---
libavcodec/x86/ac3dsp.asm | 4 +-
libavcodec/x86/ac3dsp_mmx.c | 4 +-
libavcodec/x86/dsputil_yasm.asm | 8 ++--
libavcodec/x86/dsputilenc_mmx.c | 6 +-
libavcodec/x86/dsputilenc_yasm.asm | 2 +-
libavcodec/x86/h264_chromamc.asm | 14 +++---
libavcodec/x86/h264_chromamc_10bit.asm | 4 +-
libavcodec/x86/h264_deblock.asm | 20 ++++----
libavcodec/x86/h264_deblock_10bit.asm | 4 +-
libavcodec/x86/h264_idct.asm | 46 +++++++++--------
libavcodec/x86/h264_idct_10bit.asm | 2 +-
libavcodec/x86/h264_intrapred.asm | 26 +++++-----
libavcodec/x86/h264_intrapred_init.c | 40 +++++++-------
libavcodec/x86/h264_weight.asm | 12 ++--
libavcodec/x86/pngdsp-init.c | 6 +-
libavcodec/x86/pngdsp.asm | 2 +-
libavcodec/x86/rv34dsp.asm | 6 +-
libavcodec/x86/rv34dsp_init.c | 10 ++--
libavcodec/x86/rv40dsp.asm | 4 +-
libavcodec/x86/rv40dsp_init.c | 30 ++++++------
libavcodec/x86/vc1dsp_mmx.c | 10 ++--
libavcodec/x86/vc1dsp_yasm.asm | 10 ++--
libavcodec/x86/vp3dsp.asm | 4 +-
libavcodec/x86/vp3dsp_init.c | 16 +++---
libavcodec/x86/vp8dsp-init.c | 86 ++++++++++++++++----------------
libavcodec/x86/vp8dsp.asm | 30 ++++++------
libavutil/x86/x86inc.asm | 6 +-
libavutil/x86/x86util.asm | 2 +-
libswscale/x86/output.asm | 4 +-
libswscale/x86/swscale.c | 8 ++--
30 files changed, 215 insertions(+), 211 deletions(-)
diff --git a/libavcodec/x86/ac3dsp.asm b/libavcodec/x86/ac3dsp.asm
index 5b38cff..c4b0be6 100644
--- a/libavcodec/x86/ac3dsp.asm
+++ b/libavcodec/x86/ac3dsp.asm
@@ -100,7 +100,7 @@ AC3_EXPONENT_MIN sse2
por %1, %2
pshuflw %2, %1, q0001
por %1, %2
-%elif cpuflag(mmx2)
+%elif cpuflag(mmxext)
pshufw %2, %1, q0032
por %1, %2
pshufw %2, %1, q0001
@@ -156,7 +156,7 @@ cglobal ac3_max_msb_abs_int16, 2,2,5, src, len
INIT_MMX mmx
%define ABS2 ABS2_MMX
AC3_MAX_MSB_ABS_INT16 or_abs
-INIT_MMX mmx2
+INIT_MMX mmxext
%define ABS2 ABS2_MMXEXT
AC3_MAX_MSB_ABS_INT16 min_max
INIT_XMM sse2
diff --git a/libavcodec/x86/ac3dsp_mmx.c b/libavcodec/x86/ac3dsp_mmx.c
index 0ac8685..cac51cb 100644
--- a/libavcodec/x86/ac3dsp_mmx.c
+++ b/libavcodec/x86/ac3dsp_mmx.c
@@ -28,7 +28,7 @@ extern void ff_ac3_exponent_min_mmxext(uint8_t *exp, int
num_reuse_blocks, int n
extern void ff_ac3_exponent_min_sse2 (uint8_t *exp, int num_reuse_blocks, int
nb_coefs);
extern int ff_ac3_max_msb_abs_int16_mmx (const int16_t *src, int len);
-extern int ff_ac3_max_msb_abs_int16_mmx2 (const int16_t *src, int len);
+extern int ff_ac3_max_msb_abs_int16_mmxext(const int16_t *src, int len);
extern int ff_ac3_max_msb_abs_int16_sse2 (const int16_t *src, int len);
extern int ff_ac3_max_msb_abs_int16_ssse3(const int16_t *src, int len);
@@ -67,7 +67,7 @@ av_cold void ff_ac3dsp_init_x86(AC3DSPContext *c, int
bit_exact)
}
if (mm_flags & AV_CPU_FLAG_MMXEXT && HAVE_MMXEXT) {
c->ac3_exponent_min = ff_ac3_exponent_min_mmxext;
- c->ac3_max_msb_abs_int16 = ff_ac3_max_msb_abs_int16_mmx2;
+ c->ac3_max_msb_abs_int16 = ff_ac3_max_msb_abs_int16_mmxext;
}
if (mm_flags & AV_CPU_FLAG_SSE && HAVE_SSE) {
c->float_to_fixed24 = ff_float_to_fixed24_sse;
diff --git a/libavcodec/x86/dsputil_yasm.asm b/libavcodec/x86/dsputil_yasm.asm
index 70a0aa1..64c4e32 100644
--- a/libavcodec/x86/dsputil_yasm.asm
+++ b/libavcodec/x86/dsputil_yasm.asm
@@ -109,7 +109,7 @@ cglobal scalarproduct_and_madd_int16_%1, 4,4,8, v1, v2, v3,
order, mul
%endmacro
INIT_MMX
-SCALARPRODUCT mmx2
+SCALARPRODUCT mmxext
INIT_XMM
SCALARPRODUCT sse2
@@ -328,8 +328,8 @@ APPLY_WINDOW_INT16 ssse3_atom, 0, 1
APPLY_WINDOW_INT16 ssse3, 0, 1
-; void add_hfyu_median_prediction_mmx2(uint8_t *dst, const uint8_t *top, const
uint8_t *diff, int w, int *left, int *left_top)
-cglobal add_hfyu_median_prediction_mmx2, 6,6,0, dst, top, diff, w, left,
left_top
+; void add_hfyu_median_prediction_mmxext(uint8_t *dst, const uint8_t *top,
const uint8_t *diff, int w, int *left, int *left_top)
+cglobal add_hfyu_median_prediction_mmxext, 6,6,0, dst, top, diff, w, left,
left_top
movq mm0, [topq]
movq mm2, mm0
movd mm4, [left_topq]
@@ -799,7 +799,7 @@ ALIGN 128
mov valh, vall
%if %1 >= 8
movd mm0, vald
-%if cpuflag(mmx2)
+%if cpuflag(mmxext)
pshufw mm0, mm0, 0
%else ; mmx
punpcklwd mm0, mm0
diff --git a/libavcodec/x86/dsputilenc_mmx.c b/libavcodec/x86/dsputilenc_mmx.c
index 6cffe68..f94f2d2 100644
--- a/libavcodec/x86/dsputilenc_mmx.c
+++ b/libavcodec/x86/dsputilenc_mmx.c
@@ -939,7 +939,7 @@ int ff_hadamard8_diff16_##cpu(void *s, uint8_t *src1,
uint8_t *src2, \
int stride, int h);
hadamard_func(mmx)
-hadamard_func(mmx2)
+hadamard_func(mmxext)
hadamard_func(sse2)
hadamard_func(ssse3)
@@ -1154,8 +1154,8 @@ void ff_dsputilenc_init_mmx(DSPContext* c, AVCodecContext
*avctx)
if (mm_flags & AV_CPU_FLAG_MMXEXT) {
#if HAVE_YASM
- c->hadamard8_diff[0]= ff_hadamard8_diff16_mmx2;
- c->hadamard8_diff[1]= ff_hadamard8_diff_mmx2;
+ c->hadamard8_diff[0] = ff_hadamard8_diff16_mmxext;
+ c->hadamard8_diff[1] = ff_hadamard8_diff_mmxext;
#endif
c->sum_abs_dctelem = sum_abs_dctelem_mmxext;
c->vsad[4] = vsad_intra16_mmxext;
diff --git a/libavcodec/x86/dsputilenc_yasm.asm
b/libavcodec/x86/dsputilenc_yasm.asm
index c6ed2da..6e4e1ec 100644
--- a/libavcodec/x86/dsputilenc_yasm.asm
+++ b/libavcodec/x86/dsputilenc_yasm.asm
@@ -266,7 +266,7 @@ HADAMARD8_DIFF_MMX mmx
%define ABS1 ABS1_MMXEXT
%define HSUM HSUM_MMXEXT
-HADAMARD8_DIFF_MMX mmx2
+HADAMARD8_DIFF_MMX mmxext
INIT_XMM
%define ABS2 ABS2_MMXEXT
diff --git a/libavcodec/x86/h264_chromamc.asm b/libavcodec/x86/h264_chromamc.asm
index 64a4efe..7b39c29 100644
--- a/libavcodec/x86/h264_chromamc.asm
+++ b/libavcodec/x86/h264_chromamc.asm
@@ -443,17 +443,17 @@ chroma_mc8_mmx_func put, vc1, mmx_nornd
chroma_mc8_mmx_func put, rv40, mmx
chroma_mc4_mmx_func put, h264, mmx
chroma_mc4_mmx_func put, rv40, mmx
-chroma_mc2_mmx_func put, h264, mmx2
+chroma_mc2_mmx_func put, h264, mmxext
%define CHROMAMC_AVG DIRECT_AVG
%define CHROMAMC_AVG4 COPY_AVG
%define PAVG pavgb
-chroma_mc8_mmx_func avg, h264, mmx2_rnd
-chroma_mc8_mmx_func avg, vc1, mmx2_nornd
-chroma_mc8_mmx_func avg, rv40, mmx2
-chroma_mc4_mmx_func avg, h264, mmx2
-chroma_mc4_mmx_func avg, rv40, mmx2
