On Wed, 27 Sep 2000, Dan Nelson wrote:
> In the last episode (Sep 28), Mark Powell said:
> > BTW The gcc 2.95.2 options are perfectly valid for FreeBSD as well as
> > Linux. Obviously :) Can they be copied into the FBSD section too?
> >
> > # these options for gcc-2.95.2 to produce fast code
> > # CC_OPTS = \
> > # -Wall -O9 -fomit-frame-pointer -march=pentium \
> > # -finline-functions -fexpensive-optimizations \
> > # -funroll-loops -funroll-all-loops -pipe -fschedule-insns2 \
> > # -fstrength-reduce \
> > # -malign-double -mfancy-math-387 -ffast-math
>
> I object to half of these options because whoever added obviously never
> tested his additions one at a time to verify that they actually help.
>
> 1) There is no -O level above 3
I thought that was strange. I can only assume they were thinking of
putting it as high as possible for future extra gcc optimisation modes.
Afterall -O2 was the limit at one time.
> 2) You can't assume the user is compiling on a 586
Well they are commented out by default. It was just I'd not actually
noticed them before as they only appear in the Linux section.
> 3) Here is a list of the -f options listed, and which -O level
> automatically enables them:
>
> -fomit-frame-pointer -
> -finline-functions 3
> -fexpensive-optimizations 2
> -funroll-all-loops -
> -funroll-loops enabled by -funroll-all-loops
> -fschedule-insns2 2
> -fstrength-reduce 2
> -ffast-math -
>
> So at minimum, the CC_OPTS should read
>
> -Wall -pipe -O3 -fomit-frame-pointer -funroll-all-loops -ffast-math
> -malign-double mfancy-math-387
Does -m486 still do anything useful or does -march=pentium include it?
I thought that particular optimisation was also valid for a i[56]86?
> And it's debatable whether unroll-all-loops is a win or not (depends on
> whether the unrolled code is larger than your CPU's cache).
True. Although caches are pretty large these days; 256KB, 512KB or even
large with the Xeon.
Cheers.
Mark Powell - UNIX System Administrator - The University of Salford
Academic Information Services, Clifford Whitworth Building,
Salford University, Manchester, M5 4WT, UK.
Tel: +44 161 295 5936 Fax: +44 161 295 5888 www.pgp.com for PGP key
--
MP3 ENCODER mailing list ( http://geek.rcc.se/mp3encoder/ )
