Remove unneeded tabs from makefile.

Move mcudahost.cu to its own directory, and only compile if we have CUDA.

Add support for pinned memory and use it on VecCUDA

The pinned memory is used when the requested memory size is over 128MB.

Funded-by:
Project: PETSc for GPU
Time:
Reported-by:
Thanks-to:

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

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

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-plex-sphere-mesh

* master: (2633 commits)
Bib: Added ref
SNES ex56: Update test output
add a citation to petsc.bib
Plex: Initialize variables
ini

Merge branch 'master' into knepley/feature-plex-sphere-mesh

* master: (2633 commits)
Bib: Added ref
SNES ex56: Update test output
add a citation to petsc.bib
Plex: Initialize variables
initialize ap as complained by the nightly tests
Plex ex3: Update to new point function API
initialize some variables as complained by nightly tests
SNES ex12: Mix misnaming Nf --> Nc
update tests
add tests
add MatView_SeqXAIJ_ASCII_structonly(); enable A->structure_only in MatSetValuesBlocked_SeqBAIJ
typo
fill-in some references
Bib: Fixed ref
Plex: Added PLY to file support
Plex: Adding PLY support
PetscDS: Turn off Fortran bindings for Constants
Plex tests: Updated to new point function API
SNES ex56: Moved tests to new format
DM: Allow field projection functions to take an array of scalars - This will allow us to pass in user data to kernels without void *, more efficiency than aux fields
...

Conflicts:
config/builder.py
include/petscdmplex.h
src/dm/impls/plex/examples/tests/ex1.c
src/dm/impls/plex/plexcreate.c
src/dm/impls/plex/plexgmsh.c
src/snes/examples/tutorials/ex63.c

show more ...

Merge branch 'master' into knepley/feature-plex-boundary-values

* master: (487 commits)
Bib: More refs
Fixed errors in the manuals and ts_glee source code doc
fix bug to enable '-xxx_via [alg]

Merge branch 'master' into knepley/feature-plex-boundary-values

* master: (487 commits)
Bib: More refs
Fixed errors in the manuals and ts_glee source code doc
fix bug to enable '-xxx_via [alg]' to showup in '-help' Reported-by: Mark Adams <mfadams@lbl.gov>
Plex: Remove dead code
TS tutorials ex11: remove unused functions
TSAdapt: Set TSADAPTNONE as default adapt type
DMPlex: missed viewer change in synchronized output
Revert "Merge branch 'hzhang/ts-ex9busdmnetwork'"
TSAdapt: Update changes/dev.html
TSAdapt: Set default type for TSGLEE
TS tutorials ex11: make work with complex scalars
Bib: Added ref
TS ex18: Fix initializer
TS ex11: Fix M_PI
Plex ex1f90: FUCK FORTRAN
f2cblaslapack: remove duplicate xerbla.o and xerbla_array.o from lapack [they are already in blas]
TSAdapt: Refactor handling of default adapt type
TSAdapt: Update changes/dev.html
TSAdapt: Update manual
TSAdapt: Refactor interface and implementations
...

show more ...

Merge branch 'master' into knepley/feature-plex-torus

* master: (174 commits)
Bib: Another ref
Increase patchlevel to 3.7.6
Explicitly listing the LTO library causes difficulties on MacOS so a

Merge branch 'master' into knepley/feature-plex-torus

* master: (174 commits)
Bib: Another ref
Increase patchlevel to 3.7.6
Explicitly listing the LTO library causes difficulties on MacOS so avoid it
TSBDF: Update changes/dev.html
Added Fortran TaoType and TaoLineSearchType definitions
Add Fortran TaoType and TaoLineSearchType definitions
setblocksize for iscol_sub
TSBDF: Use standard TS API and options to control adaptivity
Bib: More refs and a correction
SNESTest: Bug-fix in SNESTest when threshold printing option is turned on.
fix warning and valgrind complains reported by the nightly tests
Section+Doc: Small doc fix
pflotran: switch to using the new upstream git repo
DMPlex: Error if attempt to partition a distributed graph with Chaco
fix complex build error
fix complex build
PetscError() did not properly copy over the specific error message when error type was PETSC_ERROR_INITIAL
TSAdapt: Refactor handling of the always_accept flag
Additional clarification for MatMeshToCellGraph() manual page
optimize MatCreateSubMatrix_MPIAIJ() by merging check of sameXXXDist() into one MPI_AllReduce
...

Conflicts:
config/BuildSystem/config/packages/med.py

show more ...

