---
libavresample/x86/audio_convert.asm | 49 ++++++++++++++++++++++++++++++++
libavresample/x86/audio_convert_init.c | 9 ++++++
2 files changed, 58 insertions(+), 0 deletions(-)
diff --git a/libavresample/x86/audio_convert.asm
b/libavresample/x86/audio_convert.asm
index 26bfb9d..13777c3 100644
--- a/libavresample/x86/audio_convert.asm
+++ b/libavresample/x86/audio_convert.asm
@@ -971,3 +971,52 @@ CONV_S16_TO_FLTP_6CH
INIT_XMM avx
CONV_S16_TO_FLTP_6CH
%endif
+
+;------------------------------------------------------------------------------
+; void ff_conv_flt_to_s16p_2ch(int16_t *const *dst, float *src, int len,
+; int channels);
+;------------------------------------------------------------------------------
+
+%macro CONV_FLT_TO_S16P_2CH 0
+cglobal conv_flt_to_s16p_2ch, 3,4,7, dst0, src, len, dst1
+ lea lenq, [2*lend]
+ mov dst1q, [dst0q+gprsize]
+ mov dst0q, [dst0q ]
+ lea srcq, [srcq+4*lenq]
+ add dst0q, lenq
+ add dst1q, lenq
+ neg lenq
+ mova m6, [pf_s16_scale]
+ ALIGN 16
+.loop:
+ mova m0, [srcq+4*lenq ]
+ mova m1, [srcq+4*lenq+ mmsize]
+ mova m2, [srcq+4*lenq+2*mmsize]
+ mova m3, [srcq+4*lenq+3*mmsize]
+ shufps m4, m0, m1, q2020
+ shufps m5, m0, m1, q3131
+ shufps m0, m2, m3, q2020
+ shufps m1, m2, m3, q3131
+ mulps m4, m4, m6
+ mulps m5, m5, m6
+ mulps m0, m0, m6
+ mulps m1, m1, m6
+ cvtps2dq m4, m4
+ cvtps2dq m5, m5
+ cvtps2dq m0, m0
+ cvtps2dq m1, m1
+ packssdw m4, m0
+ packssdw m5, m1
+ mova [dst0q+lenq], m4
+ mova [dst1q+lenq], m5
+ add lenq, mmsize
+ jl .loop
+ REP_RET
+%endmacro
+
+INIT_XMM sse2
+CONV_FLT_TO_S16P_2CH
+%if HAVE_AVX
+INIT_XMM avx
+CONV_FLT_TO_S16P_2CH
+%endif
diff --git a/libavresample/x86/audio_convert_init.c
b/libavresample/x86/audio_convert_init.c
index 80b4a17..7e04dcc 100644
--- a/libavresample/x86/audio_convert_init.c
+++ b/libavresample/x86/audio_convert_init.c
@@ -100,6 +100,11 @@ extern void ff_conv_s16_to_fltp_6ch_sse4(float *const
*dst, int16_t *src,
extern void ff_conv_s16_to_fltp_6ch_avx (float *const *dst, int16_t *src,
int len, int channels);
+extern void ff_conv_flt_to_s16p_2ch_sse2(int16_t *const *dst, float *src,
+ int len, int channels);
+extern void ff_conv_flt_to_s16p_2ch_avx (int16_t *const *dst, float *src,
+ int len, int channels);
+
av_cold void ff_audio_convert_init_x86(AudioConvert *ac)
{
#if HAVE_YASM
@@ -152,6 +157,8 @@ av_cold void ff_audio_convert_init_x86(AudioConvert *ac)
2, 16, 8, "SSE2",
ff_conv_s16_to_fltp_2ch_sse2);
ff_audio_convert_set_func(ac, AV_SAMPLE_FMT_FLTP, AV_SAMPLE_FMT_S16,
6, 16, 4, "SSE2",
ff_conv_s16_to_fltp_6ch_sse2);
+ ff_audio_convert_set_func(ac, AV_SAMPLE_FMT_S16P, AV_SAMPLE_FMT_FLT,
+ 2, 16, 8, "SSE2",
ff_conv_flt_to_s16p_2ch_sse2);
}
if (mm_flags & AV_CPU_FLAG_SSE4 && HAVE_SSE) {
ff_audio_convert_set_func(ac, AV_SAMPLE_FMT_FLT, AV_SAMPLE_FMT_S16,
@@ -180,6 +187,8 @@ av_cold void ff_audio_convert_init_x86(AudioConvert *ac)
2, 16, 8, "AVX",
ff_conv_s16_to_fltp_2ch_avx);
ff_audio_convert_set_func(ac, AV_SAMPLE_FMT_FLTP, AV_SAMPLE_FMT_S16,
6, 16, 4, "AVX",
ff_conv_s16_to_fltp_6ch_avx);
+ ff_audio_convert_set_func(ac, AV_SAMPLE_FMT_S16P, AV_SAMPLE_FMT_FLT,
+ 2, 16, 8, "AVX",
ff_conv_flt_to_s16p_2ch_avx);
}
#endif
}
--
1.7.1
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel