX_StoreDiff4P		mm1, mm0, mm6, mm7, [r0], [r2]
-    MMX_StoreDiff4P		mm2, mm0, mm6, mm7, [edx+ecx], [eax+ebx]
+    MMX_StoreDiff4P		mm2, mm0, mm6, mm7, [r0+r1], [r2+r3]
 
 	WELSEMMS
-%undef	pushsize
+    LOAD_5_PARA_POP
-%undef  p_dst
+;%undef	pushsize
-%undef  i_dst
+;%undef  p_dst
-%undef  p_pred
+;%undef  i_dst
-%undef  i_pred
+;%undef  p_pred
-%undef  pDct
+;%undef  i_pred
-    pop ebx
+;%undef  pDct
+;    pop ebx
     ret
 
 
@@ -314,78 +323,88 @@ WelsIDctT4Rec_mmx:
 WELS_EXTERN WelsDctFourT4_sse2
 ALIGN 16
 WelsDctFourT4_sse2:
-    push    ebx
+    ;push    ebx
-    push	esi
+    ;push	esi
-    mov		esi, [esp+12]
+    ;mov		esi, [esp+12]
-    mov     eax, [esp+16]   ; pix1
+    ;mov     eax, [esp+16]   ; pix1
-    mov     ebx, [esp+20]   ; i_pix1
+    ;mov     ebx, [esp+20]   ; i_pix1
-    mov     ecx, [esp+24]   ; pix2
+    ;mov     ecx, [esp+24]   ; pix2
-    mov     edx, [esp+28]   ; i_pix2
+    ;mov     edx, [esp+28]   ; i_pix2
-
+    %assign push_num 0
+    LOAD_5_PARA
+	%ifndef X86_32
+	movsx r2, r2d
+	movsx r4, r4d
+	%endif
     pxor    xmm7, xmm7
-
 	;Load 4x8
-	SSE2_LoadDiff8P    xmm0, xmm6, xmm7, [eax      ], [ecx]
+	SSE2_LoadDiff8P    xmm0, xmm6, xmm7, [r1], [r3]
-    SSE2_LoadDiff8P    xmm1, xmm6, xmm7, [eax+ebx  ], [ecx+edx]
+    SSE2_LoadDiff8P    xmm1, xmm6, xmm7, [r1+r2], [r3+r4]
-	lea		eax, [eax + 2 * ebx]
+	lea		r1, [r1 + 2 * r2]
-	lea		ecx, [ecx + 2 * edx]
+	lea		r3, [r3 + 2 * r4]
-	SSE2_LoadDiff8P    xmm2, xmm6, xmm7, [eax], [ecx]
+	SSE2_LoadDiff8P    xmm2, xmm6, xmm7, [r1], [r3]
-    SSE2_LoadDiff8P    xmm3, xmm6, xmm7, [eax+ebx], [ecx+edx]
+    SSE2_LoadDiff8P    xmm3, xmm6, xmm7, [r1+r2], [r3+r4]
 
 	SSE2_DCT			xmm1, xmm2, xmm3, xmm4, xmm5, xmm0
 	SSE2_TransTwo4x4W	xmm2, xmm0, xmm3, xmm4, xmm1
 	SSE2_DCT			xmm0, xmm4, xmm1, xmm3, xmm5, xmm2
 	SSE2_TransTwo4x4W	xmm4, xmm2, xmm1, xmm3, xmm0
 
-	SSE2_Store4x8p esi, xmm4, xmm2, xmm3, xmm0, xmm5
+	SSE2_Store4x8p r0, xmm4, xmm2, xmm3, xmm0, xmm5
 
-	lea		eax, [eax + 2 * ebx]
+	lea		r1, [r1 + 2 * r2]
-	lea		ecx, [ecx + 2 * edx]
+	lea		r3, [r3 + 2 * r4]
 
 	;Load 4x8
-	SSE2_LoadDiff8P    xmm0, xmm6, xmm7, [eax      ], [ecx    ]
+	SSE2_LoadDiff8P    xmm0, xmm6, xmm7, [r1      ], [r3    ]
-    SSE2_LoadDiff8P    xmm1, xmm6, xmm7, [eax+ebx  ], [ecx+edx]
+    SSE2_LoadDiff8P    xmm1, xmm6, xmm7, [r1+r2  ], [r3+r4]
-	lea		eax, [eax + 2 * ebx]
+	lea		r1, [r1 + 2 * r2]
-	lea		ecx, [ecx + 2 * edx]
+	lea		r3, [r3 + 2 * r4]
-    SSE2_LoadDiff8P    xmm2, xmm6, xmm7, [eax], [ecx]
+    SSE2_LoadDiff8P    xmm2, xmm6, xmm7, [r1], [r3]
-    SSE2_LoadDiff8P    xmm3, xmm6, xmm7, [eax+ebx], [ecx+edx]
+    SSE2_LoadDiff8P    xmm3, xmm6, xmm7, [r1+r2], [r3+r4]
 
 	SSE2_DCT			xmm1, xmm2, xmm3, xmm4, xmm5, xmm0
 	SSE2_TransTwo4x4W	xmm2, xmm0, xmm3, xmm4, xmm1
     SSE2_DCT			xmm0, xmm4, xmm1, xmm3, xmm5, xmm2
 	SSE2_TransTwo4x4W	xmm4, xmm2, xmm1, xmm3, xmm0
 
-	lea		esi, [esi+64]
+	lea		r0, [r0+64]
-	SSE2_Store4x8p esi, xmm4, xmm2, xmm3, xmm0, xmm5
+	SSE2_Store4x8p r0, xmm4, xmm2, xmm3, xmm0, xmm5
 
-    pop esi
+    ;pop esi
-    pop ebx
+    ;pop ebx
+	LOAD_5_PARA_POP
     ret
 
 
-%define		rec			esp + pushsize + 4
+;%define		rec			esp + pushsize + 4
-%define		stride		esp + pushsize + 8
+;%define		stride		esp + pushsize + 8
-%define		pred		esp + pushsize + 12
+;%define		pred		esp + pushsize + 12
-%define		pred_stride	esp + pushsize + 16
+;%define		pred_stride	esp + pushsize + 16
-%define		rs			esp + pushsize + 20
+;%define		rs			esp + pushsize + 20
 ;***********************************************************************
 ; void WelsIDctFourT4Rec_sse2(uint8_t *rec, int32_t stride, uint8_t *pred, int32_t pred_stride, int16_t *rs);
 ;***********************************************************************
 WELS_EXTERN WelsIDctFourT4Rec_sse2
 ALIGN 16
 WelsIDctFourT4Rec_sse2:
-%define	pushsize	8
+;%define	pushsize	8
-    push		ebx
+;    push		ebx
-    push		esi
+;    push		esi
-
-    mov			eax,		[rec]
-    mov			ebx,		[stride]
-    mov			ecx,		[pred]
-    mov			edx,		[pred_stride]
-    mov			esi,		[rs]
 
+;    mov			eax,		[rec]
+;    mov			ebx,		[stride]
+;    mov			ecx,		[pred]
+;    mov			edx,		[pred_stride]
+;    mov			esi,		[rs]
+	%assign push_num 0
+	LOAD_5_PARA
+	%ifndef X86_32
+	movsx r1, r1d
+	movsx r3, r3d
+	%endif
 	;Load 4x8
-	SSE2_Load4x8p  esi, xmm0, xmm1, xmm4, xmm2, xmm5
+	SSE2_Load4x8p  r4, xmm0, xmm1, xmm4, xmm2, xmm5
 
 	SSE2_TransTwo4x4W	xmm0, xmm1, xmm4, xmm2, xmm3
   	SSE2_IDCT			xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm0
@@ -395,17 +414,17 @@ WelsIDctFourT4Rec_sse2:
 	WELS_Zero			xmm7
     WELS_DW32			xmm6
 
-	SSE2_StoreDiff8p   xmm4, xmm5, xmm6, xmm7, [eax		],	[ecx]
+	SSE2_StoreDiff8p   xmm4, xmm5, xmm6, xmm7, [r0		],	[r2]
-	SSE2_StoreDiff8p   xmm0, xmm5, xmm6, xmm7, [eax + ebx	],	[ecx + edx]
+	SSE2_StoreDiff8p   xmm0, xmm5, xmm6, xmm7, [r0 + r1	],	[r2 + r3]
-	lea		eax, [eax + 2 * ebx]
+	lea		r0, [r0 + 2 * r1]
-	lea		ecx, [ecx + 2 * edx]
+	lea		r2, [r2 + 2 * r3]
-	SSE2_StoreDiff8p   xmm1, xmm5, xmm6, xmm7, [eax],			[ecx]
+	SSE2_StoreDiff8p   xmm1, xmm5, xmm6, xmm7, [r0],			[r2]
-	SSE2_StoreDiff8p   xmm2, xmm5, xmm6, xmm7, [eax + ebx	],	[ecx + edx]
+	SSE2_StoreDiff8p   xmm2, xmm5, xmm6, xmm7, [r0 + r1	],	[r2 + r3]
 
-    add		esi, 64
+    add		r4, 64
-	lea		eax, [eax + 2 * ebx]
+	lea		r0, [r0 + 2 * r1]
-	lea		ecx, [ecx + 2 * edx]
+	lea		r2, [r2 + 2 * r3]
-   	SSE2_Load4x8p  esi, xmm0, xmm1, xmm4, xmm2, xmm5
+   	SSE2_Load4x8p  r4, xmm0, xmm1, xmm4, xmm2, xmm5
 
 	SSE2_TransTwo4x4W   xmm0, xmm1, xmm4, xmm2, xmm3
 	SSE2_IDCT			xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm0
@@ -415,15 +434,15 @@ WelsIDctFourT4Rec_sse2:
 	WELS_Zero			xmm7
     WELS_DW32			xmm6
 
-	SSE2_StoreDiff8p   xmm4, xmm5, xmm6, xmm7, [eax		],	[ecx]
+	SSE2_StoreDiff8p   xmm4, xmm5, xmm6, xmm7, [r0		],	[r2]
-	SSE2_StoreDiff8p   xmm0, xmm5, xmm6, xmm7, [eax + ebx	],	[ecx + edx]
+	SSE2_StoreDiff8p   xmm0, xmm5, xmm6, xmm7, [r0 + r1	],	[r2 + r3]
-	lea		eax, [eax + 2 * ebx]
+	lea		r0, [r0 + 2 * r1]
-	lea		ecx, [ecx + 2 * edx]
+	lea		r2, [r2 + 2 * r3]
-	SSE2_StoreDiff8p   xmm1, xmm5, xmm6, xmm7, [eax],			[ecx]
+	SSE2_StoreDiff8p   xmm1, xmm5, xmm6, xmm7, [r0],			[r2]
-	SSE2_StoreDiff8p   xmm2, xmm5, xmm6, xmm7, [eax + ebx],	[ecx + edx]
+	SSE2_StoreDiff8p   xmm2, xmm5, xmm6, xmm7, [r0 + r1],	[r2 + r3]
-
+	LOAD_5_PARA_POP
-    pop		esi
+   ; pop		esi
-    pop		ebx
+   ; pop		ebx
     ret
 
   %macro SSE2_StoreDiff4x8p 8
@@ -438,54 +457,60 @@ WelsIDctFourT4Rec_sse2:
 ;***********************************************************************
 WELS_EXTERN WelsIDctRecI16x16Dc_sse2
 ALIGN 16
-%define		pushsize	8
+;%define		pushsize	8
-%define		luma_dc		esp + pushsize + 20
+;%define		luma_dc		esp + pushsize + 20
 WelsIDctRecI16x16Dc_sse2:
-    push		esi
+	%assign push_num 0
-    push		edi
+	LOAD_5_PARA
+	%ifndef X86_32
+	movsx r1, r1d
+	movsx r3, r3d
+	%endif
+   ; push		esi
+   ; push		edi
 
-	mov			ecx,		[luma_dc]
+   ;mov			ecx,		[luma_dc] ; r4
-    mov			eax,		[rec]
+    ;mov			eax,		[rec] ; r0
-    mov			edx,		[stride]
+    ;mov			edx,		[stride] ; r1
-    mov			esi,		[pred]
+    ;mov			esi,		[pred]; r2
-    mov			edi,		[pred_stride]
+    ;mov			edi,		[pred_stride]; r3
 	pxor		xmm7,		xmm7
     WELS_DW32	xmm6
 
-	SSE2_Load8DC			xmm0, xmm1, xmm2, xmm3, xmm6, [ecx]
+	SSE2_Load8DC			xmm0, xmm1, xmm2, xmm3, xmm6, [r4]
-	SSE2_StoreDiff4x8p		xmm0, xmm1, xmm5, xmm7, eax, esi, edx, edi
+	SSE2_StoreDiff4x8p		xmm0, xmm1, xmm5, xmm7, r0, r2, r1, r3
 
-	lea			eax,		[eax + 2 * edx]
+	lea			r0,		[r0 + 2 * r1]
-	lea			esi,		[esi + 2 * edi]
+	lea			r2,		[r2 + 2 * r3]
-	SSE2_StoreDiff4x8p		xmm0, xmm1, xmm5, xmm7, eax, esi, edx, edi
+	SSE2_StoreDiff4x8p		xmm0, xmm1, xmm5, xmm7, r0, r2, r1, r3
 
-	lea			eax,		[eax + 2 * edx]
+	lea			r0,		[r0 + 2 * r1]
-	lea			esi,		[esi + 2 * edi]
+	lea			r2,		[r2 + 2 * r3]
-	SSE2_StoreDiff4x8p		xmm2, xmm3, xmm5, xmm7, eax, esi, edx, edi
+	SSE2_StoreDiff4x8p		xmm2, xmm3, xmm5, xmm7, r0, r2, r1, r3
 
-	lea			eax,		[eax + 2 * edx]
+	lea			r0,		[r0 + 2 * r1]
-	lea			esi,		[esi + 2 * edi]
+	lea			r2,		[r2 + 2 * r3]
-	SSE2_StoreDiff4x8p		xmm2, xmm3, xmm5, xmm7, eax, esi, edx, edi
+	SSE2_StoreDiff4x8p		xmm2, xmm3, xmm5, xmm7, r0, r2, r1, r3
 
-	SSE2_Load8DC			xmm0, xmm1, xmm2, xmm3, xmm6, [ecx + 16]
+	SSE2_Load8DC			xmm0, xmm1, xmm2, xmm3, xmm6, [r4 + 16]
-	lea			eax,		[eax + 2 * edx]
+	lea			r0,		[r0 + 2 * r1]
-	lea			esi,		[esi + 2 * edi]
+	lea			r2,		[r2 + 2 * r3]
-	SSE2_StoreDiff4x8p		xmm0, xmm1, xmm5, xmm7, eax, esi, edx, edi
+	SSE2_StoreDiff4x8p		xmm0, xmm1, xmm5, xmm7, r0, r2, r1, r3
 
-	lea			eax,		[eax + 2 * edx]
+	lea			r0,		[r0 + 2 * r1]
-	lea			esi,		[esi + 2 * edi]
+	lea			r2,		[r2 + 2 * r3]
-	SSE2_StoreDiff4x8p		xmm0, xmm1, xmm5, xmm7, eax, esi, edx, edi
+	SSE2_StoreDiff4x8p		xmm0, xmm1, xmm5, xmm7, r0, r2, r1, r3
 
-	lea			eax,		[eax + 2 * edx]
+	lea			r0,		[r0 + 2 * r1]
-	lea			esi,		[esi + 2 * edi]
+	lea			r2,		[r2 + 2 * r3]
-	SSE2_StoreDiff4x8p		xmm2, xmm3, xmm5, xmm7, eax, esi, edx, edi
+	SSE2_StoreDiff4x8p		xmm2, xmm3, xmm5, xmm7, r0, r2, r1, r3
 
-	lea			eax,		[eax + 2 * edx]
+	lea			r0,		[r0 + 2 * r1]
-	lea			esi,		[esi + 2 * edi]
+	lea			r2,		[r2 + 2 * r3]
-	SSE2_StoreDiff4x8p		xmm2, xmm3, xmm5, xmm7, eax, esi, edx, edi
+	SSE2_StoreDiff4x8p		xmm2, xmm3, xmm5, xmm7, r0, r2, r1, r3
-
+	LOAD_5_PARA_POP
-    pop		edi
+    ;pop		edi
-    pop		esi
+    ;pop		esi
     ret
 
 
@@ -503,17 +528,16 @@ WelsIDctRecI16x16Dc_sse2:
 	movdqa	%4, %1
 	psubd	%4, %2
 %endmacro
-
  %macro		SSE2_Load4Col	5
-	movsx		edx,		WORD[%5]
+	movsx		r2,		WORD[%5]
- 	movd		%1,			edx
+ 	movd		%1,			r2d
- 	movsx		edx,		WORD[%5 + 0x20]
+ 	movsx		r2,		WORD[%5 + 0x20]
- 	movd		%2,			edx
+ 	movd		%2,			r2d
 	punpckldq	%1,			%2
-	movsx		edx,		WORD[%5 + 0x80]
+	movsx		r2,		WORD[%5 + 0x80]
- 	movd		%3,			edx
+ 	movd		%3,			r2d
-	movsx		edx,		WORD[%5 + 0xa0]
+	movsx		r2,		WORD[%5 + 0xa0]
- 	movd		%4,			edx
+ 	movd		%4,			r2d
 	punpckldq	%3,			%4
 	punpcklqdq	%1,			%3
  %endmacro
@@ -523,13 +547,14 @@ WelsIDctRecI16x16Dc_sse2:
 ;***********************************************************************
 WELS_EXTERN WelsHadamardT4Dc_sse2
 WelsHadamardT4Dc_sse2:
-		mov			eax,		[esp + 4]	; luma_dc
+		;mov			eax,		[esp + 4]	; luma_dc
-		mov			ecx,		[esp + 8]	; pDct
+		;mov			ecx,		[esp + 8]	; pDct
-
+		%assign push_num 0
-		SSE2_Load4Col	    xmm1, xmm5, xmm6, xmm0, ecx
+		LOAD_2_PARA
-		SSE2_Load4Col	    xmm2, xmm5, xmm6, xmm0, ecx + 0x40
+		SSE2_Load4Col	    xmm1, xmm5, xmm6, xmm0, r1
-		SSE2_Load4Col	    xmm3, xmm5, xmm6, xmm0, ecx + 0x100
+		SSE2_Load4Col	    xmm2, xmm5, xmm6, xmm0, r1 + 0x40
-		SSE2_Load4Col	    xmm4, xmm5, xmm6, xmm0, ecx + 0x140
+		SSE2_Load4Col	    xmm3, xmm5, xmm6, xmm0, r1 + 0x100
+		SSE2_Load4Col	    xmm4, xmm5, xmm6, xmm0, r1 + 0x140
 
 		SSE2_SumSubD		xmm1, xmm2, xmm7
 		SSE2_SumSubD		xmm3, xmm4, xmm7
@@ -548,9 +573,7 @@ WelsHadamardT4Dc_sse2:
 
 		packssdw	xmm3,	xmm4
 		packssdw	xmm2,	xmm1
-		movdqa	[eax+ 0],   xmm3
+		movdqa	[r0+ 0],   xmm3
-		movdqa	[eax+16],   xmm2
+		movdqa	[r0+16],   xmm2
 