-chroma_mc2_mmx_func avg, h264, mmx2
+chroma_mc8_mmx_func avg, h264, mmxext_rnd
+chroma_mc8_mmx_func avg, vc1, mmxext_nornd
+chroma_mc8_mmx_func avg, rv40, mmxext
+chroma_mc4_mmx_func avg, h264, mmxext
+chroma_mc4_mmx_func avg, rv40, mmxext
+chroma_mc2_mmx_func avg, h264, mmxext
%define PAVG pavgusb
chroma_mc8_mmx_func avg, h264, 3dnow_rnd
diff --git a/libavcodec/x86/h264_chromamc_10bit.asm
b/libavcodec/x86/h264_chromamc_10bit.asm
index 370c7b5..d6e7372 100644
--- a/libavcodec/x86/h264_chromamc_10bit.asm
+++ b/libavcodec/x86/h264_chromamc_10bit.asm
@@ -256,7 +256,7 @@ CHROMA_MC8 put
INIT_XMM avx
CHROMA_MC8 put
%endif
-INIT_MMX mmx2
+INIT_MMX mmxext
CHROMA_MC4 put
CHROMA_MC2 put
@@ -268,6 +268,6 @@ CHROMA_MC8 avg
INIT_XMM avx
CHROMA_MC8 avg
%endif
-INIT_MMX mmx2
+INIT_MMX mmxext
CHROMA_MC4 avg
CHROMA_MC2 avg
diff --git a/libavcodec/x86/h264_deblock.asm b/libavcodec/x86/h264_deblock.asm
index 940a8f7..45cb7f6 100644
--- a/libavcodec/x86/h264_deblock.asm
+++ b/libavcodec/x86/h264_deblock.asm
@@ -505,7 +505,7 @@ cglobal deblock_h_luma_8, 0,5
RET
%endmacro ; DEBLOCK_LUMA
-INIT_MMX mmx2
+INIT_MMX mmxext
DEBLOCK_LUMA v8, 8
INIT_XMM sse2
DEBLOCK_LUMA v, 16
@@ -784,11 +784,11 @@ DEBLOCK_LUMA_INTRA v
INIT_XMM avx
DEBLOCK_LUMA_INTRA v
%if ARCH_X86_64 == 0
-INIT_MMX mmx2
+INIT_MMX mmxext
DEBLOCK_LUMA_INTRA v8
%endif
-INIT_MMX mmx2
+INIT_MMX mmxext
%macro CHROMA_V_START 0
dec r2d ; alpha-1
@@ -819,7 +819,7 @@ cglobal deblock_v_chroma_8, 5,6
movq m1, [t5+r1]
movq m2, [r0]
movq m3, [r0+r1]
- call ff_chroma_inter_body_mmx2
+ call ff_chroma_inter_body_mmxext
movq [t5+r1], m1
movq [r0], m2
RET
@@ -843,7 +843,7 @@ cglobal deblock_h_chroma_8, 5,7
TRANSPOSE4x8_LOAD bw, wd, dq, PASS8ROWS(t5, r0, r1, t6)
movq buf0, m0
movq buf1, m3
- call ff_chroma_inter_body_mmx2
+ call ff_chroma_inter_body_mmxext
movq m0, buf0
movq m3, buf1
TRANSPOSE8x4B_STORE PASS8ROWS(t5, r0, r1, t6)
@@ -853,7 +853,7 @@ cglobal deblock_h_chroma_8, 5,7
RET
ALIGN 16
-ff_chroma_inter_body_mmx2:
+ff_chroma_inter_body_mmxext:
LOAD_MASK r2d, r3d
movd m6, [r4] ; tc0
punpcklbw m6, m6
@@ -886,7 +886,7 @@ cglobal deblock_v_chroma_intra_8, 4,5
movq m1, [t5+r1]
movq m2, [r0]
movq m3, [r0+r1]
- call ff_chroma_intra_body_mmx2
+ call ff_chroma_intra_body_mmxext
movq [t5+r1], m1
movq [r0], m2
RET
@@ -897,12 +897,12 @@ cglobal deblock_v_chroma_intra_8, 4,5
cglobal deblock_h_chroma_intra_8, 4,6
CHROMA_H_START
TRANSPOSE4x8_LOAD bw, wd, dq, PASS8ROWS(t5, r0, r1, t6)
- call ff_chroma_intra_body_mmx2
+ call ff_chroma_intra_body_mmxext
TRANSPOSE8x4B_STORE PASS8ROWS(t5, r0, r1, t6)
RET
ALIGN 16
-ff_chroma_intra_body_mmx2:
+ff_chroma_intra_body_mmxext:
LOAD_MASK r2d, r3d
movq m5, m1
movq m6, m2
@@ -1026,7 +1026,7 @@ ff_chroma_intra_body_mmx2:
jl %%.b_idx_loop
%endmacro
-INIT_MMX mmx2
+INIT_MMX mmxext
cglobal h264_loop_filter_strength, 9, 9, 0, bs, nnz, ref, mv, bidir, edges, \
step, mask_mv0, mask_mv1, field
%define b_idxq bidirq
diff --git a/libavcodec/x86/h264_deblock_10bit.asm
b/libavcodec/x86/h264_deblock_10bit.asm
index ba2f914..123d8c4 100644
--- a/libavcodec/x86/h264_deblock_10bit.asm
+++ b/libavcodec/x86/h264_deblock_10bit.asm
@@ -792,7 +792,7 @@ cglobal deblock_h_luma_intra_10, 4,7,8*(mmsize/16)
%endmacro
%if ARCH_X86_64 == 0
-INIT_MMX mmx2
+INIT_MMX mmxext
DEBLOCK_LUMA
DEBLOCK_LUMA_INTRA
INIT_XMM sse2
@@ -907,7 +907,7 @@ cglobal deblock_v_chroma_intra_10,
4,6-(mmsize/16),8*(mmsize/16)
%endmacro
%if ARCH_X86_64 == 0
-INIT_MMX mmx2
+INIT_MMX mmxext
DEBLOCK_CHROMA
%endif
INIT_XMM sse2
diff --git a/libavcodec/x86/h264_idct.asm b/libavcodec/x86/h264_idct.asm
index 0fe5176..56d40eb 100644
--- a/libavcodec/x86/h264_idct.asm
+++ b/libavcodec/x86/h264_idct.asm
@@ -287,14 +287,14 @@ cglobal h264_idct8_add_8_sse2, 3, 4, 10
%endmacro
INIT_MMX
-; ff_h264_idct_dc_add_mmx2(uint8_t *dst, int16_t *block, int stride)
-cglobal h264_idct_dc_add_8_mmx2, 3, 3, 0
+; ff_h264_idct_dc_add_mmxext(uint8_t *dst, int16_t *block, int stride)
+cglobal h264_idct_dc_add_8_mmxext, 3, 3, 0
DC_ADD_MMXEXT_INIT r1, r2
DC_ADD_MMXEXT_OP movh, r0, r2, r1
RET
-; ff_h264_idct8_dc_add_mmx2(uint8_t *dst, int16_t *block, int stride)
-cglobal h264_idct8_dc_add_8_mmx2, 3, 3, 0
+; ff_h264_idct8_dc_add_mmxext(uint8_t *dst, int16_t *block, int stride)
+cglobal h264_idct8_dc_add_8_mmxext, 3, 3, 0
DC_ADD_MMXEXT_INIT r1, r2
DC_ADD_MMXEXT_OP mova, r0, r2, r1
lea r0, [r0+r2*4]
@@ -355,9 +355,9 @@ cglobal h264_idct8_add4_8_mmx, 5, 7 + npicregs, 0, dst,
block_offset, block, str
ADD rsp, pad
RET
-; ff_h264_idct_add16_mmx2(uint8_t *dst, const int *block_offset,
-; DCTELEM *block, int stride, const uint8_t nnzc[6*8])
-cglobal h264_idct_add16_8_mmx2, 5, 8 + npicregs, 0, dst1, block_offset, block,
stride, nnzc, cntr, coeff, dst2, picreg
+; ff_h264_idct_add16_mmxext(uint8_t *dst, const int *block_offset,
+; DCTELEM *block, int stride, const uint8_t
nnzc[6*8])
+cglobal h264_idct_add16_8_mmxext, 5, 8 + npicregs, 0, dst1, block_offset,
block, stride, nnzc, cntr, coeff, dst2, picreg
xor r5, r5
%ifdef PIC
lea picregq, [scan8_mem]
@@ -422,9 +422,10 @@ cglobal h264_idct_add16intra_8_mmx, 5, 7 + npicregs, 0,
dst, block_offset, block
jl .nextblock
REP_RET
-; ff_h264_idct_add16intra_mmx2(uint8_t *dst, const int *block_offset,
-; DCTELEM *block, int stride, const uint8_t
nnzc[6*8])
-cglobal h264_idct_add16intra_8_mmx2, 5, 8 + npicregs, 0, dst1, block_offset,
block, stride, nnzc, cntr, coeff, dst2, picreg
+; ff_h264_idct_add16intra_mmxext(uint8_t *dst, const int *block_offset,
+; DCTELEM *block, int stride,
+; const uint8_t nnzc[6*8])
+cglobal h264_idct_add16intra_8_mmxext, 5, 8 + npicregs, 0, dst1, block_offset,
block, stride, nnzc, cntr, coeff, dst2, picreg
xor r5, r5
%ifdef PIC
lea picregq, [scan8_mem]
@@ -464,9 +465,10 @@ cglobal h264_idct_add16intra_8_mmx2, 5, 8 + npicregs, 0,
dst1, block_offset, blo
jl .nextblock
REP_RET
-; ff_h264_idct8_add4_mmx2(uint8_t *dst, const int *block_offset,
-; DCTELEM *block, int stride, const uint8_t nnzc[6*8])
-cglobal h264_idct8_add4_8_mmx2, 5, 8 + npicregs, 0, dst1, block_offset, block,
stride, nnzc, cntr, coeff, dst2, picreg
