diff options
| author | James Almer <jamrial@gmail.com> | 2014-11-06 20:43:06 -0300 |
|---|---|---|
| committer | James Almer <jamrial@gmail.com> | 2014-11-06 20:54:00 -0300 |
| commit | b385c4c6a316f798b6a14418f09e545cda7fef7f (patch) | |
| tree | 2fb3571e48727956078258fe0da9325ff70c5583 /libswresample/x86/audio_convert_init.c | |
| parent | 93ab6693d8cf6e01ba415dde89860e07abed9e0b (diff) | |
| download | ffmpeg-streaming-b385c4c6a316f798b6a14418f09e545cda7fef7f.zip ffmpeg-streaming-b385c4c6a316f798b6a14418f09e545cda7fef7f.tar.gz | |
x86/swr: replace sse4 instructions in pack_6ch with sse ones
There's no benefit from using blendps here except on CPUs with AVX, where
it's faster than shufps according to Intel's documentation.
As such, rename the sse4 functions to sse/sse2 and use shufps instead.
Reviewed-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: James Almer <jamrial@gmail.com>
Diffstat (limited to 'libswresample/x86/audio_convert_init.c')
| -rw-r--r-- | libswresample/x86/audio_convert_init.c | 23 |
1 files changed, 12 insertions, 11 deletions
diff --git a/libswresample/x86/audio_convert_init.c b/libswresample/x86/audio_convert_init.c index a26cdf6..769575d 100644 --- a/libswresample/x86/audio_convert_init.c +++ b/libswresample/x86/audio_convert_init.c @@ -58,7 +58,12 @@ MULTI_CAPS_FUNC(SSE2, sse2) ac->simd_f = ff_pack_6ch_float_to_float_a_mmx; } } - + if(EXTERNAL_SSE(mm_flags)) { + 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_sse; + } + } if(EXTERNAL_SSE2(mm_flags)) { 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_sse2; @@ -105,6 +110,12 @@ MULTI_CAPS_FUNC(SSE2, sse2) if( out_fmt == AV_SAMPLE_FMT_S16P && in_fmt == AV_SAMPLE_FMT_FLT) ac->simd_f = ff_unpack_2ch_float_to_int16_a_sse2; } + if(channels == 6) { + if( out_fmt == AV_SAMPLE_FMT_FLT && in_fmt == AV_SAMPLE_FMT_S32P) + ac->simd_f = ff_pack_6ch_int32_to_float_a_sse2; + if( out_fmt == AV_SAMPLE_FMT_S32 && in_fmt == AV_SAMPLE_FMT_FLTP) + ac->simd_f = ff_pack_6ch_float_to_int32_a_sse2; + } } if(EXTERNAL_SSSE3(mm_flags)) { if(channels == 2) { @@ -116,16 +127,6 @@ MULTI_CAPS_FUNC(SSE2, sse2) ac->simd_f = ff_unpack_2ch_int16_to_float_a_ssse3; } } - if(EXTERNAL_SSE4(mm_flags)) { - 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( out_fmt == AV_SAMPLE_FMT_FLT && in_fmt == AV_SAMPLE_FMT_S32P) - ac->simd_f = ff_pack_6ch_int32_to_float_a_sse4; - if( out_fmt == AV_SAMPLE_FMT_S32 && in_fmt == AV_SAMPLE_FMT_FLTP) - ac->simd_f = ff_pack_6ch_float_to_int32_a_sse4; - } - } if(EXTERNAL_AVX(mm_flags)) { 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; |