 		ret
-
-

codec/encoder/core/asm/expand_picture.asm

@@ -1,653 +0,0 @@
-;*!
-;* \copy
-;*     Copyright (c)  2009-2013, Cisco Systems
-;*     All rights reserved.
-;*
-;*     Redistribution and use in source and binary forms, with or without
-;*     modification, are permitted provided that the following conditions
-;*     are met:
-;*
-;*        * Redistributions of source code must retain the above copyright
-;*          notice, this list of conditions and the following disclaimer.
-;*
-;*        * Redistributions in binary form must reproduce the above copyright
-;*          notice, this list of conditions and the following disclaimer in
-;*          the documentation and/or other materials provided with the
-;*          distribution.
-;*
-;*     THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
-;*     "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
-;*     LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
-;*     FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
-;*     COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
-;*     INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
-;*     BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
-;*     LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
-;*     CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
-;*     LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
-;*     ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
-;*     POSSIBILITY OF SUCH DAMAGE.
-;*
-;*
-;*  expand_picture.asm
-;*
-;*  Abstract
-;*      mmxext/sse for expand_frame
-;*
-;*  History
-;*      09/25/2009 Created
-;*
-;*
-;*************************************************************************/
-
-%include "asm_inc.asm"
-
-BITS 32
-
-;***********************************************************************
-; Macros and other preprocessor constants
-;***********************************************************************
-
-;***********************************************************************
-; Local Data (Read Only)
-;***********************************************************************
-
-;SECTION .rodata pData align=16
-
-;***********************************************************************
-; Various memory constants (trigonometric values or rounding values)
-;***********************************************************************
-;%define PADDING_SIZE_ASM 	32 	; PADDING_LENGTH
-
-;***********************************************************************
-; Code
-;***********************************************************************
-
-
-
-SECTION .text
-
-WELS_EXTERN ExpandPictureLuma_sse2
-WELS_EXTERN ExpandPictureChromaAlign_sse2	; for chroma alignment
-WELS_EXTERN ExpandPictureChromaUnalign_sse2	; for chroma unalignment
-
-;;;;;;;expanding result;;;;;;;
-
-;aaaa|attttttttttttttttb|bbbb
-;aaaa|attttttttttttttttb|bbbb
-;aaaa|attttttttttttttttb|bbbb
-;aaaa|attttttttttttttttb|bbbb
-;----------------------------
-;aaaa|attttttttttttttttb|bbbb
-;llll|l                r|rrrr
-;llll|l                r|rrrr
-;llll|l                r|rrrr
-;llll|l                r|rrrr
-;llll|l                r|rrrr
-;cccc|ceeeeeeeeeeeeeeeed|dddd
-;----------------------------
-;cccc|ceeeeeeeeeeeeeeeed|dddd
-;cccc|ceeeeeeeeeeeeeeeed|dddd
-;cccc|ceeeeeeeeeeeeeeeed|dddd
-;cccc|ceeeeeeeeeeeeeeeed|dddd
-
-%macro mov_line_8x4_mmx		3	; dst, stride, mm?
-	movq [%1], %3
-	movq [%1+%2], %3
-	lea %1, [%1+2*%2]
-	movq [%1], %3
-	movq [%1+%2], %3
-	lea %1, [%1+2*%2]
-%endmacro
-
-%macro mov_line_end8x4_mmx		3	; dst, stride, mm?
-	movq [%1], %3
-	movq [%1+%2], %3
-	lea %1, [%1+2*%2]
-	movq [%1], %3
-	movq [%1+%2], %3
-	lea %1, [%1+%2]
-%endmacro
-
-%macro mov_line_16x4_sse2	4	; dst, stride, xmm?, u/a
-	movdq%4 [%1], %3 		; top(bottom)_0
-	movdq%4 [%1+%2], %3		; top(bottom)_1
-	lea %1, [%1+2*%2]
-	movdq%4 [%1], %3 		; top(bottom)_2
-	movdq%4 [%1+%2], %3		; top(bottom)_3
-	lea %1, [%1+2*%2]
-%endmacro
-
-%macro mov_line_end16x4_sse2	4	; dst, stride, xmm?, u/a
-	movdq%4 [%1], %3 		; top(bottom)_0
-	movdq%4 [%1+%2], %3		; top(bottom)_1
-	lea %1, [%1+2*%2]
-	movdq%4 [%1], %3 		; top(bottom)_2
-	movdq%4 [%1+%2], %3		; top(bottom)_3
-	lea %1, [%1+%2]
-%endmacro
-
-%macro mov_line_32x4_sse2	3	; dst, stride, xmm?
-	movdqa [%1], %3 		; top(bottom)_0
-	movdqa [%1+16], %3 		; top(bottom)_0
-	movdqa [%1+%2], %3		; top(bottom)_1
-	movdqa [%1+%2+16], %3		; top(bottom)_1
-	lea %1, [%1+2*%2]
-	movdqa [%1], %3 		; top(bottom)_2
-	movdqa [%1+16], %3 		; top(bottom)_2
-	movdqa [%1+%2], %3		; top(bottom)_3
-	movdqa [%1+%2+16], %3		; top(bottom)_3
-	lea %1, [%1+2*%2]
-%endmacro
-
-%macro mov_line_end32x4_sse2	3	; dst, stride, xmm?
-	movdqa [%1], %3 		; top(bottom)_0
-	movdqa [%1+16], %3 		; top(bottom)_0
-	movdqa [%1+%2], %3		; top(bottom)_1
-	movdqa [%1+%2+16], %3		; top(bottom)_1
-	lea %1, [%1+2*%2]
-	movdqa [%1], %3 		; top(bottom)_2
-	movdqa [%1+16], %3 		; top(bottom)_2
-	movdqa [%1+%2], %3		; top(bottom)_3
-	movdqa [%1+%2+16], %3		; top(bottom)_3
-	lea %1, [%1+%2]
-%endmacro
-
-%macro exp_top_bottom_sse2	1	; iPaddingSize [luma(32)/chroma(16)]
-	; ebx [width/16(8)]
-	; esi [pSrc+0], edi [pSrc-1], ecx [-stride], 32(16)		; top
-	; eax [pSrc+(h-1)*stride], ebp [pSrc+(h+31)*stride], 32(16)	; bottom
-
-%if %1 == 32		; for luma
-	sar ebx, 04h 	; width / 16(8) pixels
-.top_bottom_loops:
-	; top
-	movdqa xmm0, [esi]		; first line of picture pData
-	mov_line_16x4_sse2 edi, ecx, xmm0, a	; dst, stride, xmm?
-	mov_line_16x4_sse2 edi, ecx, xmm0, a
-	mov_line_16x4_sse2 edi, ecx, xmm0, a
-	mov_line_16x4_sse2 edi, ecx, xmm0, a
-	mov_line_16x4_sse2 edi, ecx, xmm0, a	; dst, stride, xmm?
-	mov_line_16x4_sse2 edi, ecx, xmm0, a
-	mov_line_16x4_sse2 edi, ecx, xmm0, a
-	mov_line_end16x4_sse2 edi, ecx, xmm0, a
-
-	; bottom
-	movdqa xmm1, [eax] 		; last line of picture pData
-	mov_line_16x4_sse2 ebp, ecx, xmm1, a	; dst, stride, xmm?
-	mov_line_16x4_sse2 ebp, ecx, xmm1, a
-	mov_line_16x4_sse2 ebp, ecx, xmm1, a
-	mov_line_16x4_sse2 ebp, ecx, xmm1, a
-	mov_line_16x4_sse2 ebp, ecx, xmm1, a	; dst, stride, xmm?
-	mov_line_16x4_sse2 ebp, ecx, xmm1, a
-	mov_line_16x4_sse2 ebp, ecx, xmm1, a
-	mov_line_end16x4_sse2 ebp, ecx, xmm1, a
-
-	lea esi, [esi+16]		; top pSrc
-	lea edi, [edi+16]		; top dst
-	lea eax, [eax+16]		; bottom pSrc
-	lea ebp, [ebp+16]		; bottom dst
-	neg ecx 			; positive/negative stride need for next loop?
-
-	dec ebx
-	jnz near .top_bottom_loops
-%elif %1 == 16	; for chroma ??
-	mov edx, ebx
-	sar ebx, 04h 	; (width / 16) pixels
-.top_bottom_loops:
-	; top
-	movdqa xmm0, [esi]		; first line of picture pData
-	mov_line_16x4_sse2 edi, ecx, xmm0, a	; dst, stride, xmm?
-	mov_line_16x4_sse2 edi, ecx, xmm0, a
-	mov_line_16x4_sse2 edi, ecx, xmm0, a
-	mov_line_end16x4_sse2 edi, ecx, xmm0, a
-
-	; bottom
-	movdqa xmm1, [eax] 		; last line of picture pData
-	mov_line_16x4_sse2 ebp, ecx, xmm1, a	; dst, stride, xmm?
-	mov_line_16x4_sse2 ebp, ecx, xmm1, a
-	mov_line_16x4_sse2 ebp, ecx, xmm1, a
-	mov_line_end16x4_sse2 ebp, ecx, xmm1, a
-
-	lea esi, [esi+16]		; top pSrc
-	lea edi, [edi+16]		; top dst
-	lea eax, [eax+16]		; bottom pSrc
-	lea ebp, [ebp+16]		; bottom dst
-	neg ecx 			; positive/negative stride need for next loop?
-
-	dec ebx
-	jnz near .top_bottom_loops
-
-	; for remaining 8 bytes
-	and edx, 0fh		; any 8 bytes left?
-	test edx, edx
-	jz near .to_be_continued	; no left to exit here
-
-	; top
-	movq mm0, [esi]		; remained 8 byte
-	mov_line_8x4_mmx edi, ecx, mm0	; dst, stride, mm?
-	mov_line_8x4_mmx edi, ecx, mm0	; dst, stride, mm?
-	mov_line_8x4_mmx edi, ecx, mm0	; dst, stride, mm?
-	mov_line_end8x4_mmx edi, ecx, mm0	; dst, stride, mm?
-	; bottom
-	movq mm1, [eax]
-	mov_line_8x4_mmx ebp, ecx, mm1	; dst, stride, mm?
-	mov_line_8x4_mmx ebp, ecx, mm1	; dst, stride, mm?
-	mov_line_8x4_mmx ebp, ecx, mm1	; dst, stride, mm?
-	mov_line_end8x4_mmx ebp, ecx, mm1	; dst, stride, mm?
-	WELSEMMS
-
-.to_be_continued:
-%endif
-%endmacro
-
-%macro exp_left_right_sse2	2	; iPaddingSize [luma(32)/chroma(16)], u/a
-	; ecx [height]
-	; esi [pSrc+0], 	   edi [pSrc-32], edx [stride], 32(16)	; left
-	; ebx [pSrc+(w-1)], ebp [pSrc+w], 32(16)			; right
-;	xor eax, eax 	; for pixel pData (uint8_t)		; make sure eax=0 at least high 24 bits of eax = 0
-
-%if %1 == 32		; for luma
-.left_right_loops:
-	; left
-	mov al, byte [esi]		; pixel pData for left border
-	butterfly_1to16_sse	xmm0, xmm1, a				; dst, tmp, pSrc [generic register name: a/b/c/d]
-	movdqa [edi], xmm0
-	movdqa [edi+16], xmm0
-
-	; right
-	mov al, byte [ebx]
-	butterfly_1to16_sse	xmm1, xmm2, a				; dst, tmp, pSrc [generic register name: a/b/c/d]
-	movdqa [ebp], xmm1
-	movdqa [ebp+16], xmm1
-
-	lea esi, [esi+edx]		; left pSrc
-	lea edi, [edi+edx]		; left dst
-	lea ebx, [ebx+edx]		; right pSrc
-	lea ebp, [ebp+edx]		; right dst
-
-	dec ecx
-	jnz near .left_right_loops
-%elif %1 == 16	; for chroma ??
-.left_right_loops:
-	; left
-	mov al, byte [esi]		; pixel pData for left border
-	butterfly_1to16_sse	xmm0, xmm1, a				; dst, tmp, pSrc [generic register name: a/b/c/d]
-	movdqa [edi], xmm0
-
-	; right
-	mov al, byte [ebx]
-	butterfly_1to16_sse	xmm1, xmm2, a				; dst, tmp, pSrc [generic register name: a/b/c/d]
-	movdq%2 [ebp], xmm1								; might not be aligned 16 bytes in case chroma planes
-
-	lea esi, [esi+edx]		; left pSrc
-	lea edi, [edi+edx]		; left dst
-	lea ebx, [ebx+edx]		; right pSrc
-	lea ebp, [ebp+edx]		; right dst
-
-	dec ecx
-	jnz near .left_right_loops
-%endif
-%endmacro
-
-%macro exp_cross_sse2	2	; iPaddingSize [luma(32)/chroma(16)], u/a
-	; top-left: (x)mm3, top-right: (x)mm4, bottom-left: (x)mm5, bottom-right: (x)mm6
-	; edi: TL, ebp: TR, eax: BL, ebx: BR, ecx, -stride
-%if %1 == 32		; luma
-	; TL
-	mov_line_32x4_sse2	edi, ecx, xmm3	; dst, stride, xmm?
-	mov_line_32x4_sse2	edi, ecx, xmm3	; dst, stride, xmm?
-	mov_line_32x4_sse2	edi, ecx, xmm3	; dst, stride, xmm?
-	mov_line_32x4_sse2	edi, ecx, xmm3	; dst, stride, xmm?
-	mov_line_32x4_sse2	edi, ecx, xmm3	; dst, stride, xmm?
-	mov_line_32x4_sse2	edi, ecx, xmm3	; dst, stride, xmm?
-	mov_line_32x4_sse2	edi, ecx, xmm3	; dst, stride, xmm?
-	mov_line_end32x4_sse2	edi, ecx, xmm3	; dst, stride, xmm?
-
-	; TR
-	mov_line_32x4_sse2	ebp, ecx, xmm4	; dst, stride, xmm?
-	mov_line_32x4_sse2	ebp, ecx, xmm4	; dst, stride, xmm?
-	mov_line_32x4_sse2	ebp, ecx, xmm4	; dst, stride, xmm?
-	mov_line_32x4_sse2	ebp, ecx, xmm4	; dst, stride, xmm?
-	mov_line_32x4_sse2	ebp, ecx, xmm4	; dst, stride, xmm?
-	mov_line_32x4_sse2	ebp, ecx, xmm4	; dst, stride, xmm?
-	mov_line_32x4_sse2	ebp, ecx, xmm4	; dst, stride, xmm?
-	mov_line_end32x4_sse2	ebp, ecx, xmm4	; dst, stride, xmm?
-
-	; BL
-	mov_line_32x4_sse2	eax, ecx, xmm5	; dst, stride, xmm?
-	mov_line_32x4_sse2	eax, ecx, xmm5	; dst, stride, xmm?
-	mov_line_32x4_sse2	eax, ecx, xmm5	; dst, stride, xmm?
-	mov_line_32x4_sse2	eax, ecx, xmm5	; dst, stride, xmm?
-	mov_line_32x4_sse2	eax, ecx, xmm5	; dst, stride, xmm?
-	mov_line_32x4_sse2	eax, ecx, xmm5	; dst, stride, xmm?
-	mov_line_32x4_sse2	eax, ecx, xmm5	; dst, stride, xmm?
-	mov_line_end32x4_sse2	eax, ecx, xmm5	; dst, stride, xmm?
-
-	; BR
-	mov_line_32x4_sse2	ebx, ecx, xmm6	; dst, stride, xmm?
-	mov_line_32x4_sse2	ebx, ecx, xmm6	; dst, stride, xmm?
-	mov_line_32x4_sse2	ebx, ecx, xmm6	; dst, stride, xmm?
-	mov_line_32x4_sse2	ebx, ecx, xmm6	; dst, stride, xmm?
-	mov_line_32x4_sse2	ebx, ecx, xmm6	; dst, stride, xmm?
-	mov_line_32x4_sse2	ebx, ecx, xmm6	; dst, stride, xmm?
-	mov_line_32x4_sse2	ebx, ecx, xmm6	; dst, stride, xmm?
-	mov_line_end32x4_sse2	ebx, ecx, xmm6	; dst, stride, xmm?
-%elif %1 == 16	; chroma
-	; TL
-	mov_line_16x4_sse2	edi, ecx, xmm3, a	; dst, stride, xmm?
-	mov_line_16x4_sse2	edi, ecx, xmm3, a	; dst, stride, xmm?
-	mov_line_16x4_sse2	edi, ecx, xmm3, a	; dst, stride, xmm?
-	mov_line_end16x4_sse2	edi, ecx, xmm3, a	; dst, stride, xmm?
-
-	; TR
-	mov_line_16x4_sse2	ebp, ecx, xmm4, %2	; dst, stride, xmm?
-	mov_line_16x4_sse2	ebp, ecx, xmm4, %2	; dst, stride, xmm?
-	mov_line_16x4_sse2	ebp, ecx, xmm4, %2	; dst, stride, xmm?
-	mov_line_end16x4_sse2 ebp, ecx, xmm4, %2	; dst, stride, xmm?
-
-	; BL
-	mov_line_16x4_sse2	eax, ecx, xmm5, a	; dst, stride, xmm?
-	mov_line_16x4_sse2	eax, ecx, xmm5, a	; dst, stride, xmm?
-	mov_line_16x4_sse2	eax, ecx, xmm5, a	; dst, stride, xmm?
-	mov_line_end16x4_sse2	eax, ecx, xmm5, a	; dst, stride, xmm?
-
-	; BR
-	mov_line_16x4_sse2	ebx, ecx, xmm6, %2	; dst, stride, xmm?
-	mov_line_16x4_sse2	ebx, ecx, xmm6, %2	; dst, stride, xmm?
-	mov_line_16x4_sse2	ebx, ecx, xmm6, %2	; dst, stride, xmm?
-	mov_line_end16x4_sse2	ebx, ecx, xmm6, %2	; dst, stride, xmm?
-%endif
-%endmacro
-
-ALIGN 16
-;***********************************************************************----------------
-; void ExpandPictureLuma_sse2(	uint8_t *pDst,
-;									const int32_t iStride,
-;									const int32_t iWidth,
-;									const int32_t iHeight	);
-;***********************************************************************----------------
-ExpandPictureLuma_sse2:
-	push ebx
-	push edx
-	push esi
-	push edi
-	push ebp
-
-	; for both top and bottom border
-	;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
-	mov esi, [esp+24]						; p_dst
-	mov edx, [esp+28]						; stride
-	mov ebx, [esp+32]						; width
-	mov eax, [esp+36]						; height
-	; also prepare for cross border pData top-left: xmm3
-;	xor ecx, ecx
-	mov cl, byte [esi]
-	butterfly_1to16_sse xmm3, xmm4, c		; dst, tmp, pSrc [generic register name: a/b/c/d]
-	; load top border
-	mov ecx, edx							; stride
-	neg ecx 								; -stride
-	lea edi, [esi+ecx]						; last line of top border
-	; load bottom border
-	dec eax									; h-1
-	imul eax, edx 							; (h-1)*stride
-	lea eax, [esi+eax]						; last line of picture pData
-	sal edx, 05h							; 32*stride
-	lea ebp, [eax+edx]						; last line of bottom border, (h-1)*stride + 32 * stride
-	; also prepare for cross border pData: bottom-left with xmm5, bottom-right xmm6
-	dec ebx									; width-1
-	lea ebx, [eax+ebx]						; dst[w-1][h-1]
-;	xor edx, edx
-	mov dl, byte [eax]						; bottom-left
-	butterfly_1to16_sse xmm5, xmm6, d		; dst, tmp, pSrc [generic register name: a/b/c/d]
-	mov dl, byte [ebx]						; bottom-right
-	butterfly_1to16_sse xmm6, xmm4, d		; dst, tmp, pSrc [generic register name: a/b/c/d]
-	; for top & bottom expanding
-	mov ebx, [esp+32]						; width
-	exp_top_bottom_sse2	32
-
-	; for both left and right border
-	;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
-	mov esi, [esp+24]						; p_dst: left border pSrc
-	mov edx, [esp+28]						; stride
-	mov ebx, [esp+32]						; width
-	mov ecx, [esp+36]						; height
-	; load left border
-	mov eax, -32 							; luma=-32, chroma=-16
-	lea edi, [esi+eax]						; left border dst
-	dec ebx
-	lea ebx, [esi+ebx]						; right border pSrc, (p_dst + width - 1)
-	lea ebp, [ebx+1]						; right border dst
-	; prepare for cross border pData: top-right with xmm4
-;	xor eax, eax
-	mov al, byte [ebx]						; top-right
-	butterfly_1to16_sse xmm4, xmm0, a		; dst, tmp, pSrc [generic register name: a/b/c/d]
-	; for left & right border expanding
-	exp_left_right_sse2	32, a
-
-	; for cross border [top-left, top-right, bottom-left, bottom-right]
-	;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
-	mov esi, [esp+24]						; p_dst
-	mov ecx, [esp+28]						; stride
-	mov ebx, [esp+32]						; width
-	mov edx, [esp+36]						; height
-	; have done xmm3,..,xmm6 cross pData initialization above, perform pading as below, To be continued..
-	mov eax, -32							; luma=-32, chroma=-16
-	neg ecx										; -stride
-	lea edi, [esi+eax]
-	lea edi, [edi+ecx]				; last line of top-left border
-	lea ebp, [esi+ebx]
-	lea ebp, [ebp+ecx]				; last line of top-right border
-	add edx, 32								; height+32(16), luma=32, chroma=16
-	mov ecx, [esp+28]					; stride
-	imul edx, ecx							; (height+32(16)) * stride
-	lea eax, [edi+edx]						; last line of bottom-left border
-	lea ebx, [ebp+edx]						; last line of bottom-right border
-	neg ecx										; -stride
-	; for left & right border expanding
-	exp_cross_sse2		32, a
-
-;	sfence									; commit cache write back memory
-
-	pop ebp
-	pop edi
-	pop esi
-	pop edx
-	pop ebx
-
-	ret
-
-ALIGN 16
-;***********************************************************************----------------
-; void ExpandPictureChromaAlign_sse2(	uint8_t *pDst,
-;										const int32_t iStride,
-;										const int32_t iWidth,
-;										const int32_t iHeight	);
-;***********************************************************************----------------
-ExpandPictureChromaAlign_sse2:
-	push ebx
-	push edx
-	push esi
-	push edi
-	push ebp
-
-	; for both top and bottom border
-	;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
-	mov esi, [esp+24]						; p_dst
-	mov edx, [esp+28]						; stride
-	mov ebx, [esp+32]						; width
-	mov eax, [esp+36]						; height
-	; also prepare for cross border pData top-left: xmm3
-;	xor ecx, ecx
-	mov cl, byte [esi]
-	butterfly_1to16_sse xmm3, xmm4, c		; dst, tmp, pSrc [generic register name: a/b/c/d]
-	; load top border
-	mov ecx, edx							; stride
-	neg ecx 								; -stride
-	lea edi, [esi+ecx]						; last line of top border
-	; load bottom border
-	dec eax									; h-1
-	imul eax, edx 							; (h-1)*stride
-	lea eax, [esi+eax]						; last line of picture pData
-	sal edx, 04h							; 16*stride
-	lea ebp, [eax+edx]						; last line of bottom border, (h-1)*stride + 16 * stride
-	; also prepare for cross border pData: bottom-left with xmm5, bottom-right xmm6
-	dec ebx									; width-1
-	lea ebx, [eax+ebx]						; dst[w-1][h-1]
-;	xor edx, edx
-	mov dl, byte [eax]						; bottom-left
-	butterfly_1to16_sse xmm5, xmm6, d		; dst, tmp, pSrc [generic register name: a/b/c/d]
-	mov dl, byte [ebx]						; bottom-right
-	butterfly_1to16_sse xmm6, xmm4, d		; dst, tmp, pSrc [generic register name: a/b/c/d]
-	; for top & bottom expanding
-	mov ebx, [esp+32]						; width
-	exp_top_bottom_sse2	16
-
-	; for both left and right border
-	;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
-	mov esi, [esp+24]						; p_dst: left border pSrc
-	mov edx, [esp+28]						; stride
-	mov ebx, [esp+32]						; width
-	mov ecx, [esp+36]						; height
-	; load left border
-	mov eax, -16 							; luma=-32, chroma=-16
-	lea edi, [esi+eax]						; left border dst
-	dec ebx
-	lea ebx, [esi+ebx]						; right border pSrc, (p_dst + width - 1)
-	lea ebp, [ebx+1]						; right border dst
-	; prepare for cross border pData: top-right with xmm4
-;	xor eax, eax
-	mov al, byte [ebx]						; top-right
-	butterfly_1to16_sse xmm4, xmm0, a		; dst, tmp, pSrc [generic register name: a/b/c/d]
-	; for left & right border expanding
-	exp_left_right_sse2	16, a
-
-	; for cross border [top-left, top-right, bottom-left, bottom-right]
-	;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
-	mov esi, [esp+24]						; p_dst
-	mov ecx, [esp+28]						; stride
-	mov ebx, [esp+32]						; width
-	mov edx, [esp+36]						; height
-	; have done xmm3,..,xmm6 cross pData initialization above, perform pading as below, To be continued..
-	mov eax, -16							; chroma=-16
-	neg ecx										; -stride
-	lea edi, [esi+eax]
-	lea edi, [edi+ecx]				; last line of top-left border
-	lea ebp, [esi+ebx]
-	lea ebp, [ebp+ecx]				; last line of top-right border
-	mov ecx, [esp+28]						; stride
-	add edx, 16							; height+16, luma=32, chroma=16
-	imul edx, ecx							; (height+16) * stride
-	lea eax, [edi+edx]						; last line of bottom-left border
-	lea ebx, [ebp+edx]						; last line of bottom-right border
-	neg ecx										; -stride
-	; for left & right border expanding
-	exp_cross_sse2		16, a
-
-;	sfence									; commit cache write back memory
-
-	pop ebp
-	pop edi
-	pop esi
-	pop edx
-	pop ebx
-
-	ret
-
-ALIGN 16
-;***********************************************************************----------------
-; void ExpandPictureChromaUnalign_sse2(	uint8_t *pDst,
-;										const int32_t iStride,
-;										const int32_t iWidth,
-;										const int32_t iHeight	);
-;***********************************************************************----------------
-ExpandPictureChromaUnalign_sse2:
-	push ebx
-	push edx
-	push esi
-	push edi
-	push ebp
-
-	; for both top and bottom border
-	;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
-	mov esi, [esp+24]						; p_dst
-	mov edx, [esp+28]						; stride
-	mov ebx, [esp+32]						; width
-	mov eax, [esp+36]						; height
-	; also prepare for cross border pData top-left: xmm3
-;	xor ecx, ecx
-	mov cl, byte [esi]
-	butterfly_1to16_sse xmm3, xmm4, c		; dst, tmp, pSrc [generic register name: a/b/c/d]
-	; load top border
-	mov ecx, edx							; stride
-	neg ecx 								; -stride
-	lea edi, [esi+ecx]						; last line of top border
-	; load bottom border
-	dec eax									; h-1
-	imul eax, edx 							; (h-1)*stride
-	lea eax, [esi+eax]						; last line of picture pData
-	sal edx, 04h							; 16*stride
-	lea ebp, [eax+edx]						; last line of bottom border, (h-1)*stride + 16 * stride
-	; also prepare for cross border pData: bottom-left with xmm5, bottom-right xmm6
-	dec ebx									; width-1
-	lea ebx, [eax+ebx]						; dst[w-1][h-1]
-;	xor edx, edx
-	mov dl, byte [eax]						; bottom-left
-	butterfly_1to16_sse xmm5, xmm6, d		; dst, tmp, pSrc [generic register name: a/b/c/d]
-	mov dl, byte [ebx]						; bottom-right
-	butterfly_1to16_sse xmm6, xmm4, d		; dst, tmp, pSrc [generic register name: a/b/c/d]
-	; for top & bottom expanding
-	mov ebx, [esp+32]						; width
-	exp_top_bottom_sse2	16
-
-	; for both left and right border
-	;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
-	mov esi, [esp+24]						; p_dst: left border pSrc
-	mov edx, [esp+28]						; stride
-	mov ebx, [esp+32]						; width
-	mov ecx, [esp+36]						; height
-	; load left border
-	mov eax, -16 							; luma=-32, chroma=-16
-	lea edi, [esi+eax]						; left border dst
-	dec ebx
-	lea ebx, [esi+ebx]						; right border pSrc, (p_dst + width - 1)
-	lea ebp, [ebx+1]						; right border dst
-	; prepare for cross border pData: top-right with xmm4
-;	xor eax, eax
-	mov al, byte [ebx]						; top-right
-	butterfly_1to16_sse xmm4, xmm0, a		; dst, tmp, pSrc [generic register name: a/b/c/d]
-	; for left & right border expanding
-	exp_left_right_sse2	16, u
-
-	; for cross border [top-left, top-right, bottom-left, bottom-right]
-	;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
-	mov esi, [esp+24]						; p_dst
-	mov ecx, [esp+28]						; stride
-	mov ebx, [esp+32]						; width
-	mov edx, [esp+36]						; height
-	; have done xmm3,..,xmm6 cross pData initialization above, perform pading as below, To be continued..
-	neg ecx									; -stride
-	mov eax, -16							; chroma=-16
-	lea edi, [esi+eax]
-	lea edi, [edi+ecx]				; last line of top-left border
-	lea ebp, [esi+ebx]
-	lea ebp, [ebp+ecx]				; last line of top-right border
-	mov ecx, [esp+28]						; stride
-	add edx, 16							; height+16, luma=32, chroma=16
-	imul edx, ecx							; (height+16) * stride
-	lea eax, [edi+edx]						; last line of bottom-left border
-	lea ebx, [ebp+edx]						; last line of bottom-right border
-	neg ecx									; -stride
-	; for left & right border expanding
-	exp_cross_sse2		16, u
-
-;	sfence									; commit cache write back memory
-
-	pop ebp
-	pop edi
-	pop esi
-	pop edx
-	pop ebx
-
-	ret
-

