[MLton-devel] Controlling memory usage for MPI MLton processes

[Roland Olsson]

> Hi Roland.  It's nice to hear from you again after a few years.

It is good to see that you continue developing MLton, which may be the
best language/compiler combination in the world when considering
reliability, efficency and expressiveness, for example. That's what I tell
my students in their programming languages course at least.

I solved the problem by recompiling the compiler and setting fixedHeap to
170M in runtime/gc.c. The paging possibly occurred because the C part of
ADATE may use 30MB or more of heap space that MLton doesn't know about.

> I don't quite understand how your application works.   I would think
> that if you set ram-slop at the command line and then the MLton
> process forks to create other processes, then those would inherit the
> ram-slop of the parent.  Is that failing to happen?  Or is the problem
> that you would like the sum of the ram-slops for all the MLton
> processes to be constant?

Yes, it is failing to happen. The MPI standard prescribes the call 
MPI_init(&argc, &argv) to be made before any message passing. My MPI library
uses Secure Shell (ssh) to start processes on other machines and not
direct forking.

Thanks for a quick reply, just like when I found an MLton bug in 1999.

The various SML/NJ bugs that have prevented me from fully utilizing my new
cluster of 2.8GHz Pentium IV machines are now fortunately irrelevant since
I use MLton instead. Splendid!

Cheers,

Roland



-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
SourceForge.net hosts over 70,000 Open Source Projects.
See the people who have HELPED US provide better services:
Click here: http://sourceforge.net/supporters.php
_______________________________________________
MLton-devel mailing list
MLton-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mlton-devel