[x265] [PATCH] added asm code blockcopy_ps_4x16 and invoked function pointer initialization with macro

praveen at multicorewareinc.com praveen at multicorewareinc.com
Mon Nov 11 16:06:04 CET 2013


# HG changeset patch
# User Praveen Tiwari
# Date 1384182355 -19800
# Node ID 5d12397bc1572588a5f69699d7910eaf5c096b06
# Parent  e39f9ba2f96fe065dd676456f07832e8fbe8ae2d
added asm code blockcopy_ps_4x16 and invoked function pointer initialization with macro

diff -r e39f9ba2f96f -r 5d12397bc157 source/common/x86/asm-primitives.cpp
--- a/source/common/x86/asm-primitives.cpp	Mon Nov 11 20:24:33 2013 +0530
+++ b/source/common/x86/asm-primitives.cpp	Mon Nov 11 20:35:55 2013 +0530
@@ -128,7 +128,8 @@
 
 #define SETUP_CHROMA_FUNC_DEF(W, H, cpu) \
     p.chroma_hpp[CHROMA_ ## W ## x ## H] = x265_interp_4tap_horiz_pp_ ## W ## x ## H ## cpu; \
-    p.chroma_vpp[CHROMA_ ## W ## x ## H] = x265_interp_4tap_vert_pp_ ## W ## x ## H ## cpu;
+    p.chroma_vpp[CHROMA_ ## W ## x ## H] = x265_interp_4tap_vert_pp_ ## W ## x ## H ## cpu; \
+    p.chroma_copy_ps[CHROMA_ ## W ## x ## H] = x265_blockcopy_ps_ ## W ## x ## H ## cpu;
 
 #define SETUP_CHROMA_BLOCKCOPY_FUNC_DEF(W, H, cpu) \
     p.chroma_copy_pp[CHROMA_ ## W ## x ## H] = x265_blockcopy_pp_ ## W ## x ## H ## cpu;
@@ -450,35 +451,6 @@
         p.chroma_copy_sp[CHROMA_2x8] = x265_blockcopy_sp_2x8_sse4;
         p.chroma_copy_sp[CHROMA_6x8] = x265_blockcopy_sp_6x8_sse4;
 
-        // This function pointer initialization is temporary will be removed
-        // later with macro definitions.  It is used to avoid linker errors
-        // until all partitions are coded and commit smaller patches, easier to
-        // review.
-
-        p.chroma_copy_ps[CHROMA_2x4] = x265_blockcopy_ps_2x4_sse4;
-        p.chroma_copy_ps[CHROMA_2x8] = x265_blockcopy_ps_2x8_sse4;
-        p.chroma_copy_ps[CHROMA_4x2] = x265_blockcopy_ps_4x2_sse4;
-        p.chroma_copy_ps[CHROMA_4x4] = x265_blockcopy_ps_4x4_sse4;
-        p.chroma_copy_ps[CHROMA_4x8] = x265_blockcopy_ps_4x8_sse4;
-        p.chroma_copy_ps[CHROMA_6x8] = x265_blockcopy_ps_6x8_sse4;
-        p.chroma_copy_ps[CHROMA_8x2] = x265_blockcopy_ps_8x2_sse4;
-        p.chroma_copy_ps[CHROMA_8x4] = x265_blockcopy_ps_8x4_sse4;
-        p.chroma_copy_ps[CHROMA_8x6] = x265_blockcopy_ps_8x6_sse4;
-        p.chroma_copy_ps[CHROMA_8x8] = x265_blockcopy_ps_8x8_sse4;
-        p.chroma_copy_ps[CHROMA_8x16] = x265_blockcopy_ps_8x16_sse4;
-        p.chroma_copy_ps[CHROMA_8x32] = x265_blockcopy_ps_8x32_sse4;
-        p.chroma_copy_ps[CHROMA_12x16] = x265_blockcopy_ps_12x16_sse4;
-        p.chroma_copy_ps[CHROMA_16x4] = x265_blockcopy_ps_16x4_sse4;
-        p.chroma_copy_ps[CHROMA_16x8] = x265_blockcopy_ps_16x8_sse4;
-        p.chroma_copy_ps[CHROMA_16x12] = x265_blockcopy_ps_16x12_sse4;
-        p.chroma_copy_ps[CHROMA_16x16] = x265_blockcopy_ps_16x16_sse4;
-        p.chroma_copy_ps[CHROMA_16x32] = x265_blockcopy_ps_16x32_sse4;
-        p.chroma_copy_ps[CHROMA_24x32] = x265_blockcopy_ps_24x32_sse4;
-        p.chroma_copy_ps[CHROMA_32x8] = x265_blockcopy_ps_32x8_sse4;
-        p.chroma_copy_ps[CHROMA_32x16] = x265_blockcopy_ps_32x16_sse4;
-        p.chroma_copy_ps[CHROMA_32x24] = x265_blockcopy_ps_32x24_sse4;
-        p.chroma_copy_ps[CHROMA_32x32] = x265_blockcopy_ps_32x32_sse4;
-
     }
     if (cpuMask & X265_CPU_AVX)
     {
diff -r e39f9ba2f96f -r 5d12397bc157 source/common/x86/blockcopy8.asm
--- a/source/common/x86/blockcopy8.asm	Mon Nov 11 20:24:33 2013 +0530
+++ b/source/common/x86/blockcopy8.asm	Mon Nov 11 20:35:55 2013 +0530
@@ -1829,7 +1829,7 @@
 %endmacro
 
 BLOCKCOPY_PS_W4_H4 4, 8
-
+BLOCKCOPY_PS_W4_H4 4, 16
 
 ;-----------------------------------------------------------------------------
 ; void blockcopy_ps_%1x%2(int16_t *dest, intptr_t destStride, pixel *src, intptr_t srcStride);
diff -r e39f9ba2f96f -r 5d12397bc157 source/common/x86/blockcopy8.h
--- a/source/common/x86/blockcopy8.h	Mon Nov 11 20:24:33 2013 +0530
+++ b/source/common/x86/blockcopy8.h	Mon Nov 11 20:35:55 2013 +0530
@@ -99,6 +99,7 @@
     SETUP_CHROMA_BLOCKCOPY_FUNC_SSE4(4, 2, cpu); \
     SETUP_CHROMA_BLOCKCOPY_FUNC_SSE4(4, 4, cpu); \
     SETUP_CHROMA_BLOCKCOPY_FUNC_SSE4(4, 8, cpu); \
+    SETUP_CHROMA_BLOCKCOPY_FUNC_SSE4(4, 16, cpu); \
     SETUP_CHROMA_BLOCKCOPY_FUNC_SSE4(6, 8, cpu); \
     SETUP_CHROMA_BLOCKCOPY_FUNC_SSE4(8, 2, cpu); \
     SETUP_CHROMA_BLOCKCOPY_FUNC_SSE4(8, 4, cpu); \


More information about the x265-devel mailing list