+; ff_h264_idct8_add4_mmxext(uint8_t *dst, const int *block_offset,
+; DCTELEM *block, int stride,
+; const uint8_t nnzc[6*8])
+cglobal h264_idct8_add4_8_mmxext, 5, 8 + npicregs, 0, dst1, block_offset,
block, stride, nnzc, cntr, coeff, dst2, picreg
%assign pad 128+4-(stack_offset&7)
SUB rsp, pad
@@ -621,7 +623,7 @@ cglobal h264_idct_add8_8_mmx, 5, 8 + npicregs, 0, dst1,
block_offset, block, str
call h264_idct_add8_mmx_plane
RET
-h264_idct_add8_mmx2_plane
+h264_idct_add8_mmxext_plane
.nextblock
movzx r6, byte [scan8+r5]
movzx r6, byte [r4+r6]
@@ -662,9 +664,9 @@ h264_idct_add8_mmx2_plane
jnz .nextblock
rep ret
-; ff_h264_idct_add8_mmx2(uint8_t **dest, const int *block_offset,
-; DCTELEM *block, int stride, const uint8_t nnzc[6*8])
-cglobal h264_idct_add8_8_mmx2, 5, 8 + npicregs, 0, dst1, block_offset, block,
stride, nnzc, cntr, coeff, dst2, picreg
+; ff_h264_idct_add8_mmxext(uint8_t **dest, const int *block_offset,
+; DCTELEM *block, int stride, const uint8_t nnzc[6*8])
+cglobal h264_idct_add8_8_mmxext, 5, 8 + npicregs, 0, dst1, block_offset,
block, stride, nnzc, cntr, coeff, dst2, picreg
mov r5, 16
add r2, 512
%if ARCH_X86_64
@@ -673,7 +675,7 @@ cglobal h264_idct_add8_8_mmx2, 5, 8 + npicregs, 0, dst1,
block_offset, block, st
%ifdef PIC
lea picregq, [scan8_mem]
%endif
- call h264_idct_add8_mmx2_plane
+ call h264_idct_add8_mmxext_plane
mov r5, 32
add r2, 384
%if ARCH_X86_64
@@ -681,12 +683,12 @@ cglobal h264_idct_add8_8_mmx2, 5, 8 + npicregs, 0, dst1,
block_offset, block, st
%else
add r0mp, gprsize
%endif
- call h264_idct_add8_mmx2_plane
+ call h264_idct_add8_mmxext_plane
RET
INIT_MMX
; r0 = uint8_t *dst, r2 = int16_t *block, r3 = int stride, r6=clobbered
-h264_idct_dc_add8_mmx2:
+h264_idct_dc_add8_mmxext:
movd m0, [r2 ] ; 0 0 X D
punpcklwd m0, [r2+32] ; x X d D
paddsw m0, [pw_32]
@@ -780,7 +782,7 @@ cglobal h264_idct_add16_8_sse2, 5, 5 + ARCH_X86_64, 8
%else
add r0, r0m
%endif
- call h264_idct_dc_add8_mmx2
+ call h264_idct_dc_add8_mmxext
.cycle%1end
%if %1 < 7
add r2, 64
@@ -829,7 +831,7 @@ cglobal h264_idct_add16intra_8_sse2, 5, 7 + ARCH_X86_64, 8
mov r0, [r0]
add r0, dword [r1+(%1&1)*8+64*(1+(%1>>1))]
%endif
- call h264_idct_dc_add8_mmx2
+ call h264_idct_dc_add8_mmxext
.cycle%1end
%if %1 == 1
add r2, 384+64
diff --git a/libavcodec/x86/h264_idct_10bit.asm
b/libavcodec/x86/h264_idct_10bit.asm
index 2aab986..a619bcc 100644
--- a/libavcodec/x86/h264_idct_10bit.asm
+++ b/libavcodec/x86/h264_idct_10bit.asm
@@ -185,7 +185,7 @@ IDCT_ADD16_10
mova [%1+%3 ], m4
%endmacro
-INIT_MMX mmx2
+INIT_MMX mmxext
cglobal h264_idct_dc_add_10,3,3
movd m0, [r1]
paddd m0, [pd_32]
diff --git a/libavcodec/x86/h264_intrapred.asm
b/libavcodec/x86/h264_intrapred.asm
index 50a615b..9d2e5c7 100644
--- a/libavcodec/x86/h264_intrapred.asm
+++ b/libavcodec/x86/h264_intrapred.asm
@@ -119,7 +119,7 @@ cglobal pred16x16_horizontal, 2,3
INIT_MMX mmx
PRED16x16_H
-INIT_MMX mmx2
+INIT_MMX mmxext
PRED16x16_H
INIT_XMM ssse3
PRED16x16_H
@@ -180,7 +180,7 @@ cglobal pred16x16_dc, 2,7
REP_RET
%endmacro
-INIT_MMX mmx2
+INIT_MMX mmxext
PRED16x16_DC
INIT_XMM sse2
PRED16x16_DC
@@ -230,7 +230,7 @@ cglobal pred16x16_tm_vp8, 2,5
INIT_MMX mmx
PRED16x16_TM_MMX
-INIT_MMX mmx2
+INIT_MMX mmxext
PRED16x16_TM_MMX
INIT_MMX
@@ -310,14 +310,14 @@ cglobal pred16x16_plane_%1, 2,9,7
movhlps m1, m0
%endif
paddw m0, m1
-%if cpuflag(mmx2)
+%if cpuflag(mmxext)
PSHUFLW m1, m0, 0xE
%elif cpuflag(mmx)
mova m1, m0
psrlq m1, 32
%endif
paddw m0, m1
-%if cpuflag(mmx2)
+%if cpuflag(mmxext)
PSHUFLW m1, m0, 0x1
%elif cpuflag(mmx)
mova m1, m0
@@ -537,7 +537,7 @@ INIT_MMX mmx
H264_PRED16x16_PLANE h264
H264_PRED16x16_PLANE rv40
H264_PRED16x16_PLANE svq3
-INIT_MMX mmx2
+INIT_MMX mmxext
H264_PRED16x16_PLANE h264
H264_PRED16x16_PLANE rv40
H264_PRED16x16_PLANE svq3
@@ -584,7 +584,7 @@ cglobal pred8x8_plane, 2,9,7
paddw m0, m1
%if notcpuflag(ssse3)
-%if cpuflag(mmx2)
+%if cpuflag(mmxext)
PSHUFLW m1, m0, 0xE
%elif cpuflag(mmx)
mova m1, m0
@@ -593,7 +593,7 @@ cglobal pred8x8_plane, 2,9,7
paddw m0, m1
%endif ; !ssse3
-%if cpuflag(mmx2)
+%if cpuflag(mmxext)
PSHUFLW m1, m0, 0x1
%elif cpuflag(mmx)
mova m1, m0
@@ -718,7 +718,7 @@ ALIGN 16
INIT_MMX mmx
H264_PRED8x8_PLANE
-INIT_MMX mmx2
+INIT_MMX mmxext
H264_PRED8x8_PLANE
INIT_XMM sse2
H264_PRED8x8_PLANE
@@ -765,7 +765,7 @@ cglobal pred8x8_horizontal, 2,3
INIT_MMX mmx
PRED8x8_H
-INIT_MMX mmx2
+INIT_MMX mmxext
PRED8x8_H
INIT_MMX ssse3
PRED8x8_H
@@ -942,7 +942,7 @@ cglobal pred8x8_tm_vp8, 2,6
INIT_MMX mmx
PRED8x8_TM_MMX
-INIT_MMX mmx2
+INIT_MMX mmxext
PRED8x8_TM_MMX
INIT_MMX
@@ -2459,7 +2459,7 @@ cglobal pred4x4_tm_vp8, 3,6
sub r3d, r4d
movd mm2, r1d
movd mm4, r3d
-%if cpuflag(mmx2)
+%if cpuflag(mmxext)
pshufw mm2, mm2, 0
pshufw mm4, mm4, 0
%else
@@ -2482,7 +2482,7 @@ cglobal pred4x4_tm_vp8, 3,6
INIT_MMX mmx
PRED4x4_TM_MMX
-INIT_MMX mmx2
+INIT_MMX mmxext
PRED4x4_TM_MMX
INIT_MMX
diff --git a/libavcodec/x86/h264_intrapred_init.c
b/libavcodec/x86/h264_intrapred_init.c
index 08c35c7..d82a846 100644
--- a/libavcodec/x86/h264_intrapred_init.c
+++ b/libavcodec/x86/h264_intrapred_init.c
@@ -96,39 +96,39 @@ PRED16x16(horizontal, 10, sse2)
void ff_pred16x16_vertical_mmx (uint8_t *src, int stride);
void ff_pred16x16_vertical_sse (uint8_t *src, int stride);
void ff_pred16x16_horizontal_mmx (uint8_t *src, int stride);
-void ff_pred16x16_horizontal_mmx2 (uint8_t *src, int stride);
+void ff_pred16x16_horizontal_mmxext(uint8_t *src, int stride);
void ff_pred16x16_horizontal_ssse3 (uint8_t *src, int stride);
-void ff_pred16x16_dc_mmx2 (uint8_t *src, int stride);
+void ff_pred16x16_dc_mmxext (uint8_t *src, int stride);
void ff_pred16x16_dc_sse2 (uint8_t *src, int stride);
void ff_pred16x16_dc_ssse3 (uint8_t *src, int stride);
void ff_pred16x16_plane_h264_mmx (uint8_t *src, int stride);
-void ff_pred16x16_plane_h264_mmx2 (uint8_t *src, int stride);
+void ff_pred16x16_plane_h264_mmxext(uint8_t *src, int stride);
void ff_pred16x16_plane_h264_sse2 (uint8_t *src, int stride);
void ff_pred16x16_plane_h264_ssse3 (uint8_t *src, int stride);
