Yeah, I guess I need to spend more time with Massif. I tried it last week for the first time as well and was overwhelmed by the amount of output. From what I've seen just scanning over the output is a detailed log of when and how much memory is used. Is there a way to flip that tree upside-down to see what structures hold onto memory? That would be useful...
No need to dig if you don't know the answer, I'll dive into this myself in more detail later. Thanks, Cody On Thu, May 30, 2013 at 9:00 AM, Kirk, Benjamin (JSC-EG311) < [email protected]> wrote: > Here is a sample of what massif can provide: > > #----------- > snapshot=76 > #----------- > time=34259946158 > mem_heap_B=81869326 > mem_heap_extra_B=2917770 > mem_stacks_B=0 > heap_tree=detailed > n7: 81869326 (heap allocation functions) malloc/new/new[], --alloc-fns, > etc. > n3: 31253406 0xF0712EE: PetscMallocAlign (in > /software/x86_64/petsc/3.3-p4/aerolab_workstations-openmpi-1.4-gcc-4.6/lib/libpetsc.so) > n2: 24264240 0xF10F705: VecCreate_MPI_Private (in > /software/x86_64/petsc/3.3-p4/aerolab_workstations-openmpi-1.4-gcc-4.6/lib/libpetsc.so) > n1: 13865280 0xF11053C: VecCreateGhostWithArray (in > /software/x86_64/petsc/3.3-p4/aerolab_workstations-openmpi-1.4-gcc-4.6/lib/libpetsc.so) > n1: 13865280 0xF110C0A: VecCreateGhost (in > /software/x86_64/petsc/3.3-p4/aerolab_workstations-openmpi-1.4-gcc-4.6/lib/libpetsc.so) > n2: 13865280 0x692F09C: libMesh::PetscVector<double>::init(unsigned > int, unsigned int, std::vector<unsigned int, std::allocator<unsigned int> > > const&, bool, libMeshEnums::ParallelType) (petsc_vector.h:864) > n2: 10398960 0x6AA600A: libMesh::System::init_data() (system.C:274) > n1: 8665800 0x6AA6217: libMesh::System::init() (system.C:231) > n1: 8665800 0x6A5D87B: libMesh::EquationSystems::init() > (equation_systems.C:120) > n1: 8665800 0x57A21E8: FINS::Physics::system_init(std::string > const&) (physics.C:301) > n1: 8665800 0x57A65B5: FINS::Physics::run(std::string) > (physics.C:368) > n1: 8665800 0x58D85E5: FINS::CompNS<(FINS::NSTypes)3>::run() > (in /tmp/fins/without_vg/lib/libfins.so.0.0.0) > n0: 8665800 0x40BC44: main (fins.C:145) > n1: 1733160 0x6AA61F0: libMesh::System::init() (system.C:228) > n1: 1733160 0x6A5D87B: libMesh::EquationSystems::init() > (equation_systems.C:120) > n1: 1733160 0x57A21E8: FINS::Physics::system_init(std::string > const&) (physics.C:301) > n1: 1733160 0x57A65B5: FINS::Physics::run(std::string) > (physics.C:368) > n1: 1733160 0x58D85E5: FINS::CompNS<(FINS::NSTypes)3>::run() > (in /tmp/fins/without_vg/lib/libfins.so.0.0.0) > n0: 1733160 0x40BC44: main (fins.C:145) > n1: 3466320 0x6AA61B7: libMesh::System::init_data() (system.C:292) > n1: 3466320 0x6AA61F0: libMesh::System::init() (system.C:228) > n1: 3466320 0x6A5D87B: libMesh::EquationSystems::init() > (equation_systems.C:120) > n1: 3466320 0x57A21E8: FINS::Physics::system_init(std::string > const&) (physics.C:301) > n1: 3466320 0x57A65B5: FINS::Physics::run(std::string) > (physics.C:368) > n1: 3466320 0x58D85E5: FINS::CompNS<(FINS::NSTypes)3>::run() > (in /tmp/fins/without_vg/lib/libfins.so.0.0.0) > n0: 3466320 0x40BC44: main (fins.C:145) > > ------------------------------------------------------------------------------ Introducing AppDynamics Lite, a free troubleshooting tool for Java/.NET Get 100% visibility into your production application - at no cost. Code-level diagnostics for performance bottlenecks with <2% overhead Download for free and get started troubleshooting in minutes. http://p.sf.net/sfu/appdyn_d2d_ap1 _______________________________________________ Libmesh-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/libmesh-users
