I would be much happier with run-time as well, but I could also live with a reconfigure (it's just one MORE petsc build on my machine :D).
What ever can be done to fix this issue would be appreciated. Cheers, Dave On 11 September 2013 18:12, Jed Brown <[email protected]> wrote: > Dave May <[email protected]> writes: > > > Yes, being able to optionally turn off the full memory checking (either > run > > time or via a configuration flag) in MatSetValues() would be good enough. > > I'd rather make everything run-time unless it absolutely must be > configure time because it sucks to reconfigure (and possibly rebuild > downstream libraries/packages). > > > Presumably when the full error checking wasn't being used, the error > > checking should revert to using the old style memory checking (e.g. ptr > != 0) > > as was used in version 3.2 (and maybe 3.3, I cannot recall). > > Yes. As far as I'm concerned, it should also try to dereference it so > that a SEGV occurs early rather than later. The only difference between > safe and fast mode is that the SEGV in safe mode is caught by a friendly > signal handler than cleans up and returns so that a normal error can be > propagated. >
