| 47e9f906 | 17-Mar-2021 |
Vaclav Hapla <vaclav.hapla@erdw.ethz.ch> |
PTScotch.py: Add include paths of dependencies to cflags.
This fixes PT-Scotch compilation fail if zlib is in non-default location including use of --download-zlib |
| 3330e8fb | 15-Mar-2021 |
Pierre Jolivet <pierre@joliv.et> |
Bump HPDDM |
| 600dc890 | 13-Mar-2021 |
Junchao Zhang <jczhang@anl.gov> |
config: compile kokkos with host compiler |
| ead1aa40 | 13-Mar-2021 |
Junchao Zhang <jczhang@anl.gov> |
config: extend checkCxxDialect to accept compilers other than CXX |
| 694a2f0e | 13-Mar-2021 |
Junchao Zhang <jczhang@anl.gov> |
config: change HIPCC to HIPC, and HIPCCFLAGS to HIPFLAGS |
| 268256bd | 24-Dec-2020 |
Jed Brown <jed@jedbrown.org> |
config: pass hipcc and flags to Kokkos and use RPATH |
| 45aa364b | 24-Dec-2020 |
Jed Brown <jed@jedbrown.org> |
config: Kokkos needs at least C++14 (at least with HIP) |
| 904669b2 | 24-Dec-2020 |
Jed Brown <jed@jedbrown.org> |
config: hipcc -fstack-protector crashes on some source files |
| 19825c01 | 12-Mar-2021 |
Satish Balay <balay@mcs.anl.gov> |
Merge remote-tracking branch 'origin/release' |
| c803a25a | 12-Mar-2021 |
Barry Smith <bsmith@mcs.anl.gov> |
Introduce PetscGTE(), PetscLTE() to support initial conditions that depend discontinously on coordinates
This should hopefully help in reducing the need for alt test files.
Commit-type: portability
Introduce PetscGTE(), PetscLTE() to support initial conditions that depend discontinously on coordinates
This should hopefully help in reducing the need for alt test files.
Commit-type: portability-fix, testing-fix
/spend 20m
Thanks-to: Hong Zhang <hongzhang@anl.gov>
show more ...
|
| af7a0b9f | 04-Mar-2021 |
Satish Balay <balay@mcs.anl.gov> |
CI: arch-ci-linux-gcc-complex-opt.py -> arch-ci-linux-gcc-complex-opt-32bit.py
Fix bug in ex4.c Code assumed that floating point product would produce exact result
/spend 3h
Reported-by: Gitlab CI |
| a3560f20 | 15-Feb-2021 |
Barry Smith <bsmith@mcs.anl.gov> |
Save in configure.log the Intel AVX like operations supported, both by the current compiler options and the hardware
/spend 2h |
| 722672d0 | 12-Mar-2021 |
Pierre Jolivet <pierre@joliv.et> |
configure: early exit MKL w/o mkl_spblas.h + HPDDM
Reported-by: @aldaas |
| 4ef07d6f | 11-Mar-2021 |
Barry Smith <bsmith@mcs.anl.gov> |
Attempt to have configure handle -pie passed in by MPI compiler or as LDFLAG
Conda on Mac seems to provide the -pie as a linker flag that produces warnings from
the Mac linker on some systems that c
Attempt to have configure handle -pie passed in by MPI compiler or as LDFLAG
Conda on Mac seems to provide the -pie as a linker flag that produces warnings from
the Mac linker on some systems that configure interprets as errors so cannot determine the
correct linker flags.
Commit-type: portability-fix, configure
Reported-by: Fande Kong <fdkong.jd@gmail.com>
show more ...
|
| 29ae1978 | 11-Mar-2021 |
Satish Balay <balay@mcs.anl.gov> |
Merge remote-tracking branch 'origin/release' into main |
| e0dec931 | 11-Mar-2021 |
Satish Balay <balay@mcs.anl.gov> |
cuda: fix --with-cuda-dir=/usr/local/cuda-10.2 --download-openmpi, use setCudaDir() to eliminate duplicate code
Reported-by: yangym@ihep.ac.cn |
| 40564d43 | 25-Nov-2020 |
Junchao Zhang <jczhang@mcs.anl.gov> |
Add package NVSHMEM.py |
| 61d11d1e | 09-Mar-2021 |
Barry Smith <bsmith@mcs.anl.gov> |
Fix for SuiteSparse installer on Darwin
Assumed that -rpath could be passed to shared library linker
Commit-type: portability-fix, external-package
/spend 15m
Reported-by: Daniel Finn <dsfinn@buffa
Fix for SuiteSparse installer on Darwin
Assumed that -rpath could be passed to shared library linker
Commit-type: portability-fix, external-package
/spend 15m
Reported-by: Daniel Finn <dsfinn@buffalo.edu>
Reported-by: https://gitlab.com/bensworth
show more ...
|
| 1cd963cc | 07-Mar-2021 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'balay/fix-mswin-C_DOUBLE_COMPLEX-check' into 'main'
MSWIN: set HAVE_MPI_C_DOUBLE_COMPLEX on successful compile, and always compile PetscSum_Local() etc MPI utilities
See merge request
Merge branch 'balay/fix-mswin-C_DOUBLE_COMPLEX-check' into 'main'
MSWIN: set HAVE_MPI_C_DOUBLE_COMPLEX on successful compile, and always compile PetscSum_Local() etc MPI utilities
See merge request petsc/petsc!3672
show more ...
|
| 82ea9c19 | 05-Mar-2021 |
Barry Smith <bsmith@mcs.anl.gov> |
Update compilersFortran.py |
| de272c7a | 05-Mar-2021 |
Satish Balay <balay@mcs.anl.gov> |
remove PETSC_HAVE_MPI_C_DOUBLE_COMPLEX |
| e499ad34 | 04-Mar-2021 |
Satish Balay <balay@mcs.anl.gov> |
configure: icl - (mswin) fix checkC99Complex() |
| 941d8a43 | 25-Feb-2021 |
Satish Balay <balay@mcs.anl.gov> |
configure: remove --known-mpi-long-double type options and have HAVE_MPI_C_DOUBLE_COMPLEX set on successful compile [no need for run]
running binaries via configure on windows is broken as its using
configure: remove --known-mpi-long-double type options and have HAVE_MPI_C_DOUBLE_COMPLEX set on successful compile [no need for run]
running binaries via configure on windows is broken as its using cygwin path for the binary [and the checkRun code is not needed anyway]
Reported-by: Milan Pelletier <milan.pelletier@protonmail.com>
show more ...
|
| 91d58987 | 03-Mar-2021 |
Satish Balay <balay@mcs.anl.gov> |
Merge branch 'release' into main |
| 36aba219 | 03-Mar-2021 |
Pierre Jolivet <pierre@joliv.et> |
CPARDISO: stick to OpenMPI BLACS when needed |