[FFmpeg-cvslog] x86/float_dsp: zero extend offset from ff_scalarproduct_float_sse

James Almer git at videolan.org
Fri Jan 8 21:08:38 CET 2016


ffmpeg | branch: release/2.8 | James Almer <jamrial at gmail.com> | Fri Jan  8 12:08:56 2016 -0300| [3e3aa25afa1039775215882aff9d1048165e1c2a] | committer: James Almer

x86/float_dsp: zero extend offset from ff_scalarproduct_float_sse

Reviewed-by: Christophe Gisquet <christophe.gisquet at gmail.com>
Signed-off-by: James Almer <jamrial at gmail.com>
(cherry picked from commit dc79824deb6ac0ce236589c618744b33629201cd)

> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=3e3aa25afa1039775215882aff9d1048165e1c2a
---

 libavutil/x86/float_dsp.asm |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/libavutil/x86/float_dsp.asm b/libavutil/x86/float_dsp.asm
index 87229d4..2800cc6 100644
--- a/libavutil/x86/float_dsp.asm
+++ b/libavutil/x86/float_dsp.asm
@@ -332,10 +332,10 @@ VECTOR_FMUL_REVERSE
 ; float scalarproduct_float_sse(const float *v1, const float *v2, int len)
 INIT_XMM sse
 cglobal scalarproduct_float, 3,3,2, v1, v2, offset
+    shl   offsetd, 2
+    add       v1q, offsetq
+    add       v2q, offsetq
     neg   offsetq
-    shl   offsetq, 2
-    sub       v1q, offsetq
-    sub       v2q, offsetq
     xorps    xmm0, xmm0
 .loop:
     movaps   xmm1, [v1q+offsetq]



More information about the ffmpeg-cvslog mailing list