[FFmpeg-cvslog] lavfi/gradfun: remove rounding to match C and SSE code.

[Clément Bœsch]

ffmpeg | branch: master | Clément Bœsch <ubitux at gmail.com> | Fri Dec  7 00:41:04 2012 +0100| [1ae44c87c924b69a0657256fbaa8ad140df2f27c] | committer: Anton Khirnov

lavfi/gradfun: remove rounding to match C and SSE code.

There is no noticable benefit for such precision.

Signed-off-by: Anton Khirnov <anton at khirnov.net>

> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=1ae44c87c924b69a0657256fbaa8ad140df2f27c
---

 libavfilter/x86/vf_gradfun.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/libavfilter/x86/vf_gradfun.c b/libavfilter/x86/vf_gradfun.c
index a9e069a..65a8172 100644
--- a/libavfilter/x86/vf_gradfun.c
+++ b/libavfilter/x86/vf_gradfun.c
@@ -136,9 +136,9 @@ static void gradfun_filter_line_ssse3(uint8_t *dst, uint8_t *src, uint16_t *dc,
         "psubw      %%xmm6, %%xmm2 \n"
         "pminsw     %%xmm7, %%xmm2 \n" // m = -max(0, 127-m)
         "pmullw     %%xmm2, %%xmm2 \n"
-        "psllw          $1, %%xmm2 \n"
+        "psllw          $2, %%xmm1 \n"
         "paddw      %%xmm4, %%xmm0 \n" // pix += dither
-        "pmulhrsw   %%xmm2, %%xmm1 \n" // m = m*m*delta >> 14
+        "pmulhw     %%xmm2, %%xmm1 \n" // m = m*m*delta >> 14
         "paddw      %%xmm1, %%xmm0 \n" // pix += m
         "psraw          $7, %%xmm0 \n"
         "packuswb   %%xmm0, %%xmm0 \n"