diff options
| author | Justin Ruggles <justin.ruggles@gmail.com> | 2012-05-13 19:09:01 +0200 |
|---|---|---|
| committer | Michael Niedermayer <michaelni@gmx.at> | 2012-05-13 19:31:59 +0200 |
| commit | 6f67d9833b5a7d78d27c283d7e4238e6d3e5fd98 (patch) | |
| tree | 916cc36d53394a026d938f7a1e0ed69ebe4e0d43 /libswresample/x86/swresample_x86.c | |
| parent | 6108e1be2fb97bdfa63e8f9a039b8daff3caec3d (diff) | |
libswresample: Implement MMX, SSE4 and AVX 6ch float and int32 packing function.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Diffstat (limited to 'libswresample/x86/swresample_x86.c')
| -rw-r--r-- | libswresample/x86/swresample_x86.c | 20 |
1 files changed, 19 insertions, 1 deletions
diff --git a/libswresample/x86/swresample_x86.c b/libswresample/x86/swresample_x86.c index e600100deb..f2349b505a 100644 --- a/libswresample/x86/swresample_x86.c +++ b/libswresample/x86/swresample_x86.c @@ -24,9 +24,10 @@ #define PROTO(pre, in, out, cap) void ff ## pre ## _ ##in## _to_ ##out## _a_ ##cap(uint8_t **dst, const uint8_t **src, int len); #define PROTO2(pre, out, cap) PROTO(pre, int16, out, cap) PROTO(pre, int32, out, cap) PROTO(pre, float, out, cap) #define PROTO3(pre, cap) PROTO2(pre, int16, cap) PROTO2(pre, int32, cap) PROTO2(pre, float, cap) -#define PROTO4(pre) PROTO3(pre, mmx) PROTO3(pre, sse) PROTO3(pre, sse2) PROTO3(pre, ssse3) PROTO3(pre, avx) +#define PROTO4(pre) PROTO3(pre, mmx) PROTO3(pre, sse) PROTO3(pre, sse2) PROTO3(pre, ssse3) PROTO3(pre, sse4) PROTO3(pre, avx) PROTO4() PROTO4(_pack_2ch) +PROTO4(_pack_6ch) PROTO4(_unpack_2ch) void swri_audio_convert_init_x86(struct AudioConvert *ac, @@ -50,6 +51,13 @@ void swri_audio_convert_init_x86(struct AudioConvert *ac, MULTI_CAPS_FUNC(AV_CPU_FLAG_MMX, mmx) MULTI_CAPS_FUNC(AV_CPU_FLAG_SSE, sse) + if(mm_flags & AV_CPU_FLAG_MMX) { + if(channels == 6) { + if( out_fmt == AV_SAMPLE_FMT_FLT && in_fmt == AV_SAMPLE_FMT_FLTP || out_fmt == AV_SAMPLE_FMT_S32 && in_fmt == AV_SAMPLE_FMT_S32P) + ac->simd_f = ff_pack_6ch_float_to_float_a_mmx; + } + } + if(mm_flags & AV_CPU_FLAG_SSE) { if(channels == 2) { if( out_fmt == AV_SAMPLE_FMT_FLT && in_fmt == AV_SAMPLE_FMT_FLTP || out_fmt == AV_SAMPLE_FMT_S32 && in_fmt == AV_SAMPLE_FMT_S32P) @@ -111,8 +119,18 @@ MULTI_CAPS_FUNC(AV_CPU_FLAG_SSE, sse) ac->simd_f = ff_unpack_2ch_int16_to_float_a_ssse3; } } + if(mm_flags & AV_CPU_FLAG_SSE4) { + if(channels == 6) { + if( out_fmt == AV_SAMPLE_FMT_FLT && in_fmt == AV_SAMPLE_FMT_FLTP || out_fmt == AV_SAMPLE_FMT_S32 && in_fmt == AV_SAMPLE_FMT_S32P) + ac->simd_f = ff_pack_6ch_float_to_float_a_sse4; + } + } if(HAVE_AVX && mm_flags & AV_CPU_FLAG_AVX) { if( out_fmt == AV_SAMPLE_FMT_FLT && in_fmt == AV_SAMPLE_FMT_S32 || out_fmt == AV_SAMPLE_FMT_FLTP && in_fmt == AV_SAMPLE_FMT_S32P) ac->simd_f = ff_int32_to_float_a_avx; + if(channels == 6) { + if( out_fmt == AV_SAMPLE_FMT_FLT && in_fmt == AV_SAMPLE_FMT_FLTP || out_fmt == AV_SAMPLE_FMT_S32 && in_fmt == AV_SAMPLE_FMT_S32P) + ac->simd_f = ff_pack_6ch_float_to_float_a_avx; + } } } |