MrBesen/ffmpeg
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

swr: add int16_to_float_sse2()

Browse Source 
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
This commit is contained in:
Michael Niedermayer 2012-04-28 18:52:48 +02:00
parent 64943b47e7
commit b72a0f9c23
2 changed files with 39 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

36
libswresample/x86/audio_convert.asm
View File

@ -94,6 +94,40 @@ int32_to_float_u_int %+ SUFFIX
REP_RET
%endmacro
%macro INT16_TO_FLOAT 1
cglobal int16_to_float_%1, 3, 3, 4, dst, src, len
mov srcq, [srcq]
mov dstq, [dstq]
%ifidn %1, a
test dstq, mmsize-1
jne int16_to_float_u_int %+ SUFFIX
test srcq, mmsize-1
jne int16_to_float_u_int %+ SUFFIX
%else
int16_to_float_u_int %+ SUFFIX
%endif
add dstq, lenq
shr lenq, 1
add srcq, lenq
neg lenq
mova m3, [flt2pm31]
.next:
mov%1 m2, [srcq+lenq]
pxor m0, m0
pxor m1, m1
punpcklwd m0, m2
punpckhwd m1, m2
cvtdq2ps m0, m0
cvtdq2ps m1, m1
mulps m0, m3
mulps m1, m3
mov%1 [ dstq+2*lenq], m0
mov%1 [mmsize + dstq+2*lenq], m1
add lenq, mmsize
jl .next
REP_RET
%endmacro
INIT_MMX mmx
INT16_TO_INT32 u
@ -106,3 +140,5 @@ INT16_TO_INT32 a
INIT_XMM sse2
INT32_TO_FLOAT u
INT32_TO_FLOAT a
INT16_TO_FLOAT u
INT16_TO_FLOAT a

3
libswresample/x86/swresample_x86.c
View File

@ -27,6 +27,7 @@ MULTI_CAPS_FUNC_DECL(mmx)
MULTI_CAPS_FUNC_DECL(sse)
void ff_int32_to_float_a_sse2(uint8_t **dst, const uint8_t **src, int len);
void ff_int16_to_float_a_sse2(uint8_t **dst, const uint8_t **src, int len);
void swri_audio_convert_init_x86(struct AudioConvert *ac,
enum AVSampleFormat out_fmt,
@ -50,5 +51,7 @@ MULTI_CAPS_FUNC(AV_CPU_FLAG_SSE, sse)
if(mm_flags & AV_CPU_FLAG_SSE2) {
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;
if( out_fmt == AV_SAMPLE_FMT_FLT && in_fmt == AV_SAMPLE_FMT_S16 || out_fmt == AV_SAMPLE_FMT_FLTP && in_fmt == AV_SAMPLE_FMT_S16P)
ac->simd_f = ff_int16_to_float_a_sse2;
}
}