Re: Jade not using gcc

[Norman Gray <norman@xxxxxxxxxxxxxxx>]

Greetings,

On 12 May 1999, Adam Di Carlo wrote:

> >>>>> "James" == James Clark <jjc@xxxxxxxxxx> writes:
> 
> James> You could try using sunfix.sh: this avoids the preprocessor
> James> magic with Vector/NCVector.
> 
> Hmmm. Sounds like something which maybe we should autodetect and
> handle in autoconf.  Since I don't have any access to the SC5.0
> compiler set or Digital C++, someone who knows autoconf maybe could
> get chance to patch configure.in to detect and handle the situation?
> I'd be happy to help with what I can.

I asked about the (NC)Vector's processor magic - I should have found
sunfix.sh myself (blush).  As penance, I'll at least distribute notes
on how the builds go.  I had put off replying until I could provide
something a little more substantial than `doh!'.

As things stand, Jade has exposed what I _think_ is a bug in the SC5.0
compiler set - I've reported it to Sun, so we'll see if they agree.
It'll be tomorrow before I feel strong enough to face Digital C++
properly, but so far I've got as far as failing to build the shared
libraries correctly.

How does autoconf work?  Would I just edit in tweaks and tests to the
script distributed with Jade, or is there some more structured source
the configure script is, in turn, generated from?  Is that
configure.in (which doesn't appear in the distribution)?  Is there any
documentation on autoconf?  Questions, questions.

All the best,

Norman


---------------------------------------------------------------------------
Norman Gray                        http://www.astro.gla.ac.uk/users/norman/
Physics and Astronomy, University of Glasgow, UK     norman@xxxxxxxxxxxxxxx


 DSSSList info and archive:  http://www.mulberrytech.com/dsssl/dssslist