void ff_pred16x16_plane_rv40_mmx (uint8_t *src, int stride);
-void ff_pred16x16_plane_rv40_mmx2 (uint8_t *src, int stride);
+void ff_pred16x16_plane_rv40_mmxext(uint8_t *src, int stride);
void ff_pred16x16_plane_rv40_sse2 (uint8_t *src, int stride);
void ff_pred16x16_plane_rv40_ssse3 (uint8_t *src, int stride);
void ff_pred16x16_plane_svq3_mmx (uint8_t *src, int stride);
-void ff_pred16x16_plane_svq3_mmx2 (uint8_t *src, int stride);
+void ff_pred16x16_plane_svq3_mmxext(uint8_t *src, int stride);
void ff_pred16x16_plane_svq3_sse2 (uint8_t *src, int stride);
void ff_pred16x16_plane_svq3_ssse3 (uint8_t *src, int stride);
void ff_pred16x16_tm_vp8_mmx (uint8_t *src, int stride);
-void ff_pred16x16_tm_vp8_mmx2 (uint8_t *src, int stride);
+void ff_pred16x16_tm_vp8_mmxext (uint8_t *src, int stride);
void ff_pred16x16_tm_vp8_sse2 (uint8_t *src, int stride);
void ff_pred8x8_top_dc_mmxext (uint8_t *src, int stride);
void ff_pred8x8_dc_rv40_mmxext (uint8_t *src, int stride);
void ff_pred8x8_dc_mmxext (uint8_t *src, int stride);
void ff_pred8x8_vertical_mmx (uint8_t *src, int stride);
void ff_pred8x8_horizontal_mmx (uint8_t *src, int stride);
-void ff_pred8x8_horizontal_mmx2 (uint8_t *src, int stride);
+void ff_pred8x8_horizontal_mmxext (uint8_t *src, int stride);
void ff_pred8x8_horizontal_ssse3 (uint8_t *src, int stride);
void ff_pred8x8_plane_mmx (uint8_t *src, int stride);
-void ff_pred8x8_plane_mmx2 (uint8_t *src, int stride);
+void ff_pred8x8_plane_mmxext (uint8_t *src, int stride);
void ff_pred8x8_plane_sse2 (uint8_t *src, int stride);
void ff_pred8x8_plane_ssse3 (uint8_t *src, int stride);
void ff_pred8x8_tm_vp8_mmx (uint8_t *src, int stride);
-void ff_pred8x8_tm_vp8_mmx2 (uint8_t *src, int stride);
+void ff_pred8x8_tm_vp8_mmxext (uint8_t *src, int stride);
void ff_pred8x8_tm_vp8_sse2 (uint8_t *src, int stride);
void ff_pred8x8_tm_vp8_ssse3 (uint8_t *src, int stride);
void ff_pred8x8l_top_dc_mmxext (uint8_t *src, int has_topleft, int
has_topright, int stride);
@@ -163,7 +163,7 @@ void ff_pred4x4_vertical_right_mmxext(uint8_t *src, const
uint8_t *topright, int
void ff_pred4x4_horizontal_up_mmxext(uint8_t *src, const uint8_t *topright,
int stride);
void ff_pred4x4_horizontal_down_mmxext(uint8_t *src, const uint8_t *topright,
int stride);
void ff_pred4x4_tm_vp8_mmx (uint8_t *src, const uint8_t *topright, int
stride);
-void ff_pred4x4_tm_vp8_mmx2 (uint8_t *src, const uint8_t *topright, int
stride);
+void ff_pred4x4_tm_vp8_mmxext (uint8_t *src, const uint8_t *topright, int
stride);
void ff_pred4x4_tm_vp8_ssse3 (uint8_t *src, const uint8_t *topright, int
stride);
void ff_pred4x4_vertical_vp8_mmxext(uint8_t *src, const uint8_t *topright, int
stride);
@@ -199,10 +199,10 @@ void ff_h264_pred_init_x86(H264PredContext *h, int
codec_id, const int bit_depth
}
if (mm_flags & AV_CPU_FLAG_MMXEXT) {
- h->pred16x16[HOR_PRED8x8 ] =
ff_pred16x16_horizontal_mmx2;
- h->pred16x16[DC_PRED8x8 ] = ff_pred16x16_dc_mmx2;
+ h->pred16x16[HOR_PRED8x8 ] =
ff_pred16x16_horizontal_mmxext;
+ h->pred16x16[DC_PRED8x8 ] = ff_pred16x16_dc_mmxext;
if (chroma_format_idc == 1)
- h->pred8x8[HOR_PRED8x8 ] = ff_pred8x8_horizontal_mmx2;
+ h->pred8x8[HOR_PRED8x8 ] =
ff_pred8x8_horizontal_mmxext;
h->pred8x8l [TOP_DC_PRED ] = ff_pred8x8l_top_dc_mmxext;
h->pred8x8l [DC_PRED ] = ff_pred8x8l_dc_mmxext;
h->pred8x8l [HOR_PRED ] =
ff_pred8x8l_horizontal_mmxext;
@@ -232,20 +232,20 @@ void ff_h264_pred_init_x86(H264PredContext *h, int
codec_id, const int bit_depth
}
}
if (codec_id == CODEC_ID_VP8) {
- h->pred16x16[PLANE_PRED8x8 ] = ff_pred16x16_tm_vp8_mmx2;
+ h->pred16x16[PLANE_PRED8x8 ] = ff_pred16x16_tm_vp8_mmxext;
h->pred8x8 [DC_PRED8x8 ] = ff_pred8x8_dc_rv40_mmxext;
- h->pred8x8 [PLANE_PRED8x8 ] = ff_pred8x8_tm_vp8_mmx2;
- h->pred4x4 [TM_VP8_PRED ] = ff_pred4x4_tm_vp8_mmx2;
+ h->pred8x8 [PLANE_PRED8x8 ] = ff_pred8x8_tm_vp8_mmxext;
+ h->pred4x4 [TM_VP8_PRED ] = ff_pred4x4_tm_vp8_mmxext;
h->pred4x4 [VERT_PRED ] =
ff_pred4x4_vertical_vp8_mmxext;
} else {
if (chroma_format_idc == 1)
- h->pred8x8 [PLANE_PRED8x8] = ff_pred8x8_plane_mmx2;
+ h->pred8x8 [PLANE_PRED8x8] = ff_pred8x8_plane_mmxext;
if (codec_id == CODEC_ID_SVQ3) {
- h->pred16x16[PLANE_PRED8x8 ] =
ff_pred16x16_plane_svq3_mmx2;
+ h->pred16x16[PLANE_PRED8x8 ] =
ff_pred16x16_plane_svq3_mmxext;
} else if (codec_id == CODEC_ID_RV40) {
- h->pred16x16[PLANE_PRED8x8 ] =
ff_pred16x16_plane_rv40_mmx2;
+ h->pred16x16[PLANE_PRED8x8 ] =
ff_pred16x16_plane_rv40_mmxext;
} else {
- h->pred16x16[PLANE_PRED8x8 ] =
ff_pred16x16_plane_h264_mmx2;
+ h->pred16x16[PLANE_PRED8x8 ] =
ff_pred16x16_plane_h264_mmxext;
}
}
}
diff --git a/libavcodec/x86/h264_weight.asm b/libavcodec/x86/h264_weight.asm
index 22ce72d..16757c0 100644
--- a/libavcodec/x86/h264_weight.asm
+++ b/libavcodec/x86/h264_weight.asm
@@ -71,7 +71,7 @@ SECTION .text
%endmacro
INIT_MMX
-cglobal h264_weight_16_mmx2, 6, 6, 0
+cglobal h264_weight_16_mmxext, 6, 6, 0
WEIGHT_SETUP
.nextrow
WEIGHT_OP 0, 4
@@ -96,7 +96,7 @@ cglobal h264_weight_%1_%3, 6, 6, %2
%endmacro
INIT_MMX
-WEIGHT_FUNC_MM 8, 0, mmx2
+WEIGHT_FUNC_MM 8, 0, mmxext
INIT_XMM
WEIGHT_FUNC_MM 16, 8, sse2
@@ -121,7 +121,7 @@ cglobal h264_weight_%1_%3, 6, 6, %2
%endmacro
INIT_MMX
-WEIGHT_FUNC_HALF_MM 4, 0, mmx2
+WEIGHT_FUNC_HALF_MM 4, 0, mmxext
INIT_XMM
WEIGHT_FUNC_HALF_MM 8, 8, sse2
@@ -175,7 +175,7 @@ WEIGHT_FUNC_HALF_MM 8, 8, sse2
%endmacro
INIT_MMX
-cglobal h264_biweight_16_mmx2, 7, 8, 0
+cglobal h264_biweight_16_mmxext, 7, 8, 0
BIWEIGHT_SETUP
movifnidn r3d, r3m
.nextrow
@@ -210,7 +210,7 @@ cglobal h264_biweight_%1_%3, 7, 8, %2
%endmacro
INIT_MMX
-BIWEIGHT_FUNC_MM 8, 0, mmx2
+BIWEIGHT_FUNC_MM 8, 0, mmxext
INIT_XMM
BIWEIGHT_FUNC_MM 16, 8, sse2
@@ -239,7 +239,7 @@ cglobal h264_biweight_%1_%3, 7, 8, %2
%endmacro
INIT_MMX
-BIWEIGHT_FUNC_HALF_MM 4, 0, mmx2
+BIWEIGHT_FUNC_HALF_MM 4, 0, mmxext
INIT_XMM
BIWEIGHT_FUNC_HALF_MM 8, 8, sse2
diff --git a/libavcodec/x86/pngdsp-init.c b/libavcodec/x86/pngdsp-init.c
index aa21847..5ecbaef 100644
--- a/libavcodec/x86/pngdsp-init.c
+++ b/libavcodec/x86/pngdsp-init.c
@@ -23,8 +23,8 @@
