I will definately look into this tonight and see if it is a problem with my
recent patch.
On Wednesday 02 March 2005 3:39 am, Markus F.X.J. Oberhumer
wrote:> Reid Spencer wrote:
> > On Tue, 2005-03-01 at 22:07, Markus F.X.J. Oberhumer wrote:
> >>Follow up: After removing the dangling symlink the problem now
looks:
> >>
> >>-march=c((anonymous namespace)::PrintStackTrace()+0x1e)[0x847a17e]
> >>-fno-strict-aliasing: example.out.cbe.c: No such file or directory
> >>gccld: example.out.cbe.c: Can't destroy file:
> >>make: *** [example.out] Error 1
> >>At a minimum that error message is pretty confusing - does gccld
want to
> >> execute a program named "-fno-strict-aliasing" ? Have to
check later...
> >
> > No, it doesn't. I don't know what's up with the
-fno-strict-aliasing
> > output; however, the rest seems pretty straight forward to me. For
some
> > reason example.out.cbe.c didn't get created. Since its a temporary
file,
> > it tried to destroy it at program termination and failed because it
> > doesn't exist. This is most likely a bug in gccld's main
function.
>
> Short update: -native-cbe is currently broken as gccld/llc seems to
> generate assembler code instead of C. To easy debugging of such things it
> would be nice if -save-temps could get propaged from llvm-gcc to gccld.
>
> Markus
>
> > Reid
> >
> >
> >
------------------------------------------------------------------------
> >
> > _______________________________________________
> > LLVM Developers mailing list
> > LLVMdev at cs.uiuc.edu http://llvm.cs.uiuc.edu
> > http://mail.cs.uiuc.edu/mailman/listinfo/llvmdev
