Hi Thorkil,
On Sat, Sep 15, 2007 at 08:06:30PM +0200, Thorkil Naur wrote:
>
> On my x86 (SuSE) Linux, validate ends like this:
>
> > /usr/bin/ghc -Werror -H64m -Onot -fasm -istage1/utils
> > [...]
> >
> > utils/Encoding.hs:144: Warning: Defined but not used: c
> >
> > [...]
> > make[1]: *** [stage1/utils/Encoding.o] Error 1
> > make[1]: *** Waiting for unfinished jobs....
> > make: *** [stage1] Error 1
>
> This is with "version of ghc... 6.2". These (and many more) warnings are also
> reported by the tnaur-x86-Linux-head builbot builder, so the failure in
> validate is caused by the -Werror flag.
So just to check, you can reproduce the failure with the ghc -Werror
commandline, and if you remove -Werror then it works?
You have "{-# OPTIONS -w #-}" in compiler/utils/Encoding.hs, right?
If you move that to the very first line, i.e. above the comments, does
that fix it? If not, does putting it on the end of the commandline fix
it?
> > cbits/unicode.c:1:63:
> > error: floating constant in preprocessor expression
>
> The latter problem seems related to the __GHC_PATCHLEVEL__=1.20070507 which
> looks like a floating point number. Again, please indicate how I should
> proceed in this case.
Hmm, I think changing the test in compat/cbits/unicode.c to
#if __GLASGOW_HASKELL__ < 605
#if __GLASGOW_HASKELL__ != 604 || __GHC_PATCHLEVEL__ != 0
#include "WCsubst.c"
#endif
#endif
will fix it.
Thanks
Ian
_______________________________________________
Cvs-ghc mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/cvs-ghc