#include "libavutil/cpu.h"
#include "libavcodec/pngdsp.h"
-void ff_add_png_paeth_prediction_mmx2 (uint8_t *dst, uint8_t *src,
- uint8_t *top, int w, int bpp);
+void ff_add_png_paeth_prediction_mmxext(uint8_t *dst, uint8_t *src,
+ uint8_t *top, int w, int bpp);
void ff_add_png_paeth_prediction_ssse3(uint8_t *dst, uint8_t *src,
uint8_t *top, int w, int bpp);
void ff_add_bytes_l2_mmx (uint8_t *dst, uint8_t *src1,
@@ -42,7 +42,7 @@ void ff_pngdsp_init_x86(PNGDSPContext *dsp)
dsp->add_bytes_l2 = ff_add_bytes_l2_mmx;
#endif
if (flags & AV_CPU_FLAG_MMXEXT)
- dsp->add_paeth_prediction = ff_add_png_paeth_prediction_mmx2;
+ dsp->add_paeth_prediction = ff_add_png_paeth_prediction_mmxext;
if (flags & AV_CPU_FLAG_SSE2)
dsp->add_bytes_l2 = ff_add_bytes_l2_sse2;
if (flags & AV_CPU_FLAG_SSSE3)
diff --git a/libavcodec/x86/pngdsp.asm b/libavcodec/x86/pngdsp.asm
index d6e6374..8b4c42c 100644
--- a/libavcodec/x86/pngdsp.asm
+++ b/libavcodec/x86/pngdsp.asm
@@ -167,7 +167,7 @@ cglobal add_png_paeth_prediction, 5, 7, %1, dst, src, top,
w, bpp, end, cntr
RET
%endmacro
-INIT_MMX mmx2
+INIT_MMX mmxext
ADD_PAETH_PRED_FN 0
INIT_MMX ssse3
diff --git a/libavcodec/x86/rv34dsp.asm b/libavcodec/x86/rv34dsp.asm
index c43b77a..d8ec368 100644
--- a/libavcodec/x86/rv34dsp.asm
+++ b/libavcodec/x86/rv34dsp.asm
@@ -58,7 +58,7 @@ cglobal rv34_idct_%1, 1, 2, 0
REP_RET
%endmacro
-INIT_MMX mmx2
+INIT_MMX mmxext
%define IDCT_DC IDCT_DC_ROUND
rv34_idct dc
%define IDCT_DC IDCT_DC_NOROUND
@@ -134,7 +134,7 @@ cglobal rv34_idct_dc_add, 3, 3
mova mm5, [pd_512] ; 0x200
%endmacro
-; ff_rv34_idct_add_mmx2(uint8_t *dst, ptrdiff_t stride, DCTELEM *block);
+; ff_rv34_idct_add_mmxext(uint8_t *dst, ptrdiff_t stride, DCTELEM *block);
%macro COL_TRANSFORM 4
pshufw mm3, %2, 0xDD ; col. 1,3,1,3
pshufw %2, %2, 0x88 ; col. 0,2,0,2
@@ -155,7 +155,7 @@ cglobal rv34_idct_dc_add, 3, 3
packuswb %2, %2
movd %1, %2
%endmacro
-INIT_MMX mmx2
+INIT_MMX mmxext
cglobal rv34_idct_add, 3,3,0, d, s, b
ROW_TRANSFORM bq
COL_TRANSFORM [dq], mm0, [pw_col_coeffs+ 0], [pw_col_coeffs+ 8]
diff --git a/libavcodec/x86/rv34dsp_init.c b/libavcodec/x86/rv34dsp_init.c
index 7284a9b..8c43e3d 100644
--- a/libavcodec/x86/rv34dsp_init.c
+++ b/libavcodec/x86/rv34dsp_init.c
@@ -24,11 +24,11 @@
#include "libavcodec/dsputil.h"
#include "libavcodec/rv34dsp.h"
-void ff_rv34_idct_dc_mmx2(DCTELEM *block);
-void ff_rv34_idct_dc_noround_mmx2(DCTELEM *block);
+void ff_rv34_idct_dc_mmxext(DCTELEM *block);
+void ff_rv34_idct_dc_noround_mmxext(DCTELEM *block);
void ff_rv34_idct_dc_add_mmx(uint8_t *dst, ptrdiff_t stride, int dc);
void ff_rv34_idct_dc_add_sse4(uint8_t *dst, ptrdiff_t stride, int dc);
-void ff_rv34_idct_add_mmx2(uint8_t *dst, ptrdiff_t stride, DCTELEM *block);
+void ff_rv34_idct_add_mmxext(uint8_t *dst, ptrdiff_t stride, DCTELEM *block);
av_cold void ff_rv34dsp_init_x86(RV34DSPContext* c, DSPContext *dsp)
{
@@ -38,8 +38,8 @@ av_cold void ff_rv34dsp_init_x86(RV34DSPContext* c,
DSPContext *dsp)
if (mm_flags & AV_CPU_FLAG_MMX)
c->rv34_idct_dc_add = ff_rv34_idct_dc_add_mmx;
if (mm_flags & AV_CPU_FLAG_MMXEXT) {
- c->rv34_inv_transform_dc = ff_rv34_idct_dc_noround_mmx2;
- c->rv34_idct_add = ff_rv34_idct_add_mmx2;
+ c->rv34_inv_transform_dc = ff_rv34_idct_dc_noround_mmxext;
+ c->rv34_idct_add = ff_rv34_idct_add_mmxext;
}
if (mm_flags & AV_CPU_FLAG_SSE4)
c->rv34_idct_dc_add = ff_rv34_idct_dc_add_sse4;
diff --git a/libavcodec/x86/rv40dsp.asm b/libavcodec/x86/rv40dsp.asm
index ae740c2..eb7fc9f 100644
--- a/libavcodec/x86/rv40dsp.asm
+++ b/libavcodec/x86/rv40dsp.asm
@@ -241,7 +241,7 @@ INIT_MMX mmx
FILTER_V put
FILTER_H put
-INIT_MMX mmx2
+INIT_MMX mmxext
FILTER_V avg
FILTER_H avg
@@ -487,7 +487,7 @@ cglobal rv40_weight_func_%1_%2, 6, 7, 8
REP_RET
%endmacro
-INIT_MMX mmx2
+INIT_MMX mmxext
RV40_WEIGHT rnd, 8, 3
RV40_WEIGHT rnd, 16, 4
RV40_WEIGHT nornd, 8, 3
diff --git a/libavcodec/x86/rv40dsp_init.c b/libavcodec/x86/rv40dsp_init.c
index bee3365..daaa1a6 100644
--- a/libavcodec/x86/rv40dsp_init.c
+++ b/libavcodec/x86/rv40dsp_init.c
@@ -31,15 +31,15 @@
void ff_put_rv40_chroma_mc8_mmx (uint8_t *dst, uint8_t *src,
int stride, int h, int x, int y);
-void ff_avg_rv40_chroma_mc8_mmx2 (uint8_t *dst, uint8_t *src,
- int stride, int h, int x, int y);
+void ff_avg_rv40_chroma_mc8_mmxext(uint8_t *dst, uint8_t *src,
+ int stride, int h, int x, int y);
void ff_avg_rv40_chroma_mc8_3dnow(uint8_t *dst, uint8_t *src,
int stride, int h, int x, int y);
void ff_put_rv40_chroma_mc4_mmx (uint8_t *dst, uint8_t *src,
int stride, int h, int x, int y);
-void ff_avg_rv40_chroma_mc4_mmx2 (uint8_t *dst, uint8_t *src,
- int stride, int h, int x, int y);
+void ff_avg_rv40_chroma_mc4_mmxext(uint8_t *dst, uint8_t *src,
+ int stride, int h, int x, int y);
void ff_avg_rv40_chroma_mc4_3dnow(uint8_t *dst, uint8_t *src,
int stride, int h, int x, int y);
@@ -52,7 +52,7 @@ void ff_rv40_weight_func_nornd_16_##opt(uint8_t *dst, uint8_t
*src1, uint8_t *sr
int w1, int w2, ptrdiff_t stride); \
void ff_rv40_weight_func_nornd_8_##opt (uint8_t *dst, uint8_t *src1, uint8_t
*src2, \
int w1, int w2, ptrdiff_t stride);
-DECLARE_WEIGHT(mmx2)
+DECLARE_WEIGHT(mmxext)
DECLARE_WEIGHT(sse2)
DECLARE_WEIGHT(ssse3)
@@ -147,9 +147,9 @@ QPEL_MC_DECL(avg_, _sse2)
QPEL_MC_DECL(put_, _mmx)
-#define ff_put_rv40_qpel_h_mmx2 ff_put_rv40_qpel_h_mmx
-#define ff_put_rv40_qpel_v_mmx2 ff_put_rv40_qpel_v_mmx
-QPEL_MC_DECL(avg_, _mmx2)
+#define ff_put_rv40_qpel_h_mmxext ff_put_rv40_qpel_h_mmx
+#define ff_put_rv40_qpel_v_mmxext ff_put_rv40_qpel_v_mmx
+QPEL_MC_DECL(avg_, _mmxext)
#define ff_put_rv40_qpel_h_3dnow ff_put_rv40_qpel_h_mmx
#define ff_put_rv40_qpel_v_3dnow ff_put_rv40_qpel_v_mmx
@@ -201,14 +201,14 @@ void ff_rv40dsp_init_x86(RV34DSPContext *c, DSPContext
*dsp)
#endif
}
if (mm_flags & AV_CPU_FLAG_MMXEXT) {
- c->avg_chroma_pixels_tab[0] = ff_avg_rv40_chroma_mc8_mmx2;
- c->avg_chroma_pixels_tab[1] = ff_avg_rv40_chroma_mc4_mmx2;
- c->rv40_weight_pixels_tab[0][0] = ff_rv40_weight_func_rnd_16_mmx2;
- c->rv40_weight_pixels_tab[0][1] = ff_rv40_weight_func_rnd_8_mmx2;
