On Tue, Feb 10, 2009 at 09:54:09AM +0000, Garth N. Wells wrote: > > > DOLFIN wrote: > > One or more new changesets pushed to the primary dolfin repository. > > A short summary of the last three changesets is included below. > > > > changeset: 5659:16e95d7eed387c4e6fb8d1f0235d0f5eb0a0eb91 > > tag: tip > > parent: 5658:d6d661c979340701f12565329dd7019204dee78e > > parent: 5657:e893f260884275f77ce803fbf3220f140c678760 > > user: Anders Logg <[email protected]> > > date: Tue Feb 10 10:49:43 2009 +0100 > > files: dolfin/function/GenericFunction.h > > description: > > merge > > > > > > changeset: 5658:d6d661c979340701f12565329dd7019204dee78e > > parent: 5656:294c8d6cab0b74a2aefde88fce5f022cd7ffae2f > > user: Anders Logg <[email protected]> > > date: Tue Feb 10 10:49:34 2009 +0100 > > files: dolfin/ode/ODESolution.cpp dolfin/ode/TimeStepper.cpp > > dolfin/ode/TimeStepper.h dolfin/parameter/Parameter.cpp > > dolfin/parameter/Parameter.h dolfin/parameter/ParameterList.cpp > > dolfin/parameter/ParameterList.h > > description: > > Fixes for sensitive checks T - t < eps instead of t > T - eps etc. > > Still memory leaks. > > > > Have you looked at ODESolution::interpolate? > > Garth
Why is that one a problem? I think it looks good. Does valgrind mention it when you run? I think the memory leak was caused by a call to error() which just raises an exception and exits (without cleaning up). The error message was caused by a missing value not being saved at the final time due to a round-off error that only showed up with valgrind... Right now, I'm seeing a lot of "Conditional jump or move depends on uninitialised value(s)" but haven't managed to figure out why. -- Anders
signature.asc
Description: Digital signature
_______________________________________________ DOLFIN-dev mailing list [email protected] http://www.fenics.org/mailman/listinfo/dolfin-dev
