[FFmpeg-devel] [PATCH] rename the remaining instances of HAVE_MMX2 to HAVE_MMXEXT
Derek Buitenhuis
derek.buitenhuis at gmail.com
Sat Aug 18 22:15:43 CEST 2012
From: XhmikosR <xhmikosr at users.sourceforge.net>
---
configure | 1 -
libavfilter/libmpcodecs/libvo/fastmemcpy.h | 2 +-
libavfilter/libmpcodecs/vf_fspp.c | 2 +-
libavfilter/libmpcodecs/vf_noise.c | 6 +-
libavfilter/libmpcodecs/vf_pp7.c | 2 +-
libavfilter/libmpcodecs/vf_spp.c | 2 +-
libavfilter/libmpcodecs/vf_unsharp.c | 2 +-
libavfilter/libmpcodecs/vf_uspp.c | 2 +-
libavutil/utils.c | 1 -
libpostproc/postprocess.c | 22 +++----
libpostproc/postprocess_template.c | 98 ++++++++++++++--------------
libswscale/swscale.c | 2 +-
libswscale/x86/swscale.c | 2 +-
13 files changed, 71 insertions(+), 73 deletions(-)
diff --git a/configure b/configure
index 679fae0..695ad1c 100755
--- a/configure
+++ b/configure
@@ -4072,7 +4072,6 @@ cat > $TMPH <<EOF
#define EXTERN_PREFIX "${extern_prefix}"
#define EXTERN_ASM ${extern_prefix}
#define SLIBSUF "$SLIBSUF"
-#define HAVE_MMX2 HAVE_MMXEXT
EOF
test -n "$assert_level" &&
diff --git a/libavfilter/libmpcodecs/libvo/fastmemcpy.h b/libavfilter/libmpcodecs/libvo/fastmemcpy.h
index 5a17d01..f1a9127 100644
--- a/libavfilter/libmpcodecs/libvo/fastmemcpy.h
+++ b/libavfilter/libmpcodecs/libvo/fastmemcpy.h
@@ -26,7 +26,7 @@
void * fast_memcpy(void * to, const void * from, size_t len);
void * mem2agpcpy(void * to, const void * from, size_t len);
-#if ! defined(CONFIG_FASTMEMCPY) || ! (HAVE_MMX || HAVE_MMX2 || HAVE_AMD3DNOW /* || HAVE_SSE || HAVE_SSE2 */)
+#if ! defined(CONFIG_FASTMEMCPY) || ! (HAVE_MMX || HAVE_MMXEXT || HAVE_AMD3DNOW /* || HAVE_SSE || HAVE_SSE2 */)
#define mem2agpcpy(a,b,c) memcpy(a,b,c)
#define fast_memcpy(a,b,c) memcpy(a,b,c)
#endif
diff --git a/libavfilter/libmpcodecs/vf_fspp.c b/libavfilter/libmpcodecs/vf_fspp.c
index 3653187..2ef7e19 100644
--- a/libavfilter/libmpcodecs/vf_fspp.c
+++ b/libavfilter/libmpcodecs/vf_fspp.c
@@ -566,7 +566,7 @@ static int put_image(struct vf_instance *vf, mp_image_t *mpi, double pts)
#if HAVE_MMX
if(gCpuCaps.hasMMX) __asm__ volatile ("emms\n\t");
#endif
-#if HAVE_MMX2
+#if HAVE_MMXEXT
if(gCpuCaps.hasMMX2) __asm__ volatile ("sfence\n\t");
#endif
return vf_next_put_image(vf,dmpi, pts);
diff --git a/libavfilter/libmpcodecs/vf_noise.c b/libavfilter/libmpcodecs/vf_noise.c
index 9521619..9218ff8 100644
--- a/libavfilter/libmpcodecs/vf_noise.c
+++ b/libavfilter/libmpcodecs/vf_noise.c
@@ -175,7 +175,7 @@ static inline void lineNoise_MMX(uint8_t *dst, uint8_t *src, int8_t *noise, int
#endif
//duplicate of previous except movntq
-#if HAVE_MMX2
+#if HAVE_MMXEXT
static inline void lineNoise_MMX2(uint8_t *dst, uint8_t *src, int8_t *noise, int len, int shift){
x86_reg mmx_len= len&(~7);
noise+=shift;
