[mpich-discuss] MPI_Comm_size

[Pavan Balaji]

As Rajeev suggested, can you make sure you are not using some other MPI 
implementation's mpif.h?

  -- Pavan

On 07/06/2009 02:09 PM, littledddna at gmail.com wrote:
> Thank you very much for the reply,Pavan.
> I checked the RANK, and all of them are 0.......
> Xiao
> 
> On Jul 6, 2009 1:59pm, Pavan Balaji <balaji at mcs.anl.gov> wrote:
>  >
>  >
>  > Can you run examples/cpi or examples/f77/fpi?
>  >
>  >
>  >
>  > I just want to make sure you are not running into this problem: 
> http://wiki.mcs.anl.gov/mpich2/index.php/Frequently_Asked_Questions#Q:_All_my_processes_get_rank_0.
>  >
>  >
>  >
>  > -- Pavan
>  >
>  >
>  >
>  > On 07/06/2009 01:55 PM, xiao zhang wrote:
>  >
>  >
>  > HI~ Everyone: i have a little doubt about the funtion:
>  >
>  > MPI_Comm_size
>  >
>  > If I call it like:
>  >
>  >
>  >
>  > include 'mpif.h'
>  >
>  > integer :: ierr
>  >
>  > call MPI_Init(ierr)
>  >
>  > call MPI_Comm_size(MPI_COMM_WORLD,*mgroupsize*,ierr)
>  >
>  >
>  >
>  > Then I use 8 processors to run the program, the return value of 
> *mgroupsize *should be 8. Is that right?
>  >
>  > If it is right, the return value for me is just 1. What would be the 
> possible problem?
>  >
>  > I am really new to this and thank you very much
>  >
>  > Xia
>  >
>  >
>  >
>  >
>  > --
>  >
>  > Pavan Balaji
>  >
>  > http://www.mcs.anl.gov/~balaji
>  >

-- 
Pavan Balaji
http://www.mcs.anl.gov/~balaji