GPUs: Allow ViennaCL and VECCUDA to be used concurrently.

This also gets rid of the weird veccuda.py.

CUSP: Removed VECCUSP and AIJCUSP as well as preconditioners.

SA-AMG from CUSP will be re-added in a follow-up commit,
made to work with AIJCUSPARSE and VECCUDA.

Merge branch 'master' of bitbucket.org:petsc/petsc

Merge branch 'maint'

Merge remote-tracking branch 'origin/balay/fix-sprintf-warnings/maint' into maint

gcc-8: replace sprintf with PetscSNPrintf to fix -Wformat-overflow warnings

/home/balay/petsc/src/sys/objects/init.c:518:24: warning: ‘%d’ directive writing between 1 and 11 bytes into a region of s

gcc-8: replace sprintf with PetscSNPrintf to fix -Wformat-overflow warnings

/home/balay/petsc/src/sys/objects/init.c:518:24: warning: ‘%d’ directive writing between 1 and 11 bytes into a region of size between 0 and 4095 [-Wformat-overflow=]
sprintf(name,"%s.%d",mname,rank);
^~
/home/balay/petsc/src/sys/objects/init.c:518:7: note: ‘sprintf’ output between 3 and 4108 bytes into a destination of size 4096
sprintf(name,"%s.%d",mname,rank);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

show more ...

Merge branch 'stefano_zampini/feature-glvis-homesh-periodic'

Doc: Small fix for log_view
Suggested-by: Jose E. Roman <jroman@dsic.upv.es>

cosmetic fixes

Merge branch 'barry/tao-to-new-test-harness' into barry/all-new-test-harness-branches

Merge branch 'jed/tests-cxx-linker' into barry/more-to-new-test-harness

Merge remote-tracking branch 'origin/barry/utilize-hwloc-rebased-v4'

Introduced PetscCommSharedGet() and cleaned up implementation and proper automatic freeing

Add PetscProccessManagementView() and -process_view to use hwloc to display location of PETSc MPI processes on cores
This is to allow users and developers to know if the correct binding and mapping o

Add PetscProccessManagementView() and -process_view to use hwloc to display location of PETSc MPI processes on cores
This is to allow users and developers to know if the correct binding and mapping of processes to cores
on multicore systems is done as expected. Without this benchmarking to develop optimized code is not possible

Allow -process_view to be used with streams, determined optimal binding for streams on Intel 4 numa thing system

Funded-by: P-ECP
Time: 4

show more ...

Merge branch 'master' into knepley/feature-plex-adjoint-prelim

* master: (251 commits)
yaml: add a backup download URL
Fix documentation of MatSetRandom function
Ignore -malloc-hbw option if m

Merge branch 'master' into knepley/feature-plex-adjoint-prelim

* master: (251 commits)
yaml: add a backup download URL
Fix documentation of MatSetRandom function
Ignore -malloc-hbw option if malloc is already set
Plex: Fixed array initialization
pgi: compile fix
knl: update test to use memkind and remove option that worked arround a bug with intel 17
Check the return value for memkind_posix_memalign
Change the default behavour for -malloc_hbw
Bib: More refs
test: remove runex13f90 as its now converted to new test suite
fix win32 crash. This test was previously run with '-n 4' - so restoring it
Fix petscsfgetgraph_ to work for all Fortran compilers by directly using array of PetscSFNodes
bugfix for MatDuplicate_MPIAIJ_PtAP Reported-by: Samuel Lanthaler <s.lanthaler@gmail.com>
strip out dead sundials usage code; change sundials test to only run 5 timesteps to prevent different number of steps on different systems.
Remove extra lines of output file
Example uses C99 constructs so must mark with requires: c99
fix const missmatch for DMDAVecRestoreArrayRead()
remove conflict between function name in example and new PETSc TS function
builds: attempt to do maint/master/next test builds in separate clones
ex62: it turned out that CHACO partioning is not deterministic. Use a prescribed partition instead
...

Conflicts:
src/dm/interface/dm.c
src/snes/examples/tutorials/ex13.c

show more ...

