[MPICH2-dev] need-help

[Rajeev Thakur]

>     Excuse me, but I mean "cpi.c". Is it all right? Thanks.

Sorry, I didn't read your mail carefully. cpi.c is supposed to work, but I
heard it is not working currently with mpd because of some stdin/stdout
issues. It will work if you use pm=forker, which is a single-machine process
manager that just forks processes. We will fix the problem. Thanks for
bringing it to our attention.

Rajeev



> -----Original Message-----
> From: 张岳 Zhang Yue [mailto:frcchang at hpclab.cs.tsinghua.edu.cn]
> Sent: Wednesday, April 30, 2003 9:44 PM
> To: Rajeev Thakur
> Subject: Re: RE: [MPICH2-dev] need-help
>
>
> Hello, Rajeev Thakur,
>
>     Excuse me, but I mean "cpi.c". Is it all right? Thanks.
>
> ======= At 2003-04-30, 10:26:00 you wrote: =======
>
> >The examples
> >
> >fetchandadd.c:
> >fetchandadd_tree.c:
> >test3.c:
> >test4.c:
> >transpose4.c:
> >
> >are not expected to run because of lack of thread safety in the current
> >version of MPICH-2. This is explained in each of those files.
> >
> >Rajeev
> >
> >> -----Original Message-----
> >> From: owner-mpich2-dev at mcs.anl.gov
> >> [mailto:owner-mpich2-dev at mcs.anl.gov]On Behalf Of 张岳
> >> Sent: Wednesday, April 30, 2003 2:23 AM
> >> To: mpich2-dev at mcs.anl.gov
> >> Subject: [MPICH2-dev] need-help
> >>
> >>
> >> hello. when i tested the cpi in test/mpi/rma, the processes alway
> >> got blocked somewhere. are the rma options ok in mpich2-0.91 and 0.93?
> >>
> >> 张岳 Zhang Yue
> >> frcchang at hpclab.cs.tsinghua.edu.cn
> >> 2003-04-30
> >>
> >>
> >>
>
> = = = = = = = = = = = = = = = = = = = =
>
>
> Best regards.
> 张岳 Zhang Yue
> frcchang at hpclab.cs.tsinghua.edu.cn
> 2003-05-01
>
>
>
>