From febdd0b9b414b2adfe4c3ea2fa06ded2e32e510b Mon Sep 17 00:00:00 2001
From: Michael Niedermayer <michaelni@gmx.at>
Date: Fri, 21 Dec 2007 10:16:22 +0000
Subject: [PATCH] ~15% faster h264_chroma_mc2/4_c() these also prevent some
 possible out of array reads.

Originally committed as revision 11290 to svn://svn.ffmpeg.org/ffmpeg/trunk
---
 libavcodec/dsputil.c | 26 ++++++++++++++++++++++++++
 1 file changed, 26 insertions(+)

diff --git a/libavcodec/dsputil.c b/libavcodec/dsputil.c
index e3b5db74792..523921ec179 100644
--- a/libavcodec/dsputil.c
+++ b/libavcodec/dsputil.c
@@ -1440,6 +1440,7 @@ static void OPNAME ## h264_chroma_mc2_c(uint8_t *dst/*align 8*/, uint8_t *src/*a
     \
     assert(x<8 && y<8 && x>=0 && y>=0);\
 \
+    if(D){\
     for(i=0; i<h; i++)\
     {\
         OP(dst[0], (A*src[0] + B*src[1] + C*src[stride+0] + D*src[stride+1]));\
@@ -1447,6 +1448,17 @@ static void OPNAME ## h264_chroma_mc2_c(uint8_t *dst/*align 8*/, uint8_t *src/*a
         dst+= stride;\
         src+= stride;\
     }\
+    }else{\
+        const int E= B+C;\
+        const int step= C ? stride : 1;\
+        for(i=0; i<h; i++)\
+        {\
+            OP(dst[0], (A*src[0] + E*src[step+0]));\
+            OP(dst[1], (A*src[1] + E*src[step+1]));\
+            dst+= stride;\
+            src+= stride;\
+        }\
+    }\
 }\
 \
 static void OPNAME ## h264_chroma_mc4_c(uint8_t *dst/*align 8*/, uint8_t *src/*align 1*/, int stride, int h, int x, int y){\
@@ -1458,6 +1470,7 @@ static void OPNAME ## h264_chroma_mc4_c(uint8_t *dst/*align 8*/, uint8_t *src/*a
     \
     assert(x<8 && y<8 && x>=0 && y>=0);\
 \
+    if(D){\
     for(i=0; i<h; i++)\
     {\
         OP(dst[0], (A*src[0] + B*src[1] + C*src[stride+0] + D*src[stride+1]));\
@@ -1467,6 +1480,19 @@ static void OPNAME ## h264_chroma_mc4_c(uint8_t *dst/*align 8*/, uint8_t *src/*a
         dst+= stride;\
         src+= stride;\
     }\
+    }else{\
+        const int E= B+C;\
+        const int step= C ? stride : 1;\
+        for(i=0; i<h; i++)\
+        {\
+            OP(dst[0], (A*src[0] + E*src[step+0]));\
+            OP(dst[1], (A*src[1] + E*src[step+1]));\
+            OP(dst[2], (A*src[2] + E*src[step+2]));\
+            OP(dst[3], (A*src[3] + E*src[step+3]));\
+            dst+= stride;\
+            src+= stride;\
+        }\
+    }\
 }\
 \
 static void OPNAME ## h264_chroma_mc8_c(uint8_t *dst/*align 8*/, uint8_t *src/*align 1*/, int stride, int h, int x, int y){\
-- 
2.39.2