[m-rev.] Trivial diff: fix hlc.par segfaults

Ralph Becket rafe at cs.mu.OZ.AU
Mon Nov 22 11:55:52 AEDT 2004

Simon Taylor, Monday, 22 November 2004:
> On 22-Nov-2004, Fergus Henderson <fjh at cs.mu.OZ.AU> wrote:
> > 
> > Well, I still read this newsgroup on occaision, and I still have a
> > reasonable understanding of the .par code.  If there are bugs in the
> > .par grades, I'm at least willing to have a look at them when I get
> > a chance.  But I'm not aware of any such bugs.
> If there are no known bugs, that message shouldn't be in the BUGS
> file either, because it just creates FUD in the minds of the users.

We would like to say something to the effect that, because of changes in
the set of developers, addressing problems that arise with .par grades
may take some time and that some such solutions may be beyond our reach.

That said, I think there are two bugs with .par grades:

(1) the program terminates when main/2 returns, regardless of whether
there are other running threads;

(2) if I understand Zoltan correctly, nobody managed to get concurrent
code running on more than three processors on a four processor machine
(the code ran fine, just ~25% slower than expected).

-- Ralph
mercury-reviews mailing list
post:  mercury-reviews at cs.mu.oz.au
administrative address: owner-mercury-reviews at cs.mu.oz.au
unsubscribe: Address: mercury-reviews-request at cs.mu.oz.au Message: unsubscribe
subscribe:   Address: mercury-reviews-request at cs.mu.oz.au Message: subscribe

More information about the reviews mailing list