[MLton-devel] NetBSD support for MLton (!)

[Stephen Weeks]

> > http://j.mongers.org/pub/misc/MLton-netbsd-20030821.tar.gz
> 
> I've grabbed it and will try it out tonight or tomorrow.

I tried this version on my NetBSD 1.6.1 machine.  It is pretty
unstable for me.  Many of the regressions give segfaults.  Even worse,
they are nondeterministic, sometimes working and sometimes failing. We
must keep in mind that my NetBSD is running on VMWare, and they don't
claim to support it, so that could be the problem.  But, Jesper, I am
interested to know if you are seeing any problems.  For example, the
array5 regression fails about half the time.  When it does fail, the
error is always in the same place -- I see

13 13 13 13 13 
13 13 13 13 13 
13 13 13 13 13 
13 13 13 13 13 
0 1 2Segmentation fault (core dumped)

The problem appears to be GC related, since if I run with @MLton
fixed-heap 10m --, which is enough to prevent any GCs, the program
always succeeds.  The failure appears with both the C and native
codegens, and at the same place in the code.

Jesper, if you're not able to see this problem with array5 or any of
the other regressions, I will investigate further.


-------------------------------------------------------
This SF.net email is sponsored by: VM Ware
With VMware you can run multiple operating systems on a single machine.
WITHOUT REBOOTING! Mix Linux / Windows / Novell virtual machines
at the same time. Free trial click here:http://www.vmware.com/wl/offer/358/0
_______________________________________________
MLton-devel mailing list
MLton-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mlton-devel