>From 253 to 51 cycles on Arrandale and Win64.
44 cycles on SandyBridge.
---
libavcodec/x86/sbrdsp.asm | 30 ++++++++++++++++++++++++++++++
libavcodec/x86/sbrdsp_init.c | 5 +++++
2 files changed, 35 insertions(+)
diff --git a/libavcodec/x86/sbrdsp.asm b/libavcodec/x86/sbrdsp.asm
index 5e54505..5c21b11 100644
--- a/libavcodec/x86/sbrdsp.asm
+++ b/libavcodec/x86/sbrdsp.asm
@@ -245,3 +245,33 @@ cglobal sbr_neg_odd_64, 1,2,4,z
cmp zq, r1q
jne .loop
REP_RET
+
+INIT_XMM sse2
+cglobal sbr_qmf_pre_shuffle, 1,4,7,z
+%define OFFSET (32*4-2*mmsize)
+ mov r3q, OFFSET
+ lea r1q, [zq + (32+1)*4]
+ lea r2q, [zq + 64*4]
+ mova m6, [ps_neg]
+.loop:
+ movu m0, [r1q]
+ movu m2, [r1q + mmsize]
+ movu m1, [zq + r3q + 4 + mmsize]
+ movu m3, [zq + r3q + 4]
+
+ pxor m2, m6
+ pxor m0, m6
+ pshufd m2, m2, q0123
+ pshufd m0, m0, q0123
+ SBUTTERFLY dq, 2, 3, 5
+ SBUTTERFLY dq, 0, 1, 4
+ mova [r2q + 2*r3q + 0*mmsize], m2
+ mova [r2q + 2*r3q + 1*mmsize], m3
+ mova [r2q + 2*r3q + 2*mmsize], m0
+ mova [r2q + 2*r3q + 3*mmsize], m1
+ add r1q, 2*mmsize
+ sub r3q, 2*mmsize
+ jge .loop
+ mova m2, [zq]
+ movq [r2q], m2
+ REP_RET
diff --git a/libavcodec/x86/sbrdsp_init.c b/libavcodec/x86/sbrdsp_init.c
index cb65a23..46e183a 100644
--- a/libavcodec/x86/sbrdsp_init.c
+++ b/libavcodec/x86/sbrdsp_init.c
@@ -34,6 +34,7 @@ void ff_sbr_hf_gen_sse(float (*X_high)[2], const float
(*X_low)[2],
float bw, int start, int end);
void ff_sbr_neg_odd_64_sse(float *z);
void ff_sbr_qmf_post_shuffle_sse(float W[32][2], const float *z);
+void ff_sbr_qmf_pre_shuffle_sse2(float *z);
av_cold void ff_sbrdsp_init_x86(SBRDSPContext *s)
{
@@ -47,4 +48,8 @@ av_cold void ff_sbrdsp_init_x86(SBRDSPContext *s)
s->hf_gen = ff_sbr_hf_gen_sse;
s->qmf_post_shuffle = ff_sbr_qmf_post_shuffle_sse;
}
+
+ if (EXTERNAL_SSE2(mm_flags)) {
+ s->qmf_pre_shuffle = ff_sbr_qmf_pre_shuffle_sse2;
+ }
}
--
1.8.0.msysgit.0
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel
