On 14 May 2014 00:36, Luca Barbato <[email protected]> wrote: > On 13/05/14 18:20, James Yu wrote: > > I try benchmark by using Linaro toolchain for this patch, And run below > > commend > > "avconv -i mp4_movie -an -f null EMPTY-OUTPUT" > > to get the fps. it doesn't have obvious different performance. > > You can use our microbenchmark > > # avconv -benchmark -i sample -an -f null - > > Might be interesting checking if our current set of supported compilers > groks intrinsics properly so we could not have to write everything in > assembly. > > I run this command on ARMv7 cortex-a8, 1 GHz, single core SoC.
master: NEON asm bench: utime=312.580s maxrss=18500kB With my patch: using neon intrinsics bench: utime=310.880s maxrss=18504kB Seems we have a bit better performance with this patch. -- Regards, James > lu > _______________________________________________ > libav-devel mailing list > [email protected] > https://lists.libav.org/mailman/listinfo/libav-devel > _______________________________________________ libav-devel mailing list [email protected] https://lists.libav.org/mailman/listinfo/libav-devel
