Apelete Seketeli via cfe-commits <cfe-commits@lists.llvm.org> writes:
> apelete created this revision.
> apelete added reviewers: kevin.qin, rsmith.
> apelete added a subscriber: cfe-commits.
>
> Signed-off-by: Apelete Seketeli <apel...@seketeli.net>
>
> http://reviews.llvm.org/D19959
>
> Files:
>   lib/Driver/Tools.cpp
>
> Index: lib/Driver/Tools.cpp
> ===================================================================
> --- lib/Driver/Tools.cpp
> +++ lib/Driver/Tools.cpp
> @@ -2346,7 +2346,7 @@
>      success = getAArch64MicroArchFeaturesFromMcpu(D, 
> getAArch64TargetCPU(Args),
>                                                    Args, Features);
>  
> -  if (!success)
> +  if (!success && A)
>      D.Diag(diag::err_drv_clang_unsupported) << A->getAsString(Args);
>  
>    if (Args.getLastArg(options::OPT_mgeneral_regs_only)) {

This doesn't look right. The case where (!Success && !A) needs *some*
kind of diagnosic, doesn't it? I guess the value of OPT_arch was invalid
in that case.

I think this change just papers over a real diagnostic bug.
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

Reply via email to