[FFmpeg-devel] [flamefest-start] A little something on MMX/SSE intrinsics

Michael Niedermayer michaelni
Thu Feb 28 02:53:43 CET 2008


On Thu, Feb 28, 2008 at 01:43:33AM +0000, M?ns Rullg?rd wrote:
> Rich Felker <dalias at aerifal.cx> writes:
> 
> > On Wed, Feb 27, 2008 at 10:59:52PM +0000, M?ns Rullg?rd wrote:
> >> > I do know that gcc does quite stupid things on x86 be it when compiling C
> >> > code or intrinsics. And i know that gcc is generally better at compiling
> >> > x86 code than code for other less common architecures. Combining these
> >> > does strongly point toward that the gap between intrinsics and asm will
> >> > be bigger on ppc than x86 not smaller.
> >> > Of course you are correct that i do not strictly "know" it. Its just VERY
> >> > likely.
> >> 
> >> It is wrong to talk about the probability of something that has a
> >> definite value, even if it has not yet been measured.  Probabilities
> >> would be appropriate if speculating over the abilities of future GCC
> >> versions.
> >
> > Welcome to the world of QM. Unfortunately that's how all facts are..
> 
> If the trend in gcc code size keeps up, they won't be in the QM domain
> for quite some time to come.

QM is not less true for large things, unless you want to claim that
gravitational interactions are becoming significant for gcc. In that
case we of course should apply general relativity. If so it would be
interresting to calculate how long it will take for it to collaps into
a black hole if its code size keeps growing ...
We probably should switch to a different compiler before that happens.

[...]
-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

I have never wished to cater to the crowd; for what I know they do not
approve, and what they approve I do not know. -- Epicurus
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20080228/722aa632/attachment.pgp>



More information about the ffmpeg-devel mailing list