@@ -360,7 +360,7 @@ static int put_image(struct vf_instance *vf, mp_image_t *mpi, double pts){
#if HAVE_MMX
if(gCpuCaps.hasMMX) __asm__ volatile ("emms\n\t");
#endif
-#if HAVE_MMX2
+#if HAVE_MMXEXT
if(gCpuCaps.hasMMX2) __asm__ volatile ("sfence\n\t");
#endif
@@ -454,7 +454,7 @@ static int vf_open(vf_instance_t *vf, char *args){
lineNoiseAvg= lineNoiseAvg_MMX;
}
#endif
-#if HAVE_MMX2
+#if HAVE_MMXEXT
if(gCpuCaps.hasMMX2) lineNoise= lineNoise_MMX2;
// if(gCpuCaps.hasMMX) lineNoiseAvg= lineNoiseAvg_MMX2;
#endif
diff --git a/libavfilter/libmpcodecs/vf_pp7.c b/libavfilter/libmpcodecs/vf_pp7.c
index eae30bf..9bcc7ab 100644
--- a/libavfilter/libmpcodecs/vf_pp7.c
+++ b/libavfilter/libmpcodecs/vf_pp7.c
@@ -402,7 +402,7 @@ static int put_image(struct vf_instance *vf, mp_image_t *mpi, double pts){
#if HAVE_MMX
if(gCpuCaps.hasMMX) __asm__ volatile ("emms\n\t");
#endif
-#if HAVE_MMX2
+#if HAVE_MMXEXT
if(gCpuCaps.hasMMX2) __asm__ volatile ("sfence\n\t");
#endif
diff --git a/libavfilter/libmpcodecs/vf_spp.c b/libavfilter/libmpcodecs/vf_spp.c
index 0b4b230..e747d74 100644
--- a/libavfilter/libmpcodecs/vf_spp.c
+++ b/libavfilter/libmpcodecs/vf_spp.c
@@ -510,7 +510,7 @@ static int put_image(struct vf_instance *vf, mp_image_t *mpi, double pts){
#if HAVE_MMX
if(gCpuCaps.hasMMX) __asm__ volatile ("emms\n\t");
#endif
-#if HAVE_MMX2
+#if HAVE_MMXEXT
if(gCpuCaps.hasMMX2) __asm__ volatile ("sfence\n\t");
#endif
diff --git a/libavfilter/libmpcodecs/vf_unsharp.c b/libavfilter/libmpcodecs/vf_unsharp.c
index db22f78..75d6e5b 100644
--- a/libavfilter/libmpcodecs/vf_unsharp.c
+++ b/libavfilter/libmpcodecs/vf_unsharp.c
@@ -196,7 +196,7 @@ static int put_image( struct vf_instance *vf, mp_image_t *mpi, double pts) {
if(gCpuCaps.hasMMX)
__asm__ volatile ("emms\n\t");
#endif
-#if HAVE_MMX2
+#if HAVE_MMXEXT
if(gCpuCaps.hasMMX2)
__asm__ volatile ("sfence\n\t");
#endif
diff --git a/libavfilter/libmpcodecs/vf_uspp.c b/libavfilter/libmpcodecs/vf_uspp.c
index cc9a3db..239f671 100644
--- a/libavfilter/libmpcodecs/vf_uspp.c
+++ b/libavfilter/libmpcodecs/vf_uspp.c
@@ -291,7 +291,7 @@ static int put_image(struct vf_instance *vf, mp_image_t *mpi, double pts){
#if HAVE_MMX
if(gCpuCaps.hasMMX) __asm__ volatile ("emms\n\t");
#endif
-#if HAVE_MMX2
+#if HAVE_MMXEXT
if(gCpuCaps.hasMMX2) __asm__ volatile ("sfence\n\t");
#endif
diff --git a/libavutil/utils.c b/libavutil/utils.c
index 127c4b1..971b48b 100644
--- a/libavutil/utils.c
+++ b/libavutil/utils.c
@@ -33,7 +33,6 @@ unsigned avutil_version(void)
av_assert0(AVMEDIA_TYPE_ATTACHMENT == 4);
av_assert0(AV_PICTURE_TYPE_BI == 7);
av_assert0(LIBAVUTIL_VERSION_MICRO >= 100);
- av_assert0(HAVE_MMX2 == HAVE_MMXEXT);
return LIBAVUTIL_VERSION_INT;
}
diff --git a/libpostproc/postprocess.c b/libpostproc/postprocess.c
index 2234ba8..3c3f68c 100644
