On 12 April 2011 00:50, Satish Balay <balay at mcs.anl.gov> wrote: > On Tue, 12 Apr 2011, Lisandro Dalcin wrote: > >> After adding this test in petsc4py-dev/test/test_ksp.py : >> >> ? ? def testResetAndSolve(self): >> ? ? ? ? self.ksp.reset() >> ? ? ? ? self.testSolve() >> ? ? ? ? self.ksp.reset() >> ? ? ? ? self.testSolve() >> ? ? ? ? self.ksp.reset() >> >> it seems there is some memory bleeding: >> >> [0]Total space allocated 256 bytes >> [ 0]80 bytes KSPSetUp_BCGSL() line 476 in >> /home/devel/petsc/dev/src/ksp/ksp/impls/bcgsl/bcgsl.c >> ? ? ? [0] ?KSPSetUp() line 184 in >> /home/devel/petsc/dev/src/ksp/ksp/interface/itfunc.c >> ? ? ? [0] ?KSPSolve() line 332 in >> /home/devel/petsc/dev/src/ksp/ksp/interface/itfunc.c >> >> These are workspace arrays that are re-allocated at every KSPSetup_BCGSL >> call. >> >> Not sure about the proper way to fix this, but I would deallocate the >> arrays at KSPReset_BCGSL. Can anyone help me with this? >>
Never mind, I've pushed the fix I proposed above. >> >> PS: Will we ever have petsc4py in the nightly builds? Is there >> something to be done from my side? > > Sorry - will check on adding one.. > I was not complaining. But I think petsc4py do help to catch bugs in core PETSc. BTW, take a look here: http://code.google.com/p/petsc4py/source/browse/misc/makefile.dev . Ideally, you should test many PETSC_ARCH's as possible. PS: Could you add mpiuni+single+complex to the nighties? I recently fixed an mpiuni issue for this case. -- Lisandro Dalcin --------------- CIMEC (INTEC/CONICET-UNL) Predio CONICET-Santa Fe Colectora RN 168 Km 472, Paraje El Pozo 3000 Santa Fe, Argentina Tel: +54-342-4511594 (ext 1011) Tel/Fax: +54-342-4511169
