Re: [FFmpeg-devel] [PATCH] hevc_mc: reduce stride for bidir temp buffers

2014-07-28 Thread Christophe Gisquet
Hi, 2014-07-27 19:18 GMT+02:00 Michael Niedermayer : >> It is unconditionally set to 64, which is quite higher than the >> actual block size. > > is this faster? No, which surprises me, but so be it. On the other hand, if this stride is PB_MAX_SIZE throughout the code, then maybe its value could

Re: [FFmpeg-devel] [PATCH] hevc_mc: reduce stride for bidir temp buffers

2014-07-27 Thread Mickaƫl Raulet
Hi Christophe >> hevc.c |9 + >> 1 file changed, 5 insertions(+), 4 deletions(-) >> 2445ba15d38b2472f8f1b24aa75e63c089971480 >> 0012-hevc_mc-reduce-stride-for-bidir-temp-buffers.patch >> From 126adf820bc54c2d00f794629595ad6310fbfc37 Mon Sep 17 00:00:00 2001 >> From: Christophe Gisquet

Re: [FFmpeg-devel] [PATCH] hevc_mc: reduce stride for bidir temp buffers

2014-07-27 Thread Michael Niedermayer
On Sun, Jul 27, 2014 at 10:24:43AM +0200, Christophe Gisquet wrote: > The stride is sometimes way bigger than actually needed. ptrdiff_t is > the actual type used. Aligning to 8 is needed as the lines must have > aligned addresses for SIMD. > > -- > Christophe > hevc.c |9 + > 1 fil

[FFmpeg-devel] [PATCH] hevc_mc: reduce stride for bidir temp buffers

2014-07-27 Thread Christophe Gisquet
The stride is sometimes way bigger than actually needed. ptrdiff_t is the actual type used. Aligning to 8 is needed as the lines must have aligned addresses for SIMD. -- Christophe From 126adf820bc54c2d00f794629595ad6310fbfc37 Mon Sep 17 00:00:00 2001 From: Christophe Gisquet Date: Sat, 26 Jul 2