--- a/libpostproc/postprocess.c
+++ b/libpostproc/postprocess.c
@@ -80,7 +80,7 @@ try to unroll inner for(x=0 ... loop to avoid these damn if(x ... checks
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
-//#undef HAVE_MMX2
+//#undef HAVE_MMXEXT
//#define HAVE_AMD3DNOW
//#undef HAVE_MMX
//#undef ARCH_X86
@@ -546,23 +546,23 @@ static av_always_inline void do_a_deblock_C(uint8_t *src, int step, int stride,
#if ARCH_X86
-#if (HAVE_MMX && !HAVE_AMD3DNOW && !HAVE_MMX2) || CONFIG_RUNTIME_CPUDETECT
+#if (HAVE_MMX && !HAVE_AMD3DNOW && !HAVE_MMXEXT) || CONFIG_RUNTIME_CPUDETECT
#define COMPILE_MMX
#endif
-#if HAVE_MMX2 || CONFIG_RUNTIME_CPUDETECT
+#if HAVE_MMXEXT || CONFIG_RUNTIME_CPUDETECT
#define COMPILE_MMX2
#endif
-#if (HAVE_AMD3DNOW && !HAVE_MMX2) || CONFIG_RUNTIME_CPUDETECT
+#if (HAVE_AMD3DNOW && !HAVE_MMXEXT) || CONFIG_RUNTIME_CPUDETECT
#define COMPILE_3DNOW
#endif
#endif /* ARCH_X86 */
#undef HAVE_MMX
#define HAVE_MMX 0
-#undef HAVE_MMX2
-#define HAVE_MMX2 0
+#undef HAVE_MMXEXT
+#define HAVE_MMXEXT 0
#undef HAVE_AMD3DNOW
#define HAVE_AMD3DNOW 0
#undef HAVE_ALTIVEC
@@ -595,9 +595,9 @@ static av_always_inline void do_a_deblock_C(uint8_t *src, int step, int stride,
#ifdef COMPILE_MMX2
#undef RENAME
#undef HAVE_MMX
-#undef HAVE_MMX2
+#undef HAVE_MMXEXT
#define HAVE_MMX 1
-#define HAVE_MMX2 1
+#define HAVE_MMXEXT 1
#define RENAME(a) a ## _MMX2
#include "postprocess_template.c"
#endif
@@ -606,10 +606,10 @@ static av_always_inline void do_a_deblock_C(uint8_t *src, int step, int stride,
#ifdef COMPILE_3DNOW
#undef RENAME
#undef HAVE_MMX
-#undef HAVE_MMX2
+#undef HAVE_MMXEXT
#undef HAVE_AMD3DNOW
#define HAVE_MMX 1
-#define HAVE_MMX2 0
+#define HAVE_MMXEXT 0
#define HAVE_AMD3DNOW 1
#define RENAME(a) a ## _3DNow
#include "postprocess_template.c"
@@ -652,7 +652,7 @@ static inline void postProcess(const uint8_t src[], int srcStride, uint8_t dst[]
postProcess_C(src, srcStride, dst, dstStride, width, height, QPs, QPStride, isColor, c);
#endif
#else /* CONFIG_RUNTIME_CPUDETECT */
-#if HAVE_MMX2
+#if HAVE_MMXEXT
postProcess_MMX2(src, srcStride, dst, dstStride, width, height, QPs, QPStride, isColor, c);
#elif HAVE_AMD3DNOW
postProcess_3DNow(src, srcStride, dst, dstStride, width, height, QPs, QPStride, isColor, c);
diff --git a/libpostproc/postprocess_template.c b/libpostproc/postprocess_template.c
index 0fb2da4..a6752df 100644
--- a/libpostproc/postprocess_template.c
+++ b/libpostproc/postprocess_template.c
@@ -30,14 +30,14 @@
#undef PMINUB
#undef PMAXUB
-#if HAVE_MMX2
+#if HAVE_MMXEXT
#define REAL_PAVGB(a,b) "pavgb " #a ", " #b " \n\t"
#elif HAVE_AMD3DNOW
#define REAL_PAVGB(a,b) "pavgusb " #a ", " #b " \n\t"
#endif
#define PAVGB(a,b) REAL_PAVGB(a,b)
-#if HAVE_MMX2
+#if HAVE_MMXEXT
#define PMINUB(a,b,t) "pminub " #a ", " #b " \n\t"
#elif HAVE_MMX
