Re: gcc, march and sse

[Johnny Billquist <bqt%softjar.se@localhost>]

Joel Carnat wrote:
> Le Mer 3 décembre 2008 16:40, Dennis den Brok a écrit :
> > Joel Carnat <joel%carnat.net@localhost> schrieb:
> > > Hello,
> > >
> > > I read that Slackware Linux compiles most of its packages with
> > > "-march=i486 -mtune=i686". As far as I understood gcc's manpage, this
> > > means "keep code compatible with 486 CPU but allow the use pentiumpro
> > > special features", right ?
> > >
> > > Still from the man page, MMX is supported since "pentium2 (or
> > > pentium-mmx
> > > only)", SSE since "pentium3", SSE2 since "pentium-m", SSE3 since
> > > "prescott".
> > >
> > > I tweaked my mk.conf to compile pkgsrc (on a domU) using "-O2 -march=486
> > > -mtune=686" but I'm now wondering if any on those compiled programs will
> > > use MMX or SSE when running on my laptop or intel-based servers (intel
> > > core2duo).
> > >
> > > Then second question, in case the previous thought is correct, if I use
> > > "-march=i486 -mtune=nocona", is there a way to see which instruction the
> > > program uses ? For example, if I compile xorg/firefox/mplayer and then
> > > run
> > > it in a special manner, can I see that the core2duo uses MMX/SSE*
> > > instruction while the C3 only use MMX instruction (from the same binary)
> > > ?
> > -march specifies the instruction set the compiler may use; thus, neither
> > MMX
> > nor SSE will be used. There may be exceptions, however, because some
> > pkgsrc
>
> hum... then what would be the point in setting "mtune" ?

Isn't that obvious? Tune for optimal performance on that variant.
But you cannot use instructions that aren't allowed by march, because 
then it won't run on march (if so, what was the point of march?).

But different instruction sequences can be the optimal on different 
implementations of an instruction set.

        Johnny