On Thu, 12 Mar 2009, Roy Stogner wrote: > adjoint_solve() stubs (and QOI function attachment, etc) are in now.
And now adjoint_solve() implementations for non-transient problems are in. I may or may not have time to futz with this again next week, so Derek, if you or someone working with you is willing to be the guinea pig who determines whether or not any of this actually works, now's the time to start. There's definitely some room for debate on the APIs, both older and new, too. The NonlinearSolver, in particular, has jacobian, matvec, etc. as function pointers which probably ought to be in NonlinearImplicitSystem. And I'm not sure whether the DiffSolver ought to have separate solve() and adjoint_solve() functions; it makes some of the code simpler, and could allow for some optimizations later, but feels redundant. --- Roy ------------------------------------------------------------------------------ Apps built with the Adobe(R) Flex(R) framework and Flex Builder(TM) are powering Web 2.0 with engaging, cross-platform capabilities. Quickly and easily build your RIAs with Flex Builder, the Eclipse(TM)based development software that enables intelligent coding and step-through debugging. Download the free 60 day trial. http://p.sf.net/sfu/www-adobe-com _______________________________________________ Libmesh-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/libmesh-devel
