[FFmpeg-devel] [PATCH] WMA Voice postfilter

Ronald S. Bultje rsbultje
Tue Apr 6 22:37:40 CEST 2010


Hi,

On Thu, Apr 1, 2010 at 12:39 PM, Ronald S. Bultje <rsbultje at gmail.com> wrote:
> On Thu, Apr 1, 2010 at 11:52 AM, Reimar D?ffinger
> <Reimar.Doeffinger at gmx.de> wrote:
>> On Thu, Apr 01, 2010 at 11:25:25AM -0400, Ronald S. Bultje wrote:
>>> I'm probably asking silly questions here, but can I allocate aligned
>>> memory on the stack? Or do they have to be in the struct?
>>
>> You better avoid it on the stack, it won't work reliably on all
>> supported configurations.
>
> Thanks for the advice, this patch is tested with yasm and does not
> crash, so I think alignment is OK now. By putting synth buffer in the
> context, I was also able to remove one memcpy(). (One or two more of
> those are possible in the original code, previously already suggested
> by Vitor, so I might go ahead and do that in a future patch also.)

Ping (I guess Vitor, Reynaldo, anyone interested in voice...?)

Ronald



More information about the ffmpeg-devel mailing list