[ET Trac] [Einstein Toolkit] #1743: Reduce number of output files per directory

[Einstein Toolkit]

#1743: Reduce number of output files per directory
-----------------------+----------------------------------------------------
  Reporter:  eschnett  |       Owner:                     
      Type:  defect    |      Status:  review             
  Priority:  unset     |   Milestone:                     
 Component:  Other     |     Version:  development version
Resolution:            |    Keywords:                     
-----------------------+----------------------------------------------------

Comment (by rhaas):

 The extra argument of file_nioproc seems gone, so could you use
 cGH->nioprocs for this? I added a minor comment on using strcpy/malloc vs
 strdup (in a C++ file stdup is somewhat safer since there is no issue of
 accidentally implicitly declaring strdup as returning int due to missing
 _BSD defines).

 This will affect checkpoint files and out_vars 3d HDF5 output but not eg
 out3d_vars 3d HDF5 output I think (though I may be wrong and the latter
 may also use AssembleFilename).

 This is a step in the right direction but not quite enough I believe since
 we have other output that is produced with one-file-per-process (all of
 which is optional eg TimerReport, carept-timing-statistics etc) and that
 faces the same issues. So it should be applied ideally along with code
 that makes the other places that produce per-process output use this
 function as well.

-- 
Ticket URL: <https://trac.einsteintoolkit.org/ticket/1743#comment:22>
Einstein Toolkit <http://einsteintoolkit.org>
The Einstein Toolkit