Optimizations

Dan Osterrath do3 at mail.inf.tu-dresden.de
Wed Nov 27 14:06:18 PST 2002


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

> 1) Does configure script detect my machine and let gcc optimize the code
> for my processor type? (i686 for example)

Partly yes.

> 2) If the answer (1) is "yes" why should I specify additional
> CFLAGS="-march=processor_type" for optimization?

Because -march=i686 is not equal to -march=pentium3. The last one implies 
- -mmmx -msse and probably some other. At least this is what I understood of 
some mailings and the gcc man page.

> 3) If the answer (1) is "yes" why should I specify -march for
> cross-compiling in addition to host/target argument to configure?

see 2)

> 4) Why optimized applications may not work correctly? optimization in gcc
> is buggy?

Hmm, good question. I think the gcc developers blame it on the application 
developers and they blame it on the gcc team. Probably its the combination of 
both.

> 5) You say that for example glibc shouldn't be compiled with
> CFLAGS, but I must set it for cross-compiling!

You can set these flags. I compiled mine with "-s -O2 -fomit-frame-pointer 
- -march=pentium3" and it works fine. I think this warning is a rudiment from 
old books.

- -- 
- ----------------------------------------------------------------------
%> ln -s /dev/null /dev/brain
%> ln -s /dev/urandom /dev/world
%> dd if=/dev/world of=/dev/brain
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.7 (GNU/Linux)

iD8DBQE95UHc9NbB8EM160MRArUaAJ40J8pzb71rgYGoQbrseDbfR7JWbgCg5OHF
/AQ3VS3n5gVUkGX9CVkV2d4=
=O2rC
-----END PGP SIGNATURE-----

-- 
Unsubscribe: send email to listar at linuxfromscratch.org
and put 'unsubscribe lfs-support' in the subject header of the message



More information about the lfs-support mailing list