Merge remote-tracking branch 'origin/balay/knl-test-use-memkind-and-intel-18' into next-tmp

Ignore -malloc-hbw option if malloc is already set

When this option takes effect and PETSc is configured with memkind, HBW allocator is used for the whole running period.
This suffices for normal us

Ignore -malloc-hbw option if malloc is already set

When this option takes effect and PETSc is configured with memkind, HBW allocator is used for the whole running period.
This suffices for normal users. For some use cases where one wants to control HBW memory explicitly,
(e.g. some checkpoint strategy needs to switch between MCDRAM and DRAM on KNL) do not use this option.

Funded-by:
Project: PETSc for Theta
Time: 1.0 hours
Reported-by:
Thanks-to:

show more ...

Merge branch 'master' into knepley/fix-plex-bc-multiple

* master: (347 commits)
tests: add 'PETSC_HAVE_C99' flag - so that 'requires: c99' can be used in the test suite
PGI: fix build error with

Merge branch 'master' into knepley/fix-plex-bc-multiple

* master: (347 commits)
tests: add 'PETSC_HAVE_C99' flag - so that 'requires: c99' can be used in the test suite
PGI: fix build error with updated 17.10 compilers
Fix for test dirs that are not tutorials/tests
fix use of double in PetscAbsReal() which produces compiler warnings with single precision build
elemental: remove info on conflict of -with-cxx-dialect=C++11 on osx - it works now.
Cleanup .bib files
Fix allfortranstubs rule to rm generated files from correct directory, not outdated location
ierr set but not checked
Remove initial refine so test runs faster and does not time out on some machines
Gail's copyedits for bibtex items
Docs: manual: add further description of 3 different current testing options
DM: check for ops->localtolocalbegin/end before calling
PetscSF: fix typo in tutorial ex1
Vec+HDF5: Fix output for complex - Need to check for empty group name
SNES: SNESDestroy() can be called with snes->dm == NULL
SNES: SNESDestroy() can be called with snes->dm == NULL
Vec+HDF5: Fix output for complex - Need to check for empty group name
Docs: Manual: Update tests section
gmakegentest: also exclude .#filenames created by emacs
DMPlex: Fix output of test ex1 after changes to Gmsh reader
...

show more ...

Merge commit 'd47bf9aaf1e5266cc3f0ff499f934c85788965a9' into jed/fix-matcreatempibaij/maint

Obtain changed (v3.5) handling of MatMPIBAIJSetPreallocationCSR to match
documentation.

Merge remote-tracking branch 'origin/jed/variadic-malloc'

PetscMallocA: implement the Malloc[1-7] suite using variadic function

The macros had become increasingly complicated and were opaque to debug.
It also prevented run-time choice of coalesced malloc,

PetscMallocA: implement the Malloc[1-7] suite using variadic function

The macros had become increasingly complicated and were opaque to debug.
It also prevented run-time choice of coalesced malloc, which is now
available via -malloc_coalesce which is useful to debug memory leaks due
to mismatched PetscMalloc3/PetscFree2 and the like. This implementation
dispatches to a single variadic function that handles allocation for
both PetscMalloc[1-7] and PetscCalloc[1-7], as well as a matching
case for freeing memory.

show more ...

Merge branch 'master' into knepley/fix-fe-default

* master: (297 commits)
Bib: Added refs and fixed one
Doc: Fixed formatting problems for MatCreateAIJ() manual page.
Add to PCSOR manual page

Merge branch 'master' into knepley/fix-fe-default

* master: (297 commits)
Bib: Added refs and fixed one
Doc: Fixed formatting problems for MatCreateAIJ() manual page.
Add to PCSOR manual page information about convergence testing when using KSPRICHARDSON
SNES ex5: Restore old output since this breaks an example I have
Noted that some Cray XC systems will use 'aprun' instead of 'srun'. Thanks to Herr Doktor Hong Zhang for this suggestion.
Added example configure script for Cray XC40 with KNL processors (e.g., NERSC Cori).
Fix possible sprintf -Wformat-overflow
PetscMalloc2-7: fix -Wint-in-bool-context for GCC 7
SNES ex5: make boundary condition handling consistent for symmetry
SNES ex5: use L2 norm and *_monitor_short for output
SNES ex5: the L^2 norm scales with the l^2 norm over sqrt(N)
SNES ex5: refactor MMS capability to avoid extreme code duplication
Documentation fix: typo in PetscLogStageRegister manual page.
nodes -> vertices
rename DMNetworkGetComponentTypeOffset -> DMNetworkGetComponentKeyOffset
rename DMNetworkGetConnectedNodes() -> DMNetworkGetConnectedVertices()
Jacobians for adjoints for explicit methods and constant Jacobians never got initialized
fix complex build error
fix build failure with fortran interface
fix compile error with cdc753b64c658d532d7813e942002e67025667a6
...

Conflicts:
src/snes/examples/tutorials/ex13.c
src/snes/examples/tutorials/ex62.c
src/snes/examples/tutorials/ex69.c

show more ...

Merge branch 'master' into knepley/feature-rigid-bodies

* master: (204 commits)
windows fixes
revert this change as a relavent fix is already in origin/knepley/fix-ex-types
regenerate output l

Merge branch 'master' into knepley/feature-rigid-bodies

* master: (204 commits)
windows fixes
revert this change as a relavent fix is already in origin/knepley/fix-ex-types
regenerate output log for snes_tutorials-ex62_fetidp_3d_tet that got corrupted in 188e10d6f1cffa866ec0197053a0e55fa4e7da41
fix opensolaris build error with --with-log=0 Undefined first referenced symbol in file PetscLogEventGetPerfInfo arch-opensolaris-misc/tests/src/dm/impls/plex/examples/tests/ex9.o
fix warnings on windows C:\cygwin64\home\petsc\PETSC~1.CLO\src\tao\examples\TUTORI~1\ex1.c(217): warning C4113: 'void (__cdecl *)()' differs in parameter lists from 'void (__cdecl *)(void)'
disable example with absoft (due to fixed/free form limit)
c89 fixes
metis: setself.downloaddirnames for change in 6d90219d8fa4f8b2fd242cc6ff9008358d87c26f
Plex ex19: Remove unneeded code
Plex: Fix projection when we have a dmAux which is a domain boundary - If we are projecting on a lower dimension thing (effectiveHeight > 0), then we need to get the space and dual space on this subcell - The tabulation needs to be in the right dimension for both dm and dmAux
PetscFE: Added height subspaces to PetscFE - Added PetscFEGetHeightSubspace() - Added array in struct
PetscDualSpace: Small cleanup of parameter name
PetscSpace: Added height subspaces for PetscSpace - Added PetscSapceGetHeightSubspace() and member function - Added subspaces array
fix compile warning: /sandbox/petsc/petsc.clone/src/snes/examples/tests/ex2.c:250:8: warning: ‘pointsAllProcs’ may be used uninitialized in this function [-Wmaybe-uninitialized] ierr = DMInterpolationSetUp(interpolator, dm, pointsAllProcs);CHKERRQ(ierr);
migrate 'build requires: !mpiuni' to new format
migrate 'build requires: !complex' to new format
workarround 'division by zero' compiler warning on windows C:\cygwin64\home\petsc\PETSC~3.CLO\src\ts\examples\TUTORI~1\ex11.c(416): warning #39: division by zero if (uL->h < 0 || uR->h < 0) {for (i=0; i<1+dim; i++) flux[i] = 0./0.; return;} /* SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Reconstructed thickness is negative"); */
dos2unix to prevent warnings on opensolaris "/export/home/petsc/petsc.clone/src/ksp/ksp/examples/tests/ex53.c", line 6: warning: invalid white space character in directive
fix some minor style issues
Extend AMS/ADS to support high-order element.
...

show more ...

Merge branch 'master' into stefano_zampini/feature-gmsh-periodic

* master: (1016 commits)
metis: setself.downloaddirnames for change in 6d90219d8fa4f8b2fd242cc6ff9008358d87c26f
fix compile warni