- c->rv40_weight_pixels_tab[1][0] = ff_rv40_weight_func_nornd_16_mmx2;
- c->rv40_weight_pixels_tab[1][1] = ff_rv40_weight_func_nornd_8_mmx2;
+ c->avg_chroma_pixels_tab[0] = ff_avg_rv40_chroma_mc8_mmxext;
+ c->avg_chroma_pixels_tab[1] = ff_avg_rv40_chroma_mc4_mmxext;
+ c->rv40_weight_pixels_tab[0][0] = ff_rv40_weight_func_rnd_16_mmxext;
+ c->rv40_weight_pixels_tab[0][1] = ff_rv40_weight_func_rnd_8_mmxext;
+ c->rv40_weight_pixels_tab[1][0] = ff_rv40_weight_func_nornd_16_mmxext;
+ c->rv40_weight_pixels_tab[1][1] = ff_rv40_weight_func_nornd_8_mmxext;
#if ARCH_X86_32
- QPEL_MC_SET(avg_, _mmx2)
+ QPEL_MC_SET(avg_, _mmxext)
#endif
} else if (mm_flags & AV_CPU_FLAG_3DNOW) {
c->avg_chroma_pixels_tab[0] = ff_avg_rv40_chroma_mc8_3dnow;
diff --git a/libavcodec/x86/vc1dsp_mmx.c b/libavcodec/x86/vc1dsp_mmx.c
index 05a8dd2..e5dc219 100644
--- a/libavcodec/x86/vc1dsp_mmx.c
+++ b/libavcodec/x86/vc1dsp_mmx.c
@@ -712,7 +712,7 @@ static void vc1_h_loop_filter16_ ## EXT(uint8_t *src, int
stride, int pq) \
}
#if HAVE_YASM
-LOOP_FILTER(mmx2)
+LOOP_FILTER(mmxext)
LOOP_FILTER(sse2)
LOOP_FILTER(ssse3)
@@ -728,8 +728,8 @@ static void vc1_h_loop_filter16_sse4(uint8_t *src, int
stride, int pq)
void ff_put_vc1_chroma_mc8_mmx_nornd (uint8_t *dst, uint8_t *src,
int stride, int h, int x, int y);
-void ff_avg_vc1_chroma_mc8_mmx2_nornd (uint8_t *dst, uint8_t *src,
- int stride, int h, int x, int y);
+void ff_avg_vc1_chroma_mc8_mmxext_nornd(uint8_t *dst, uint8_t *src,
+ int stride, int h, int x, int y);
void ff_avg_vc1_chroma_mc8_3dnow_nornd(uint8_t *dst, uint8_t *src,
int stride, int h, int x, int y);
void ff_put_vc1_chroma_mc8_ssse3_nornd(uint8_t *dst, uint8_t *src,
@@ -806,8 +806,8 @@ void ff_vc1dsp_init_mmx(VC1DSPContext *dsp)
}
if (mm_flags & AV_CPU_FLAG_MMXEXT) {
- ASSIGN_LF(mmx2);
- dsp->avg_no_rnd_vc1_chroma_pixels_tab[0]=
ff_avg_vc1_chroma_mc8_mmx2_nornd;
+ ASSIGN_LF(mmxext);
+ dsp->avg_no_rnd_vc1_chroma_pixels_tab[0] =
ff_avg_vc1_chroma_mc8_mmxext_nornd;
} else if (mm_flags & AV_CPU_FLAG_3DNOW) {
dsp->avg_no_rnd_vc1_chroma_pixels_tab[0]=
ff_avg_vc1_chroma_mc8_3dnow_nornd;
}
diff --git a/libavcodec/x86/vc1dsp_yasm.asm b/libavcodec/x86/vc1dsp_yasm.asm
index d78e1b5..adfb977 100644
--- a/libavcodec/x86/vc1dsp_yasm.asm
+++ b/libavcodec/x86/vc1dsp_yasm.asm
@@ -239,19 +239,19 @@ cglobal vc1_h_loop_filter_internal_%1
VC1_H_LOOP_FILTER 4, r4
ret
-; void ff_vc1_v_loop_filter4_mmx2(uint8_t *src, int stride, int pq)
+; void ff_vc1_v_loop_filter4_mmxext(uint8_t *src, int stride, int pq)
cglobal vc1_v_loop_filter4_%1, 3,5,0
START_V_FILTER
call vc1_v_loop_filter_internal_%1
RET
-; void ff_vc1_h_loop_filter4_mmx2(uint8_t *src, int stride, int pq)
+; void ff_vc1_h_loop_filter4_mmxext(uint8_t *src, int stride, int pq)
cglobal vc1_h_loop_filter4_%1, 3,5,0
START_H_FILTER 4
call vc1_h_loop_filter_internal_%1
RET
-; void ff_vc1_v_loop_filter8_mmx2(uint8_t *src, int stride, int pq)
+; void ff_vc1_v_loop_filter8_mmxext(uint8_t *src, int stride, int pq)
cglobal vc1_v_loop_filter8_%1, 3,5,0
START_V_FILTER
call vc1_v_loop_filter_internal_%1
@@ -260,7 +260,7 @@ cglobal vc1_v_loop_filter8_%1, 3,5,0
call vc1_v_loop_filter_internal_%1
RET
-; void ff_vc1_h_loop_filter8_mmx2(uint8_t *src, int stride, int pq)
+; void ff_vc1_h_loop_filter8_mmxext(uint8_t *src, int stride, int pq)
cglobal vc1_h_loop_filter8_%1, 3,5,0
START_H_FILTER 4
call vc1_h_loop_filter_internal_%1
@@ -270,7 +270,7 @@ cglobal vc1_h_loop_filter8_%1, 3,5,0
%endmacro
%define PABSW PABSW_MMXEXT
-VC1_LF_MMX mmx2
+VC1_LF_MMX mmxext
INIT_XMM
; void ff_vc1_v_loop_filter8_sse2(uint8_t *src, int stride, int pq)
diff --git a/libavcodec/x86/vp3dsp.asm b/libavcodec/x86/vp3dsp.asm
index 7a88892..ebe763f 100644
--- a/libavcodec/x86/vp3dsp.asm
+++ b/libavcodec/x86/vp3dsp.asm
@@ -102,7 +102,7 @@ SECTION .text
mov [r0+r3 -1], r2w
%endmacro
-INIT_MMX mmx2
+INIT_MMX mmxext
cglobal vp3_v_loop_filter, 3, 4
%if ARCH_X86_64
movsxd r1, r1d
@@ -634,7 +634,7 @@ vp3_idct_funcs
movq [r0+r3 ], m5
%endmacro
-INIT_MMX mmx2
+INIT_MMX mmxext
cglobal vp3_idct_dc_add, 3, 4
%if ARCH_X86_64
movsxd r1, r1d
diff --git a/libavcodec/x86/vp3dsp_init.c b/libavcodec/x86/vp3dsp_init.c
index 45af041..91ac777 100644
--- a/libavcodec/x86/vp3dsp_init.c
+++ b/libavcodec/x86/vp3dsp_init.c
@@ -30,11 +30,13 @@ void ff_vp3_idct_add_mmx(uint8_t *dest, int line_size,
DCTELEM *block);
void ff_vp3_idct_put_sse2(uint8_t *dest, int line_size, DCTELEM *block);
void ff_vp3_idct_add_sse2(uint8_t *dest, int line_size, DCTELEM *block);
-void ff_vp3_idct_dc_add_mmx2(uint8_t *dest, int line_size,
- const DCTELEM *block);
+void ff_vp3_idct_dc_add_mmxext(uint8_t *dest, int line_size,
+ const DCTELEM *block);
-void ff_vp3_v_loop_filter_mmx2(uint8_t *src, int stride, int *bounding_values);
-void ff_vp3_h_loop_filter_mmx2(uint8_t *src, int stride, int *bounding_values);
+void ff_vp3_v_loop_filter_mmxext(uint8_t *src, int stride,
+ int *bounding_values);
+void ff_vp3_h_loop_filter_mmxext(uint8_t *src, int stride,
+ int *bounding_values);
av_cold void ff_vp3dsp_init_x86(VP3DSPContext *c, int flags)
{
@@ -50,11 +52,11 @@ av_cold void ff_vp3dsp_init_x86(VP3DSPContext *c, int flags)
#endif
if (HAVE_MMXEXT && cpuflags & AV_CPU_FLAG_MMXEXT) {
- c->idct_dc_add = ff_vp3_idct_dc_add_mmx2;
+ c->idct_dc_add = ff_vp3_idct_dc_add_mmxext;
if (!(flags & CODEC_FLAG_BITEXACT)) {
- c->v_loop_filter = ff_vp3_v_loop_filter_mmx2;
- c->h_loop_filter = ff_vp3_h_loop_filter_mmx2;
+ c->v_loop_filter = ff_vp3_v_loop_filter_mmxext;
+ c->h_loop_filter = ff_vp3_h_loop_filter_mmxext;
}
}
diff --git a/libavcodec/x86/vp8dsp-init.c b/libavcodec/x86/vp8dsp-init.c
index 64dd8ce..4db1bd3 100644
--- a/libavcodec/x86/vp8dsp-init.c
+++ b/libavcodec/x86/vp8dsp-init.c
@@ -29,16 +29,16 @@
/*
* MC functions
*/
-extern void ff_put_vp8_epel4_h4_mmx2 (uint8_t *dst, ptrdiff_t dststride,
+extern void ff_put_vp8_epel4_h4_mmxext(uint8_t *dst, ptrdiff_t dststride,
uint8_t *src, ptrdiff_t srcstride,
int height, int mx, int my);
-extern void ff_put_vp8_epel4_h6_mmx2 (uint8_t *dst, ptrdiff_t dststride,
