On 21 July 2004 02:31, Ian Lynagh wrote:
> test.T in testsuite/tests/ghc-regress/programs/okeefe_neural says
>
> # Skip the test on alpha-dec-osf3, where we
> # run into an infinite loop on some builds
> if config.platform == "alpha-dec-osf3":
> opts.skip = 1;
>
> Does anyone know any more details on this?
>
> The same thing seems to happen on Linux IA64 (unregisterised) - I
> caught the ghc process having used just over 200 minutes of CPU time.
> Has anyone else seen (or not seen) this on IA64?
>
> Is there any more information you would like or should or should I
> just skip it on IA64 too?
>
> Does anyone know which arches I can expect it to also happen on?
> "on some builds" implies it is a bit random, which worries me...
I don't have anything to add, I'm afraid. Someone should investigate
why the test runs out of heap on other platforms - it might be caused by
the same thing.
Simon: this is a good example of a program with a large static data
structure that causes the simplifier to wander off for ever.
Cheers,
Simon
_______________________________________________
Cvs-ghc mailing list
[EMAIL PROTECTED]
http://www.haskell.org/mailman/listinfo/cvs-ghc