Merge branch 'master' into stefano_zampini/feature-gmsh-periodic

* master: (1016 commits)
metis: setself.downloaddirnames for change in 6d90219d8fa4f8b2fd242cc6ff9008358d87c26f
fix compile warning: /sandbox/petsc/petsc.clone/src/snes/examples/tests/ex2.c:250:8: warning: ‘pointsAllProcs’ may be used uninitialized in this function [-Wmaybe-uninitialized] ierr = DMInterpolationSetUp(interpolator, dm, pointsAllProcs);CHKERRQ(ierr);
migrate 'build requires: !mpiuni' to new format
migrate 'build requires: !complex' to new format
workarround 'division by zero' compiler warning on windows C:\cygwin64\home\petsc\PETSC~3.CLO\src\ts\examples\TUTORI~1\ex11.c(416): warning #39: division by zero if (uL->h < 0 || uR->h < 0) {for (i=0; i<1+dim; i++) flux[i] = 0./0.; return;} /* SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Reconstructed thickness is negative"); */
dos2unix to prevent warnings on opensolaris "/export/home/petsc/petsc.clone/src/ksp/ksp/examples/tests/ex53.c", line 6: warning: invalid white space character in directive
fix output file - corresponding to code changes in efd4aadf157bf1ba2d80c2be092fcf4247860003
disable test on windows
update log from opensolaris so that the diff is clean on linux aswell
update grep in sync with change in efd4aadf157bf1ba2d80c2be092fcf4247860003
Fixed code that did not correctly set current color for block insertion
Update output files that did not get fixed after previous merges
Change MPIUni to not immediately abort on operations it cannot perform, instead return error code
Add filter for example that can produce 1 different iteration count
Example add system calloc, bad formating of %d in print, commented out code
Always use (void) for functions that take no arguments, never ()
always need to caste PetscReal variables to (double) in print statements
updates to manual page for clarity
update help menu for MatRARt() -- MatRARt_mpiaij_mpiaij is supported now, but not efficient.
fix nightlybuild error with c1cbb1def3f8ddca04c3998b9c06618fd1ffef35 on arch-linux-IPL64 src/ksp/pc/examples/tests ex5_1
...

Conflicts:
include/petscdm.h
src/dm/impls/plex/plexcreate.c
src/dm/impls/plex/plexgmsh.c
src/dm/impls/plex/plexhdf5.c

show more ...

Merge branch 'master' into knepley/feature-dm-bc-components

* master: (320 commits)
fix compile warning: /sandbox/petsc/petsc.clone/src/snes/examples/tests/ex2.c:250:8: warning: ‘pointsAllProcs’ m

Merge branch 'master' into knepley/feature-dm-bc-components

* master: (320 commits)
fix compile warning: /sandbox/petsc/petsc.clone/src/snes/examples/tests/ex2.c:250:8: warning: ‘pointsAllProcs’ may be used uninitialized in this function [-Wmaybe-uninitialized] ierr = DMInterpolationSetUp(interpolator, dm, pointsAllProcs);CHKERRQ(ierr);
migrate 'build requires: !mpiuni' to new format
migrate 'build requires: !complex' to new format
workarround 'division by zero' compiler warning on windows C:\cygwin64\home\petsc\PETSC~3.CLO\src\ts\examples\TUTORI~1\ex11.c(416): warning #39: division by zero if (uL->h < 0 || uR->h < 0) {for (i=0; i<1+dim; i++) flux[i] = 0./0.; return;} /* SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Reconstructed thickness is negative"); */
dos2unix to prevent warnings on opensolaris "/export/home/petsc/petsc.clone/src/ksp/ksp/examples/tests/ex53.c", line 6: warning: invalid white space character in directive
fix output file - corresponding to code changes in efd4aadf157bf1ba2d80c2be092fcf4247860003
disable test on windows
update log from opensolaris so that the diff is clean on linux aswell
update grep in sync with change in efd4aadf157bf1ba2d80c2be092fcf4247860003
Fixed code that did not correctly set current color for block insertion
Update output files that did not get fixed after previous merges
Change MPIUni to not immediately abort on operations it cannot perform, instead return error code
Add filter for example that can produce 1 different iteration count
Example add system calloc, bad formating of %d in print, commented out code
Always use (void) for functions that take no arguments, never ()
always need to caste PetscReal variables to (double) in print statements
updates to manual page for clarity
update help menu for MatRARt() -- MatRARt_mpiaij_mpiaij is supported now, but not efficient.
fix nightlybuild error with c1cbb1def3f8ddca04c3998b9c06618fd1ffef35 on arch-linux-IPL64 src/ksp/pc/examples/tests ex5_1
output has changed with e6d0a238963c2a97dd04845ea512b529992c7cdb so update the output file
...

show more ...