LAM/MPI logo

LAM/MPI General User's Mailing List Archives

  |   Home   |   Download   |   Documentation   |   FAQ   |   all just in this list

From: Yanni Batsiolas (jbats_at_[hidden])
Date: 2003-05-31 02:36:24


Hi Jeff

You are correct, Mandrake 9.1 has glibc-devel-2.3.1-10mdk. I have
submitted an error report to Intel, but I believe their response would
be that they do not yet support Mandrake 9.1.

In any event, it appears that when using the intel compilers __WORDSIZE
does not get defined at all. If I define that (to be 32) then lam
compiles just fine. Defining__WORDSIZE helps also with compiling romio.

I still have to run the test suite to see if it actually works though
:-)

Cheers, Yannis

> -----Original Message-----
> From: lam-bounces_at_[hidden]
> [mailto:lam-bounces_at_[hidden]] On Behalf Of Jeff Squyres
> Sent: May 30, 2003 8:37 PM
> To: General LAM/MPI mailing list
> Subject: Re: LAM: Problems building lab with intel 7.1
>
>
> On Fri, 30 May 2003, Yanni Batsiolas wrote:
>
> > Well, it seems that
> >
> > /usr/include/bits/ipctypes.h has:
> >
> > /* Used in `struct shmid_ds'. */
> > # if __WORDSIZE == 32
> > typedef unsigned short int __ipc_pid_t;
> > # else
> > typedef int __ipc_pid_t;
> > # endif
> >
> > and the intel supplied
> > intel/compiler70/ia32/substitute_headers/bits/types.h:
> >
> > /* Used in `struct shmid_ds'. */
> > typedef unsigned short int __ipc_pid_t;
> >
> > So, which file should I change, the intel supplied or the
> system one??
> > Any suggestions?
>
> Yikes -- tough call. Hmm. Interestingly enough, I don't
> have the /usr/include/bits/ipctypes.h file on my MDK 9.0
> system, but I do see it on a RH 8.0 system as part of the
> glibc-devel-2.3.2 RPM (my MDK 9.0 system has glibc-2.2.5).
>
> So it looks like this is a recent addition to glibc. You
> might want to ping the Intel folks and ask them for advice... :-\
>
> --
> {+} Jeff Squyres
> {+} jsquyres_at_[hidden]
> {+} http://www.lam-mpi.org/
> _______________________________________________
> This list is archived at http://www.lam-mpi.org/MailArchives/lam/
>