Thanks, Jeff
Do you know such MPICH related mailing list?
Does LAM-MPI excludes such limitations on the number of MPI processes running
on one node?
Best regards,
Anton
--- Jeff Squyres <jsquyres_at_[hidden]> wrote:
> This list is for support of the LAM implementation of MPI. If you're
> using MPICH, you should ask the MPICH maintainers for support.
>
> Good luck.
>
>
> On Sep 2, 2005, at 7:36 AM, a17567182 wrote:
>
> > Hello,
> >
> > Could anyone explain the limit of MPI processes allowed to be run?
> > I have a Linux RedHat box with MPICH 1.2.6 using shmem installed. I
> > cannot run
> > 80 MPI processes (aborted with the message "exceeding max num of p4
> > semids
> > 16"), 79 processes fail to finalise and 78 processes run fine. In
> > mpid/ch_shmem/p2psemop.c there is a definition
> > #define P2_MAX_SYSV_SEMIDS 8
> > which is used later to print out this error message:
> > setnum = p2_global->sysv_next_lock / 10;
> > if (setnum > P2_MAX_SYSV_SEMIDS)
> > {
> > p2p_error("exceeding max num of p4 semids\n",P2_MAX_SYSV_SEMIDS);
> > }
> > I cannot find where this p2_global->sysv_next_lock is initialised by
> > 78 or 79
> > and why this value is set?
> >
> > Thanks,
> >
> > Anton
> >
> >
> > __________________________________________________
> > Do You Yahoo!?
> > Tired of spam? Yahoo! Mail has the best spam protection around
> > http://mail.yahoo.com
> > _______________________________________________
> > This list is archived at http://www.lam-mpi.org/MailArchives/lam/
> >
>
> --
> {+} Jeff Squyres
> {+} jsquyres_at_[hidden]
> {+} http://www.lam-mpi.org/
>
> _______________________________________________
> This list is archived at http://www.lam-mpi.org/MailArchives/lam/
>
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
|
