On Wed, 2002-05-15 at 11:11, Raymond Toy wrote:
> >>>>> "James" == James Amundson <amundson@fnal.gov> writes:
>
> James> On Sat, 2002-05-11 at 01:20, Vadim V. Zhytnikov wrote:
> >> I don't think so. Last weekend I've changed my ASUS P2B with PIII on
> >> Soltek 75DRV5 with AthlonXP. It works to me fine on both systems.
> >> I have got other problem. CMUCL 18c now coredumps.
>
> James> CMUCL 18c is known (by me) not to work. I just added the README.lisps
> James> file in attempt to clear up that and other related issues.
>
> I find this surprising. Is it important that 18c work?
No. We can require 18d.
> 18d has many
> bug fixes, and the platforms supported by 18d are the same as 18c, so
> I don't think it's important. However, 18a runs on several more
> platforms than 18c/d, so that might be important.
>
> Is the core caused by the new numeric code?
I don't think so. I didn't look at the problem very hard. I've seen
hangs with 18c and bignums since long before the new numeric code. Since
18d just works, I didn't bother to investigate further.
--Jim