#define PMINUB(b,a,t) \
@@ -46,7 +46,7 @@
"psubb " #t ", " #a " \n\t"
#endif
-#if HAVE_MMX2
+#if HAVE_MMXEXT
#define PMAXUB(a,b) "pmaxub " #a ", " #b " \n\t"
#elif HAVE_MMX
#define PMAXUB(a,b) \
@@ -135,7 +135,7 @@ static inline int RENAME(vertClassify)(uint8_t src[], int stride, PPContext *c){
"psubusb %%mm3, %%mm4 \n\t"
" \n\t"
-#if HAVE_MMX2
+#if HAVE_MMXEXT
"pxor %%mm7, %%mm7 \n\t"
"psadbw %%mm7, %%mm0 \n\t"
#else
@@ -178,7 +178,7 @@ static inline int RENAME(vertClassify)(uint8_t src[], int stride, PPContext *c){
#if !HAVE_ALTIVEC
static inline void RENAME(doVertLowPass)(uint8_t *src, int stride, PPContext *c)
{
-#if HAVE_MMX2 || HAVE_AMD3DNOW
+#if HAVE_MMXEXT || HAVE_AMD3DNOW
src+= stride*3;
__asm__ volatile( //"movv %0 %1 %2\n\t"
"movq %2, %%mm0 \n\t" // QP,..., QP
@@ -305,7 +305,7 @@ static inline void RENAME(doVertLowPass)(uint8_t *src, int stride, PPContext *c)
: "r" (src), "r" ((x86_reg)stride), "m" (c->pQPb)
: "%"REG_a, "%"REG_c
);
-#else //HAVE_MMX2 || HAVE_AMD3DNOW
+#else //HAVE_MMXEXT || HAVE_AMD3DNOW
const int l1= stride;
const int l2= stride + l1;
const int l3= stride + l2;
@@ -344,7 +344,7 @@ static inline void RENAME(doVertLowPass)(uint8_t *src, int stride, PPContext *c)
src++;
}
-#endif //HAVE_MMX2 || HAVE_AMD3DNOW
+#endif //HAVE_MMXEXT || HAVE_AMD3DNOW
}
#endif //HAVE_ALTIVEC
@@ -357,7 +357,7 @@ static inline void RENAME(doVertLowPass)(uint8_t *src, int stride, PPContext *c)
*/
static inline void RENAME(vertX1Filter)(uint8_t *src, int stride, PPContext *co)
{
-#if HAVE_MMX2 || HAVE_AMD3DNOW
+#if HAVE_MMXEXT || HAVE_AMD3DNOW
src+= stride*3;
__asm__ volatile(
@@ -443,7 +443,7 @@ static inline void RENAME(vertX1Filter)(uint8_t *src, int stride, PPContext *co)
: "r" (src), "r" ((x86_reg)stride), "m" (co->pQPb)
: "%"REG_a, "%"REG_c
);
-#else //HAVE_MMX2 || HAVE_AMD3DNOW
+#else //HAVE_MMXEXT || HAVE_AMD3DNOW
const int l1= stride;
const int l2= stride + l1;
@@ -477,13 +477,13 @@ static inline void RENAME(vertX1Filter)(uint8_t *src, int stride, PPContext *co)
}
src++;
}
-#endif //HAVE_MMX2 || HAVE_AMD3DNOW
+#endif //HAVE_MMXEXT || HAVE_AMD3DNOW
}
#if !HAVE_ALTIVEC
static inline void RENAME(doVertDefFilter)(uint8_t src[], int stride, PPContext *c)
{
-#if HAVE_MMX2 || HAVE_AMD3DNOW
+#if HAVE_MMXEXT || HAVE_AMD3DNOW
/*
uint8_t tmp[16];
const int l1= stride;
@@ -872,7 +872,7 @@ static inline void RENAME(doVertDefFilter)(uint8_t src[], int stride, PPContext
"movq (%3), %%mm2 \n\t" // 2L0 - 5L1 + 5L2 - 2L3
"movq 8(%3), %%mm3 \n\t" // 2H0 - 5H1 + 5H2 - 2H3
-#if HAVE_MMX2
+#if HAVE_MMXEXT
"movq %%mm7, %%mm6 \n\t" // 0
"psubw %%mm0, %%mm6 \n\t"
"pmaxsw %%mm6, %%mm0 \n\t" // |2L4 - 5L5 + 5L6 - 2L7|
@@ -904,7 +904,7 @@ static inline void RENAME(doVertDefFilter)(uint8_t src[], int stride, PPContext
"psubw %%mm6, %%mm3 \n\t" // |2H0 - 5H1 + 5H2 - 2H3|