codec/encoder/core/asm/intra_pred_util.asm

@@ -1,156 +0,0 @@
-;*!
-;* \copy
-;*     Copyright (c)  2009-2013, Cisco Systems
-;*     All rights reserved.
-;*
-;*     Redistribution and use in source and binary forms, with or without
-;*     modification, are permitted provided that the following conditions
-;*     are met:
-;*
-;*        * Redistributions of source code must retain the above copyright
-;*          notice, this list of conditions and the following disclaimer.
-;*
-;*        * Redistributions in binary form must reproduce the above copyright
-;*          notice, this list of conditions and the following disclaimer in
-;*          the documentation and/or other materials provided with the
-;*          distribution.
-;*
-;*     THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
-;*     "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
-;*     LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
-;*     FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
-;*     COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
-;*     INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
-;*     BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
-;*     LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
-;*     CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
-;*     LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
-;*     ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
-;*     POSSIBILITY OF SUCH DAMAGE.
-;*
-;*
-;*  intra_pred_util.asm
-;*
-;*  Abstract
-;*      mmxext/sse for WelsFillingPred8to16, WelsFillingPred8x2to16 and
-;*		WelsFillingPred1to16 etc.
-;*
-;*  History
-;*      09/29/2009 Created
-;*
-;*
-;*************************************************************************/
-
-%include "asm_inc.asm"
-
-BITS 32
-
-;***********************************************************************
-; Macros and other preprocessor constants
-;***********************************************************************
-
-;***********************************************************************
-; Local Data (Read Only)
-;***********************************************************************
-
-;SECTION .rodata pData align=16
-
-;***********************************************************************
-; Various memory constants (trigonometric values or rounding values)
-;***********************************************************************
-
-;***********************************************************************
-; Code
-;***********************************************************************
-
-
-SECTION .text
-
-WELS_EXTERN WelsFillingPred8to16_mmx
-WELS_EXTERN WelsFillingPred8x2to16_mmx
-WELS_EXTERN WelsFillingPred1to16_mmx
-WELS_EXTERN WelsFillingPred8x2to16_sse2
-WELS_EXTERN WelsFillingPred1to16_sse2
-
-
-ALIGN 16
-;***********************************************************************----------------
-; void WelsFillingPred8to16_mmx( uint8_t *pred, uint8_t *v );
-;***********************************************************************----------------
-WelsFillingPred8to16_mmx:
-	mov eax, [esp+4]	; pred
-	mov ecx, [esp+8]	; v
-
-	movq mm0, [ecx]
-	movq [eax  ], mm0
-	movq [eax+8], mm0
-
-	WELSEMMS
-	ret
-
-ALIGN 16
-;***********************************************************************----------------
-; void WelsFillingPred8x2to16_mmx( uint8_t *pred, uint8_t *v );
-;***********************************************************************----------------
-WelsFillingPred8x2to16_mmx:
-	mov eax, [esp+4]	; pred
-	mov ecx, [esp+8]	; v
-
-	movq mm0, [ecx  ]
-	movq mm1, [ecx+8]
-	movq [eax  ], mm0
-	movq [eax+8], mm1
-
-	WELSEMMS
-
-	ret
-
-%macro butterfly_1to8_mmx	3	; mm? for dst, mm? for tmp, one byte for pSrc [generic register name: a/b/c/d]
-	mov %3h, %3l
-	movd %2, e%3x		; i.e, 1% = eax (=b0)
-	pshufw %1, %2, 00h	; b0 b0 b0 b0, b0 b0 b0 b0
-%endmacro
-
-ALIGN 16
-;***********************************************************************----------------
-; void WelsFillingPred1to16_mmx( uint8_t *pred, const uint8_t v );
-;***********************************************************************----------------
-WelsFillingPred1to16_mmx:
-	mov eax, [esp+4]		; pred
-
-	mov cl, byte [esp+8]	; v
-	butterfly_1to8_mmx	mm0, mm1, c	; mm? for dst, mm? for tmp, one byte for pSrc [generic register name: a/b/c/d]
-
-	movq [eax  ], mm0
-	movq [eax+8], mm0
-
-	WELSEMMS
-
-	ret
-
-ALIGN 16
-;***********************************************************************----------------
-; void WelsFillingPred8x2to16_sse2( uint8_t *pred, uint8_t *v );
-;***********************************************************************----------------
-WelsFillingPred8x2to16_sse2:
-	mov eax, [esp+4]	; pred
-	mov ecx, [esp+8]	; v
-
-	movdqa xmm0, [ecx]
-	movdqa [eax], xmm0
-
-	ret
-
-ALIGN 16
-;***********************************************************************----------------
-; void WelsFillingPred1to16_sse2( uint8_t *pred, const uint8_t v );
-;***********************************************************************----------------
-WelsFillingPred1to16_sse2:
-	mov eax, [esp+4]		; pred
-
-	mov cl, byte [esp+8]	; v
-	butterfly_1to16_sse	xmm0, xmm1, c		; dst, tmp, pSrc [generic register name: a/b/c/d]
-
-	movdqa [eax], xmm0
-
-	ret

codec/encoder/core/asm/mb_copy.asm

@@ -1,687 +0,0 @@
-;*!
-;* \copy
-;*     Copyright (c)  2009-2013, Cisco Systems
-;*     All rights reserved.
-;*
-;*     Redistribution and use in source and binary forms, with or without
-;*     modification, are permitted provided that the following conditions
-;*     are met:
-;*
-;*        * Redistributions of source code must retain the above copyright
-;*          notice, this list of conditions and the following disclaimer.
-;*
-;*        * Redistributions in binary form must reproduce the above copyright
-;*          notice, this list of conditions and the following disclaimer in
-;*          the documentation and/or other materials provided with the
-;*          distribution.
-;*
-;*     THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
-;*     "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
-;*     LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
-;*     FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
-;*     COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
-;*     INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
-;*     BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
-;*     LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
-;*     CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
-;*     LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
-;*     ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
-;*     POSSIBILITY OF SUCH DAMAGE.
-;*
-;*
-;*  mb_copy.asm
-;*
-;*  Abstract
-;*      mb_copy
-;*
-;*
-;*********************************************************************************************/
-%include "asm_inc.asm"
-BITS 32
-
-;***********************************************************************
-; Macros and other preprocessor constants
-;***********************************************************************
-
-;***********************************************************************
-; Code
-;***********************************************************************
-
-SECTION .text
-
-WELS_EXTERN WelsCopy16x16_sse2
-WELS_EXTERN WelsCopy16x16NotAligned_sse2
-WELS_EXTERN WelsCopy8x8_mmx
-WELS_EXTERN WelsCopy16x8NotAligned_sse2	;
-WELS_EXTERN WelsCopy8x16_mmx		;
-WELS_EXTERN UpdateMbMv_sse2		;
-
-;***********************************************************************
-; void WelsCopy16x16_sse2(	uint8_t* Dst,
-;							int32_t  iStrideD,
-;							uint8_t* Src,
-;							int32_t  iStrideS )
-;***********************************************************************
-ALIGN 16
-WelsCopy16x16_sse2:
-	push esi
-	push edi
-	push ebx
-
-	mov edi, [esp+16]	; Dst
-	mov eax, [esp+20]	; iStrideD
-	mov esi, [esp+24]	; Src
-	mov ecx, [esp+28]	; iStrideS
-
-	lea ebx, [eax+2*eax]	; x3
-	lea edx, [ecx+2*ecx]	; x3
-
-	movdqa xmm0, [esi]
-	movdqa xmm1, [esi+ecx]
-	movdqa xmm2, [esi+2*ecx]
-	movdqa xmm3, [esi+edx]
-	lea esi, [esi+4*ecx]
-	movdqa xmm4, [esi]
-	movdqa xmm5, [esi+ecx]
-	movdqa xmm6, [esi+2*ecx]
-	movdqa xmm7, [esi+edx]
-	lea esi, [esi+4*ecx]
-
-	movdqa [edi], xmm0
-	movdqa [edi+eax], xmm1
-	movdqa [edi+2*eax], xmm2
-	movdqa [edi+ebx], xmm3
-	lea edi, [edi+4*eax]
-	movdqa [edi], xmm4
-	movdqa [edi+eax], xmm5
-	movdqa [edi+2*eax], xmm6
-	movdqa [edi+ebx], xmm7
-	lea edi, [edi+4*eax]
-
-	movdqa xmm0, [esi]
-	movdqa xmm1, [esi+ecx]
-	movdqa xmm2, [esi+2*ecx]
-	movdqa xmm3, [esi+edx]
-	lea esi, [esi+4*ecx]
-	movdqa xmm4, [esi]
-	movdqa xmm5, [esi+ecx]
-	movdqa xmm6, [esi+2*ecx]
-	movdqa xmm7, [esi+edx]
-
-	movdqa [edi], xmm0
-	movdqa [edi+eax], xmm1
-	movdqa [edi+2*eax], xmm2
-	movdqa [edi+ebx], xmm3
-	lea edi, [edi+4*eax]
-	movdqa [edi], xmm4
-	movdqa [edi+eax], xmm5
-	movdqa [edi+2*eax], xmm6
-	movdqa [edi+ebx], xmm7
-
-	pop ebx
-	pop edi
-	pop esi
-	ret
-
-;***********************************************************************
-; void WelsCopy16x16NotAligned_sse2(	uint8_t* Dst,
-;							int32_t  iStrideD,
-;							uint8_t* Src,
-;							int32_t  iStrideS )
-;***********************************************************************
-ALIGN 16
-; dst can be align with 16 bytes, but not sure about pSrc, 12/29/2011
-WelsCopy16x16NotAligned_sse2:
-	push esi
-	push edi
-	push ebx
-
-	mov edi, [esp+16]	; Dst
-	mov eax, [esp+20]	; iStrideD
-	mov esi, [esp+24]	; Src
-	mov ecx, [esp+28]	; iStrideS
-
-	lea ebx, [eax+2*eax]	; x3
-	lea edx, [ecx+2*ecx]	; x3
-
-	movdqu xmm0, [esi]
-	movdqu xmm1, [esi+ecx]
-	movdqu xmm2, [esi+2*ecx]
-	movdqu xmm3, [esi+edx]
-	lea esi, [esi+4*ecx]
-	movdqu xmm4, [esi]
-	movdqu xmm5, [esi+ecx]
-	movdqu xmm6, [esi+2*ecx]
-	movdqu xmm7, [esi+edx]
-	lea esi, [esi+4*ecx]
-
-	movdqa [edi], xmm0
-	movdqa [edi+eax], xmm1
-	movdqa [edi+2*eax], xmm2
-	movdqa [edi+ebx], xmm3
-	lea edi, [edi+4*eax]
-	movdqa [edi], xmm4
-	movdqa [edi+eax], xmm5
-	movdqa [edi+2*eax], xmm6
-	movdqa [edi+ebx], xmm7
-	lea edi, [edi+4*eax]
-
-	movdqu xmm0, [esi]
-	movdqu xmm1, [esi+ecx]
-	movdqu xmm2, [esi+2*ecx]
-	movdqu xmm3, [esi+edx]
-	lea esi, [esi+4*ecx]
-	movdqu xmm4, [esi]
-	movdqu xmm5, [esi+ecx]
-	movdqu xmm6, [esi+2*ecx]
-	movdqu xmm7, [esi+edx]
-
-	movdqa [edi], xmm0
-	movdqa [edi+eax], xmm1
-	movdqa [edi+2*eax], xmm2
-	movdqa [edi+ebx], xmm3
-	lea edi, [edi+4*eax]
-	movdqa [edi], xmm4
-	movdqa [edi+eax], xmm5
-	movdqa [edi+2*eax], xmm6
-	movdqa [edi+ebx], xmm7
-
-	pop ebx
-	pop edi
-	pop esi
-	ret
-
-; , 12/29/2011
-;***********************************************************************
-; void WelsCopy16x8NotAligned_sse2(uint8_t* Dst,
-;							int32_t  iStrideD,
-;							uint8_t* Src,
-;							int32_t  iStrideS )
-;***********************************************************************
-ALIGN 16
-WelsCopy16x8NotAligned_sse2:
-	push esi
-	push edi
-	push ebx
-
-	mov edi, [esp+16]	; Dst
-	mov eax, [esp+20]	; iStrideD
-	mov esi, [esp+24]	; Src
-	mov ecx, [esp+28]	; iStrideS
-
-	lea ebx, [eax+2*eax]	; x3
-	lea edx, [ecx+2*ecx]	; x3
-
-	movdqu xmm0, [esi]
-	movdqu xmm1, [esi+ecx]
-	movdqu xmm2, [esi+2*ecx]
-	movdqu xmm3, [esi+edx]
-	lea esi, [esi+4*ecx]
-	movdqu xmm4, [esi]
-	movdqu xmm5, [esi+ecx]
-	movdqu xmm6, [esi+2*ecx]
-	movdqu xmm7, [esi+edx]
-
-	movdqa [edi], xmm0
-	movdqa [edi+eax], xmm1
-	movdqa [edi+2*eax], xmm2
-	movdqa [edi+ebx], xmm3
-	lea edi, [edi+4*eax]
-	movdqa [edi], xmm4
-	movdqa [edi+eax], xmm5
-	movdqa [edi+2*eax], xmm6
-	movdqa [edi+ebx], xmm7
-
-	pop ebx
-	pop edi
-	pop esi
-	ret
-
-
-;***********************************************************************
-; void WelsCopy8x16_mmx(uint8_t* Dst,
-;                       int32_t  iStrideD,
-;                       uint8_t* Src,
-;                       int32_t  iStrideS )
-;***********************************************************************
-ALIGN 16
-WelsCopy8x16_mmx:
-	push ebx
-
-	mov eax, [esp + 8 ]           ;Dst
-	mov ecx, [esp + 12]           ;iStrideD
-	mov ebx, [esp + 16]           ;Src
-	mov edx, [esp + 20]           ;iStrideS
-
-	movq mm0, [ebx]
-	movq mm1, [ebx+edx]
-	lea ebx, [ebx+2*edx]
-	movq mm2, [ebx]
-	movq mm3, [ebx+edx]
-	lea ebx, [ebx+2*edx]
-	movq mm4, [ebx]
-	movq mm5, [ebx+edx]
-	lea ebx, [ebx+2*edx]
-	movq mm6, [ebx]
-	movq mm7, [ebx+edx]
-	lea ebx, [ebx+2*edx]
-
-	movq [eax], mm0
-	movq [eax+ecx], mm1
-	lea eax, [eax+2*ecx]
-	movq [eax], mm2
-	movq [eax+ecx], mm3
-	lea eax, [eax+2*ecx]
-	movq [eax], mm4
-	movq [eax+ecx], mm5
-	lea eax, [eax+2*ecx]
-	movq [eax], mm6
-	movq [eax+ecx], mm7
-	lea eax, [eax+2*ecx]
-
-	movq mm0, [ebx]
-	movq mm1, [ebx+edx]
-	lea ebx, [ebx+2*edx]
-	movq mm2, [ebx]
-	movq mm3, [ebx+edx]
-	lea ebx, [ebx+2*edx]
-	movq mm4, [ebx]
-	movq mm5, [ebx+edx]
-	lea ebx, [ebx+2*edx]
-	movq mm6, [ebx]
-	movq mm7, [ebx+edx]
-
-	movq [eax], mm0
-	movq [eax+ecx], mm1
-	lea eax, [eax+2*ecx]
-	movq [eax], mm2
-	movq [eax+ecx], mm3
-	lea eax, [eax+2*ecx]
-	movq [eax], mm4
-	movq [eax+ecx], mm5
-	lea eax, [eax+2*ecx]
-	movq [eax], mm6
-	movq [eax+ecx], mm7
-
-	WELSEMMS
-	pop ebx
-	ret
-
-;***********************************************************************
-; void WelsCopy8x8_mmx(  uint8_t* Dst,
-;                        int32_t  iStrideD,
-;                        uint8_t* Src,
-;                        int32_t  iStrideS )
-;***********************************************************************
-ALIGN 16
-WelsCopy8x8_mmx:
-	push ebx
-	push esi
-	mov eax, [esp + 12]           ;Dst
-	mov ecx, [esp + 16]           ;iStrideD
-	mov esi, [esp + 20]           ;Src
-	mov ebx, [esp + 24]           ;iStrideS
-	lea edx, [ebx+2*ebx]
-
-	; to prefetch next loop
-	prefetchnta [esi+2*ebx]
-	prefetchnta [esi+edx]
-	movq mm0, [esi]
-	movq mm1, [esi+ebx]
-	lea esi, [esi+2*ebx]
-	; to prefetch next loop
-	prefetchnta [esi+2*ebx]
-	prefetchnta [esi+edx]
-	movq mm2, [esi]
-	movq mm3, [esi+ebx]
-	lea esi, [esi+2*ebx]
-	; to prefetch next loop
-	prefetchnta [esi+2*ebx]
-	prefetchnta [esi+edx]
-	movq mm4, [esi]
-	movq mm5, [esi+ebx]
-	lea esi, [esi+2*ebx]
-	movq mm6, [esi]
-	movq mm7, [esi+ebx]
-
-	movq [eax], mm0
-	movq [eax+ecx], mm1
-	lea eax, [eax+2*ecx]
-	movq [eax], mm2
-	movq [eax+ecx], mm3
-	lea eax, [eax+2*ecx]
-	movq [eax], mm4
-	movq [eax+ecx], mm5
-	lea eax, [eax+2*ecx]
-	movq [eax], mm6
-	movq [eax+ecx], mm7
-
-	WELSEMMS
-	pop esi
-	pop ebx
-	ret
-
-; (dunhuang@cisco), 12/21/2011
-;***********************************************************************
-; void UpdateMbMv_sse2( SMVUnitXY *pMvBuffer, const SMVUnitXY sMv )
-;***********************************************************************
-ALIGN 16
-UpdateMbMv_sse2:
-	mov eax, [esp+4]	; mv_buffer
-	movd xmm0, [esp+8]	; _mv
-	pshufd xmm1, xmm0, $0
-	movdqa [eax     ], xmm1
-	movdqa [eax+0x10], xmm1
-	movdqa [eax+0x20], xmm1
-	movdqa [eax+0x30], xmm1
-	ret
-
-
-
-;***********************************************************************
-; Macros and other preprocessor constants
-;***********************************************************************
-
-;***********************************************************************
-; Local Data (Read Only)
-;***********************************************************************
-
-;SECTION .rodata pData align=16
-
-;***********************************************************************
-; Various memory constants (trigonometric values or rounding values)
-;***********************************************************************
-;read unaligned memory
-%macro SSE2_READ_UNA 2
-	movq	%1, [%2]
-	movhps	%1,	[%2+8]
-%endmacro
-
-;write unaligned memory
-%macro SSE2_WRITE_UNA 2
-	movq	[%1],	%2
-	movhps	[%1+8], %2
-%endmacro
-
-ALIGN 16
-
-;***********************************************************************
-; Code
-;***********************************************************************
-
-SECTION .text
-
-WELS_EXTERN PixelAvgWidthEq8_mmx
-WELS_EXTERN PixelAvgWidthEq16_sse2
-
-WELS_EXTERN McCopyWidthEq4_mmx
-WELS_EXTERN McCopyWidthEq8_mmx
-WELS_EXTERN McCopyWidthEq16_sse2
-
-
-ALIGN 16
-;***********************************************************************
-; void PixelAvgWidthEq8_mmx( uint8_t *dst,  int32_t iDstStride,
-;                           uint8_t *pSrc1, int32_t iSrc1Stride,
-;                           uint8_t *pSrc2, int32_t iSrc2Stride,
-;                           int32_t iHeight );
-;***********************************************************************
-PixelAvgWidthEq8_mmx:
-    push        ebp
-    push        ebx
-    push        esi
-    push        edi
-
-    mov         edi, [esp+20]
-    mov         esi, [esp+28]
-    mov         edx, [esp+36]
-    mov         ebp, [esp+24]
-    mov         eax, [esp+32]
-    mov         ebx, [esp+40]
-    mov         ecx, [esp+44]
-	sar			ecx, 2
-.height_loop:
-	movq        mm0, [esi]
-    pavgb       mm0, [edx]
-    movq        [edi], mm0
-	movq		mm1, [esi+eax]
-	pavgb		mm1, [edx+ebx]
-	movq		[edi+ebp], mm1
-	lea         edi, [edi+2*ebp]
-	lea         esi, [esi+2*eax]
-	lea         edx, [edx+2*ebx]
-
-	movq        mm2, [esi]
-	pavgb       mm2, [edx]
-    movq        [edi], mm2
-	movq		mm3, [esi+eax]
-	pavgb		mm3, [edx+ebx]
-	movq		[edi+ebp], mm3
-	lea         edi, [edi+2*ebp]
-	lea         esi, [esi+2*eax]
-	lea         edx, [edx+2*ebx]
-
-	dec         ecx
-    jne         .height_loop
-
-	WELSEMMS
-    pop         edi
-    pop         esi
-    pop         ebx
-    pop         ebp
-    ret
-
-
-ALIGN 16
-;***********************************************************************
-; void PixelAvgWidthEq16_sse2( uint8_t *dst,  int32_t iDstStride,
-;                          uint8_t *pSrc1, int32_t iSrc1Stride,
-;                          uint8_t *pSrc2, int32_t iSrc2Stride,
-;                          int32_t iHeight );
-;***********************************************************************
-PixelAvgWidthEq16_sse2:
-	push        ebp
-    push        ebx
-    push        esi
-    push        edi
-
-    mov         edi, [esp+20]
-    mov         esi, [esp+28]
-    mov         edx, [esp+36]
-    mov         ebp, [esp+24]
-    mov         eax, [esp+32]
-    mov         ebx, [esp+40]
-    mov         ecx, [esp+44]
-	sar			ecx, 2
-.height_loop:
-	movdqu      xmm0, [esi]
-	movdqu      xmm1, [edx]
-	movdqu      xmm2, [esi+eax]
-	movdqu      xmm3, [edx+ebx]
-	pavgb       xmm0, xmm1
-	pavgb       xmm2, xmm3
-	movdqu      [edi], xmm0
-	movdqu      [edi+ebp], xmm2
-	lea			edi, [edi+2*ebp]
-	lea			esi, [esi+2*eax]
-	lea			edx, [edx+2*ebx]
-
-	movdqu      xmm4, [esi]
-	movdqu      xmm5, [edx]
-	movdqu      xmm6, [esi+eax]
-	movdqu      xmm7, [edx+ebx]
-	pavgb       xmm4, xmm5
-	pavgb       xmm6, xmm7
-	movdqu      [edi], xmm4
-	movdqu      [edi+ebp], xmm6
-	lea         edi, [edi+2*ebp]
-	lea         esi, [esi+2*eax]
-    lea         edx, [edx+2*ebx]
-
-	dec         ecx
-	jne         .height_loop
-
-    pop         edi
-    pop         esi
-    pop         ebx
-    pop         ebp
-    ret
-
-
-ALIGN 64
-avg_w16_align_0_ssse3:
-    movdqa  xmm1, [ebx]
-    movdqu  xmm2, [ecx]
-    pavgb   xmm1, xmm2
-    movdqa  [edi], xmm1
-    add    ebx, eax
-    add    ecx, ebp
-    add    edi, esi
-    dec    dword [esp+4]
-    jg     avg_w16_align_0_ssse3
-    ret
-
-    ALIGN 64
-avg_w16_align_1_ssse3:
-    movdqa  xmm1, [ebx+16]
-    movdqu  xmm2, [ecx]
-    palignr xmm1, [ebx], 1
-    pavgb   xmm1, xmm2
-    movdqa  [edi], xmm1
-    add    ebx, eax
-    add    ecx, ebp
-    add    edi, esi
-    dec    dword [esp+4]
-    jg     avg_w16_align_1_ssse3
-    ret
-
-
-ALIGN 16
-;***********************************************************************
-; void PixelAvgWidthEq16_ssse3(uint8_t *pDst,  int32_t iDstStride,
-;                          uint8_t *pSrc1, int32_t iSrc1Stride,
-;                          uint8_t *pSrc2, int32_t iSrc2Stride,
-;                          int32_t iHeight );
-;***********************************************************************
-WELS_EXTERN PixelAvgWidthEq16_ssse3
-PixelAvgWidthEq16_ssse3:
-    push        ebp
-    push        ebx
-    push        esi
-    push        edi
-
-    mov         edi, [esp+20]       ; dst
-    mov         ebx, [esp+28]       ; src1
-    mov         ecx, [esp+36]       ; src2
-    mov         esi, [esp+24]       ; i_dst_stride
-
-     %define avg_w16_offset (avg_w16_align_1_ssse3-avg_w16_align_0_ssse3)
-    mov edx, ebx
-    and edx, 0x01
-    lea eax, [avg_w16_align_0_ssse3]
-    lea ebp, [avg_w16_offset]
-    imul ebp, edx
-    lea edx, [ebp+eax]
-
-    mov eax, [esp+32]
-    mov ebp, [esp+44]
-    push ebp
-    mov ebp, [esp+44]
-    and ebx, 0xfffffff0
-    call edx
-	pop		   ebp
-    pop         edi
-    pop         esi
-    pop         ebx
-    pop         ebp
-    ret
-
-
-ALIGN 16
-;*******************************************************************************
-;  void McCopyWidthEq4_mmx( uint8_t *pSrc, int32_t iSrcStride,
-;                          uint8_t *pDst, int32_t iDstStride, int32_t iHeight )
-;*******************************************************************************
-McCopyWidthEq4_mmx:
-    push    esi
-    push    edi
-    push    ebx
-
-
-    mov esi,  [esp+16]
-    mov eax, [esp+20]
-    mov edi,  [esp+24]
-    mov ecx,  [esp+28]
-    mov edx,  [esp+32]
-ALIGN 4
-.height_loop:
-	mov ebx, [esi]
-	mov [edi], ebx
-
-	add esi, eax
-	add edi, ecx
-	dec edx
-	jnz .height_loop
-	WELSEMMS
-	pop	   ebx
-    pop     edi
-    pop     esi
-    ret
-
-ALIGN 16
-;*******************************************************************************
-;   void McCopyWidthEq8_mmx( uint8_t *pSrc, int32_t iSrcStride,
-;                           uint8_t *pDst, int32_t iDstStride, int32_t iHeight )
-;*******************************************************************************
-McCopyWidthEq8_mmx:
-    push  esi
-    push  edi
-	mov  esi, [esp+12]
-	mov eax, [esp+16]
-	mov edi, [esp+20]
-	mov ecx, [esp+24]
-	mov edx, [esp+28]
-
-ALIGN 4
-.height_loop:
-	movq mm0, [esi]
-	movq [edi], mm0
-	add esi, eax
-	add edi, ecx
-	dec edx
-	jnz .height_loop
-
-	WELSEMMS
-    pop     edi
-    pop     esi
-    ret
-
-ALIGN 16
-;***********************************************************************
-;   void McCopyWidthEq16_sse2( uint8_t *pSrc, int32_t iSrcStride, uint8_t *pDst, int32_t iDstStride, int32_t iHeight )
-;***********************************************************************
-McCopyWidthEq16_sse2:
-    push    esi
-    push    edi
-
-    mov     esi, [esp+12]
-    mov     eax, [esp+16]
-    mov     edi, [esp+20]
-    mov     edx, [esp+24]
-    mov     ecx, [esp+28]
-
-ALIGN 4
-.height_loop:
-    SSE2_READ_UNA	xmm0, esi
-    SSE2_READ_UNA	xmm1, esi+eax
-    SSE2_WRITE_UNA	edi, xmm0
-    SSE2_WRITE_UNA	edi+edx, xmm1
-
-	sub		ecx,	2
-    lea     esi, [esi+eax*2]
-    lea     edi, [edi+edx*2]
-    jnz     .height_loop
-
-    pop     edi
-    pop     esi
-    ret

codec/encoder/core/asm/memzero.asm

@@ -40,8 +40,6 @@
 ;*
 ;*************************************************************************/
 
-BITS 32
-
 %include "asm_inc.asm"
 ;***********************************************************************
 ; Code
@@ -55,8 +53,10 @@ ALIGN 16
 ;***********************************************************************
 WELS_EXTERN WelsPrefetchZero_mmx
 WelsPrefetchZero_mmx:
-	mov  eax,[esp+4]
+	%assign  push_num 0
-	prefetchnta [eax]
+	LOAD_1_PARA
+	;mov  eax,[esp+4]
+	prefetchnta [r0]
 	ret 			
 
 
@@ -66,19 +66,21 @@ ALIGN 16
 ;***********************************************************************
 WELS_EXTERN WelsSetMemZeroAligned64_sse2
 WelsSetMemZeroAligned64_sse2:
-		mov		eax,	[esp + 4]          ; dst
+
-		mov		ecx,	[esp + 8]
+		%assign  push_num 0
-		neg		ecx
+		LOAD_2_PARA
+		SIGN_EXTENTION r1, r1d
+		neg		r1
 			
 		pxor	xmm0,		xmm0
 .memzeroa64_sse2_loops:
-		movdqa	[eax],		xmm0
+		movdqa	[r0],		xmm0
-		movdqa	[eax+16],	xmm0
+		movdqa	[r0+16],	xmm0
-		movdqa	[eax+32],	xmm0
+		movdqa	[r0+32],	xmm0
-		movdqa	[eax+48],	xmm0
+		movdqa	[r0+48],	xmm0
-		add		eax, 0x40
+		add		r0, 0x40
 		
-		add ecx, 0x40
+		add r1, 0x40
 		jnz near .memzeroa64_sse2_loops
 			
 		ret	
@@ -89,23 +91,25 @@ ALIGN 16
 ;***********************************************************************
 WELS_EXTERN WelsSetMemZeroSize64_mmx
 WelsSetMemZeroSize64_mmx:
-		mov		eax,	[esp + 4]          ; dst
+
-		mov		ecx,	[esp + 8]
+		%assign  push_num 0
-		neg		ecx
+		LOAD_2_PARA
+		SIGN_EXTENTION r1, r1d
+		neg		r1
 			
 		pxor	mm0,		mm0
 .memzero64_mmx_loops:
-		movq	[eax],		mm0
+		movq	[r0],		mm0
-		movq	[eax+8],	mm0
+		movq	[r0+8],	mm0
-		movq	[eax+16],	mm0
+		movq	[r0+16],	mm0
-		movq	[eax+24],	mm0
+		movq	[r0+24],	mm0
-		movq	[eax+32],	mm0
+		movq	[r0+32],	mm0
-		movq	[eax+40],	mm0
+		movq	[r0+40],	mm0
-		movq	[eax+48],	mm0
+		movq	[r0+48],	mm0
-		movq	[eax+56],	mm0
+		movq	[r0+56],	mm0		
-		add		eax,		0x40
+		add		r0,		0x40
 		
-		add ecx, 0x40
+		add r1, 0x40
 		jnz near .memzero64_mmx_loops
 			
 		WELSEMMS	
@@ -117,16 +121,18 @@ ALIGN 16
 ;***********************************************************************
 WELS_EXTERN WelsSetMemZeroSize8_mmx
 WelsSetMemZeroSize8_mmx:
-		mov		eax,	[esp + 4]		; dst
+
-		mov		ecx,	[esp + 8]		; size
+		%assign  push_num 0
-		neg		ecx
+		LOAD_2_PARA
+		SIGN_EXTENTION r1, r1d
+		neg		r1			
 		pxor	mm0,		mm0
 		
 .memzero8_mmx_loops:
-		movq	[eax],		mm0
+		movq	[r0],		mm0
-		add		eax,		0x08
+		add		r0,		0x08
 	
-		add		ecx,		0x08
+		add		r1,		0x08
 		jnz near .memzero8_mmx_loops
 		
 		WELSEMMS	

codec/encoder/core/asm/quant.asm

@@ -42,7 +42,6 @@
 
 %include "asm_inc.asm"
 
-BITS 32
 
 SECTION .text
 ;************************************************
@@ -86,14 +85,16 @@ SECTION .text
 WELS_EXTERN WelsQuant4x4_sse2
 align 16
 WelsQuant4x4_sse2:
-		mov		eax,  [ff]
+		%assign push_num 0
-		mov		ecx,  [mf]
+                LOAD_3_PARA
-		MOVDQ	xmm2, [eax]
+		;mov		eax,  [ff]
-		MOVDQ	xmm3, [ecx]
+		;mov		ecx,  [mf]
+		movdqa	xmm2, [r1]
+		movdqa	xmm3, [r2]
 
-		mov		edx,  [pDct]
+		;mov		edx,  [pDct]
-		SSE2_Quant8	xmm0, xmm1, xmm2, xmm3, [edx]
+		SSE2_Quant8	xmm0, xmm1, xmm2, xmm3, [r0]
-		SSE2_Quant8	xmm0, xmm1, xmm2, xmm3, [edx + 0x10]
+		SSE2_Quant8	xmm0, xmm1, xmm2, xmm3, [r0 + 0x10]
 
 		ret
 
@@ -103,15 +104,21 @@ WelsQuant4x4_sse2:
 WELS_EXTERN WelsQuant4x4Dc_sse2
 align 16
 WelsQuant4x4Dc_sse2:
-		mov		ax,		[mf]
+ 		%assign push_num 0
-		SSE2_Copy8Times xmm3, eax
+		LOAD_3_PARA
+		%ifndef X86_32
+		movsx r1, r1w
+		movsx r2, r2w
+		%endif
+		;mov		ax,		[mf]
+		SSE2_Copy8Times xmm3, r2d
 
-		mov		cx, [ff]
+		;mov		cx, [ff]
-		SSE2_Copy8Times xmm2, ecx
+		SSE2_Copy8Times xmm2, r1d
 
-		mov		edx,  [pDct]
+		;mov		edx,  [pDct]
-		SSE2_Quant8	xmm0, xmm1, xmm2, xmm3, [edx]
+		SSE2_Quant8	xmm0, xmm1, xmm2, xmm3, [r0]
-		SSE2_Quant8	xmm0, xmm1, xmm2, xmm3, [edx + 0x10]
+		SSE2_Quant8	xmm0, xmm1, xmm2, xmm3, [r0 + 0x10]
 
 		ret
 
@@ -121,20 +128,22 @@ WelsQuant4x4Dc_sse2:
 WELS_EXTERN WelsQuantFour4x4_sse2
 align 16
 WelsQuantFour4x4_sse2:
-		mov		eax,  [ff]
+		%assign push_num 0
-		mov		ecx,  [mf]
+		LOAD_3_PARA
-		MOVDQ	xmm2, [eax]
+		;mov		eax,  [ff]
-		MOVDQ	xmm3, [ecx]
+		;mov		ecx,  [mf]
+		MOVDQ	xmm2, [r1]
+		MOVDQ	xmm3, [r2]
 
-		mov		edx,  [pDct]
+		;mov		edx,  [pDct]
-		SSE2_Quant8	xmm0, xmm1, xmm2, xmm3, [edx]
+		SSE2_Quant8	xmm0, xmm1, xmm2, xmm3, [r0]
-		SSE2_Quant8	xmm0, xmm1, xmm2, xmm3, [edx + 0x10]
+		SSE2_Quant8	xmm0, xmm1, xmm2, xmm3, [r0 + 0x10]
-		SSE2_Quant8	xmm0, xmm1, xmm2, xmm3, [edx + 0x20]
+		SSE2_Quant8	xmm0, xmm1, xmm2, xmm3, [r0 + 0x20]
-		SSE2_Quant8	xmm0, xmm1, xmm2, xmm3, [edx + 0x30]
+		SSE2_Quant8	xmm0, xmm1, xmm2, xmm3, [r0 + 0x30]
-		SSE2_Quant8	xmm0, xmm1, xmm2, xmm3, [edx + 0x40]
+		SSE2_Quant8	xmm0, xmm1, xmm2, xmm3, [r0 + 0x40]
-		SSE2_Quant8	xmm0, xmm1, xmm2, xmm3, [edx + 0x50]
+		SSE2_Quant8	xmm0, xmm1, xmm2, xmm3, [r0 + 0x50]
-		SSE2_Quant8	xmm0, xmm1, xmm2, xmm3, [edx + 0x60]
+		SSE2_Quant8	xmm0, xmm1, xmm2, xmm3, [r0 + 0x60]
-		SSE2_Quant8	xmm0, xmm1, xmm2, xmm3, [edx + 0x70]
+		SSE2_Quant8	xmm0, xmm1, xmm2, xmm3, [r0 + 0x70]
 
 		ret
 
@@ -144,24 +153,26 @@ WelsQuantFour4x4_sse2:
 WELS_EXTERN WelsQuantFour4x4Max_sse2
 align 16
 WelsQuantFour4x4Max_sse2:
-		mov		eax,  [ff]
+		%assign push_num 0
-		mov		ecx,  [mf]
+		LOAD_4_PARA
-		MOVDQ	xmm2, [eax]
+		;mov		eax,  [ff]
-		MOVDQ	xmm3, [ecx]
+		;mov		ecx,  [mf]
+		MOVDQ	xmm2, [r1]
+		MOVDQ	xmm3, [r2]
 
-		mov		edx,  [pDct]
+		;mov		edx,  [pDct]
 		pxor	xmm4, xmm4
 		pxor	xmm5, xmm5
 		pxor	xmm6, xmm6
 		pxor	xmm7, xmm7
-		SSE2_QuantMax8	xmm0, xmm1, xmm2, xmm3, [edx	   ], xmm4
+		SSE2_QuantMax8	xmm0, xmm1, xmm2, xmm3, [r0	  ], xmm4
-		SSE2_QuantMax8	xmm0, xmm1, xmm2, xmm3, [edx + 0x10], xmm4
+		SSE2_QuantMax8	xmm0, xmm1, xmm2, xmm3, [r0 + 0x10], xmm4
-		SSE2_QuantMax8	xmm0, xmm1, xmm2, xmm3, [edx + 0x20], xmm5
+		SSE2_QuantMax8	xmm0, xmm1, xmm2, xmm3, [r0 + 0x20], xmm5
-		SSE2_QuantMax8	xmm0, xmm1, xmm2, xmm3, [edx + 0x30], xmm5
+		SSE2_QuantMax8	xmm0, xmm1, xmm2, xmm3, [r0 + 0x30], xmm5
-		SSE2_QuantMax8	xmm0, xmm1, xmm2, xmm3, [edx + 0x40], xmm6
+		SSE2_QuantMax8	xmm0, xmm1, xmm2, xmm3, [r0 + 0x40], xmm6
-		SSE2_QuantMax8	xmm0, xmm1, xmm2, xmm3, [edx + 0x50], xmm6
+		SSE2_QuantMax8	xmm0, xmm1, xmm2, xmm3, [r0 + 0x50], xmm6
-		SSE2_QuantMax8	xmm0, xmm1, xmm2, xmm3, [edx + 0x60], xmm7
+		SSE2_QuantMax8	xmm0, xmm1, xmm2, xmm3, [r0 + 0x60], xmm7
-		SSE2_QuantMax8	xmm0, xmm1, xmm2, xmm3, [edx + 0x70], xmm7
+		SSE2_QuantMax8	xmm0, xmm1, xmm2, xmm3, [r0 + 0x70], xmm7
 
 		SSE2_TransTwo4x4W xmm4, xmm5, xmm6, xmm7, xmm0
 		pmaxsw  xmm0,  xmm4
@@ -171,9 +182,9 @@ WelsQuantFour4x4Max_sse2:
 		punpckhqdq	xmm0, xmm1
 		pmaxsw	xmm0, xmm1
 
-		mov		edx,  [max]
+		;mov		r0,  [r3]
-		movq	[edx], xmm0
+		movq	[r3], xmm0
-
+		LOAD_4_PARA_POP
 		ret
 
 %macro  MMX_Copy4Times 2
@@ -203,21 +214,20 @@ SECTION .text
 WELS_EXTERN WelsHadamardQuant2x2_mmx
 align 16
 WelsHadamardQuant2x2_mmx:
-
+		%assign push_num 0
-		mov			eax,			[pDct]
+		LOAD_5_PARA
-		movd		mm0,			[eax]
+		%ifndef X86_32
-		movd		mm1,			[eax + 0x20]
+		movsx r1, r1w
+		movsx r2, r2w
+		%endif
+		;mov			eax,			[pDct]
+		movd		mm0,			[r0]
+		movd		mm1,			[r0 + 0x20]
 		punpcklwd	mm0,			mm1
-		movd		mm3,			[eax + 0x40]
+		movd		mm3,			[r0 + 0x40]
-		movd		mm1,			[eax + 0x60]
+		movd		mm1,			[r0 + 0x60]
 		punpcklwd	mm3,			mm1
 
-		mov			cx,				0
-		mov			[eax],			cx
-		mov			[eax + 0x20],	cx
-		mov			[eax + 0x40],	cx
-		mov			[eax + 0x60],	cx
-
 		;hdm_2x2,	mm0 = dct0 dct1, mm3 = dct2 dct3
 		movq		mm5,			mm3
 		paddw		mm3,			mm0
@@ -231,17 +241,17 @@ WelsHadamardQuant2x2_mmx:
 		punpcklwd	mm1,			mm3
 
 		;quant_2x2_dc
-		mov			ax,				[mf]
+		;mov			ax,				[mf]
-		MMX_Copy4Times	mm3,		eax
+		MMX_Copy4Times	mm3,		r2d
-		mov			cx,				[ff]
+		;mov			cx,				[ff]
-		MMX_Copy4Times	mm2,		ecx
+		MMX_Copy4Times	mm2,		r1d
 		MMX_Quant4		mm1,	mm0,	mm2,	mm3
 
 		; store dct_2x2
-		mov			edx,			[dct2x2]
+		;mov			edx,			[dct2x2]
-		movq		[edx],			mm1
+		movq		[r3],			mm1
-		mov			ecx,			[iChromaDc]
+		;mov			ecx,			[iChromaDc]
-		movq		[ecx],			mm1
+		movq		[r4],			mm1
 
 		; pNonZeroCount of dct_2x2
 		pcmpeqb		mm2,			mm2		; mm2 = FF
@@ -250,9 +260,17 @@ WelsHadamardQuant2x2_mmx:
 		pcmpeqb		mm1,			mm3		; set FF if equal, 0 if not equal
 		psubsb		mm1,			mm2		; set 0 if equal, 1 if not equal
 		psadbw		mm1,			mm3		;
-		movd		eax,			mm1
+		mov			r1w,				0
+		mov			[r0],			r1w
+		mov			[r0 + 0x20],	r1w
+		mov			[r0 + 0x40],	r1w
+		mov			[r0 + 0x60],	r1w
+
+
+		movd		retrd,		mm1
 
 		WELSEMMS
+		LOAD_5_PARA_POP
 		ret
 
 ;***********************************************************************
@@ -261,13 +279,18 @@ WelsHadamardQuant2x2_mmx:
 WELS_EXTERN WelsHadamardQuant2x2Skip_mmx
 align 16
 WelsHadamardQuant2x2Skip_mmx:
-
+		%assign push_num 0
-		mov			eax,			[pDct]
+		LOAD_3_PARA
-		movd		mm0,			[eax]
+		%ifndef X86_32
-		movd		mm1,			[eax + 0x20]
+		movsx r1, r1w
+		movsx r2, r2w
+		%endif
+		;mov			eax,			[pDct]
+		movd		mm0,			[r0]
+		movd		mm1,			[r0 + 0x20]
 		punpcklwd	mm0,			mm1
-		movd		mm3,			[eax + 0x40]
+		movd		mm3,			[r0 + 0x40]
-		movd		mm1,			[eax + 0x60]
+		movd		mm1,			[r0 + 0x60]
 		punpcklwd	mm3,			mm1
 
 		;hdm_2x2,	mm0 = dct0 dct1, mm3 = dct2 dct3
@@ -283,10 +306,10 @@ WelsHadamardQuant2x2Skip_mmx:
 		punpcklwd	mm1,			mm3
 
 		;quant_2x2_dc
-		mov			ax,				[mf]
+		;mov			ax,				[mf]
-		MMX_Copy4Times	mm3,		eax
+		MMX_Copy4Times	mm3,		r2d
-		mov			cx,				[ff]
+		;mov			cx,				[ff]
-		MMX_Copy4Times	mm2,		ecx
+		MMX_Copy4Times	mm2,		r1d
 		MMX_Quant4		mm1,	mm0,	mm2,	mm3
 
 		; pNonZeroCount of dct_2x2
@@ -296,7 +319,7 @@ WelsHadamardQuant2x2Skip_mmx:
 		pcmpeqb		mm1,			mm3		; set FF if equal, 0 if not equal
 		psubsb		mm1,			mm2		; set 0 if equal, 1 if not equal
 		psadbw		mm1,			mm3		;
-		movd		eax,			mm1
+		movd		retrd,			mm1
 
 		WELSEMMS
 		ret
@@ -317,12 +340,14 @@ align 16
 WELS_EXTERN WelsDequant4x4_sse2
 WelsDequant4x4_sse2:
 	;ecx = dequant_mf[qp], edx = pDct
-	mov		ecx,  [esp + 8]
+	%assign push_num 0
-	mov		edx,  [esp + 4]
+	LOAD_2_PARA
+	;mov		ecx,  [esp + 8]
+	;mov		edx,  [esp + 4]
 
-	movdqa  xmm1, [ecx]
+	movdqa  xmm1, [r1]
-	SSE2_DeQuant8 [edx		],  xmm0, xmm1
+	SSE2_DeQuant8 [r0	],  xmm0, xmm1
-	SSE2_DeQuant8 [edx+0x10	],  xmm0, xmm1
+	SSE2_DeQuant8 [r0 + 0x10],  xmm0, xmm1
 
     ret
 
@@ -335,18 +360,20 @@ align 16
 WELS_EXTERN WelsDequantFour4x4_sse2
 WelsDequantFour4x4_sse2:
     ;ecx = dequant_mf[qp], edx = pDct
-	mov		ecx,  [esp + 8]
+	%assign push_num 0
-	mov		edx,  [esp + 4]
+	LOAD_2_PARA
+	;mov		ecx,  [esp + 8]
+	;mov		edx,  [esp + 4]
 
-	movdqa  xmm1, [ecx]
+	movdqa  xmm1, [r1]
-	SSE2_DeQuant8 [edx		],  xmm0, xmm1
+	SSE2_DeQuant8 [r0	],  xmm0, xmm1
-	SSE2_DeQuant8 [edx+0x10	],  xmm0, xmm1
+	SSE2_DeQuant8 [r0+0x10	],  xmm0, xmm1
-	SSE2_DeQuant8 [edx+0x20	],  xmm0, xmm1
+	SSE2_DeQuant8 [r0+0x20	],  xmm0, xmm1
-	SSE2_DeQuant8 [edx+0x30	],  xmm0, xmm1
+	SSE2_DeQuant8 [r0+0x30	],  xmm0, xmm1
-	SSE2_DeQuant8 [edx+0x40	],  xmm0, xmm1
+	SSE2_DeQuant8 [r0+0x40	],  xmm0, xmm1
-	SSE2_DeQuant8 [edx+0x50	],  xmm0, xmm1
+	SSE2_DeQuant8 [r0+0x50	],  xmm0, xmm1
-	SSE2_DeQuant8 [edx+0x60	],  xmm0, xmm1
+	SSE2_DeQuant8 [r0+0x60	],  xmm0, xmm1
-	SSE2_DeQuant8 [edx+0x70	],  xmm0, xmm1
+	SSE2_DeQuant8 [r0+0x70	],  xmm0, xmm1
 
     ret
 
@@ -356,14 +383,19 @@ WelsDequantFour4x4_sse2:
 WELS_EXTERN WelsDequantIHadamard4x4_sse2
 align 16
 WelsDequantIHadamard4x4_sse2:
-		mov			eax,			[esp + 4]
+		%assign push_num 0
-		mov			cx,				[esp + 8]
+		LOAD_2_PARA
+		%ifndef X86_32
+		movzx r1, r1w
+		%endif
+		;mov			eax,			[esp + 4]
+		;mov			cx,				[esp + 8]
 
 		; WelsDequantLumaDc4x4
-		SSE2_Copy8Times	xmm1,		ecx
+		SSE2_Copy8Times	xmm1,		r1d
 		;psrlw		xmm1,		2		; for the (>>2) in ihdm
-		MOVDQ		xmm0,		[eax]
+		MOVDQ		xmm0,		[r0]
-		MOVDQ		xmm2,		[eax+0x10]
+		MOVDQ		xmm2,		[r0+0x10]
 		pmullw		xmm0,		xmm1
 		pmullw		xmm2,		xmm1
 
@@ -386,8 +418,8 @@ WelsDequantIHadamard4x4_sse2:
 		SSE2_TransTwo4x4W	xmm0, xmm1, xmm4, xmm2, xmm3
 
 		punpcklqdq	xmm0,		xmm1
-		MOVDQ		[eax],		xmm0
+		MOVDQ		[r0],		xmm0
 
 		punpcklqdq	xmm2,		xmm3
-		MOVDQ		[eax+16],	xmm2
+		MOVDQ		[r0+16],	xmm2
 		ret

codec/encoder/core/asm/score.asm

@@ -42,8 +42,6 @@
 
 %include "asm_inc.asm"
 
-bits 32
-
 ;***********************************************************************
 ; Macros
 ;***********************************************************************
@@ -171,25 +169,34 @@ SECTION .text
 ALIGN 16
 WELS_EXTERN WelsScan4x4DcAc_sse2
 WelsScan4x4DcAc_sse2:
-
+	%ifdef X86_32
-	mov        eax, [esp+8]
+	push r3
-	movdqa     xmm0, [eax]			; 7 6 5 4 3 2 1 0
+	%assign push_num 1
-	movdqa     xmm1, [eax+16]		; f e d c b a 9 8
+	%else
-	pextrw     ecx, xmm0, 7			; ecx = 7
+	%assign push_num 0
-	pextrw     edx, xmm1, 2			; edx = a
+	%endif
-	pextrw     eax, xmm0, 5			; eax = 5
+	LOAD_2_PARA
-	pinsrw     xmm1, ecx, 2			; f e d c b 7 9 8
+	;mov        eax, [esp+8]
-	pinsrw     xmm0, eax, 7			; 5 6 5 4 3 2 1 0
+	movdqa     xmm0, [r1]			; 7 6 5 4 3 2 1 0
-	pextrw     ecx, xmm1, 0			; ecx = 8
+	movdqa     xmm1, [r1+16]		; f e d c b a 9 8
-	pinsrw     xmm0, ecx, 5			; 5 6 8 4 3 2 1 0
+	pextrw     r2d, xmm0, 7			; ecx = 7
-	pinsrw     xmm1, edx, 0			; f e d c b 7 9 a
+	pextrw     r3d, xmm1, 2			; edx = a
+	pextrw     r1d, xmm0, 5			; eax = 5
+	pinsrw     xmm1, r2d, 2			; f e d c b 7 9 8
+	pinsrw     xmm0, r1d, 7			; 5 6 5 4 3 2 1 0
+	pextrw     r2d, xmm1, 0			; ecx = 8
+	pinsrw     xmm0, r2d, 5			; 5 6 8 4 3 2 1 0
+	pinsrw     xmm1, r3d, 0			; f e d c b 7 9 a
 	pshufd     xmm2, xmm0, 0xd8		; 5 6 3 2 8 4 1 0
 	pshufd     xmm3, xmm1, 0xd8		; f e b 7 d c 9 a
 	pshufhw    xmm0, xmm2, 0x93		; 6 3 2 5 8 4 1 0
 	pshuflw    xmm1, xmm3, 0x39		; f e b 7 a d c 9
-	mov        eax,  [esp+4]
+	;mov        eax,  [esp+4]
-	movdqa     [eax],xmm0
+	movdqa     [r0],xmm0
-	movdqa     [eax+16], xmm1
+	movdqa     [r0+16], xmm1
+	%ifdef X86_32
+	pop r3
+	%endif
 	ret
 
 ;***********************************************************************
@@ -198,19 +205,21 @@ WelsScan4x4DcAc_sse2:
 ALIGN 16
 WELS_EXTERN WelsScan4x4DcAc_ssse3
 WelsScan4x4DcAc_ssse3:
-	mov        eax, [esp+8]
+	%assign push_num 0
-	movdqa     xmm0, [eax]
+	LOAD_2_PARA
-	movdqa     xmm1, [eax+16]
+	;mov        eax, [esp+8]
-	pextrw		ecx,  xmm0, 7			; ecx = [7]
+	movdqa     xmm0, [r1]
-	pextrw		eax,  xmm1, 0			; eax = [8]
+	movdqa     xmm1, [r1+16]
-	pinsrw		xmm0, eax, 7			; xmm0[7]	=	[8]
+	pextrw		r2d,  xmm0, 7			; ecx = [7]
-	pinsrw		xmm1, ecx, 0			; xmm1[0]	=	[7]
+	pextrw		r1d,  xmm1, 0			; eax = [8]
+	pinsrw		xmm0, r1d, 7			; xmm0[7]	=	[8]
+	pinsrw		xmm1, r2d, 0			; xmm1[0]	=	[7]
 	pshufb		xmm1, [pb_scanacdc_maskb]
 	pshufb		xmm0, [pb_scanacdc_maska]
 
-	mov        eax,  [esp+4]
+	;mov        eax,  [esp+4]
-	movdqa     [eax],xmm0
+	movdqa     [r0],xmm0
-	movdqa     [eax+16], xmm1
+	movdqa     [r0+16], xmm1
 	ret
 ;***********************************************************************
 ;void WelsScan4x4Ac_sse2( int16_t* zig_value, int16_t* pDct )
@@ -218,9 +227,11 @@ WelsScan4x4DcAc_ssse3:
 ALIGN 16
 WELS_EXTERN WelsScan4x4Ac_sse2
 WelsScan4x4Ac_sse2:
-	mov        eax, [esp+8]
+	%assign push_num 0
-	movdqa     xmm0, [eax]
+	LOAD_2_PARA
-	movdqa     xmm1, [eax+16]
+	;mov        eax, [esp+8]
+	movdqa     xmm0, [r1]
+	movdqa     xmm1, [r1+16]
 	movdqa     xmm2, xmm0
 	punpcklqdq xmm0, xmm1
 	punpckhqdq xmm2, xmm1
@@ -228,14 +239,14 @@ WelsScan4x4Ac_sse2:
 	movdqa     xmm3, xmm0
 	punpckldq  xmm0, xmm2
 	punpckhdq  xmm3, xmm2
-	pextrw     eax , xmm0, 3
+	pextrw     r1d , xmm0, 3
-	pextrw     edx , xmm0, 7
+	pextrw     r2d , xmm0, 7
-	pinsrw     xmm0, eax,  7
+	pinsrw     xmm0, r1d,  7
-	pextrw     eax,  xmm3, 4
+	pextrw     r1d,  xmm3, 4
-	pinsrw     xmm3, edx,  4
+	pinsrw     xmm3, r2d,  4
-	pextrw     edx,  xmm3, 0
+	pextrw     r2d,  xmm3, 0
-	pinsrw     xmm3, eax,  0
+	pinsrw     xmm3, r1d,  0
-	pinsrw     xmm0, edx,  3
+	pinsrw     xmm0, r2d,  3
 
 	pshufhw    xmm1, xmm0, 0x93
 	pshuflw    xmm2, xmm3, 0x39
@@ -245,9 +256,9 @@ WelsScan4x4Ac_sse2:
     pslldq     xmm3, 14
     por        xmm1, xmm3
     psrldq     xmm2, 2
-	mov        eax,  [esp+4]
+	;mov        eax,  [esp+4]
-	movdqa     [eax],xmm1
+	movdqa     [r0],xmm1
-	movdqa     [eax+16], xmm2
+	movdqa     [r0+16], xmm2
 	ret
 
 
@@ -257,44 +268,60 @@ WelsScan4x4Ac_sse2:
 ALIGN 16
 WELS_EXTERN WelsCalculateSingleCtr4x4_sse2
 WelsCalculateSingleCtr4x4_sse2:
-	push      ebx
+	;push      ebx
-	mov       eax,  [esp+8]
+	;mov       eax,  [esp+8]
-	movdqa    xmm0, [eax]
+	%ifdef X86_32
-	movdqa    xmm1, [eax+16]
+	push r3
+	%assign push_num 1
+	%else
+	%assign push_num 0
+	%endif
+	LOAD_1_PARA
+	movdqa    xmm0, [r0]
+	movdqa    xmm1, [r0+16]
 
 	packsswb  xmm0, xmm1
-
+	; below is the register map: r0 - eax, r1 - ebx, r2 - ecx, r3 - edx
+	xor r3, r3
     pxor      xmm3, xmm3
     pcmpeqb   xmm0, xmm3
-    pmovmskb  edx,  xmm0
+    pmovmskb  r3d,  xmm0
 
-    xor       edx,  0xffff
+    xor       r3,  0xffff
 
-	xor       eax,  eax
+	xor       r0,  r0
-	mov       ecx,  7
+	mov       r2,  7
-	mov       ebx,  8
+	mov       r1,  8
 .loop_low8_find1:
-	bt        edx,  ecx
+	bt        r3,  r2
 	jc        .loop_high8_find1
-	loop      .loop_low8_find1
+	dec		  r2
+	jnz      .loop_low8_find1
 .loop_high8_find1:
-	bt        edx, ebx
+	bt        r3, r1
 	jc        .find1end
-	inc       ebx
+	inc       r1
-	cmp       ebx,16
+	cmp       r1,16
 	jb        .loop_high8_find1
 .find1end:
-	sub       ebx, ecx
+	sub       r1, r2
-	sub       ebx, 1
+	sub       r1, 1
-	add       al,  [i_ds_table+ebx]
+	lea	  r2,  [i_ds_table]
-	mov       ebx, edx
+	add       r0b,  [r2+r1]
-	and       edx, 0xff
+	mov       r1, r3
-	shr       ebx, 8
+	and       r3, 0xff
-	and       ebx, 0xff
+	shr       r1, 8
-	add       al,  [low_mask_table +edx]
+	and       r1, 0xff
-	add       al,  [high_mask_table+ebx]
+	lea	  r2 , [low_mask_table]
-
+	add       r0b,  [r2 +r3]
-	pop       ebx
+	lea	  r2, [high_mask_table]
+	add       r0b,  [r2+r1]
+	%ifdef X86_32
+	pop r3
+	%else
+	mov retrd, r0d
+	%endif
+	;pop       ebx
 	ret
 
 
@@ -304,21 +331,29 @@ WelsCalculateSingleCtr4x4_sse2:
 ALIGN 16
 WELS_EXTERN WelsGetNoneZeroCount_sse2
 WelsGetNoneZeroCount_sse2:
-	mov       eax,  [esp+4]
+	%assign push_num 0
-	movdqa    xmm0, [eax]
+	LOAD_1_PARA
-	movdqa    xmm1, [eax+16]
+	;mov       eax,  [esp+4]
+	movdqa    xmm0, [r0]
+	movdqa    xmm1, [r0+16]
 	pxor      xmm2, xmm2
 	pcmpeqw   xmm0, xmm2
 	pcmpeqw   xmm1, xmm2
 	packsswb  xmm1, xmm0
-	pmovmskb  edx,  xmm1
+	xor r1, r1
-	xor       edx,  0xffff
+	pmovmskb  r1d,  xmm1
-	mov       ecx,  edx
+	xor       r1d,  0xffff
-	and       edx,  0xff
+	mov       r2,  r1
-	shr       ecx,  8
+	and       r1,  0xff
+	shr       r2,  8
 ;	and       ecx,  0xff	; we do not need this due to high 16bits equal to 0 yet
-	xor       eax,  eax
+;	xor       retr,  retr
-	add       al,  [nozero_count_table+ecx]
+	;add       al,  [nozero_count_table+r2]
-	add       al,  [nozero_count_table+edx]
+	lea 	  r0 , [nozero_count_table]
+	movzx	  r2, byte [r0+r2]
+	movzx	  r1,   byte [r0+r1]
+	mov	  retrq, r2
+	add	  retrq, r1
+	;add       al,  [nozero_count_table+r1]
 	ret
 

codec/encoder/core/inc/mc.h

@@ -61,15 +61,15 @@ void McCopyWidthEq4_mmx (uint8_t*, int32_t, uint8_t*, int32_t, int32_t);
 void McCopyWidthEq8_mmx (uint8_t*, int32_t, uint8_t*, int32_t, int32_t);
 void PixelAvgWidthEq8_mmx (uint8_t*,  int32_t, uint8_t*, int32_t, uint8_t*, int32_t, int32_t);
 
-void McHorVer20_sse2 (uint8_t* pSrc, int32_t iSrcStride, uint8_t* pDst, int32_t iDstStride, int32_t iWidth,
+void McHorVer20Width9Or17_sse2 (uint8_t* pSrc, int32_t iSrcStride, uint8_t* pDst, int32_t iDstStride, int32_t iWidth,
                       int32_t iHeight);
-void McHorVer02_sse2 (uint8_t* pSrc, int32_t iSrcStride, uint8_t* pDst, int32_t iDstStride, int32_t iWidth,
+void McHorVer02Height9Or17_sse2 (uint8_t* pSrc, int32_t iSrcStride, uint8_t* pDst, int32_t iDstStride, int32_t iWidth,
                       int32_t iHeight);
 void McHorVer22HorFirst_sse2 (uint8_t* pSrc, int32_t iSrcStride, uint8_t* pTap, int32_t iTapStride, int32_t iWidth,
                               int32_t iHeight);
-void McHorVer22VerLastAlign_sse2 (uint8_t* pTap, int32_t iTapStride, uint8_t* pDst, int32_t iDstStride, int32_t iWidth,
+void McHorVer22Width8VerLastAlign_sse2 (uint8_t* pTap, int32_t iTapStride, uint8_t* pDst, int32_t iDstStride, int32_t iWidth,
                                   int32_t iHeight);
-void McHorVer22VerLastUnAlign_sse2 (uint8_t* pTap, int32_t iTapStride, uint8_t* pDst, int32_t iDstStride,
+void McHorVer22Width8VerLastUnAlign_sse2 (uint8_t* pTap, int32_t iTapStride, uint8_t* pDst, int32_t iDstStride,
                                     int32_t iWidth, int32_t iHeight);
 void McChromaWidthEq8_sse2 (uint8_t* pSrc, int32_t iSrcStride, uint8_t* pDst, int32_t iDstStride, const uint8_t* kpABCD,
                             int32_t iHeigh);
@@ -80,8 +80,6 @@ void McHorVer22Width8HorFirst_sse2 (uint8_t* pSrc, int32_t iSrcStride, uint8_t*
                                     int32_t iHeight);
 void PixelAvgWidthEq16_sse2 (uint8_t*,  int32_t, uint8_t*, int32_t, uint8_t*, int32_t, int32_t);
 
-
-void PixelAvgWidthEq16_ssse3 (uint8_t*,  int32_t, uint8_t*, int32_t, uint8_t*, int32_t, int32_t);
 void McChromaWidthEq8_ssse3 (uint8_t* pSrc, int32_t iSrcStride, uint8_t* pDst, int32_t iDstStride,
                              const uint8_t* kpABCD, int32_t iHeigh);
 

codec/encoder/core/inc/sample.h

@@ -107,9 +107,6 @@ int32_t WelsIntra16x16Combined3Satd_sse41 (uint8_t*, int32_t, uint8_t*, int32_t,
 int32_t WelsIntra16x16Combined3Sad_ssse3 (uint8_t*, int32_t, uint8_t*, int32_t, int32_t*, int32_t, uint8_t*);
 int32_t WelsIntraChroma8x8Combined3Satd_sse41 (uint8_t*, int32_t, uint8_t*, int32_t, int32_t*, int32_t, uint8_t*,
     uint8_t*, uint8_t*);
-int32_t WelsIntraChroma8x8Combined3Sad_ssse3 (uint8_t*, int32_t, uint8_t*, int32_t, int32_t*, int32_t, uint8_t*,
-    uint8_t*, uint8_t*);
-
 
 #endif//X86_ASM
 

codec/encoder/core/src/decode_mb_aux.cpp

@@ -261,15 +261,15 @@ void WelsInitReconstructionFuncs (SWelsFuncPtrList* pFuncList, uint32_t  uiCpuFl
 
 #if defined(X86_ASM)
   if (uiCpuFlag & WELS_CPU_MMXEXT) {
-    pFuncList->pfIDctT4		= WelsIDctT4Rec_mmx;
+  //  pFuncList->pfIDctT4		= WelsIDctT4Rec_mmx;
   }
   if (uiCpuFlag & WELS_CPU_SSE2) {
-    pFuncList->pfDequantization4x4			= WelsDequant4x4_sse2;
+   /* pFuncList->pfDequantization4x4			= WelsDequant4x4_sse2;
     pFuncList->pfDequantizationFour4x4		= WelsDequantFour4x4_sse2;
     pFuncList->pfDequantizationIHadamard4x4	= WelsDequantIHadamard4x4_sse2;
 
     pFuncList->pfIDctFourT4		= WelsIDctFourT4Rec_sse2;
-    pFuncList->pfIDctI16x16Dc = WelsIDctRecI16x16Dc_sse2;
+    pFuncList->pfIDctI16x16Dc = WelsIDctRecI16x16Dc_sse2;*/
   }
 #endif//X86_ASM
 }

codec/encoder/core/src/encode_mb_aux.cpp

@@ -569,7 +569,7 @@ void WelsInitEncodingFuncs (SWelsFuncPtrList* pFuncList, uint32_t  uiCpuFlag) {
   }
 //#ifndef MACOS
   if (uiCpuFlag & WELS_CPU_SSSE3) {
-    pFuncList->pfScan4x4				= WelsScan4x4DcAc_ssse3;
+  //  pFuncList->pfScan4x4				= WelsScan4x4DcAc_ssse3;
   }
 
 //#endif//MACOS

codec/encoder/core/src/encoder_ext.cpp

@@ -1919,7 +1919,7 @@ void OutputCpuFeaturesLog (uint32_t uiCpuFeatureFlags, uint32_t uiCpuCores, int3
            uiCpuCores,
            iCacheLineSize);
 
-#ifdef _DEBUG	// output at console & _debug
+//#ifdef _DEBUG	// output at console & _debug
   fprintf (stderr, "WELS CPU features/capacities (0x%x) detected: \n"	\
            "HTT:      %c, "	\
            "MMX:      %c, "	\
@@ -1962,7 +1962,7 @@ void OutputCpuFeaturesLog (uint32_t uiCpuFeatureFlags, uint32_t uiCpuCores, int3
            (uiCpuFeatureFlags & WELS_CPU_AES) ? 'Y' : 'N',
            uiCpuCores,
            iCacheLineSize);
-#endif//_DEBUG
+//#endif//_DEBUG
 }
 
 /*!

codec/encoder/core/src/expand_pic.cpp

@@ -29,14 +29,12 @@
  *     POSSIBILITY OF SUCH DAMAGE.
  *
  */
-
 #include <string.h>
 #include "expand_pic.h"
 #include "cpu_core.h"
 #include "wels_func_ptr_def.h"
 
 namespace WelsSVCEnc{
-
 // rewrite it (split into luma & chroma) that is helpful for mmx/sse2 optimization perform, 9/27/2009
 static inline void ExpandPictureLuma_c (uint8_t* pDst, const int32_t kiStride, const int32_t kiPicW,
                                         const int32_t kiPicH) {
@@ -144,6 +142,8 @@ void ExpandReferencingPicture (SPicture* pPic, PExpandPictureFunc pExpLuma, PExp
   const int32_t kiWidthUV	= kiWidthY >> 1;
   const int32_t kiHeightUV	= kiHeightY >> 1;
 
+
+
   pExpLuma (pPicY, pPic->iLineSize[0], kiWidthY, kiHeightY);
   if (kiWidthUV >= 16) {
     // fix coding picture size as 16x16
@@ -155,6 +155,7 @@ void ExpandReferencingPicture (SPicture* pPic, PExpandPictureFunc pExpLuma, PExp
     ExpandPictureChroma_c (pPicCb, pPic->iLineSize[1], kiWidthUV, kiHeightUV);
     ExpandPictureChroma_c (pPicCr, pPic->iLineSize[2], kiWidthUV, kiHeightUV);
   }
+
 }
 
 }

codec/encoder/core/src/get_intra_predictor.cpp

@@ -80,13 +80,13 @@ void WelsInitFillingPredFuncs (const uint32_t kuiCpuFlag) {
 
 #if defined(X86_ASM)
   if (kuiCpuFlag & WELS_CPU_MMXEXT) {
-    WelsFillingPred8to16		= WelsFillingPred8to16_mmx;
+  //  WelsFillingPred8to16		= WelsFillingPred8to16_mmx;
-    WelsFillingPred8x2to16	    = WelsFillingPred8x2to16_mmx;
+  //  WelsFillingPred8x2to16	    = WelsFillingPred8x2to16_mmx;
-    WelsFillingPred1to16		= WelsFillingPred1to16_mmx;
+  //  WelsFillingPred1to16		= WelsFillingPred1to16_mmx;
   }
   if (kuiCpuFlag & WELS_CPU_SSE2) {
-    WelsFillingPred8x2to16	    = WelsFillingPred8x2to16_sse2;
+   // WelsFillingPred8x2to16	    = WelsFillingPred8x2to16_sse2;
-    WelsFillingPred1to16		= WelsFillingPred1to16_sse2;
+   // WelsFillingPred1to16		= WelsFillingPred1to16_sse2;
   }
 #endif//X86_ASM
 }

codec/encoder/core/src/mc.cpp

@@ -426,7 +426,7 @@ static inline void McHorVer22WidthEq8_sse2 (uint8_t* pSrc, int32_t iSrcStride, u
     int32_t iHeight) {
   ENFORCE_STACK_ALIGN_2D (int16_t, pTap, 21, 8, 16)
   McHorVer22Width8HorFirst_sse2 (pSrc - 2, iSrcStride, (uint8_t*)pTap, 16, iHeight + 5);
-  McHorVer22VerLastAlign_sse2 ((uint8_t*)pTap, 16, pDst, iDstStride, 8, iHeight);
+  McHorVer22Width8VerLastAlign_sse2 ((uint8_t*)pTap, 16, pDst, iDstStride, 8, iHeight);
 }
 
 //2010.2.5
@@ -441,13 +441,13 @@ static inline void McHorVer22WidthEq16_sse2 (uint8_t* pSrc, int32_t iSrcStride,
   McHorVer22WidthEq8_sse2 (pSrc,     iSrcStride, pDst,     iDstStride, iHeight);
   McHorVer22WidthEq8_sse2 (&pSrc[8], iSrcStride, &pDst[8], iDstStride, iHeight);
 }
-void McHorVer22_sse2 (uint8_t* pSrc, int32_t iSrcStride, uint8_t* pDst, int32_t iDstStride, int32_t iWidth,
+void McHorVer22Width9Or17Height9Or17_sse2 (uint8_t* pSrc, int32_t iSrcStride, uint8_t* pDst, int32_t iDstStride, int32_t iWidth,
                       int32_t iHeight) {
   ENFORCE_STACK_ALIGN_2D (int16_t, pTap, 22, 24, 16)
   int32_t tmp1 = 2 * (iWidth - 8);
   McHorVer22HorFirst_sse2 (pSrc - 2, iSrcStride, (uint8_t*)pTap, 48, iWidth, iHeight + 5);
-  McHorVer22VerLastAlign_sse2 ((uint8_t*)pTap,  48, pDst, iDstStride, iWidth - 1, iHeight);
+  McHorVer22Width8VerLastAlign_sse2 ((uint8_t*)pTap,  48, pDst, iDstStride, iWidth - 1, iHeight);
-  McHorVer22VerLastUnAlign_sse2 ((uint8_t*)pTap + tmp1,  48, pDst + iWidth - 8, iDstStride, 8, iHeight);
+  McHorVer22Width8VerLastUnAlign_sse2 ((uint8_t*)pTap + tmp1,  48, pDst + iWidth - 8, iDstStride, 8, iHeight);
 }
 
 typedef void (*McChromaWidthEqx) (uint8_t* pSrc, int32_t iSrcStride, uint8_t* pDst, int32_t iDstStride,
@@ -523,9 +523,9 @@ void WelsInitMcFuncs (SWelsFuncPtrList* pFuncList, uint32_t uiCpuFlag) {
   pFuncList->sMcFuncs.pfLumaQuarpelMc = pWelsMcFuncWidthEq16;
 #if defined (X86_ASM)
   if (uiCpuFlag & WELS_CPU_SSE2) {
-    pFuncList->sMcFuncs.pfLumaHalfpelHor = McHorVer20_sse2;
+    pFuncList->sMcFuncs.pfLumaHalfpelHor = McHorVer20Width9Or17_sse2;
-    pFuncList->sMcFuncs.pfLumaHalfpelVer = McHorVer02_sse2;
+    pFuncList->sMcFuncs.pfLumaHalfpelVer = McHorVer02Height9Or17_sse2;
-    pFuncList->sMcFuncs.pfLumaHalfpelCen = McHorVer22_sse2;
+    pFuncList->sMcFuncs.pfLumaHalfpelCen = McHorVer22Width9Or17Height9Or17_sse2;
     pFuncList->sMcFuncs.pfSampleAveraging[0] = PixelAvgWidthEq8_mmx;
     pFuncList->sMcFuncs.pfSampleAveraging[1] = PixelAvgWidthEq16_sse2;
     pFuncList->sMcFuncs.pfChromaMc = McChroma_sse2;
@@ -541,7 +541,6 @@ void WelsInitMcFuncs (SWelsFuncPtrList* pFuncList, uint32_t uiCpuFlag) {
 
   if (uiCpuFlag & WELS_CPU_SSSE3) {
     pFuncList->sMcFuncs.pfChromaMc = McChroma_ssse3;
-    pFuncList->sMcFuncs.pfSampleAveraging[1] = PixelAvgWidthEq16_ssse3;
   }
 
 #endif //(X86_ASM)

codec/encoder/core/src/md.cpp

@@ -439,7 +439,7 @@ uint8_t MdInterAnalysisVaaInfo_c (int32_t* pSad8x8) {
   return (uiMbSign);
 }
 
-static inline int32_t AnalysisVaaInfoIntra_c (uint8_t* pDataY, const int32_t kiLineSize) {
+int32_t AnalysisVaaInfoIntra_c (uint8_t* pDataY, const int32_t kiLineSize) {
   ENFORCE_STACK_ALIGN_1D (uint16_t, uiAvgBlock, 16, 16)
   uint16_t* pBlock = &uiAvgBlock[0];
   uint8_t* pEncData	= pDataY;

codec/encoder/core/src/sample.cpp

@@ -465,11 +465,11 @@ void WelsInitSampleSadFunc (SWelsFuncPtrList* pFuncList, uint32_t uiCpuFlag) {
     pFuncList->sSampleDealingFuncs.pfSampleSatd[BLOCK_8x16 ] = WelsSampleSatd8x16_sse2;
     pFuncList->sSampleDealingFuncs.pfSampleSatd[BLOCK_16x8 ] = WelsSampleSatd16x8_sse2;
     pFuncList->sSampleDealingFuncs.pfSampleSatd[BLOCK_16x16] = WelsSampleSatd16x16_sse2;
-    pFuncList->sSampleDealingFuncs.pfIntra4x4Combined3Satd =  WelsSmpleSatdThree4x4_sse2;
+    //pFuncList->sSampleDealingFuncs.pfIntra4x4Combined3Satd =  WelsSmpleSatdThree4x4_sse2;
   }
 
   if (uiCpuFlag & WELS_CPU_SSSE3) {
-    pFuncList->sSampleDealingFuncs.pfIntra16x16Combined3Sad = WelsIntra16x16Combined3Sad_ssse3;
+    //pFuncList->sSampleDealingFuncs.pfIntra16x16Combined3Sad = WelsIntra16x16Combined3Sad_ssse3;
   }
 
   if (uiCpuFlag & WELS_CPU_SSE41) {
@@ -478,8 +478,8 @@ void WelsInitSampleSadFunc (SWelsFuncPtrList* pFuncList, uint32_t uiCpuFlag) {
     pFuncList->sSampleDealingFuncs.pfSampleSatd[BLOCK_8x16] = WelsSampleSatd8x16_sse41;
     pFuncList->sSampleDealingFuncs.pfSampleSatd[BLOCK_8x8] = WelsSampleSatd8x8_sse41;
     pFuncList->sSampleDealingFuncs.pfSampleSatd[BLOCK_4x4] = WelsSampleSatd4x4_sse41;
-    pFuncList->sSampleDealingFuncs.pfIntra16x16Combined3Satd = WelsIntra16x16Combined3Satd_sse41;
+    //pFuncList->sSampleDealingFuncs.pfIntra16x16Combined3Satd = WelsIntra16x16Combined3Satd_sse41;
-    pFuncList->sSampleDealingFuncs.pfIntra8x8Combined3Satd = WelsIntraChroma8x8Combined3Satd_sse41;
+    //pFuncList->sSampleDealingFuncs.pfIntra8x8Combined3Satd = WelsIntraChroma8x8Combined3Satd_sse41;
   }
 
 #endif //(X86_ASM)

codec/encoder/core/src/set_mb_syn_cavlc.cpp

@@ -207,7 +207,7 @@ void InitCoeffFunc (const uint32_t uiCpuFlag) {
 
 #if defined(X86_ASM)
   if (uiCpuFlag & WELS_CPU_SSE2) {
-    sCoeffFunc.pfCavlcParamCal = CavlcParamCal_sse2;
+   // sCoeffFunc.pfCavlcParamCal = CavlcParamCal_sse2;
   }
 #endif
 }

codec/encoder/targets.mk

@@ -38,22 +38,13 @@ ENCODER_CPP_SRCS=\
 ENCODER_OBJS += $(ENCODER_CPP_SRCS:.cpp=.o)
 ifeq ($(USE_ASM), Yes)
 ENCODER_ASM_SRCS=\
-	$(ENCODER_SRCDIR)/./core/asm/asm_inc.asm\
 	$(ENCODER_SRCDIR)/./core/asm/coeff.asm\
-	$(ENCODER_SRCDIR)/./core/asm/cpuid.asm\
 	$(ENCODER_SRCDIR)/./core/asm/dct.asm\
-	$(ENCODER_SRCDIR)/./core/asm/deblock.asm\
-	$(ENCODER_SRCDIR)/./core/asm/expand_picture.asm\
 	$(ENCODER_SRCDIR)/./core/asm/intra_pred.asm\
-	$(ENCODER_SRCDIR)/./core/asm/intra_pred_util.asm\
-	$(ENCODER_SRCDIR)/./core/asm/mb_copy.asm\
-	$(ENCODER_SRCDIR)/./core/asm/mc_chroma.asm\
-	$(ENCODER_SRCDIR)/./core/asm/mc_luma.asm\
 	$(ENCODER_SRCDIR)/./core/asm/memzero.asm\
 	$(ENCODER_SRCDIR)/./core/asm/quant.asm\
 	$(ENCODER_SRCDIR)/./core/asm/satd_sad.asm\
 	$(ENCODER_SRCDIR)/./core/asm/score.asm\
-	$(ENCODER_SRCDIR)/./core/asm/vaa.asm\
 
 ENCODER_OBJS += $(ENCODER_ASM_SRCS:.asm=.o)
 endif
@@ -158,39 +149,15 @@ $(ENCODER_SRCDIR)/./plus/src/welsCodecTrace.o: $(ENCODER_SRCDIR)/./plus/src/wels
 $(ENCODER_SRCDIR)/./plus/src/welsEncoderExt.o: $(ENCODER_SRCDIR)/./plus/src/welsEncoderExt.cpp
 	$(CXX) $(CFLAGS) $(CXXFLAGS) $(INCLUDES) $(ENCODER_CFLAGS) $(ENCODER_INCLUDES) -c -o $(ENCODER_SRCDIR)/./plus/src/welsEncoderExt.o $(ENCODER_SRCDIR)/./plus/src/welsEncoderExt.cpp
 
-$(ENCODER_SRCDIR)/./core/asm/asm_inc.o: $(ENCODER_SRCDIR)/./core/asm/asm_inc.asm
-	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(ENCODER_ASMFLAGS) $(ENCODER_ASM_INCLUDES) -o $(ENCODER_SRCDIR)/./core/asm/asm_inc.o $(ENCODER_SRCDIR)/./core/asm/asm_inc.asm
-
 $(ENCODER_SRCDIR)/./core/asm/coeff.o: $(ENCODER_SRCDIR)/./core/asm/coeff.asm
 	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(ENCODER_ASMFLAGS) $(ENCODER_ASM_INCLUDES) -o $(ENCODER_SRCDIR)/./core/asm/coeff.o $(ENCODER_SRCDIR)/./core/asm/coeff.asm
 
-$(ENCODER_SRCDIR)/./core/asm/cpuid.o: $(ENCODER_SRCDIR)/./core/asm/cpuid.asm
-	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(ENCODER_ASMFLAGS) $(ENCODER_ASM_INCLUDES) -o $(ENCODER_SRCDIR)/./core/asm/cpuid.o $(ENCODER_SRCDIR)/./core/asm/cpuid.asm
-
 $(ENCODER_SRCDIR)/./core/asm/dct.o: $(ENCODER_SRCDIR)/./core/asm/dct.asm
 	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(ENCODER_ASMFLAGS) $(ENCODER_ASM_INCLUDES) -o $(ENCODER_SRCDIR)/./core/asm/dct.o $(ENCODER_SRCDIR)/./core/asm/dct.asm
 
-$(ENCODER_SRCDIR)/./core/asm/deblock.o: $(ENCODER_SRCDIR)/./core/asm/deblock.asm
-	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(ENCODER_ASMFLAGS) $(ENCODER_ASM_INCLUDES) -o $(ENCODER_SRCDIR)/./core/asm/deblock.o $(ENCODER_SRCDIR)/./core/asm/deblock.asm
-
-$(ENCODER_SRCDIR)/./core/asm/expand_picture.o: $(ENCODER_SRCDIR)/./core/asm/expand_picture.asm
-	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(ENCODER_ASMFLAGS) $(ENCODER_ASM_INCLUDES) -o $(ENCODER_SRCDIR)/./core/asm/expand_picture.o $(ENCODER_SRCDIR)/./core/asm/expand_picture.asm
-
 $(ENCODER_SRCDIR)/./core/asm/intra_pred.o: $(ENCODER_SRCDIR)/./core/asm/intra_pred.asm
 	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(ENCODER_ASMFLAGS) $(ENCODER_ASM_INCLUDES) -o $(ENCODER_SRCDIR)/./core/asm/intra_pred.o $(ENCODER_SRCDIR)/./core/asm/intra_pred.asm
 
-$(ENCODER_SRCDIR)/./core/asm/intra_pred_util.o: $(ENCODER_SRCDIR)/./core/asm/intra_pred_util.asm
-	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(ENCODER_ASMFLAGS) $(ENCODER_ASM_INCLUDES) -o $(ENCODER_SRCDIR)/./core/asm/intra_pred_util.o $(ENCODER_SRCDIR)/./core/asm/intra_pred_util.asm
-
-$(ENCODER_SRCDIR)/./core/asm/mb_copy.o: $(ENCODER_SRCDIR)/./core/asm/mb_copy.asm
-	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(ENCODER_ASMFLAGS) $(ENCODER_ASM_INCLUDES) -o $(ENCODER_SRCDIR)/./core/asm/mb_copy.o $(ENCODER_SRCDIR)/./core/asm/mb_copy.asm
-
-$(ENCODER_SRCDIR)/./core/asm/mc_chroma.o: $(ENCODER_SRCDIR)/./core/asm/mc_chroma.asm
-	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(ENCODER_ASMFLAGS) $(ENCODER_ASM_INCLUDES) -o $(ENCODER_SRCDIR)/./core/asm/mc_chroma.o $(ENCODER_SRCDIR)/./core/asm/mc_chroma.asm
-
-$(ENCODER_SRCDIR)/./core/asm/mc_luma.o: $(ENCODER_SRCDIR)/./core/asm/mc_luma.asm
-	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(ENCODER_ASMFLAGS) $(ENCODER_ASM_INCLUDES) -o $(ENCODER_SRCDIR)/./core/asm/mc_luma.o $(ENCODER_SRCDIR)/./core/asm/mc_luma.asm
-
 $(ENCODER_SRCDIR)/./core/asm/memzero.o: $(ENCODER_SRCDIR)/./core/asm/memzero.asm
 	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(ENCODER_ASMFLAGS) $(ENCODER_ASM_INCLUDES) -o $(ENCODER_SRCDIR)/./core/asm/memzero.o $(ENCODER_SRCDIR)/./core/asm/memzero.asm
 
@@ -203,9 +170,6 @@ $(ENCODER_SRCDIR)/./core/asm/satd_sad.o: $(ENCODER_SRCDIR)/./core/asm/satd_sad.a
 $(ENCODER_SRCDIR)/./core/asm/score.o: $(ENCODER_SRCDIR)/./core/asm/score.asm
 	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(ENCODER_ASMFLAGS) $(ENCODER_ASM_INCLUDES) -o $(ENCODER_SRCDIR)/./core/asm/score.o $(ENCODER_SRCDIR)/./core/asm/score.asm
 
