[petsc-dev] Hijacked MPI calls...

[Satish Balay]

merged the fix to 'next'

https://bitbucket.org/petsc/petsc/commits/1dac896b5164bd7681661a63f944dd271192ecc1

Satish

On Thu, 15 Sep 2016, Barry Smith wrote:

> 
> > On Sep 15, 2016, at 11:06 PM, Jed Brown <jed at jedbrown.org> wrote:
> > 
> > Satish Balay <balay at mcs.anl.gov> writes:
> > 
> >> On Tue, 13 Sep 2016, Satish Balay wrote:
> >> 
> >>> On Tue, 13 Sep 2016, Satish Balay wrote:
> >>> 
> >>>> And this one is perhaps for Matt..
> >>> 
> >>> Ops - perhaps Jed needs to look at this..
> >>>> 
> >>>>>>>>>>>>>>>>>>>>>>> 
> >>>>          CC arch-clang-cmplx/obj/src/dm/impls/composite/pack.o
> >>>> /home/balay/petsc/src/dm/impls/composite/pack.c:657:17: warning: passing an object that undergoes default argument promotion to 'va_start' has undefined behavior [-Wvarargs]
> >>>>  va_start(Argp,imode);
> >>>>                ^
> >>>> /home/balay/petsc/src/dm/impls/composite/pack.c:641:61: note: parameter of type 'InsertMode' is declared here
> >>>> PetscErrorCode  DMCompositeGather(DM dm,Vec gvec,InsertMode imode,...)
> >>>>                                                            ^
> >>>> 1 warning generated.
> >>>> <<<<<<<<<<<<<<<<<
> >> 
> >> The fix here is:
> >> 
> >> -PetscErrorCode  DMCompositeGather(DM dm,Vec gvec,InsertMode imode,...)
> >> +PetscErrorCode  DMCompositeGather(DM dm,InsertMode imode,Vec gvec,...)
> >> 
> >> [Barry is ok - but Jed,Matt might object?]
> > 
> > It's annoying, but it appears to be needed.  It's screwy because enum
> > ought to be handled no differently than int for this purpose.
> 
>    Yeah, we checked and sizeof(enum type) = 4 so there is no need for promotion so I blame it on compiler flakiness.
> 
>