#endif
-#if HAVE_MMX2
+#if HAVE_MMXEXT
"pminsw %%mm2, %%mm0 \n\t"
"pminsw %%mm3, %%mm1 \n\t"
#else
@@ -968,7 +968,7 @@ static inline void RENAME(doVertDefFilter)(uint8_t src[], int stride, PPContext
"pand %%mm2, %%mm4 \n\t"
"pand %%mm3, %%mm5 \n\t"
-#if HAVE_MMX2
+#if HAVE_MMXEXT
"pminsw %%mm0, %%mm4 \n\t"
"pminsw %%mm1, %%mm5 \n\t"
#else
@@ -995,7 +995,7 @@ static inline void RENAME(doVertDefFilter)(uint8_t src[], int stride, PPContext
: "r" ((x86_reg)stride), "m" (c->pQPb), "r"(tmp)
: "%"REG_a
);
-#else //HAVE_MMX2 || HAVE_AMD3DNOW
+#else //HAVE_MMXEXT || HAVE_AMD3DNOW
const int l1= stride;
const int l2= stride + l1;
const int l3= stride + l2;
@@ -1033,14 +1033,14 @@ static inline void RENAME(doVertDefFilter)(uint8_t src[], int stride, PPContext
}
src++;
}
-#endif //HAVE_MMX2 || HAVE_AMD3DNOW
+#endif //HAVE_MMXEXT || HAVE_AMD3DNOW
}
#endif //HAVE_ALTIVEC
#if !HAVE_ALTIVEC
static inline void RENAME(dering)(uint8_t src[], int stride, PPContext *c)
{
-#if HAVE_MMX2 || HAVE_AMD3DNOW
+#if HAVE_MMXEXT || HAVE_AMD3DNOW
DECLARE_ALIGNED(8, uint64_t, tmp)[3];
__asm__ volatile(
"pxor %%mm6, %%mm6 \n\t"
@@ -1060,7 +1060,7 @@ static inline void RENAME(dering)(uint8_t src[], int stride, PPContext *c)
#undef REAL_FIND_MIN_MAX
#undef FIND_MIN_MAX
-#if HAVE_MMX2
+#if HAVE_MMXEXT
#define REAL_FIND_MIN_MAX(addr)\
"movq " #addr ", %%mm0 \n\t"\
"pminub %%mm0, %%mm7 \n\t"\
@@ -1087,7 +1087,7 @@ FIND_MIN_MAX((%0, %1, 8))
"movq %%mm7, %%mm4 \n\t"
"psrlq $8, %%mm7 \n\t"
-#if HAVE_MMX2
+#if HAVE_MMXEXT
"pminub %%mm4, %%mm7 \n\t" // min of pixels
"pshufw $0xF9, %%mm7, %%mm4 \n\t"
"pminub %%mm4, %%mm7 \n\t" // min of pixels
@@ -1112,7 +1112,7 @@ FIND_MIN_MAX((%0, %1, 8))
"movq %%mm6, %%mm4 \n\t"
"psrlq $8, %%mm6 \n\t"
-#if HAVE_MMX2
+#if HAVE_MMXEXT
"pmaxub %%mm4, %%mm6 \n\t" // max of pixels
"pshufw $0xF9, %%mm6, %%mm4 \n\t"
"pmaxub %%mm4, %%mm6 \n\t"
@@ -1266,7 +1266,7 @@ DERING_CORE((%0, %1, 8) ,(%%REGd, %1, 4),%%mm2,%%mm4,%%mm0,%%mm3,%%mm5,%%mm1,
: : "r" (src), "r" ((x86_reg)stride), "m" (c->pQPb), "m"(c->pQPb2), "q"(tmp)
: "%"REG_a, "%"REG_d
);
-#else //HAVE_MMX2 || HAVE_AMD3DNOW
+#else //HAVE_MMXEXT || HAVE_AMD3DNOW
int y;
int min=255;
int max=0;
@@ -1383,7 +1383,7 @@ DERING_CORE((%0, %1, 8) ,(%%REGd, %1, 4),%%mm2,%%mm4,%%mm0,%%mm3,%%mm5,%%mm1,
// src[0] = src[7]=src[stride*7]=src[stride*7 + 7]=255;
}
#endif
-#endif //HAVE_MMX2 || HAVE_AMD3DNOW
+#endif //HAVE_MMXEXT || HAVE_AMD3DNOW
}
#endif //HAVE_ALTIVEC
@@ -1395,7 +1395,7 @@ DERING_CORE((%0, %1, 8) ,(%%REGd, %1, 4),%%mm2,%%mm4,%%mm0,%%mm3,%%mm5,%%mm1,
*/
static inline void RENAME(deInterlaceInterpolateLinear)(uint8_t src[], int stride)
{
-#if HAVE_MMX2 || HAVE_AMD3DNOW
+#if HAVE_MMXEXT || HAVE_AMD3DNOW
src+= 4*stride;
__asm__ volatile(
"lea (%0, %1), %%"REG_a" \n\t"
