[petsc-dev] BG hang still broken in petsc-maint!

[Satish Balay]

Works for me on vesta with [the following on sys/examples/tutorials/ex1]

runjob --np 8192  --ranks-per-node 16 --cwd $PWD --block  VST-00440-33771-512 : $PWD/ex1 -log_summary

More details on the error case would be useful.

BTW: one of the fixes is the following - so ~/.petscrc nolonger is
read on BGL or similar batch systems where HOME env variable is not
set.

Satish

-------

commit 099d3e56c9b6e2de54268b943fc5bf31a542b4a9
Merge: 4cce697 e2e8085
Author: Jed Brown <jedbrown at mcs.anl.gov>
Date:   Wed Oct 9 08:46:35 2013 -0500

    Merge branch 'jed/avoid-getpwuid' into maint
    
     The file path routines were broken because the current directory was
     placed in front of the given path before checking for leading ~ or
     ~user. This is now fixed, along with the removal of getpwuid (a
     liability on BG/Q and Cray). We still attempt to use getpwnam when
     encountering ~user.
    
    * jed/avoid-getpwuid:
      Sys ex13: avoid non-portable realpath(1) in verification test
      cleanup includes that are no longer needed. Also remove check for _fullpath() as the corresponding code is removed.
      Sys: fix PetscGetFullPath expansion of ~ and ~user, ensure NULL termination
      Sys: avoid getpwuid in file path routines
      Sys: add failing test expanding ~/file-name and ~user/file-name
    
    Conflicts:
        config/PETSc/Configure.py



On Tue, 17 Dec 2013, Barry Smith wrote:

> 
> On Dec 17, 2013, at 10:03 AM, Satish Balay <balay at mcs.anl.gov> wrote:
> 
> > On Tue, 17 Dec 2013, Barry Smith wrote:
> > 
> >> 
> >>   Derek reported that the maint branch on Blue-Gene still hangs due to the .petscrc business! WTF? Come on guys, this was supposed to have been fixed months ago.
> >> 
> > 
> > Works with master - but not maint?
> 
>   He has not tried master
> 
> > 
> > The relavent commits are in maint (and petsc-3.4.3) - so I'm not sure
> > what the issue is.
> 
>    The issue is that the “commits” in maint don’t work?  Please try a build with maint ….
> 
> 
>    Barry
> 
> > 
> > Satish
> > 
> > -----
> > 
> > balay at asterix /home/balay/petsc (maint)
> > $ git log --merges --grep=bgq
> > commit 3b778056deceb36e04d69cacb7fdbb47c6629cc8
> > Merge: 099d3e5 3a01836
> > Author: Jed Brown <jedbrown at mcs.anl.gov>
> > Date:   Wed Oct 9 08:48:47 2013 -0500
> > 
> >    Merge branch 'jed/options-insert-bcast-bgq' into maint
> > 
> >    Attempt to work around BG/Q MPI_Bcast bug.  It is unknown whether it is
> >    successful, but this uses fewer communications so it should not be
> >    worse.
> > 
> >    * jed/options-insert-bcast-bgq:
> >      PetscOptionsInsertFile: combine Bcasts for alias and values
> > 
> > commit 2a08145f31bf26ef86d11056c9adcbe6b71cd7b9
> > Merge: d96a3a3 1ea3218
> > Author: Satish Balay <balay at mcs.anl.gov>
> > Date:   Thu Aug 29 10:41:24 2013 -0500
> > 
> >    Merge branch 'balay/fix-deadlock-on-bgq' into maint
> > balay at asterix /home/balay/petsc (maint)
> 
>