[FFmpeg-devel] [PATCH] update doc/optimization.txt

Måns Rullgård mans
Mon Sep 20 20:33:57 CEST 2010


Michael Niedermayer <michaelni at gmx.at> writes:

> On Tue, Sep 21, 2010 at 01:23:24AM +1000, Peter Ross wrote:
>> On Mon, Sep 20, 2010 at 03:31:31PM +0200, Michael Niedermayer wrote:
>> > On Fri, Sep 17, 2010 at 03:53:45PM -0400, Ronald S. Bultje wrote:
>> > > Hi,
>> > > 
>> > > On Fri, Sep 17, 2010 at 2:49 PM, Michael Niedermayer <michaelni at gmx.at> wrote:
>> > > > On Fri, Sep 17, 2010 at 06:23:16PM +0100, M?ns Rullg?rd wrote:
>> > > >> "Ronald S. Bultje" <rsbultje at gmail.com> writes:
>> > > >>
>> > > >> > Hi,
>> > > >> >
>> > > >> > $subj, fixes a typo and mentions yasm.
>> > > >> >
>> > > >> > I could word it stronger ("if you write new code, yasm is preferred
>> > > >> > for non-inlined functions") if we can agree on that (which I don't
>> > > >> > think we do yet).
>> > > >>
>> > > >> Does anyone who actually writes any asm nowadays disagree?
>> > > >
>> > > > i dont know, but the one maintaining the x86 asm does and the project
>> > > > leader disagrees too
>> > > 
>> > > So is the patch as posted ok then?
>> > 
>> > ive no oppinon on this it feels a bit moving toward inconsistent asm mix
>> > though
>> 
>> FWIW:
>> 
>> Xvid uses nasm/yasm to provide fast routines across gcc and msvc
>> and friends.
>> 
>> The decision to was made during the Instruction Set War. Back then
>> msvc6 did not support the latest opcodes.
>
> and i remember suggesting that ffmpeg used nasm too so that code
> could be shared with xvid but people back then where against it

I dug into the archives, and the closest things resembling a discussion
on the topic I could find are these:

http://article.gmane.org/gmane.comp.video.ffmpeg.devel/18410
http://thread.gmane.org/gmane.comp.video.ffmpeg.devel/9628

-- 
M?ns Rullg?rd
mans at mansr.com



More information about the ffmpeg-devel mailing list