+extern void ff_put_vp8_epel4_h6_mmxext(uint8_t *dst, ptrdiff_t dststride,
uint8_t *src, ptrdiff_t srcstride,
int height, int mx, int my);
-extern void ff_put_vp8_epel4_v4_mmx2 (uint8_t *dst, ptrdiff_t dststride,
+extern void ff_put_vp8_epel4_v4_mmxext(uint8_t *dst, ptrdiff_t dststride,
uint8_t *src, ptrdiff_t srcstride,
int height, int mx, int my);
-extern void ff_put_vp8_epel4_v6_mmx2 (uint8_t *dst, ptrdiff_t dststride,
+extern void ff_put_vp8_epel4_v6_mmxext(uint8_t *dst, ptrdiff_t dststride,
uint8_t *src, ptrdiff_t srcstride,
int height, int mx, int my);
@@ -80,7 +80,7 @@ extern void ff_put_vp8_epel8_v6_ssse3 (uint8_t *dst,
ptrdiff_t dststride,
uint8_t *src, ptrdiff_t srcstride,
int height, int mx, int my);
-extern void ff_put_vp8_bilinear4_h_mmx2 (uint8_t *dst, ptrdiff_t dststride,
+extern void ff_put_vp8_bilinear4_h_mmxext(uint8_t *dst, ptrdiff_t dststride,
uint8_t *src, ptrdiff_t srcstride,
int height, int mx, int my);
extern void ff_put_vp8_bilinear8_h_sse2 (uint8_t *dst, ptrdiff_t dststride,
@@ -93,7 +93,7 @@ extern void ff_put_vp8_bilinear8_h_ssse3 (uint8_t *dst,
ptrdiff_t dststride,
uint8_t *src, ptrdiff_t srcstride,
int height, int mx, int my);
-extern void ff_put_vp8_bilinear4_v_mmx2 (uint8_t *dst, ptrdiff_t dststride,
+extern void ff_put_vp8_bilinear4_v_mmxext(uint8_t *dst, ptrdiff_t dststride,
uint8_t *src, ptrdiff_t srcstride,
int height, int mx, int my);
extern void ff_put_vp8_bilinear8_v_sse2 (uint8_t *dst, ptrdiff_t dststride,
@@ -139,16 +139,16 @@ static void ff_put_vp8_ ## FILTERTYPE ## 8_ ## TAPTYPE ##
_ ## OPT( \
}
#if ARCH_X86_32
-TAP_W8 (mmx2, epel, h4)
-TAP_W8 (mmx2, epel, h6)
-TAP_W16(mmx2, epel, h6)
-TAP_W8 (mmx2, epel, v4)
-TAP_W8 (mmx2, epel, v6)
-TAP_W16(mmx2, epel, v6)
-TAP_W8 (mmx2, bilinear, h)
-TAP_W16(mmx2, bilinear, h)
-TAP_W8 (mmx2, bilinear, v)
-TAP_W16(mmx2, bilinear, v)
+TAP_W8 (mmxext, epel, h4)
+TAP_W8 (mmxext, epel, h6)
+TAP_W16(mmxext, epel, h6)
+TAP_W8 (mmxext, epel, v4)
+TAP_W8 (mmxext, epel, v6)
+TAP_W16(mmxext, epel, v6)
+TAP_W8 (mmxext, bilinear, h)
+TAP_W16(mmxext, bilinear, h)
+TAP_W8 (mmxext, bilinear, v)
+TAP_W16(mmxext, bilinear, v)
#endif
TAP_W16(sse2, epel, h6)
@@ -177,13 +177,13 @@ static void ff_put_vp8_epel ## SIZE ## _h ## TAPNUMX ## v
## TAPNUMY ## _ ## OPT
#if ARCH_X86_32
#define HVTAPMMX(x, y) \
-HVTAP(mmx2, 8, x, y, 4, 8) \
-HVTAP(mmx2, 8, x, y, 8, 16)
+HVTAP(mmxext, 8, x, y, 4, 8) \
+HVTAP(mmxext, 8, x, y, 8, 16)
-HVTAP(mmx2, 8, 6, 6, 16, 16)
+HVTAP(mmxext, 8, 6, 6, 16, 16)
#else
#define HVTAPMMX(x, y) \
-HVTAP(mmx2, 8, x, y, 4, 8)
+HVTAP(mmxext, 8, x, y, 4, 8)
#endif
HVTAPMMX(4, 4)
@@ -218,10 +218,10 @@ static void ff_put_vp8_bilinear ## SIZE ## _hv_ ## OPT( \
dst, dststride, tmp, SIZE, height, mx, my); \
}
-HVBILIN(mmx2, 8, 4, 8)
+HVBILIN(mmxext, 8, 4, 8)
#if ARCH_X86_32
-HVBILIN(mmx2, 8, 8, 16)
-HVBILIN(mmx2, 8, 16, 16)
+HVBILIN(mmxext, 8, 8, 16)
+HVBILIN(mmxext, 8, 16, 16)
#endif
HVBILIN(sse2, 8, 8, 16)
HVBILIN(sse2, 8, 16, 16)
@@ -283,7 +283,7 @@ extern void ff_vp8_h_loop_filter8uv_mbedge_ ## NAME(uint8_t
*dstU, \
int e, int i, int hvt);
DECLARE_LOOP_FILTER(mmx)
-DECLARE_LOOP_FILTER(mmx2)
+DECLARE_LOOP_FILTER(mmxext)
DECLARE_LOOP_FILTER(sse2)
DECLARE_LOOP_FILTER(ssse3)
DECLARE_LOOP_FILTER(sse4)
@@ -351,26 +351,26 @@ av_cold void ff_vp8dsp_init_x86(VP8DSPContext* c)
/* note that 4-tap width=16 functions are missing because w=16
* is only used for luma, and luma is always a copy or sixtap. */
if (mm_flags & AV_CPU_FLAG_MMXEXT) {
- VP8_MC_FUNC(2, 4, mmx2);
- VP8_BILINEAR_MC_FUNC(2, 4, mmx2);
+ VP8_MC_FUNC(2, 4, mmxext);
+ VP8_BILINEAR_MC_FUNC(2, 4, mmxext);
#if ARCH_X86_32
- VP8_LUMA_MC_FUNC(0, 16, mmx2);
- VP8_MC_FUNC(1, 8, mmx2);
- VP8_BILINEAR_MC_FUNC(0, 16, mmx2);
- VP8_BILINEAR_MC_FUNC(1, 8, mmx2);
-
- c->vp8_v_loop_filter_simple = ff_vp8_v_loop_filter_simple_mmx2;
- c->vp8_h_loop_filter_simple = ff_vp8_h_loop_filter_simple_mmx2;
-
- c->vp8_v_loop_filter16y_inner = ff_vp8_v_loop_filter16y_inner_mmx2;
- c->vp8_h_loop_filter16y_inner = ff_vp8_h_loop_filter16y_inner_mmx2;
- c->vp8_v_loop_filter8uv_inner = ff_vp8_v_loop_filter8uv_inner_mmx2;
- c->vp8_h_loop_filter8uv_inner = ff_vp8_h_loop_filter8uv_inner_mmx2;
-
- c->vp8_v_loop_filter16y = ff_vp8_v_loop_filter16y_mbedge_mmx2;
- c->vp8_h_loop_filter16y = ff_vp8_h_loop_filter16y_mbedge_mmx2;
- c->vp8_v_loop_filter8uv = ff_vp8_v_loop_filter8uv_mbedge_mmx2;
- c->vp8_h_loop_filter8uv = ff_vp8_h_loop_filter8uv_mbedge_mmx2;
+ VP8_LUMA_MC_FUNC(0, 16, mmxext);
+ VP8_MC_FUNC(1, 8, mmxext);
+ VP8_BILINEAR_MC_FUNC(0, 16, mmxext);
+ VP8_BILINEAR_MC_FUNC(1, 8, mmxext);
+
+ c->vp8_v_loop_filter_simple = ff_vp8_v_loop_filter_simple_mmxext;
+ c->vp8_h_loop_filter_simple = ff_vp8_h_loop_filter_simple_mmxext;
+
+ c->vp8_v_loop_filter16y_inner = ff_vp8_v_loop_filter16y_inner_mmxext;
+ c->vp8_h_loop_filter16y_inner = ff_vp8_h_loop_filter16y_inner_mmxext;
+ c->vp8_v_loop_filter8uv_inner = ff_vp8_v_loop_filter8uv_inner_mmxext;
+ c->vp8_h_loop_filter8uv_inner = ff_vp8_h_loop_filter8uv_inner_mmxext;
+
+ c->vp8_v_loop_filter16y = ff_vp8_v_loop_filter16y_mbedge_mmxext;
+ c->vp8_h_loop_filter16y = ff_vp8_h_loop_filter16y_mbedge_mmxext;
+ c->vp8_v_loop_filter8uv = ff_vp8_v_loop_filter8uv_mbedge_mmxext;
+ c->vp8_h_loop_filter8uv = ff_vp8_h_loop_filter8uv_mbedge_mmxext;
#endif
}
diff --git a/libavcodec/x86/vp8dsp.asm b/libavcodec/x86/vp8dsp.asm
index 531b205..3b5fd18 100644
--- a/libavcodec/x86/vp8dsp.asm
+++ b/libavcodec/x86/vp8dsp.asm
@@ -339,7 +339,7 @@ INIT_XMM ssse3
FILTER_SSSE3 8
; 4x4 block, H-only 4-tap filter
-INIT_MMX mmx2
+INIT_MMX mmxext
cglobal put_vp8_epel4_h4, 6, 6 + npicregs, 0, dst, dststride, src, srcstride,
height, mx, picreg
shl mxd, 4
%ifdef PIC
@@ -387,7 +387,7 @@ cglobal put_vp8_epel4_h4, 6, 6 + npicregs, 0, dst,
dststride, src, srcstride, he
REP_RET
; 4x4 block, H-only 6-tap filter
-INIT_MMX mmx2
+INIT_MMX mmxext
cglobal put_vp8_epel4_h6, 6, 6 + npicregs, 0, dst, dststride, src, srcstride,
height, mx, picreg
lea mxd, [mxq*3]
%ifdef PIC
@@ -674,7 +674,7 @@ cglobal put_vp8_epel%1_v6, 7, 7, 8, dst, dststride, src,