@@ -1448,7 +1448,7 @@ static inline void RENAME(deInterlaceInterpolateLinear)(uint8_t src[], int strid
*/
static inline void RENAME(deInterlaceInterpolateCubic)(uint8_t src[], int stride)
{
-#if HAVE_MMX2 || HAVE_AMD3DNOW
+#if HAVE_MMXEXT || HAVE_AMD3DNOW
src+= stride*3;
__asm__ volatile(
"lea (%0, %1), %%"REG_a" \n\t"
@@ -1490,7 +1490,7 @@ DEINT_CUBIC((%%REGd, %1), (%0, %1, 8) , (%%REGd, %1, 4), (%%REGc) , (%%REGc,
: : "r" (src), "r" ((x86_reg)stride)
: "%"REG_a, "%"REG_d, "%"REG_c
);
-#else //HAVE_MMX2 || HAVE_AMD3DNOW
+#else //HAVE_MMXEXT || HAVE_AMD3DNOW
int x;
src+= stride*3;
for(x=0; x<8; x++){
@@ -1500,7 +1500,7 @@ DEINT_CUBIC((%%REGd, %1), (%0, %1, 8) , (%%REGd, %1, 4), (%%REGc) , (%%REGc,
src[stride*9] = CLIP((-src[stride*6] + 9*src[stride*8] + 9*src[stride*10] - src[stride*12])>>4);
src++;
}
-#endif //HAVE_MMX2 || HAVE_AMD3DNOW
+#endif //HAVE_MMXEXT || HAVE_AMD3DNOW
}
/**
@@ -1512,7 +1512,7 @@ DEINT_CUBIC((%%REGd, %1), (%0, %1, 8) , (%%REGd, %1, 4), (%%REGc) , (%%REGc,
*/
static inline void RENAME(deInterlaceFF)(uint8_t src[], int stride, uint8_t *tmp)
{
-#if HAVE_MMX2 || HAVE_AMD3DNOW
+#if HAVE_MMXEXT || HAVE_AMD3DNOW
src+= stride*4;
__asm__ volatile(
"lea (%0, %1), %%"REG_a" \n\t"
@@ -1561,7 +1561,7 @@ DEINT_FF((%%REGd, %1), (%%REGd, %1, 2), (%0, %1, 8) , (%%REGd, %1, 4))
: : "r" (src), "r" ((x86_reg)stride), "r"(tmp)
: "%"REG_a, "%"REG_d
);
-#else //HAVE_MMX2 || HAVE_AMD3DNOW
+#else //HAVE_MMXEXT || HAVE_AMD3DNOW
int x;
src+= stride*4;
for(x=0; x<8; x++){
@@ -1579,7 +1579,7 @@ DEINT_FF((%%REGd, %1), (%%REGd, %1, 2), (%0, %1, 8) , (%%REGd, %1, 4))
src++;
}
-#endif //HAVE_MMX2 || HAVE_AMD3DNOW
+#endif //HAVE_MMXEXT || HAVE_AMD3DNOW
}
/**
@@ -1591,7 +1591,7 @@ DEINT_FF((%%REGd, %1), (%%REGd, %1, 2), (%0, %1, 8) , (%%REGd, %1, 4))
*/
static inline void RENAME(deInterlaceL5)(uint8_t src[], int stride, uint8_t *tmp, uint8_t *tmp2)
{
-#if HAVE_MMX2 || HAVE_AMD3DNOW
+#if HAVE_MMXEXT || HAVE_AMD3DNOW
src+= stride*4;
__asm__ volatile(
"lea (%0, %1), %%"REG_a" \n\t"
@@ -1651,7 +1651,7 @@ DEINT_L5(%%mm1, %%mm0, (%%REGd, %1, 2), (%0, %1, 8) , (%%REGd, %1, 4))
: : "r" (src), "r" ((x86_reg)stride), "r"(tmp), "r"(tmp2)
: "%"REG_a, "%"REG_d
);
-#else //HAVE_MMX2 || HAVE_AMD3DNOW
+#else //HAVE_MMXEXT || HAVE_AMD3DNOW
int x;
src+= stride*4;
for(x=0; x<8; x++){
@@ -1680,7 +1680,7 @@ DEINT_L5(%%mm1, %%mm0, (%%REGd, %1, 2), (%0, %1, 8) , (%%REGd, %1, 4))
src++;
}
-#endif //HAVE_MMX2 || HAVE_AMD3DNOW
+#endif //HAVE_MMXEXT || HAVE_AMD3DNOW
}
/**
@@ -1692,7 +1692,7 @@ DEINT_L5(%%mm1, %%mm0, (%%REGd, %1, 2), (%0, %1, 8) , (%%REGd, %1, 4))
*/
static inline void RENAME(deInterlaceBlendLinear)(uint8_t src[], int stride, uint8_t *tmp)
{
-#if HAVE_MMX2 || HAVE_AMD3DNOW
+#if HAVE_MMXEXT || HAVE_AMD3DNOW
src+= 4*stride;
__asm__ volatile(
"lea (%0, %1), %%"REG_a" \n\t"
@@ -1739,7 +1739,7 @@ static inline void RENAME(deInterlaceBlendLinear)(uint8_t src[], int stride, uin
: : "r" (src), "r" ((x86_reg)stride), "r" (tmp)
: "%"REG_a, "%"REG_d
);
-#else //HAVE_MMX2 || HAVE_AMD3DNOW
+#else //HAVE_MMXEXT || HAVE_AMD3DNOW
int a, b, c, x;
src+= 4*stride;
@@ -1782,7 +1782,7 @@ static inline void RENAME(deInterlaceBlendLinear)(uint8_t src[], int stride, uin
src += 4;
tmp += 4;
}
-#endif //HAVE_MMX2 || HAVE_AMD3DNOW
+#endif //HAVE_MMXEXT || HAVE_AMD3DNOW
}
/**
@@ -1795,7 +1795,7 @@ static inline void RENAME(deInterlaceMedian)(uint8_t src[], int stride)
{
#if HAVE_MMX
src+= 4*stride;
-#if HAVE_MMX2
+#if HAVE_MMXEXT
__asm__ volatile(
"lea (%0, %1), %%"REG_a" \n\t"
"lea (%%"REG_a", %1, 4), %%"REG_d" \n\t"
@@ -1885,7 +1885,7 @@ MEDIAN((%%REGd, %1), (%%REGd, %1, 2), (%0, %1, 8))
: : "r" (src), "r" ((x86_reg)stride)
: "%"REG_a, "%"REG_d
);
-#endif //HAVE_MMX2
+#endif //HAVE_MMXEXT
#else //HAVE_MMX
int x, y;
src+= 4*stride;
@@ -2087,7 +2087,7 @@ static inline void RENAME(tempNoiseReducer)(uint8_t *src, int stride,
#define FAST_L2_DIFF
//#define L1_DIFF //u should change the thresholds too if u try that one
-#if HAVE_MMX2 || HAVE_AMD3DNOW
+#if HAVE_MMXEXT || HAVE_AMD3DNOW
__asm__ volatile(
"lea (%2, %2, 2), %%"REG_a" \n\t" // 3*stride
"lea (%2, %2, 4), %%"REG_d" \n\t" // 5*stride
@@ -2375,7 +2375,7 @@ L2_DIFF_CORE((%0, %%REGc) , (%1, %%REGc))
:: "r" (src), "r" (tempBlurred), "r"((x86_reg)stride), "m" (tempBlurredPast)
: "%"REG_a, "%"REG_d, "%"REG_c, "memory"
);
-#else //HAVE_MMX2 || HAVE_AMD3DNOW
+#else //HAVE_MMXEXT || HAVE_AMD3DNOW
{
int y;
int d=0;
@@ -2458,7 +2458,7 @@ Switch between
}
}
}
-#endif //HAVE_MMX2 || HAVE_AMD3DNOW
+#endif //HAVE_MMXEXT || HAVE_AMD3DNOW
}
#endif //HAVE_ALTIVEC
@@ -2865,7 +2865,7 @@ static av_always_inline void RENAME(do_a_deblock)(uint8_t *src, int step, int st
"movq (%4), %%mm2 \n\t" // 2L0 - 5L1 + 5L2 - 2L3
"movq 8(%4), %%mm3 \n\t" // 2H0 - 5H1 + 5H2 - 2H3
-#if HAVE_MMX2
+#if HAVE_MMXEXT
"movq %%mm7, %%mm6 \n\t" // 0
"psubw %%mm0, %%mm6 \n\t"
"pmaxsw %%mm6, %%mm0 \n\t" // |2L4 - 5L5 + 5L6 - 2L7|
@@ -2897,7 +2897,7 @@ static av_always_inline void RENAME(do_a_deblock)(uint8_t *src, int step, int st
"psubw %%mm6, %%mm3 \n\t" // |2H0 - 5H1 + 5H2 - 2H3|
#endif
-#if HAVE_MMX2
+#if HAVE_MMXEXT
"pminsw %%mm2, %%mm0 \n\t"
"pminsw %%mm3, %%mm1 \n\t"
#else
@@ -2961,7 +2961,7 @@ static av_always_inline void RENAME(do_a_deblock)(uint8_t *src, int step, int st
"pand %%mm2, %%mm4 \n\t"
"pand %%mm3, %%mm5 \n\t"
-#if HAVE_MMX2
+#if HAVE_MMXEXT
"pminsw %%mm0, %%mm4 \n\t"
"pminsw %%mm1, %%mm5 \n\t"
#else
@@ -3024,7 +3024,7 @@ static inline void RENAME(blockCopy)(uint8_t dst[], int dstStride, const uint8_t
"lea (%2,%4), %%"REG_a" \n\t"
"lea (%3,%5), %%"REG_d" \n\t"
"pxor %%mm4, %%mm4 \n\t"
-#if HAVE_MMX2
+#if HAVE_MMXEXT
#define REAL_SCALED_CPY(src1, src2, dst1, dst2) \
"movq " #src1 ", %%mm0 \n\t"\
"movq " #src1 ", %%mm5 \n\t"\
@@ -3047,7 +3047,7 @@ static inline void RENAME(blockCopy)(uint8_t dst[], int dstStride, const uint8_t
"movq %%mm0, " #dst1 " \n\t"\
"movq %%mm1, " #dst2 " \n\t"\
-#else //HAVE_MMX2
+#else //HAVE_MMXEXT
#define REAL_SCALED_CPY(src1, src2, dst1, dst2) \
"movq " #src1 ", %%mm0 \n\t"\
"movq " #src1 ", %%mm5 \n\t"\
@@ -3074,7 +3074,7 @@ static inline void RENAME(blockCopy)(uint8_t dst[], int dstStride, const uint8_t
"movq %%mm0, " #dst1 " \n\t"\
"movq %%mm1, " #dst2 " \n\t"\
-#endif //HAVE_MMX2
+#endif //HAVE_MMXEXT
#define SCALED_CPY(src1, src2, dst1, dst2)\
REAL_SCALED_CPY(src1, src2, dst1, dst2)
@@ -3248,7 +3248,7 @@ static void RENAME(postProcess)(const uint8_t src[], int srcStride, uint8_t dst[
scale= (double)(c.ppMode.maxAllowedY - c.ppMode.minAllowedY) / (double)(white-black);
-#if HAVE_MMX2
+#if HAVE_MMXEXT
c.packedYScale= (uint16_t)(scale*256.0 + 0.5);
c.packedYOffset= (((black*c.packedYScale)>>8) - c.ppMode.minAllowedY) & 0xFFFF;
#else
@@ -3281,7 +3281,7 @@ static void RENAME(postProcess)(const uint8_t src[], int srcStride, uint8_t dst[
// with the L1 Cache of the P4 ... or only a few blocks at a time or soemthing
for(x=0; x<width; x+=BLOCK_SIZE){
-#if HAVE_MMX2
+#if HAVE_MMXEXT
/*
prefetchnta(srcBlock + (((x>>2)&6) + 5)*srcStride + 32);
prefetchnta(srcBlock + (((x>>2)&6) + 6)*srcStride + 32);
@@ -3417,7 +3417,7 @@ static void RENAME(postProcess)(const uint8_t src[], int srcStride, uint8_t dst[
#endif
-#if HAVE_MMX2
+#if HAVE_MMXEXT
/*
prefetchnta(srcBlock + (((x>>2)&6) + 5)*srcStride + 32);
prefetchnta(srcBlock + (((x>>2)&6) + 6)*srcStride + 32);
diff --git a/libswscale/swscale.c b/libswscale/swscale.c
index ae79eb6..07443db 100644
--- a/libswscale/swscale.c
+++ b/libswscale/swscale.c
@@ -422,7 +422,7 @@ static int swScale(SwsContext *c, const uint8_t *src[],
) {
static int warnedAlready=0;
int cpu_flags = av_get_cpu_flags();
- if (HAVE_MMX2 && (cpu_flags & AV_CPU_FLAG_SSE2) && !warnedAlready){
+ if (HAVE_MMXEXT && (cpu_flags & AV_CPU_FLAG_SSE2) && !warnedAlready){
av_log(c, AV_LOG_WARNING, "Warning: data is not aligned! This can lead to a speedloss\n");
warnedAlready=1;
}
diff --git a/libswscale/x86/swscale.c b/libswscale/x86/swscale.c
index 67cda51..b2e06e3 100644
--- a/libswscale/x86/swscale.c
+++ b/libswscale/x86/swscale.c
@@ -204,7 +204,7 @@ void updateMMXDitherTables(SwsContext *c, int dstY, int lumBufIndex, int chrBufI
}
}
-#if HAVE_MMX2
+#if HAVE_MMXEXT
static void yuv2yuvX_sse3(const int16_t *filter, int filterSize,
const int16_t **src, uint8_t *dest, int dstW,
const uint8_t *dither, int offset)
--
1.7.10.4
More information about the ffmpeg-devel
mailing list