Merge branch 'barry/2020-09-30/retry-other-cuda-errors/release' into 'master'

Report all GPU errors as PETSC_ERR_GPU so that the test harness can retry any GPU failure

See merge request petsc/petsc

Merge branch 'barry/2020-09-30/retry-other-cuda-errors/release' into 'master'

Report all GPU errors as PETSC_ERR_GPU so that the test harness can retry any GPU failure

See merge request petsc/petsc!3283

show more ...

Report all GPU errors as PETSC_ERR_GPU so that the test harness can retry any GPU failure

We still get occassional errors:

ot ok ksp_ksp_tutorials-ex71_bddc_cusparse # Error code: 180

That occur a

Report all GPU errors as PETSC_ERR_GPU so that the test harness can retry any GPU failure

We still get occassional errors:

ot ok ksp_ksp_tutorials-ex71_bddc_cusparse # Error code: 180

That occur after the cuSPARSE library must have been loaded since the load command already succeeded

Commit-type: error-checking, CI, enhancement
/spend 15m

show more ...

Merge branch 'barry/2020-09-24/validatemalloconerror' into 'master'

Call PetscMallocValidate() on signal and error conditions to check if the...

See merge request petsc/petsc!3246

Call PetscMallocValidate() on signal and error conditions to check if the final memory state is correct

Commit-type: error-checking
/spend 10m

Merge branch 'barry/2020-09-26/error-mans' into 'master'

Improve manual pages for PetscError() and the various error handlers including...

See merge request petsc/petsc!3249

Improve manual pages for PetscError() and the various error handlers including providing cross links in the manual pages

Commit-type: documentation
/spend 45m

Merge branch 'barry/2020-09-23/gpu-resource-error' into 'master'

Add specific PETSc error flag for GPU not having resources available

See merge request petsc/petsc!3241

Add specific PETSc error flag for GPU not having resources available

Try three times to initialize CUDA/cuBLAS and cuSPARSE before erroring. This is for GPU tests that sometimes fail since the GPU
i

Add specific PETSc error flag for GPU not having resources available

Try three times to initialize CUDA/cuBLAS and cuSPARSE before erroring. This is for GPU tests that sometimes fail since the GPU
is busy with other tests when a new test is started.

Commit-type: error-checking, testing
/spend 1h

show more ...

Merge branch 'barry/09-08-2020/fix-debug-mpiabort-on-other-ranks' into 'master'

Barry/09 08 2020/fix debug mpiabort on other ranks

See merge request petsc/petsc!3179

Fix -debugger_nodes x,y,z so that ranks that are not in the debugger sleep before calling MPI_Abort so they do
not bring down the user's debug session.

Commit-type: bug-fix, feature
/spend 1.5h
Repo

Fix -debugger_nodes x,y,z so that ranks that are not in the debugger sleep before calling MPI_Abort so they do
not bring down the user's debug session.

Commit-type: bug-fix, feature
/spend 1.5h
Reported-by: Ed Bueler <elbueler@alaska.edu>

show more ...

Merge branch 'barry/2020-08-16/fix-source-formatting' into 'master'

Turn on checkbadSource test to generate an error when found; fix all source code that causes errors

See merge request petsc/petsc

Merge branch 'barry/2020-08-16/fix-source-formatting' into 'master'

Turn on checkbadSource test to generate an error when found; fix all source code that causes errors

See merge request petsc/petsc!3064

show more ...

Turn on checkbadSource test to generate an error when found; fix all source code that causes errors

Commit-type: portability-fix, testing-fix, style-fix, feature, maintainability
/spend 1.5h

Merge branch 'dalcinl/binary-io' into 'master'

Binary IO Enhancements

See merge request petsc/petsc!2594

PetscBinaryWrite: Remove `istemp` boolean flag

Merge branch 'dalcinl/fix-null' into 'master'

Sys: Replace 0 -> NULL for pointers

See merge request petsc/petsc!2482

Sys: Replace 0 -> NULL for pointers

Merge branch 'tisaac/feature-alt-v' into 'master'

Tisaac/feature alt v

See merge request petsc/petsc!2357

Start PetscDTAltV

Added reference implementation of alternating multilinear forms

But I'm going to clean up the interface. I don't think an actual object is necessary
and we want to optimize the c

Start PetscDTAltV

Added reference implementation of alternating multilinear forms

But I'm going to clean up the interface. I don't think an actual object is necessary
and we want to optimize the cases N = 2, 3

PetscDTAltV: change representation order

so that when k != N / 2, there is no reordering in hodge star

PetscDTAltv: New basis for alterting multilinear forms.

For $k$-forms with $2 k \leq N$, the standard lexicographic
basis of $\{dx_{\sigma(1)} \wedge \dots \wedge dx_{\sigma(k)}\}$
is used.

For $2 k > N$, we modify this by first reversing the lexicographic
order, and then swapping $\sigma(1)$ and $\sigma(k)$ if $\sigma$ is odd.
While this involves more sign flips than the basic approach, it has two
good properties:

- it preserves the case as before that the dofs of a row vector /
covector are the dofs of the equivalent 1-form, but now
- the dofs of a contravariant vector (like an H(div)-conforming flux)
are the dofs of the equivalent (N-1)-form (importantly, this means
that the contravariant pullback is the same as the alternating form
pullback)

The only exception is 2D, for which the case is the same as before: the
1-form pullback is the covariant pullback, the contravariant pullback
can be obtained by Hodge star isometry before and afer the pullback.

Remove bad file from rename during merge

PetscDTAltV: interface simplifaction, undo specialized basis.

The fact that the N=2 case wasn't handle bothered me. Now I think the
thing to do is to let PetscDTAltVPullback() and PetscDTAltVInterior() take
negative form degrees, indicating that the pullback/interior product should be
conjugated with the hodge star.

Fix c++ PetscBool issues with PetscDT static inlines

PetscDTAltV: negative form index in pullback...

indicates $\star^{-1} \circ L* \circ$. This gives the contravariant
pullback by $ k=-(N-1)$ in all dimensions.

PetscDTAltV: stop using altv data structures

I'm going to remove the object

PetscDTAltV: Remove object

PetscDTAltV: avoid mallocs/frees for N <= 3

show more ...

Move PetscCUBLASGetErrorName from vec to sys for --with-single-library=0 --with-visibility=1

Ref: https://gitlab.com/petsc/petsc/merge_requests/2282

Merge branch 'barry/2019-10-16/use-error-codes' into 'master'

Replace use of 1 with appropriate error code; use PetscObjecComm() instead of...

See merge request petsc/petsc!2182

Merge branch 'barry/2019-10-16/mpi_abort-main' into 'master'

SETERRQ(), CHKERRQ(), PetscError() and friends provide better information when...

See merge request petsc/petsc!2181

Fix some typos

Replace use of 1 with appropriate error code; use PetscObjecComm() instead of PETSC_COMM_SELF for errors when appropriate

Commit-type: error-checking, testing-fix, style-fix, examples

SETERRQ(), CHKERRQ(), PetscError() and friends provide better information when MPI_Abort() is called in main()

Previously PetscError() called PETSCABORT() directly which lost the line number informa

SETERRQ(), CHKERRQ(), PetscError() and friends provide better information when MPI_Abort() is called in main()

Previously PetscError() called PETSCABORT() directly which lost the line number information and fact that PETSCABORT() was called from within main()
Added new error index for a PETSCABORT() in main and provide the correct line number of the problem

Commit-type: error-checking, testing-fix, feature

show more ...

Merge branch 'jczhang/feature-nice-mpi-abort' into 'master'

Provide a nicer MPI_Abort

See merge request petsc/petsc!2137