On Tue, 2019-03-05 at 15:42 +0100, Martin Liška wrote:
> On 3/5/19 3:21 PM, David Malcolm wrote:
> > On Tue, 2019-03-05 at 15:14 +0100, Martin Liška wrote:
> > > Hi.
> > >
> > > The patch extends option suggestion for both -Werror and
> > > corresponding
> > > pragram.
> > >
> > > Patch can bootstrap on x86_64-linux-gnu and survives regression
> > > tests.
> > >
> > > Ready to be installed after stage1 opens?
> > > Thanks,
> > > Martin
> >
> > Good idea - thanks.
> >
> > The patch also fixes some quoting issues.
>
> Yep. Btw. I have a huge patch that fixes very many quoting issues.
> I'll send as stage1 material once I'll build all targets.
Great!
(BTW, did you see my nit-pick about it being better to use "'" rather
than "." in DejaGnu directives for quote characters? That might apply
to that patch also)
> >
> > Ideally we'd also provide fix-it hints for the unknown options -
> > but
> > I'm guessing we probably don't have accurate enough location_t
> > values
> > for that - are the location_t values just for the whole of the line
> > containing the pragma, or do they have fine-grained token
> > information?
>
> It's fine-grained if I see correctly:
>
> ./xgcc -B. /home/marxin/Programming/gcc/gcc/testsuite/gcc.dg/pragma-
> diag-6.c
> /home/marxin/Programming/gcc/gcc/testsuite/gcc.dg/pragma-diag-
> 6.c:2:30: warning: option ‘-Wnoexcept’ is valid for C++/ObjC++ but
> not for C [-Wpragmas]
> 2 | #pragma GCC diagnostic error "-Wnoexcept" /* { dg-warning "is
> valid for C../ObjC.. but not for C" } */
> | ^~~~~~~~~~~~
>
> That said, will you please do a follow up patch?
Sure; please ping me once you've committed your patch.
Thanks
Dave
