[petsc-dev] petsc release plan for Sept/2020

All, We are to make a petsc release by the end of September. For this release [3.14], will work with the following dates: - feature freeze: Sept 27 say 5PM EST - release: Sept 29 Merges after freeze should contain only fixes that would normally be acceptable to maint workflow. I've created a

Re: [petsc-dev] DMPlex + Cohesive Cells Setup

On Wed, Sep 2, 2020 at 3:19 PM Jacob Faibussowitsch wrote: > Isn't the whole point of cohesive elements that they are not everywhere? > All the papers they are citing in the one you link use them only along > the crack path. > > > don’t have any reliable way of determining the origin of the crac

Re: [petsc-dev] DMPlex + Cohesive Cells Setup

> Isn't the whole point of cohesive elements that they are not everywhere? All > the papers they are citing in the one you link use them only along > the crack path. don’t have any reliable way of determining the origin of the crack path and hence can’t selectively put them in places. The mesh

Re: [petsc-dev] DMPlex + Cohesive Cells Setup

On Wed, Sep 2, 2020 at 2:26 PM Jacob Faibussowitsch wrote: > It is a complicated topological problem and I do not have a general > solution. > > > Definitely. I’ve been writing my own impls in python this past month that > follows the algorithm laid out here > https://www.sciencedirect.com/scienc

Re: [petsc-dev] DMPlex + Cohesive Cells Setup

> It is a complicated topological problem and I do not have a general solution. Definitely. I’ve been writing my own impls in python this past month that follows the algorithm laid out here https://www.sciencedirect.com/science/article/pii/S2352711018300797

Re: [petsc-dev] DMPlex + Cohesive Cells Setup

On Wed, Sep 2, 2020 at 1:43 PM Jacob Faibussowitsch wrote: > I’m getting an argument out of bounds error (attached) but I suspect it’s > because I’m feeding in bad input to either calls (I am marking __all__ > vertices in the mesh in my input label to DMPlexLabelCohesiveComplete). Are > the cohes

Re: [petsc-dev] PetscMallocAlign for Cuda

OK good to know. I will now worry even less about making this very complete. On Wed, Sep 2, 2020 at 1:33 PM Barry Smith wrote: > > Mark, > >Currently you use directly the Nvidia provided mallocs cudaMalloc for > all mallocs on the GPU. See for example aijcusparse.cu. > >I will be using

Re: [petsc-dev] DMPlex + Cohesive Cells Setup

I’m getting an argument out of bounds error (attached) but I suspect it’s because I’m feeding in bad input to either calls (I am marking __all__ vertices in the mesh in my input label to DMPlexLabelCohesiveComplete). Are the cohesive grains allowed to intersect? Does a cohesive region have to be cl

Re: [petsc-dev] PetscMallocAlign for Cuda

Mark, Currently you use directly the Nvidia provided mallocs cudaMalloc for all mallocs on the GPU. See for example aijcusparse.cu. I will be using Stefano's work to start developing a unified PETSc based system for all memory management but don't wait for that. Barry > On Sep

Re: [petsc-dev] PetscMallocAlign for Cuda

I believe there are a few PetscMallocCuda impls in src/sys/memory/cuda/mcudahost.cu that seem to do what you are describing. If you are creating mats you can also consider cudaMallocPitch, but I’m not sure how that plays with the sparse storage impls that petsc mat uses. Seems more useful for d

[petsc-dev] petsc-3.13.5.tar.gz now available

Dear PETSc users, The patch release petsc-3.13.5 is now available for download, with change list at 'PETSc-3.13 Changelog' http://www.mcs.anl.gov/petsc/download/index.html Satish

[petsc-dev] PetscMallocAlign for Cuda

PETSc mallocs seem to boil down to PetscMallocAlign. There are switches in here but I don't see a Cuda malloc. THis would seem to be convenient if I want to create an Object entirely on Cuda or any device. Are there any thoughts along these lines or should I just duplicate Mat creation, for instan

Re: [petsc-dev] DMPlex + Cohesive Cells Setup

On Mon, Aug 31, 2020 at 5:39 PM Jacob Faibussowitsch wrote: > Hello All, > > What is the simplest possible configuration required to insert cohesive > cells into a plex? Plex/tests/ex5.c would be the example to ape here but > there is a lot there that I don’t (think) I need like a DS or any kind