srcstride, height, picr
REP_RET
%endmacro
-INIT_MMX mmx2
+INIT_MMX mmxext
FILTER_V 4
INIT_XMM sse2
FILTER_V 8
@@ -770,7 +770,7 @@ cglobal put_vp8_bilinear%1_h, 6, 6 + npicregs, 7, dst,
dststride, src, srcstride
REP_RET
%endmacro
-INIT_MMX mmx2
+INIT_MMX mmxext
FILTER_BILINEAR 4
INIT_XMM sse2
FILTER_BILINEAR 8
@@ -1612,7 +1612,7 @@ cglobal vp8_%1_loop_filter_simple, 3, %2, 8, dst, stride,
flim, cntr
INIT_MMX mmx
SIMPLE_LOOPFILTER v, 4
SIMPLE_LOOPFILTER h, 5
-INIT_MMX mmx2
+INIT_MMX mmxext
SIMPLE_LOOPFILTER v, 4
SIMPLE_LOOPFILTER h, 5
%endif
@@ -1836,7 +1836,7 @@ cglobal vp8_%1_loop_filter16y_inner, 5, 5, 13, dst,
stride, flimE, flimI, hevthr
psubusb m6, m5 ; q2-q1
por m6, m4 ; abs(q2-q1)
-%if notcpuflag(mmx2)
+%if notcpuflag(mmxext)
mova m4, m_flimI
pxor m3, m3
psubusb m0, m4
@@ -1876,7 +1876,7 @@ cglobal vp8_%1_loop_filter16y_inner, 5, 5, 13, dst,
stride, flimE, flimI, hevthr
psubusb m1, m3 ; p1-p0
psubusb m6, m2 ; p0-p1
por m1, m6 ; abs(p1-p0)
-%if notcpuflag(mmx2)
+%if notcpuflag(mmxext)
mova m6, m1
psubusb m1, m4
psubusb m6, m_hevthr
@@ -1907,7 +1907,7 @@ cglobal vp8_%1_loop_filter16y_inner, 5, 5, 13, dst,
stride, flimE, flimI, hevthr
psubusb m1, m5 ; q0-q1
psubusb m7, m4 ; q1-q0
por m1, m7 ; abs(q1-q0)
-%if notcpuflag(mmx2)
+%if notcpuflag(mmxext)
mova m7, m1
psubusb m1, m6
psubusb m7, m_hevthr
@@ -2015,14 +2015,14 @@ cglobal vp8_%1_loop_filter16y_inner, 5, 5, 13, dst,
stride, flimE, flimI, hevthr
%else
mova m6, m_maskres
%endif
-%if notcpuflag(mmx2)
+%if notcpuflag(mmxext)
mova m7, [pb_1]
%else ; mmxext/sse2
pxor m7, m7
%endif
pand m0, m6
pand m1, m6
-%if notcpuflag(mmx2)
+%if notcpuflag(mmxext)
paddusb m0, m7
pand m1, [pb_FE]
pandn m7, m0
@@ -2098,7 +2098,7 @@ INNER_LOOPFILTER h, 16
INNER_LOOPFILTER v, 8
INNER_LOOPFILTER h, 8
-INIT_MMX mmx2
+INIT_MMX mmxext
INNER_LOOPFILTER v, 16
INNER_LOOPFILTER h, 16
INNER_LOOPFILTER v, 8
@@ -2344,7 +2344,7 @@ cglobal vp8_%1_loop_filter16y_mbedge, 5, 5, 15, dst1,
stride, flimE, flimI, hevt
psubusb m6, m5 ; q2-q1
por m6, m4 ; abs(q2-q1)
-%if notcpuflag(mmx2)
+%if notcpuflag(mmxext)
mova m4, m_flimI
pxor m3, m3
psubusb m0, m4
@@ -2384,7 +2384,7 @@ cglobal vp8_%1_loop_filter16y_mbedge, 5, 5, 15, dst1,
stride, flimE, flimI, hevt
psubusb m1, m3 ; p1-p0
psubusb m6, m2 ; p0-p1
por m1, m6 ; abs(p1-p0)
-%if notcpuflag(mmx2)
+%if notcpuflag(mmxext)
mova m6, m1
psubusb m1, m4
psubusb m6, m_hevthr
@@ -2415,7 +2415,7 @@ cglobal vp8_%1_loop_filter16y_mbedge, 5, 5, 15, dst1,
stride, flimE, flimI, hevt
psubusb m1, m5 ; q0-q1
psubusb m7, m4 ; q1-q0
por m1, m7 ; abs(q1-q0)
-%if notcpuflag(mmx2)
+%if notcpuflag(mmxext)
mova m7, m1
psubusb m1, m6
psubusb m7, m_hevthr
@@ -2756,7 +2756,7 @@ MBEDGE_LOOPFILTER h, 16
MBEDGE_LOOPFILTER v, 8
MBEDGE_LOOPFILTER h, 8
-INIT_MMX mmx2
+INIT_MMX mmxext
MBEDGE_LOOPFILTER v, 16
MBEDGE_LOOPFILTER h, 16
MBEDGE_LOOPFILTER v, 8
diff --git a/libavutil/x86/x86inc.asm b/libavutil/x86/x86inc.asm
index 03e6c07..9dbd9b0 100644
--- a/libavutil/x86/x86inc.asm
+++ b/libavutil/x86/x86inc.asm
@@ -555,10 +555,10 @@ SECTION .note.GNU-stack noalloc noexec nowrite progbits
; cpuflags
%assign cpuflags_mmx (1<<0)
-%assign cpuflags_mmx2 (1<<1) | cpuflags_mmx
+%assign cpuflags_mmxext (1<<1) | cpuflags_mmx
%assign cpuflags_3dnow (1<<2) | cpuflags_mmx
%assign cpuflags_3dnowext (1<<3) | cpuflags_3dnow
-%assign cpuflags_sse (1<<4) | cpuflags_mmx2
+%assign cpuflags_sse (1<<4) | cpuflags_mmxext
%assign cpuflags_sse2 (1<<5) | cpuflags_sse
%assign cpuflags_sse2slow (1<<6) | cpuflags_sse2
%assign cpuflags_sse3 (1<<7) | cpuflags_sse2
@@ -611,7 +611,7 @@ SECTION .note.GNU-stack noalloc noexec nowrite progbits
%elifidn %1, sse3
%define movu lddqu
%endif
- %if notcpuflag(mmx2)
+ %if notcpuflag(mmxext)
CPU basicnop
%endif
%else
diff --git a/libavutil/x86/x86util.asm b/libavutil/x86/x86util.asm
index 4c80809..1125e09 100644
--- a/libavutil/x86/x86util.asm
+++ b/libavutil/x86/x86util.asm
@@ -540,7 +540,7 @@
%if mmsize == 16
pshuflw %1, %2, (%3)*0x55
punpcklqdq %1, %1
-%elif cpuflag(mmx2)
+%elif cpuflag(mmxext)
pshufw %1, %2, (%3)*0x55
%else
%ifnidn %1, %2
diff --git a/libswscale/x86/output.asm b/libswscale/x86/output.asm
index 1c68113..2aac55d 100644
--- a/libswscale/x86/output.asm
+++ b/libswscale/x86/output.asm
@@ -249,7 +249,7 @@ cglobal yuv2planeX_%1, %3, 8, %2, filter, fltsize, src,
dst, w, dither, offset
%define PALIGNR PALIGNR_MMX
%if ARCH_X86_32
-INIT_MMX mmx2
+INIT_MMX mmxext
yuv2planeX_fn 8, 0, 7
yuv2planeX_fn 9, 0, 5
yuv2planeX_fn 10, 0, 5
@@ -391,7 +391,7 @@ INIT_MMX mmx
yuv2plane1_fn 8, 0, 5
yuv2plane1_fn 16, 0, 3
-INIT_MMX mmx2
+INIT_MMX mmxext
yuv2plane1_fn 9, 0, 3
yuv2plane1_fn 10, 0, 3
%endif
diff --git a/libswscale/x86/swscale.c b/libswscale/x86/swscale.c
index 5ed76f5..767b39d 100644
--- a/libswscale/x86/swscale.c
+++ b/libswscale/x86/swscale.c
@@ -249,7 +249,7 @@ extern void ff_yuv2planeX_ ## size ## _ ## opt(const
int16_t *filter, int filter
VSCALEX_FUNC(10, opt)
#if ARCH_X86_32
-VSCALEX_FUNCS(mmx2);
+VSCALEX_FUNCS(mmxext);
#endif
VSCALEX_FUNCS(sse2);
VSCALEX_FUNCS(sse4);
@@ -266,7 +266,7 @@ extern void ff_yuv2plane1_ ## size ## _ ## opt(const
int16_t *src, uint8_t *dst,
VSCALE_FUNC(16, opt1)
#if ARCH_X86_32
-VSCALE_FUNCS(mmx, mmx2);
+VSCALE_FUNCS(mmx, mmxext);
#endif
VSCALE_FUNCS(sse2, sse2);
VSCALE_FUNC(16, sse4);
@@ -360,7 +360,7 @@ switch(c->dstBpc){ \
if (cpu_flags & AV_CPU_FLAG_MMX) {
ASSIGN_MMX_SCALE_FUNC(c->hyScale, c->hLumFilterSize, mmx, mmx);
ASSIGN_MMX_SCALE_FUNC(c->hcScale, c->hChrFilterSize, mmx, mmx);
- ASSIGN_VSCALE_FUNC(c->yuv2plane1, mmx, mmx2, cpu_flags &
AV_CPU_FLAG_MMXEXT);
+ ASSIGN_VSCALE_FUNC(c->yuv2plane1, mmx, mmxext, cpu_flags &
AV_CPU_FLAG_MMXEXT);
switch (c->srcFormat) {
case PIX_FMT_Y400A:
@@ -393,7 +393,7 @@ switch(c->dstBpc){ \
}
}
if (cpu_flags & AV_CPU_FLAG_MMXEXT) {
- ASSIGN_VSCALEX_FUNC(c->yuv2planeX, mmx2, , 1);
+ ASSIGN_VSCALEX_FUNC(c->yuv2planeX, mmxext, , 1);
}
#endif
#define ASSIGN_SSE_SCALE_FUNC(hscalefn, filtersize, opt1, opt2) \
--
1.7.2.5
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel
