You are right, the machine I compile on has 256 MB. I guess I
didn't expect the change in memory requirements to be so drastic. I
never had any problems with 20041109, but with the new version my
machine starts swapping so much that it is practically
unusable. Thanks for looking at the issue.
-Ivan
Stephen Weeks <[EMAIL PROTECTED]> writes:
>> I upgraded mlton to version 20051109-1 from 20041109, and I can no
>> longer compile my ML projects, because when the compiler reaches the
>> defunctorize stage it takes up all available memory, and I have to
>> kill the process.
>
> I haven't been able to reproduce the problem. When I compiled
> dynamo.mlb on my machine with 2G RAM, defunctorize completed in less
> than seven seconds, and the entire compile completed in less than two
> minutes. The log is below. The max amount of live data was 200M, so
> I would expect no trouble for the compile to complete on a machine
> with 512M RAM. As a test, I compiled dynamo with "@MLton fixed-heap
> 400M --" and the compile still completed in under two minutes.
>
> Does your machine have less than 512M? I did notice in your log the
> following, which indicates a lack of memory.
>
> parseAndElaborate finished in 3.62 + 34.85 (91% GC)
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