-$(ENCODER_SRCDIR)/./core/asm/vaa.o: $(ENCODER_SRCDIR)/./core/asm/vaa.asm
-	$(ASM) $(ASMFLAGS) $(ASM_INCLUDES) $(ENCODER_ASMFLAGS) $(ENCODER_ASM_INCLUDES) -o $(ENCODER_SRCDIR)/./core/asm/vaa.o $(ENCODER_SRCDIR)/./core/asm/vaa.asm
-
 $(LIBPREFIX)encoder.$(LIBSUFFIX): $(ENCODER_OBJS)
 	rm -f $(LIBPREFIX)encoder.$(LIBSUFFIX)
 	$(AR) cr $@ $(ENCODER_OBJS)

codec/processing/build/win32/WelsVP_2008.vcproj

@@ -106,6 +106,91 @@
 				CommandLine=""
 			/>
 		</Configuration>
+		<Configuration
+			Name="Debug|x64"
+			OutputDirectory=".\..\..\..\..\bin\win64\Debug"
+			IntermediateDirectory=".\..\..\..\obj\vp\Debug"
+			ConfigurationType="2"
+			CharacterSet="1"
+			WholeProgramOptimization="0"
+			>
+			<Tool
+				Name="VCPreBuildEventTool"
+			/>
+			<Tool
+				Name="VCCustomBuildTool"
+				CommandLine=""
+			/>
+			<Tool
+				Name="MASM"
+			/>
+			<Tool
+				Name="VCXMLDataGeneratorTool"
+			/>
+			<Tool
+				Name="VCWebServiceProxyGeneratorTool"
+			/>
+			<Tool
+				Name="VCMIDLTool"
+				TargetEnvironment="3"
+			/>
+			<Tool
+				Name="VCCLCompilerTool"
+				Optimization="0"
+				AdditionalIncludeDirectories=""
+				PreprocessorDefinitions="WIN64;_DEBUG;_WINDOWS;_USRDLL;WELSVP_EXPORTS;X86_ASM"
+				MinimalRebuild="true"
+				BasicRuntimeChecks="3"
+				RuntimeLibrary="1"
+				UsePrecompiledHeader="0"
+				AssemblerListingLocation=""
+				WarningLevel="3"
+				DebugInformationFormat="3"
+			/>
+			<Tool
+				Name="VCManagedResourceCompilerTool"
+			/>
+			<Tool
+				Name="VCResourceCompilerTool"
+			/>
+			<Tool
+				Name="VCPreLinkEventTool"
+			/>
+			<Tool
+				Name="VCLinkerTool"
+				LinkLibraryDependencies="true"
+				OutputFile="$(OutDir)\welsvp.dll"
+				LinkIncremental="2"
+				ModuleDefinitionFile="../../src/common/WelsVP.def"
+				GenerateDebugInformation="true"
+				GenerateMapFile="true"
+				MapFileName="$(OutDir)\welsvp.map"
+				SubSystem="2"
+				TargetMachine="17"
+			/>
+			<Tool
+				Name="VCALinkTool"
+			/>
+			<Tool
+				Name="VCManifestTool"
+			/>
+			<Tool
+				Name="VCXDCMakeTool"
+			/>
+			<Tool
+				Name="VCBscMakeTool"
+			/>
+			<Tool
+				Name="VCFxCopTool"
+			/>
+			<Tool
+				Name="VCAppVerifierTool"
+			/>
+			<Tool
+				Name="VCPostBuildEventTool"
+				CommandLine=""
+			/>
+		</Configuration>
 		<Configuration
 			Name="Release|Win32"
 			OutputDirectory=".\..\..\..\bin\win32\Release"
@@ -194,94 +279,9 @@
 				CommandLine=""
 			/>
 		</Configuration>
-		<Configuration
-			Name="Debug|x64"
-			OutputDirectory=".\..\..\..\bin\win32\Debug"
-			IntermediateDirectory=".\..\..\..\obj\vp\Debug"
-			ConfigurationType="2"
-			CharacterSet="1"
-			WholeProgramOptimization="0"
-			>
-			<Tool
-				Name="VCPreBuildEventTool"
-			/>
-			<Tool
-				Name="VCCustomBuildTool"
-				CommandLine=""
-			/>
-			<Tool
-				Name="MASM"
-			/>
-			<Tool
-				Name="VCXMLDataGeneratorTool"
-			/>
-			<Tool
-				Name="VCWebServiceProxyGeneratorTool"
-			/>
-			<Tool
-				Name="VCMIDLTool"
-				TargetEnvironment="3"
-			/>
-			<Tool
-				Name="VCCLCompilerTool"
-				Optimization="0"
-				AdditionalIncludeDirectories=""
-				PreprocessorDefinitions="WIN64;_DEBUG;_WINDOWS;_USRDLL;WELSVP_EXPORTS"
-				MinimalRebuild="true"
-				BasicRuntimeChecks="3"
-				RuntimeLibrary="1"
-				UsePrecompiledHeader="0"
-				AssemblerListingLocation=""
-				WarningLevel="3"
-				DebugInformationFormat="3"
-			/>
-			<Tool
-				Name="VCManagedResourceCompilerTool"
-			/>
-			<Tool
-				Name="VCResourceCompilerTool"
-			/>
-			<Tool
-				Name="VCPreLinkEventTool"
-			/>
-			<Tool
-				Name="VCLinkerTool"
-				LinkLibraryDependencies="true"
-				OutputFile="$(OutDir)\welsvp.dll"
-				LinkIncremental="2"
-				ModuleDefinitionFile="../../src/common/WelsVP.def"
-				GenerateDebugInformation="true"
-				GenerateMapFile="true"
-				MapFileName="$(OutDir)\welsvp.map"
-				SubSystem="2"
-				TargetMachine="17"
-			/>
-			<Tool
-				Name="VCALinkTool"
-			/>
-			<Tool
-				Name="VCManifestTool"
-			/>
-			<Tool
-				Name="VCXDCMakeTool"
-			/>
-			<Tool
-				Name="VCBscMakeTool"
-			/>
-			<Tool
-				Name="VCFxCopTool"
-			/>
-			<Tool
-				Name="VCAppVerifierTool"
-			/>
-			<Tool
-				Name="VCPostBuildEventTool"
-				CommandLine=""
-			/>
-		</Configuration>
 		<Configuration
 			Name="Release|x64"
-			OutputDirectory=".\..\..\..\bin\win64\Release"
+			OutputDirectory=".\..\..\..\..\bin\win64\Release"
 			IntermediateDirectory=".\..\..\..\obj\vp\Release"
 			ConfigurationType="2"
 			CharacterSet="1"
@@ -313,7 +313,7 @@
 				Optimization="3"
 				EnableIntrinsicFunctions="false"
 				FavorSizeOrSpeed="1"
-				PreprocessorDefinitions="WIN64;NDEBUG;_WINDOWS;_USRDLL;WELSVP_EXPORTS"
+				PreprocessorDefinitions="WIN64;NDEBUG;_WINDOWS;_USRDLL;WELSVP_EXPORTS;X86_ASM"
 				RuntimeLibrary="0"
 				EnableFunctionLevelLinking="false"
 				UsePrecompiledHeader="0"
@@ -480,14 +480,23 @@
 			Name="ASM"
 			>
 			<File
-				RelativePath="..\..\src\asm\asm_inc.asm"
+				RelativePath="..\..\..\common\cpuid.asm"
 				>
 				<FileConfiguration
 					Name="Debug|Win32"
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="nasm  -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+						CommandLine="nasm  -I$(InputDir) -f win32 -DPREFIX -DX86_32 -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+						Outputs="$(IntDir)\$(InputName).obj"
+					/>
+				</FileConfiguration>
+				<FileConfiguration
+					Name="Debug|x64"
+					>
+					<Tool
+						Name="VCCustomBuildTool"
+						CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)"
 						Outputs="$(IntDir)\$(InputName).obj"
 					/>
 				</FileConfiguration>
@@ -496,69 +505,16 @@
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="nasm  -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+						CommandLine="nasm  -I$(InputDir) -f win32 -DPREFIX -DX86_32 -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
-						Outputs="$(IntDir)\$(InputName).obj"
-					/>
-				</FileConfiguration>
-				<FileConfiguration
-					Name="Debug|x64"
-					ExcludedFromBuild="true"
-					>
-					<Tool
-						Name="VCCustomBuildTool"
-						CommandLine="nasm  -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
 						Outputs="$(IntDir)\$(InputName).obj"
 					/>
 				</FileConfiguration>
 				<FileConfiguration
 					Name="Release|x64"
-					ExcludedFromBuild="true"
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="nasm  -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+						CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)"
-						Outputs="$(IntDir)\$(InputName).obj"
-					/>
-				</FileConfiguration>
-			</File>
-			<File
-				RelativePath="..\..\src\asm\cpuid.asm"
-				>
-				<FileConfiguration
-					Name="Debug|Win32"
-					>
-					<Tool
-						Name="VCCustomBuildTool"
-						CommandLine="nasm  -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
-						Outputs="$(IntDir)\$(InputName).obj"
-					/>
-				</FileConfiguration>
-				<FileConfiguration
-					Name="Release|Win32"
-					>
-					<Tool
-						Name="VCCustomBuildTool"
-						CommandLine="nasm -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
-						Outputs="$(IntDir)\$(InputName).obj"
-					/>
-				</FileConfiguration>
-				<FileConfiguration
-					Name="Debug|x64"
-					ExcludedFromBuild="true"
-					>
-					<Tool
-						Name="VCCustomBuildTool"
-						CommandLine="nasm  -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
-						Outputs="$(IntDir)\$(InputName).obj"
-					/>
-				</FileConfiguration>
-				<FileConfiguration
-					Name="Release|x64"
-					ExcludedFromBuild="true"
-					>
-					<Tool
-						Name="VCCustomBuildTool"
-						CommandLine="nasm -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
 						Outputs="$(IntDir)\$(InputName).obj"
 					/>
 				</FileConfiguration>
@@ -571,7 +527,16 @@
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="nasm   -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+						CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win32 -DPREFIX  -DX86_32 -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+						Outputs="$(IntDir)\$(InputName).obj"
+					/>
+				</FileConfiguration>
+				<FileConfiguration
+					Name="Debug|x64"
+					>
+					<Tool
+						Name="VCCustomBuildTool"
+						CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
 						Outputs="$(IntDir)\$(InputName).obj"
 					/>
 				</FileConfiguration>
@@ -580,27 +545,16 @@
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="nasm -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+						CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win32 -DPREFIX  -DX86_32 -o $(IntDir)\$(InputName).obj $(InputPath)"
-						Outputs="$(IntDir)\$(InputName).obj"
-					/>
-				</FileConfiguration>
-				<FileConfiguration
-					Name="Debug|x64"
-					ExcludedFromBuild="true"
-					>
-					<Tool
-						Name="VCCustomBuildTool"
-						CommandLine="nasm   -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
 						Outputs="$(IntDir)\$(InputName).obj"
 					/>
 				</FileConfiguration>
 				<FileConfiguration
 					Name="Release|x64"
-					ExcludedFromBuild="true"
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="nasm -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+						CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
 						Outputs="$(IntDir)\$(InputName).obj"
 					/>
 				</FileConfiguration>
@@ -613,7 +567,16 @@
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="nasm  -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+						CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+						Outputs="$(IntDir)\$(InputName).obj"
+					/>
+				</FileConfiguration>
+				<FileConfiguration
+					Name="Debug|x64"
+					>
+					<Tool
+						Name="VCCustomBuildTool"
+						CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)"
 						Outputs="$(IntDir)\$(InputName).obj"
 					/>
 				</FileConfiguration>
@@ -622,27 +585,16 @@
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="nasm -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+						CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
-						Outputs="$(IntDir)\$(InputName).obj"
-					/>
-				</FileConfiguration>
-				<FileConfiguration
-					Name="Debug|x64"
-					ExcludedFromBuild="true"
-					>
-					<Tool
-						Name="VCCustomBuildTool"
-						CommandLine="nasm  -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
 						Outputs="$(IntDir)\$(InputName).obj"
 					/>
 				</FileConfiguration>
 				<FileConfiguration
 					Name="Release|x64"
-					ExcludedFromBuild="true"
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="nasm -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+						CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)"
 						Outputs="$(IntDir)\$(InputName).obj"
 					/>
 				</FileConfiguration>
@@ -655,7 +607,16 @@
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="nasm  -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+						CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win32 -DPREFIX  -DX86_32 -o $(IntDir)\$(InputName).obj $(InputPath)"
+						Outputs="$(IntDir)\$(InputName).obj"
+					/>
+				</FileConfiguration>
+				<FileConfiguration
+					Name="Debug|x64"
+					>
+					<Tool
+						Name="VCCustomBuildTool"
+						CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)"
 						Outputs="$(IntDir)\$(InputName).obj"
 					/>
 				</FileConfiguration>
@@ -664,27 +625,16 @@
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="nasm -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+						CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win32 -DPREFIX  -DX86_32 -o $(IntDir)\$(InputName).obj $(InputPath)"
-						Outputs="$(IntDir)\$(InputName).obj"
-					/>
-				</FileConfiguration>
-				<FileConfiguration
-					Name="Debug|x64"
-					ExcludedFromBuild="true"
-					>
-					<Tool
-						Name="VCCustomBuildTool"
-						CommandLine="nasm  -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
 						Outputs="$(IntDir)\$(InputName).obj"
 					/>
 				</FileConfiguration>
 				<FileConfiguration
 					Name="Release|x64"
-					ExcludedFromBuild="true"
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="nasm -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+						CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)"
 						Outputs="$(IntDir)\$(InputName).obj"
 					/>
 				</FileConfiguration>
@@ -697,7 +647,16 @@
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="nasm  -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+						CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win32 -DPREFIX  -DX86_32 -o $(IntDir)\$(InputName).obj $(InputPath)"
+						Outputs="$(IntDir)\$(InputName).obj"
+					/>
+				</FileConfiguration>
+				<FileConfiguration
+					Name="Debug|x64"
+					>
+					<Tool
+						Name="VCCustomBuildTool"
+						CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)"
 						Outputs="$(IntDir)\$(InputName).obj"
 					/>
 				</FileConfiguration>
@@ -706,27 +665,16 @@
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="nasm -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+						CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win32 -DPREFIX  -DX86_32 -o $(IntDir)\$(InputName).obj $(InputPath)"
-						Outputs="$(IntDir)\$(InputName).obj"
-					/>
-				</FileConfiguration>
-				<FileConfiguration
-					Name="Debug|x64"
-					ExcludedFromBuild="true"
-					>
-					<Tool
-						Name="VCCustomBuildTool"
-						CommandLine="nasm  -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
 						Outputs="$(IntDir)\$(InputName).obj"
 					/>
 				</FileConfiguration>
 				<FileConfiguration
 					Name="Release|x64"
-					ExcludedFromBuild="true"
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="nasm -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+						CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)"
 						Outputs="$(IntDir)\$(InputName).obj"
 					/>
 				</FileConfiguration>
@@ -739,7 +687,16 @@
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="nasm  -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+						CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+						Outputs="$(IntDir)\$(InputName).obj"
+					/>
+				</FileConfiguration>
+				<FileConfiguration
+					Name="Debug|x64"
+					>
+					<Tool
+						Name="VCCustomBuildTool"
+						CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)"
 						Outputs="$(IntDir)\$(InputName).obj"
 					/>
 				</FileConfiguration>
@@ -748,27 +705,16 @@
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="nasm  -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+						CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
-						Outputs="$(IntDir)\$(InputName).obj"
-					/>
-				</FileConfiguration>
-				<FileConfiguration
-					Name="Debug|x64"
-					ExcludedFromBuild="true"
-					>
-					<Tool
-						Name="VCCustomBuildTool"
-						CommandLine="nasm  -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
 						Outputs="$(IntDir)\$(InputName).obj"
 					/>
 				</FileConfiguration>
 				<FileConfiguration
 					Name="Release|x64"
-					ExcludedFromBuild="true"
 					>
 					<Tool
 						Name="VCCustomBuildTool"
-						CommandLine="nasm  -I$(InputDir) -f win32 -DPREFIX -o $(IntDir)\$(InputName).obj $(InputPath)&#x0D;&#x0A;"
+						CommandLine="nasm -I$(InputDir) -I$(InputDir)/../../../common/ -f win64 -O3 -DWIN64 -o $(IntDir)\$(InputName).obj $(InputPath)"
 						Outputs="$(IntDir)\$(InputName).obj"
 					/>
 				</FileConfiguration>

codec/processing/build/win32/WelsVP_2010.vcxproj

@@ -63,16 +63,16 @@
   <PropertyGroup Label="UserMacros" />
   <PropertyGroup>
     <_ProjectFileVersion>10.0.40219.1</_ProjectFileVersion>
-    <OutDir Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">.\..\..\..\bin\win32\Debug\</OutDir>
+    <OutDir Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">..\..\..\..\bin\win32\Debug\</OutDir>
-    <OutDir Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">.\..\..\..\bin\win64\Debug\</OutDir>
+    <OutDir Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">..\..\..\..\bin\win64\Debug\</OutDir>
-    <IntDir Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">.\..\..\..\obj\vp\Debug\</IntDir>
+    <IntDir Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">..\..\..\..\obj\vp\Debug\</IntDir>
-    <IntDir Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">.\..\..\..\obj\vp\Debug\</IntDir>
+    <IntDir Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">..\..\..\..\obj\vp\Debug\</IntDir>
     <LinkIncremental Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">true</LinkIncremental>
     <LinkIncremental Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</LinkIncremental>
-    <OutDir Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">.\..\..\..\bin\win32\Release\</OutDir>
+    <OutDir Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">..\..\..\..\bin\win32\Release\</OutDir>
-    <OutDir Condition="'$(Configuration)|$(Platform)'=='Release|x64'">.\..\..\..\bin\win64\Release\</OutDir>
+    <OutDir Condition="'$(Configuration)|$(Platform)'=='Release|x64'">..\..\..\..\bin\win64\Release\</OutDir>
-    <IntDir Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">.\..\..\..\obj\vp\Release\</IntDir>
+    <IntDir Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">..\..\..\..\obj\vp\Release\</IntDir>
-    <IntDir Condition="'$(Configuration)|$(Platform)'=='Release|x64'">.\..\..\..\obj\vp\Release\</IntDir>
+    <IntDir Condition="'$(Configuration)|$(Platform)'=='Release|x64'">..\..\..\..\obj\vp\Release\</IntDir>
     <LinkIncremental Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">false</LinkIncremental>
     <LinkIncremental Condition="'$(Configuration)|$(Platform)'=='Release|x64'">false</LinkIncremental>
     <GenerateManifest Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">false</GenerateManifest>
@@ -144,7 +144,7 @@
     <ClCompile>
       <Optimization>Disabled</Optimization>
       <AdditionalIncludeDirectories>%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories>
-      <PreprocessorDefinitions>WIN64;_DEBUG;_WINDOWS;_USRDLL;WELSVP_EXPORTS;%(PreprocessorDefinitions)</PreprocessorDefinitions>
+      <PreprocessorDefinitions>WIN64;_DEBUG;X86_ASM;_WINDOWS;_USRDLL;WELSVP_EXPORTS;%(PreprocessorDefinitions)</PreprocessorDefinitions>
       <BasicRuntimeChecks>EnableFastChecks</BasicRuntimeChecks>
       <RuntimeLibrary>MultiThreadedDebug</RuntimeLibrary>
       <PrecompiledHeader>
@@ -235,7 +235,7 @@
       <Optimization>Full</Optimization>
       <IntrinsicFunctions>false</IntrinsicFunctions>
       <FavorSizeOrSpeed>Speed</FavorSizeOrSpeed>
-      <PreprocessorDefinitions>WIN64;NDEBUG;_WINDOWS;_USRDLL;WELSVP_EXPORTS;%(PreprocessorDefinitions)</PreprocessorDefinitions>
+      <PreprocessorDefinitions>WIN64;NDEBUG;X86_ASM;_WINDOWS;_USRDLL;WELSVP_EXPORTS;%(PreprocessorDefinitions)</PreprocessorDefinitions>
       <RuntimeLibrary>MultiThreaded</RuntimeLibrary>
       <FunctionLevelLinking>false</FunctionLevelLinking>
       <PrecompiledHeader>
@@ -315,117 +315,68 @@
     <ResourceCompile Include="..\..\src\common\WelsVP.rc" />
   </ItemGroup>
   <ItemGroup>
-    <CustomBuild Include="..\..\src\asm\asm_inc.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
-    </CustomBuild>
-    <CustomBuild Include="..\..\src\asm\cpuid.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
-    </CustomBuild>
     <CustomBuild Include="..\..\src\asm\denoisefilter.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm   -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm   -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
     </CustomBuild>
     <CustomBuild Include="..\..\src\asm\downsample_bilinear.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
     </CustomBuild>
     <CustomBuild Include="..\..\src\asm\intra_pred.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
     </CustomBuild>
     <CustomBuild Include="..\..\src\asm\sad.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
     </CustomBuild>
     <CustomBuild Include="..\..\src\asm\vaa.asm">
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-</Command>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
-      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm  -I%(RootDir)%(Directory) -f win32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)
-</Command>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
       <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">true</ExcludedFromBuild>
+    </CustomBuild>
-      <ExcludedFromBuild Condition="'$(Configuration)|$(Platform)'=='Release|x64'">true</ExcludedFromBuild>
+  </ItemGroup>
+  <ItemGroup>
+    <CustomBuild Include="..\..\..\common\cpuid.asm">
+      <FileType>Document</FileType>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win32 -DX86_32 -DPREFIX -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Release|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Release|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
+      <Command Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">nasm -I ..\..\..\common\  -I%(RootDir)%(Directory) -f win64 -DWIN64 -o $(IntDir)%(Filename).obj %(FullPath)</Command>
+      <Outputs Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">$(IntDir)%(Filename).obj;%(Outputs)</Outputs>
     </CustomBuild>
   </ItemGroup>
   <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />

codec/processing/build/win32/WelsVP_2010.vcxproj.filters

@@ -116,12 +116,6 @@
     </ClInclude>
   </ItemGroup>
   <ItemGroup>
-    <CustomBuild Include="..\..\src\asm\asm_inc.asm">
-      <Filter>ASM</Filter>
-    </CustomBuild>
-    <CustomBuild Include="..\..\src\asm\cpuid.asm">
-      <Filter>ASM</Filter>
-    </CustomBuild>
     <CustomBuild Include="..\..\src\asm\denoisefilter.asm">
       <Filter>ASM</Filter>
     </CustomBuild>
@@ -137,6 +131,9 @@
     <CustomBuild Include="..\..\src\asm\vaa.asm">
       <Filter>ASM</Filter>
     </CustomBuild>
+    <CustomBuild Include="..\..\..\common\cpuid.asm">
+      <Filter>ASM</Filter>
+    </CustomBuild>
   </ItemGroup>
   <ItemGroup>
     <Filter Include="ASM">

codec/processing/src/adaptivequantization/AdaptiveQuantization.cpp

@@ -223,7 +223,7 @@ void CAdaptiveQuantization::WelsInitVarFunc (PVarFunc& pfVar,  int32_t iCpuFlag)
 
 #ifdef X86_ASM
   if (iCpuFlag & WELS_CPU_SSE2) {
-    pfVar = SampleVariance16x16_sse2;
+   // pfVar = SampleVariance16x16_sse2;
   }
 #endif
 }

codec/processing/src/asm/denoisefilter.asm

@@ -50,7 +50,7 @@ sse2_32 times 8 dw 32
 sse2_20 times 8 dw 20
 
 
-BITS 32
+
 ;***********************************************************************
 ; Code
 ;***********************************************************************
@@ -173,35 +173,40 @@ WELS_EXTERN BilateralLumaFilter8_sse2
 ;	6	7	8
 ;	0:	the center point
 %define		pushsize	4
-%define		pixel		esp + pushsize + 4
+;%define		pixel		esp + pushsize + 4
-%define		stride		esp + pushsize + 8
+;%define		stride		esp + pushsize + 8
+;%define         pixel  r0
+;%define         stride r1
+
 BilateralLumaFilter8_sse2:
-		push		ebx
+       
+        push r3 
+        %assign push_num 1
+        LOAD_2_PARA
 
 		pxor		xmm7,	xmm7
-		mov			eax,	[pixel]
+	
-		mov			ebx,	eax
+		mov         r3,     r0
-		movq		xmm6,	[eax]
+		
+		movq        xmm6,   [r0]
 		punpcklbw	xmm6,	xmm7
 		movdqa		xmm3,	[sse2_32]
 		pxor		xmm4,	xmm4		; nTotWeight
 		pxor		xmm5,	xmm5		; nSum
 
-		dec			eax
+        dec         r0
-		mov			ecx,	[stride]
+		WEIGHT_LINE	xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm0,  [r0]			; pixel 4
+		WEIGHT_LINE	xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm0,  [r0 + 2]		; pixel 5
 
-		WEIGHT_LINE	xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm0,  [eax]			; pixel 4
+		sub			r0,	r1
-		WEIGHT_LINE	xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm0,  [eax + 2]		; pixel 5
+		WEIGHT_LINE	xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm0,  [r0]			; pixel 1
+		WEIGHT_LINE	xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm0,  [r0 + 1]		; pixel 2
+		WEIGHT_LINE	xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm0,  [r0 + 2]		; pixel 3
 
-		sub			eax,	ecx
+		lea			r0,	[r0 + r1 * 2]
-		WEIGHT_LINE	xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm0,  [eax]			; pixel 1
+		WEIGHT_LINE	xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm0,  [r0]			; pixel 6
-		WEIGHT_LINE	xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm0,  [eax + 1]		; pixel 2
+		WEIGHT_LINE	xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm0,  [r0 + 1]		; pixel 7
-		WEIGHT_LINE	xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm0,  [eax + 2]		; pixel 3
+		WEIGHT_LINE	xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm0,  [r0 + 2]		; pixel 8
-
-		lea			eax,	[eax + ecx * 2]
-		WEIGHT_LINE	xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm0,  [eax]			; pixel 6
-		WEIGHT_LINE	xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm0,  [eax + 1]		; pixel 7
-		WEIGHT_LINE	xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm0,  [eax + 2]		; pixel 8
 
 		pcmpeqw		xmm0,	xmm0
 		psrlw		xmm0,	15
@@ -211,9 +216,12 @@ BilateralLumaFilter8_sse2:
 		paddusw		xmm5,	xmm0
 		psrlw		xmm5,	8
 		packuswb	xmm5,	xmm5
-		movq		[ebx],	xmm5
+		movq		[r3],	xmm5
+
+       
+		pop r3
+		%assign push_num 0
 		
-		pop ebx
 		ret
 
 WELS_EXTERN WaverageChromaFilter8_sse2
@@ -229,35 +237,43 @@ WELS_EXTERN WaverageChromaFilter8_sse2
 
 ALIGN 16
 WaverageChromaFilter8_sse2:
-		mov		edx,	[esp + 4]	; pixels
-		mov		ecx,	[esp + 8]	; stride
 
-		mov		eax,	ecx
+        push r3
-		add		eax,	eax
+       
-		sub		edx,	eax			; pixels - 2 * stride
+        %assign push_num 1
-		sub		edx,	2
+        
+        LOAD_2_PARA
+        
+        mov		r3,	r1
+		add		r3,	r3
+		sub		r0,	r3			; pixels - 2 * stride
+		sub		r0,	2
 
 		pxor	xmm0,	xmm0
 		pxor	xmm3,	xmm3
 
-		movdqu		xmm1,	[edx]
+		movdqu		xmm1,	[r0]
 		WEIGHT_LINE1_UV	xmm1,	xmm2,	xmm3,	xmm0
 
-		movdqu		xmm1,	[edx + ecx]
+		movdqu		xmm1,	[r0 + r1]
 		WEIGHT_LINE2_UV	xmm1,	xmm2,	xmm3,	xmm0
 
-		add		edx,	eax
+		add		r0,	r3
-		movdqu		xmm1,	[edx]
+		movdqu		xmm1,	[r0]
 		WEIGHT_LINE3_UV	xmm1,	xmm2,	xmm3,	xmm0
 
-		movdqu		xmm1,	[edx + ecx]
+		movdqu		xmm1,	[r0 + r1]
 		WEIGHT_LINE2_UV	xmm1,	xmm2,	xmm3,	xmm0
 
-		movdqu		xmm1,	[edx + ecx * 2]
+		movdqu		xmm1,	[r0 + r1 * 2]
 		WEIGHT_LINE1_UV	xmm1,	xmm2,	xmm3,	xmm0
 
 		psrlw		xmm3,		6
 		packuswb	xmm3,		xmm3
-		movq		[edx + 2],		xmm3
+		movq		[r0 + 2],		xmm3
 
+              
+        pop r3
+        
+        %assign push_num 0
 		ret

codec/processing/src/asm/downsample_bilinear.asm

@@ -39,8 +39,7 @@
 ;*
 ;*************************************************************************/
 %include "asm_inc.asm"
-BITS 32
+%ifdef X86_32
-
 ;***********************************************************************
 ; Macros and other preprocessor constants
 ;***********************************************************************
@@ -1223,3 +1222,4 @@ FAST_LAST_ROW_END:
 %undef		xInverse
 %undef		dstStep
 	ret
+%endif

codec/processing/src/asm/sad.asm

@@ -29,7 +29,7 @@
 ;*     POSSIBILITY OF SUCH DAMAGE.
 ;*
 ;*
-;*  pixel_sse2.asm
+;*  sad.asm
 ;*
 ;*  Abstract
 ;*      WelsSampleSad8x8_sse21
@@ -42,50 +42,26 @@
 
 %include "asm_inc.asm"
 
-BITS 32
-
 ;***********************************************************************
 ; Macros and other preprocessor constants
 ;***********************************************************************
-
-%macro SAD_8x4 0
-	movq   xmm0,   [eax]
-	movq   xmm1,   [eax+ebx]
-	lea    eax,    [eax+2*ebx]
-	movhps xmm0,   [eax]
-	movhps xmm1,   [eax+ebx]
-
-	movq   xmm2,   [ecx]
-	movq   xmm3,   [ecx+edx]
-	lea    ecx,    [ecx+2*edx]
-	movhps xmm2,   [ecx]
-	movhps xmm3,   [ecx+edx]
-	psadbw xmm0,   xmm2
-	psadbw xmm1,   xmm3
-	paddw  xmm6,   xmm0
-	paddw  xmm6,   xmm1
-%endmacro
-
-
-
 %macro CACHE_SPLIT_CHECK 3 ; address, width, cacheline
 and    %1,  0x1f|(%3>>1)
 cmp    %1,  (32-%2)|(%3>>1)
 %endmacro
 
-
 %macro SSE2_GetSad8x4 0
-	movq   xmm0,   [eax]
+	movq   xmm0,   [r0]
-	movq   xmm1,   [eax+ebx]
+	movq   xmm1,   [r0+r1]
-	lea    eax,    [eax+2*ebx]
+	lea    r0,     [r0+2*r1]
-	movhps xmm0,   [eax]
+	movhps xmm0,   [r0]
-	movhps xmm1,   [eax+ebx]
+	movhps xmm1,   [r0+r1]
 
-	movq   xmm2,   [ecx]
+	movq   xmm2,   [r2]
-	movq   xmm3,   [ecx+edx]
+	movq   xmm3,   [r2+r3]
-	lea    ecx,    [ecx+2*edx]
+	lea    r2,     [r2+2*r3]
-	movhps xmm2,   [ecx]
+	movhps xmm2,   [r2]
-	movhps xmm3,   [ecx+edx]
+	movhps xmm3,   [r2+r3]
 	psadbw xmm0,   xmm2
 	psadbw xmm1,   xmm3
 	paddw  xmm6,   xmm0
@@ -100,38 +76,55 @@ SECTION .text
 
 WELS_EXTERN WelsSampleSad8x8_sse21
 WelsSampleSad8x8_sse21:
-    mov    ecx,    [esp+12]
+    ;mov    ecx,    [esp+12]
-	mov    edx,    ecx
+	;mov    edx,    ecx
-    CACHE_SPLIT_CHECK edx, 8, 64
+    ;CACHE_SPLIT_CHECK edx, 8, 64
-	jle    near   .pixel_sad_8x8_nsplit
+	;jle    near   .pixel_sad_8x8_nsplit
-	push   ebx
+	;push   ebx
-	push   edi
+	;push   edi
-	mov    eax,    [esp+12]
+	;mov    eax,    [esp+12]
-	mov    ebx,    [esp+16]
+	;mov    ebx,    [esp+16]
 	
+	%assign  push_num 0
+	mov		r2,  arg3
+	push	r2
+	CACHE_SPLIT_CHECK r2, 8, 64
+	jle    near   .pixel_sad_8x8_nsplit
+	pop		r2
+%ifdef X86_32	
+	push	r3
+	push	r4
+	push	r5
+%endif
+	%assign  push_num 3
+	mov		r0,  arg1
+	mov		r1,  arg2	
+	SIGN_EXTENTION r1, r1d
     pxor   xmm7,   xmm7
     
-    mov    edi,    ecx
+    ;ecx r2, edx r4, edi r5
-    and    edi,    0x07
-    sub    ecx,    edi
-    mov    edx,    8
-    sub    edx,    edi
 
-    shl    edi,    3
+    mov    r5,    r2
-    shl    edx,    3
+    and    r5,    0x07
-    movd   xmm5,   edi
+    sub    r2,    r5
-    movd   xmm6,   edx
+    mov    r4,    8
-	mov    edi,    8
+    sub    r4,    r5
-	add    edi,    ecx
-    mov    edx,    [esp+24]
 
-    movq   xmm0,   [eax]
+    shl    r5,    3
-	movhps xmm0,   [eax+ebx]
+    shl    r4,    3
+    movd   xmm5,   r5d
+    movd   xmm6,   r4d
+	mov    r5,    8
+	add    r5,    r2
+    mov    r3,    arg4
+	SIGN_EXTENTION r3, r3d
+    movq   xmm0,   [r0]
+	movhps xmm0,   [r0+r1]
 
-	movq   xmm1,   [ecx]
+	movq   xmm1,   [r2]
-	movq   xmm2,   [edi]
+	movq   xmm2,   [r5]
-	movhps xmm1,   [ecx+edx]
+	movhps xmm1,   [r2+r3]
-	movhps xmm2,   [edi+edx]
+	movhps xmm2,   [r5+r3]
 	psrlq  xmm1,   xmm5
 	psllq  xmm2,   xmm6
 	por    xmm1,   xmm2
@@ -139,17 +132,17 @@ WelsSampleSad8x8_sse21:
 	psadbw xmm0,   xmm1
 	paddw  xmm7,   xmm0
 
-	lea    eax,    [eax+2*ebx]
+	lea    r0,    [r0+2*r1]
-	lea    ecx,    [ecx+2*edx]
+	lea    r2,    [r2+2*r3]
-	lea    edi,    [edi+2*edx]
+	lea    r5,    [r5+2*r3]
 
-    movq   xmm0,   [eax]
+    movq   xmm0,   [r0]
-	movhps xmm0,   [eax+ebx]
+	movhps xmm0,   [r0+r1]
 
-	movq   xmm1,   [ecx]
+	movq   xmm1,   [r2]
-	movq   xmm2,   [edi]
+	movq   xmm2,   [r5]
-	movhps xmm1,   [ecx+edx]
+	movhps xmm1,   [r2+r3]
-	movhps xmm2,   [edi+edx]
+	movhps xmm2,   [r5+r3]
 	psrlq  xmm1,   xmm5
 	psllq  xmm2,   xmm6
 	por    xmm1,   xmm2
@@ -157,17 +150,17 @@ WelsSampleSad8x8_sse21:
 	psadbw xmm0,   xmm1
 	paddw  xmm7,   xmm0
 
-	lea    eax,    [eax+2*ebx]
+	lea    r0,    [r0+2*r1]
-	lea    ecx,    [ecx+2*edx]
+	lea    r2,    [r2+2*r3]
-	lea    edi,    [edi+2*edx]
+	lea    r5,    [r5+2*r3]
 
-    movq   xmm0,   [eax]
+    movq   xmm0,   [r0]
-	movhps xmm0,   [eax+ebx]
+	movhps xmm0,   [r0+r1]
 
-	movq   xmm1,   [ecx]
+	movq   xmm1,   [r2]
-	movq   xmm2,   [edi]
+	movq   xmm2,   [r5]
-	movhps xmm1,   [ecx+edx]
+	movhps xmm1,   [r2+r3]
-	movhps xmm2,   [edi+edx]
+	movhps xmm2,   [r5+r3]
 	psrlq  xmm1,   xmm5
 	psllq  xmm2,   xmm6
 	por    xmm1,   xmm2
@@ -175,17 +168,17 @@ WelsSampleSad8x8_sse21:
 	psadbw xmm0,   xmm1
 	paddw  xmm7,   xmm0
 
-	lea    eax,    [eax+2*ebx]
+	lea    r0,    [r0+2*r1]
-	lea    ecx,    [ecx+2*edx]
+	lea    r2,    [r2+2*r3]
-	lea    edi,    [edi+2*edx]
+	lea    r5,    [r5+2*r3]
 
-    movq   xmm0,   [eax]
+    movq   xmm0,   [r0]
-	movhps xmm0,   [eax+ebx]
+	movhps xmm0,   [r0+r1]
 
-	movq   xmm1,   [ecx]
+	movq   xmm1,   [r2]
-	movq   xmm2,   [edi]
+	movq   xmm2,   [r5]
-	movhps xmm1,   [ecx+edx]
+	movhps xmm1,   [r2+r3]
-	movhps xmm2,   [edi+edx]
+	movhps xmm2,   [r5+r3]
 	psrlq  xmm1,   xmm5
 	psllq  xmm2,   xmm6
 	por    xmm1,   xmm2
@@ -195,22 +188,33 @@ WelsSampleSad8x8_sse21:
 
     movhlps    xmm0, xmm7
 	paddw      xmm0, xmm7
-	movd       eax,  xmm0
+	movd       retrd,  xmm0
-	pop        edi
+%ifdef X86_32
+	pop	 r5
+	pop	 r4
+	pop	 r3
+%endif
 	jmp        .return
+	
 .pixel_sad_8x8_nsplit:
-    push   ebx
+    ;push   ebx
-    mov    eax,    [esp+8]
+    ;mov    eax,    [esp+8]
-	mov    ebx,    [esp+12]
+	;mov    ebx,    [esp+12]
-	mov    edx,    [esp+20]
+	;mov    edx,    [esp+20]
+	
+	pop r2
+	%assign  push_num 0
+	LOAD_4_PARA
+	SIGN_EXTENTION r1, r1d
+	SIGN_EXTENTION r3, r3d	
 	pxor   xmm6,   xmm6
 	SSE2_GetSad8x4
-    lea    eax,    [eax+2*ebx]
+    lea    r0,    [r0+2*r1]
-	lea    ecx,    [ecx+2*edx]
+	lea    r2,    [r2+2*r3]
     SSE2_GetSad8x4
     movhlps    xmm0, xmm6
 	paddw      xmm0, xmm6
-	movd       eax,  xmm0
+	movd       retrd,  xmm0
+	LOAD_4_PARA_POP
 .return:
-	pop        ebx
 	ret

codec/processing/src/asm/vaa.asm

@@ -39,8 +39,7 @@
 ;*
 ;*************************************************************************/
 %include "asm_inc.asm"
-BITS 32
+%ifdef X86_32
-
 ;***********************************************************************
 ; Macros and other preprocessor constants
 ;***********************************************************************
@@ -481,181 +480,6 @@ SampleVariance16x16_sse2:
 
 ; , 6/7/2010
 
-%ifndef NO_DYNAMIC_VP
-WELS_EXTERN AnalysisVaaInfoIntra_sse2
-;***********************************************************************
-;	int32_t AnalysisVaaInfoIntra_sse2(	uint8_t *pDataY, const int32_t linesize );
-;***********************************************************************
-ALIGN 16
-AnalysisVaaInfoIntra_sse2:
-	push ebx
-	push edx
-	push esi
-	push edi
-	push ebp
-
-	mov ebp, esp
-	and ebp, 0fh
-	sub esp, ebp
-	sub esp, 32
-	%define PUSH_SIZE	52	; 20 + 32
-
-	mov esi, [esp+ebp+PUSH_SIZE+4]	; data_y
-	mov ecx, [esp+ebp+PUSH_SIZE+8]	; linesize
-
-	mov ebx, ecx
-	sal ebx, $1			; linesize x 2 [ebx]
-	mov edx, ebx
-	add edx, ecx		; linesize x 3 [edx]
-	mov eax, ebx
-	sal eax, $1			; linesize x 4 [eax]
-
-	pxor xmm7, xmm7
-
-	; loops
-	VAA_AVG_BLOCK_SSE2 xmm0, xmm1, xmm2, xmm3, xmm4, xmm5
-	movq [esp], xmm0
-
-	lea esi, [esi+eax]
-	VAA_AVG_BLOCK_SSE2 xmm0, xmm1, xmm2, xmm3, xmm4, xmm5
-	movq [esp+8], xmm0
-
-	lea esi, [esi+eax]
-	VAA_AVG_BLOCK_SSE2 xmm0, xmm1, xmm2, xmm3, xmm4, xmm5
-	movq [esp+16], xmm0
-
-	lea esi, [esi+eax]
-	VAA_AVG_BLOCK_SSE2 xmm0, xmm1, xmm2, xmm3, xmm4, xmm5
-	movq [esp+24], xmm0
-
-	movdqa xmm0, [esp]		; block 0~7
-	movdqa xmm1, [esp+16]	; block 8~15
-	movdqa xmm2, xmm0
-	paddw xmm0, xmm1
-	SUM_WORD_8x2_SSE2 xmm0, xmm3
-
-	pmullw xmm1, xmm1
-	pmullw xmm2, xmm2
-	movdqa xmm3, xmm1
-	movdqa xmm4, xmm2
-	punpcklwd xmm1, xmm7
-	punpckhwd xmm3, xmm7
-	punpcklwd xmm2, xmm7
-	punpckhwd xmm4, xmm7
-	paddd xmm1, xmm2
-	paddd xmm3, xmm4
-	paddd xmm1, xmm3
-	pshufd xmm2, xmm1, 01Bh
-	paddd xmm1, xmm2
-	pshufd xmm2, xmm1, 0B1h
-	paddd xmm1, xmm2
-
-	movd ebx, xmm0
-	and ebx, 0ffffh		; effective low word truncated
-	mov ecx, ebx
-	imul ebx, ecx
-	sar ebx, $4
-	movd eax, xmm1
-	sub eax, ebx
-
-	%undef PUSH_SIZE
-	add esp, 32
-	add esp, ebp
-	pop ebp
-	pop edi
-	pop esi
-	pop edx
-	pop ebx
-	ret
-
-WELS_EXTERN AnalysisVaaInfoIntra_ssse3
-;***********************************************************************
-;	int32_t AnalysisVaaInfoIntra_ssse3(	uint8_t *pDataY, const int32_t linesize );
-;***********************************************************************
-ALIGN 16
-AnalysisVaaInfoIntra_ssse3:
-	push ebx
-	push edx
-	push esi
-	push edi
-	push ebp
-
-	mov ebp, esp
-	and ebp, 0fh
-	sub esp, ebp
-	sub esp, 32
-	%define PUSH_SIZE	52	; 20 + 32
-
-	mov esi, [esp+ebp+PUSH_SIZE+4]	; data_y
-	mov ecx, [esp+ebp+PUSH_SIZE+8]	; linesize
-
-	mov ebx, ecx
-	sal ebx, $1			; linesize x 2 [ebx]
-	mov edx, ebx
-	add edx, ecx		; linesize x 3 [edx]
-	mov eax, ebx
-	sal eax, $1			; linesize x 4 [eax]
-
-	pxor xmm7, xmm7
-
-	; loops
-	VAA_AVG_BLOCK_SSSE3 xmm0, xmm1, xmm2, xmm3, xmm4, xmm5
-	movq [esp], xmm0
-
-	lea esi, [esi+eax]
-	VAA_AVG_BLOCK_SSSE3 xmm1, xmm2, xmm3, xmm4, xmm5, xmm6
-	movq [esp+8], xmm1
-
-	lea esi, [esi+eax]
-	VAA_AVG_BLOCK_SSSE3 xmm0, xmm1, xmm2, xmm3, xmm4, xmm5
-	movq [esp+16], xmm0
-
-	lea esi, [esi+eax]
-	VAA_AVG_BLOCK_SSSE3 xmm1, xmm2, xmm3, xmm4, xmm5, xmm6
-	movq [esp+24], xmm1
-
-	movdqa xmm0, [esp]		; block 0~7
-	movdqa xmm1, [esp+16]	; block 8~15
-	movdqa xmm2, xmm0
-	paddw xmm0, xmm1
-	SUM_WORD_8x2_SSE2 xmm0, xmm3	; better performance than that of phaddw sets
-
-	pmullw xmm1, xmm1
-	pmullw xmm2, xmm2
-	movdqa xmm3, xmm1
-	movdqa xmm4, xmm2
-	punpcklwd xmm1, xmm7
-	punpckhwd xmm3, xmm7
-	punpcklwd xmm2, xmm7
-	punpckhwd xmm4, xmm7
-	paddd xmm1, xmm2
-	paddd xmm3, xmm4
-	paddd xmm1, xmm3
-	pshufd xmm2, xmm1, 01Bh
-	paddd xmm1, xmm2
-	pshufd xmm2, xmm1, 0B1h
-	paddd xmm1, xmm2
-
-	movd ebx, xmm0
-	and ebx, 0ffffh		; effective low work truncated
-	mov ecx, ebx
-	imul ebx, ecx
-	sar ebx, $4
-	movd eax, xmm1
-	sub eax, ebx
-
-	%undef PUSH_SIZE
-	add esp, 32
-	add esp, ebp
-	pop ebp
-	pop edi
-	pop esi
-	pop edx
-	pop ebx
-	ret
-%endif
-
-
 
 WELS_EXTERN abs_difference_mbrow_sse2
 ;*************************************************************************************************************
@@ -1587,3 +1411,4 @@ sqdiff_bgd_width_loop:
 %undef		pushsize
 %undef		localsize
 	ret
+%endif