Re: [petsc-dev] GPU regression tests

2017-07-28 Thread Karl Rupp
Hi Alejandro, I have tested the branch and it seems to fix the problem. Our software runs fine with it. great, thanks for the quick feedback. Best regards, Karli

Re: [petsc-dev] GPU regression tests

2017-07-28 Thread Alejandro Lamas Daviña
On viernes, 28 de julio de 2017 14:45:09 (CEST) Karl Rupp wrote: > Hi Jose, > > a fix for MatDuplicate() for AIJCUSPARSE matrices is in the branch > karlrupp/fix-MatDuplicate-GPU > ex_duplicate.c now runs correctly in both cases. > > Could you please report back whether it fixes the problems yo

Re: [petsc-dev] GPU regression tests

2017-07-28 Thread Karl Rupp
Hi Jose, a fix for MatDuplicate() for AIJCUSPARSE matrices is in the branch karlrupp/fix-MatDuplicate-GPU ex_duplicate.c now runs correctly in both cases. Could you please report back whether it fixes the problems you encountered in your enclosing application? Thanks and best regards, Karli

Re: [petsc-dev] GPU regression tests

2017-07-27 Thread Karl Rupp
Thanks, Satish, fixes are now in next. Best regards, Karli On 07/27/2017 09:52 AM, Satish Balay wrote: There are a bunch of nightlybuild regressions with the GPU stuff - perhaps you can take a look. http://ftp.mcs.anl.gov/pub/petsc/nightlylogs/archive/2017/07/27/examples_next_arch-cuda-double

Re: [petsc-dev] GPU regression tests

2017-07-27 Thread Satish Balay
There are a bunch of nightlybuild regressions with the GPU stuff - perhaps you can take a look. http://ftp.mcs.anl.gov/pub/petsc/nightlylogs/archive/2017/07/27/examples_next_arch-cuda-double_es.log Satish On Thu, 27 Jul 2017, Karl Rupp wrote: > Hi Jose, > > > We have detected another problem.

Re: [petsc-dev] GPU regression tests

2017-07-27 Thread Karl Rupp
Hi Jose, We have detected another problem. Could you take care of it? MatDuplicate() does not work for MATSEQAIJCUSPARSE (probably also for MATMPIAIJCUSPARSE). The attached example creates a matrix and duplicates it. There are two cases: 1) With a diagonal matrix it fails on GPU because MatDup

Re: [petsc-dev] GPU regression tests

2017-07-27 Thread Jose E. Roman
Karl, We have detected another problem. Could you take care of it? MatDuplicate() does not work for MATSEQAIJCUSPARSE (probably also for MATMPIAIJCUSPARSE). The attached example creates a matrix and duplicates it. There are two cases: 1) With a diagonal matrix it fails on GPU because MatDuplicat

Re: [petsc-dev] GPU regression tests

2017-07-26 Thread Karl Rupp
Hi Jose, With pull request #719 we have finished a set of fixes to VECCUDA stuff. With these changes it is now possible to run many tests in SLEPc's testsuite on GPU (AIJCUSPARSE+VECCUDA). These tests will be included in the nightly tests from now on. great! However, PETSc nightly tests r

[petsc-dev] GPU regression tests

2017-07-26 Thread Jose E. Roman
Hi. With pull request #719 we have finished a set of fixes to VECCUDA stuff. With these changes it is now possible to run many tests in SLEPc's testsuite on GPU (AIJCUSPARSE+VECCUDA). These tests will be included in the nightly tests from now on. However, PETSc nightly tests related to VECCUDA

[petsc-dev] GPU regression tests

2016-05-24 Thread Jose E. Roman
Hi. We recently noticed that MatMultTranspose_MPIAIJCUSPARSE does not work. We tried to debug it but it is difficult because we don't know the internal details of Mat. There might be other cases. It would be convenient to add nightly tests for VECCUDA and MATAIJCUSPARSE (in real and complex sc