[x264-devel] Add fenc prefetching to adaptive quant

[Jason Garrett-Glaser]

x264 | branch: master | Jason Garrett-Glaser <jason at x264.com> | Tue Oct 18 14:30:26 2011 -0700| [07efeb45db224b7757880d4d63bb549fb454f6db] | committer: Jason Garrett-Glaser

Add fenc prefetching to adaptive quant
Many fewer cache misses, faster adaptive quant.

> http://git.videolan.org/gitweb.cgi/x264.git/?a=commit;h=07efeb45db224b7757880d4d63bb549fb454f6db
---

 encoder/ratecontrol.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/encoder/ratecontrol.c b/encoder/ratecontrol.c
index 25d2bb3..4853432 100644
--- a/encoder/ratecontrol.c
+++ b/encoder/ratecontrol.c
@@ -247,6 +247,7 @@ static NOINLINE uint32_t x264_ac_energy_mb( x264_t *h, int mb_x, int mb_y, x264_
      * function and make sure that its always called before the float math.  Noinline makes
      * sure no reordering goes on. */
     uint32_t var;
+    x264_prefetch_fenc( h, frame, mb_x, mb_y );
     if( h->mb.b_adaptive_mbaff )
     {
         /* We don't know the super-MB mode we're going to pick yet, so