On 12/04/2010, at 4:30 PM, Matthias Kilian wrote:
> 
> And here we have that validation problem again. You can't expect
> that everyone who sends patches has access to all those types of
> operating systems. Nor can one expect all the GHC developers to
> have access to other, more `exotic' systems.



Oh, that's exactly what I am expecting! A project as serious as GHC should have 
no trouble providing build resources to its developers. My comment was more 
about shell scripting being fundamentally broken.

I'm still think a "please validate this bundle please" feature that tests 
across all platforms is the only way to go. I see reports from two separate 
build systems in cvs-ghc now too, so I'm still hopeful.

Ah, and I see the new instructions are also up on the wiki page. I'll install 
it on the Solaris machine here...

Ben.




_______________________________________________
Cvs-ghc mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/cvs-ghc

Reply via email to