Re: rpmbuild --rebuild ends up in /i386 on ahtlon SMP

On Fri, 2002-09-20 at 05:49, Jef Spaleta wrote:
> I don't think  common applications typically can get compiled to
> understand multiple processor usage.  the kernel level stuff handles the
> smp, applications typically don't worry about consuming multiple
> processors, as far as I understand it...and I don't know whats out there
> in the application space so take that "typically" comment with a large
> grain of salt.  But it would be interesting to note which applications
> out there in the OSS space actually can be compiled to use multiple
> processing...or better yet clustered processors.   I have to work with a
> lot of scientific codes that use MPI to get access to clustered cpus and
> I'd like to peruse some better written and more functional code than the
> stuff I have acces to. I've been toying with the idea of building my own
> little cluster to play with different clustering technologies...mosix
> looks like a good starting point for a home cluster of outdated pc's.
> 
> -jef

I'm wondering if the poster was asking for the use of SMP in
compilation, ie parallelizing make? 

In other words, making sure that rpmbuild understands you want make
executed with a -j4 or -j3? 

Doing an XFree86 rebuild with only one CPU just sucks when you paid for
two.


--Chris



-- 
Homepage: http://interclypse.net
Registered Linux user #215241 (http://counter.li.org/)

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.7 (GNU/Linux)

iD8DBQA9i038R6ATq7AtBW4RAs5VAJ9MThZIi51ujYZ26tn6BweE0GM0vgCeM9zt
RT4uLiWxAu0gesJxYMklBtU=
=4hgb
-----END PGP SIGNATURE-----