On Jan 6, 2012, at 1:37 PM, Jed Brown wrote: > On Fri, Jan 6, 2012 at 13:31, Barry Smith <bsmith at mcs.anl.gov> wrote: > I agree. VecValidValues() could simply return without checking if getarray > is not support for it. > > Done. > > Also, what should we do about VecGetArrayRead() for non-petscnative? Maybe > this is not a priority, but right now I'm copying back into the nested Vec in > VecRestoreArrayRead().
It needs to be added to the Vec function table just like VecGetArray() has a slot in the function table. Barry