Merge branch 'master' into knepley/feature-plex-tests-new

* master: (108 commits)
Bib: Added ref
Bib: Updated reference
Bib: Updated reference
Work around problems with Apple Xcode 8.3 and h

Merge branch 'master' into knepley/feature-plex-tests-new

* master: (108 commits)
Bib: Added ref
Bib: Updated reference
Bib: Updated reference
Work around problems with Apple Xcode 8.3 and handling of rpath argument with gfortran
help menu update for MatCreateSubMatrices() Reported-by: Brad Aagaard <baagaard@usgs.gov>
fix flops for PCApply_PBJacobi_1 and PCApply_PBJacobi_7 Reported-by: Pierre Jolivet <pierre.jolivet@enseeiht.fr>
TSRK: Fix FSAL optimization to not skip update and check or stages.
Travis-CI: Workaround too long TMPDIR in macOS with Open MPI
test: Fix type mismatch in KSP test.
test: Fix warnings from gfortran -Wconversion (OS X, Homebrew build)
MPIUNI: Silence GCC -Wconversion warnings
Fortran: Add error checking for incorrect PETSC_NULL_XXX argument when expecting function
fix test output per changes in 517c9301b2e9239931bfd644803fd6bb859387ea
MUMPS: update to version 5.1.1 with opensolaris compile fixes
TSRK: Mention 5BS in the changelog.
TSRK: Mention the new 5BS scheme in the manual.
fix 'unused variable' warning
Remove unused code
added hbw_realloc
fixed undefined ptr
...

Conflicts:
config/BuildSystem/config/packages/med.py

show more ...

Merge remote-tracking branch 'origin/hongzh/knl'

add HBW allocator and a special use case for adjoint checkpointing on KNL
Funded-by:
Project: PETSc for Theta
Time: 8 hours
Reported-by:
Thanks-to:

Merge branch 'master' into pr263/PierreBdR/ts-domain-error-in-stage/master

Merge branch 'barry/fix-nonew-notcollective/maint' into jed/mat-assembly-perf

Jed this fucking 6+ month running thing in next that is not in master is a royal fucking pain in the ass and
an abuse of

Merge branch 'barry/fix-nonew-notcollective/maint' into jed/mat-assembly-perf

Jed this fucking 6+ month running thing in next that is not in master is a royal fucking pain in the ass and
an abuse of git. I think it is not justified to ever have anything in next for more than a few weeks at most.
Either take it out of next if it is broken or put it in master if it is not broken. Hanging around in next but
not master for ever is not a good policy.

show more ...

Merge branch 'master' into knepley/solkx

* master: (6933 commits)
Bib: Added reference
Includegraph: Updated for compatibility with new folder layout.
Add SNESLineSearchReason to fortran inclu

Merge branch 'master' into knepley/solkx

* master: (6933 commits)
Bib: Added reference
Includegraph: Updated for compatibility with new folder layout.
Add SNESLineSearchReason to fortran includes Also truncate names to fit 32 char fortran limit
fix compile warnings /usr/home/balay/petsc.clone-2/src/vec/vec/interface/vector.c:1944: warning: division by zero in '1.0e+0 / 0.' /usr/home/balay/petsc.clone-2/src/ksp/ksp/interface/itfunc.c:508: warning: 'vec_rhs' may be used uninitialized in this function
updated output for new trust region initial size
initial size of the trust region is set as a percent of the norm of the initial guess, it should not be related to the initial norm of the function (which is kind of nuts).
The default bounds for SNESVISetVariableBounds() in the manual page were reversed from correct values
Bib: Added refs
Bib: Update MPICH webpage
Plex: Forgot to close file
removed nonexistent ex46f from makefile test rule
Revert "fixed bad merge into master"
fixed bad merge into master
fix memory leak in mkl_pardiso fix some formatting in mkl_pardiso code
mv runex111 from TESTEXAMPLES_C to TESTEXAMPLES_DATAFILESPATH
Mat+Doc: More Fortran docs
Plex ex3: Run tests with the correct number of field components
Bib: Added TetGen cite
parmetis: make sure 'ldd libparmetis.so' points to the correct libmetis.so that its linked with.
small fix as reported by the nightly test
...

Conflicts:
config/builder.py

show more ...

Merge branch 'master' into tisaac/dmforest

directory layout changed again: merging to switching branches back and forth
doesn't cause reconfigure to have a hard time

* master: (357 commits)
remov

Merge branch 'master' into tisaac/dmforest

directory layout changed again: merging to switching branches back and forth
doesn't cause reconfigure to have a hard time

* master: (357 commits)
removed nonexistent ex46f from makefile test rule
Revert "fixed bad merge into master"
fixed bad merge into master
fix memory leak in mkl_pardiso fix some formatting in mkl_pardiso code
mv runex111 from TESTEXAMPLES_C to TESTEXAMPLES_DATAFILESPATH
Mat+Doc: More Fortran docs
Bib: Added TetGen cite
parmetis: make sure 'ldd libparmetis.so' points to the correct libmetis.so that its linked with.
small fix as reported by the nightly test
add a test for MatPtAP()
small cleanup; add tests
cleanup MatPtAPSymbolic_MPIAIJ_MPIAIJ()
Fix nonzerostate tracking in all MATMPI types.
update location of test matrices in faq.html file
unused variable still declared
Fix for MatLoad_MPIDense() from sparsely stored matrix also fix for MatLoad_MPIDense_DenseInFile() when matrix has not had its local sizes preset by user
Preserve the B nonzerostate during disassembly.
TSEvent: Fix arch-mswin nightly errors due to declaring variables in the middle of a function.
Fix error in parallel loading of dense format matrix
removed KSPSetNullSpace() since mat already carries that information and having two locations for it is confusing, unnecessary and prone to error added MatSetTransposeNullSpace() for providing the null space of the transpose of the matrix and added support in KSPSolve() to remove this null space from the right handside of the linear system thus making the system consistent.
...

show more ...

Merge branch 'master' into knepley/feature-composable-callbacks

* master: (151 commits)
PetscMalloc: Fix handling of zero-sized allocations
add note to DMGetLocal/GlobalVector() about intended s

Merge branch 'master' into knepley/feature-composable-callbacks

* master: (151 commits)
PetscMalloc: Fix handling of zero-sized allocations
add note to DMGetLocal/GlobalVector() about intended short-time usage of the vectors.
sf: use patched openmpi tarball that has a potential fix for sf example hang
SNES problems that set VI bounds must use a VI solver since the other solvers ignore the bounds
pounders viewer did not cast PetscReal to double in printf nor use %D instead of %d
fix ts ex20adj
fix ts theta
fixed another variable declaration in the middle of a code block
example main() had PetscInt return type and argument that should be int
DMDAVTKWriteAll_VTS: Fix handling of higher-dimension coordinates
fixed example that declared variables in the middle of a function
C programs do not return PetscInt nor do they take an argument that is PetscInt, fixed example that did
You cannot declare variables in the middle of functions in PETSc.
add ierr checking to PetscOptionsBegin/End() in example
configure: add error check if user specifies PETSC_ARCH=-dbg [i.e starts with a '-']
scale norm of vector difference by norm of vector to insure it is small enough to consider the two vectors equal
missing PetscViewerDestroy() added
MatNestGetISs: use custom fortranstub to handle PETSC_NULL_OBJECT
update location of open-mx website on FAQ page
If the user sets bounds and calls a solver that does not support them, throw an error.
...

show more ...

Merge branch 'pr192/Fande-Kong/scalable-matincreaseoverlap/master' of https://bitbucket.org/petsc/petsc

Merged petsc/petsc into master

Merge branch 'master' into stefano_zampini/pcbddc-adaptive-selection

Conflicts:
src/ksp/pc/impls/bddc/bddcprivate.c
src/ksp/pc/impls/bddc/bddcscalingbasic.c
src/mat/impls/aij/mpi/mumps/mumps.c
s

Merge branch 'master' into stefano_zampini/pcbddc-adaptive-selection

Conflicts:
src/ksp/pc/impls/bddc/bddcprivate.c
src/ksp/pc/impls/bddc/bddcscalingbasic.c
src/mat/impls/aij/mpi/mumps/mumps.c
src/mat/impls/is/matis.h

show more ...

Merged master into semihozmen/ksp-fcg-eigen-comp-support

Merge branch 'master' into pr192/Fande-Kong/scalable-matincreaseoverlap/master

Needed to merge master to update to new calling sequence PetscCommBuildTwoSided

Merged petsc/petsc into master

Merge branch 'barry/fix-directory-layout'

updated PETSc directory layout to match standard packaging strategies

include/petsc finclude,private,mpiuni
lib/petsc conf
bin/petsc*

Merge branch 'master' into tisaac/dmforest

If we're going to start developing on this again, the recent changes to the
location of configuration files will make it annoying to switch to and from
thi

Merge branch 'master' into tisaac/dmforest

If we're going to start developing on this again, the recent changes to the
location of configuration files will make it annoying to switch to and from
this branch. Matt's and my fixes in DMPlex will also be relevant.

* master: (1557 commits)
MatLoad_MPIDense() always allocated array space even if it was previously provided
Clarified lumping options for PCFieldSplitSchurPreType SELFP.
fixed up PetscInfo in GAMG
TSErrorWeightedNorm[2|Infinity]: API changes
MatLoad_SeqDense() always allocated new space for the data even if the user already provided it
Fixed DMCompositeGetAccessArray() to support readonly vectors
Fix obviously-wrong aspects of documentation on SNESVINEWTONRSLS and SNESVINEWTONSSLS.
fixed example to compile - demonstrates a bug
var name 'len' was F90 key word
For user provided Jacobian functions they should always assembly into the second matrix, the one from which the preconditioner is built because the first may be used as matrix-free with -snes_mf_operator
Suggest --with-64-bit-indices if users mistakenly use --with-64bit-indices
Fix configure logging of integer size.
missing underscore in #define of vecgetarrayread translation of fortran stub for VECGETARRAYREAD
updated test output files to reflect differences introduced previously in PCView output
-ksp_type preonly does not support -ksp_monitor_short so remove from test line to prevent extranious warning output
missing CHKERRQ; and ugly double declaration of ierr
Plex ex3: For want of a semicolon, the kingdom was lost
MatSetValuesMPISBAIJ will now print correct global indices when error generated due to new nonzero location
completed basic use of MatSetValuesBlocked_SeqBAIJ_Inlined() by MatSetValuesBlocked_MPIBAIJ() to allow proper global indices being printed when error generated by new nonzero locations.
Added inline version of MatSetValuesBlocked_SeqBAIJ() to be used by MatSetValuesBlocked_MPIBAIJ() to print error message for nonzero block with global indices
...

show more ...