* Make release_90 preferred version.
* Be more explicity about CUDA dependencies.
* Remove duplicate CUDA dependency in Flang package and introduce nvptx variant.
* Fix nvptx variant message.
* Fixed wrong link to version 0.0.0 and add hash for version 0.1.4
* Fix failing build for neovim@master and neovim@stable and add hash for version 0.4.0
* Fix flake8 issues
* Removed unnecessary newline
* Depedency conditions restriction to neovim >= 0.2.0 as previous versions fail to compile
* Removed build dependency on git
* Removed master from all conditions
* autotools: add attribute to delete libtool archives .la files
According to Autotools Mythbuster (https://autotools.io/libtool/lafiles.html)
libtool archive files are mostly vestigial, but they might create issues
when relocating binary packages as shown in #18694.
For GCC specifically, most distributions remove these files with
explicit commands:
https://git.stg.centos.org/rpms/gcc/blob/master/f/gcc.spec#_1303
Considered all of that, this commit adds an easy way for each
AutotoolsPackage to remove every .la file that has been installed.
The default, for the time being, is to maintain them - to be consistent
with what Spack was doing previously.
* autotools: delete libtool archive files by default
Following review this commit changes the default for
libtool archive files deletion and adds test to verify
the behavior.
* Add new package: py-rbtools
* Update var/spack/repos/builtin/packages/py-rbtools/package.py
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
* Update var/spack/repos/builtin/packages/py-rbtools/package.py
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
* Update var/spack/repos/builtin/packages/py-rbtools/package.py
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
* Update var/spack/repos/builtin/packages/py-rbtools/package.py
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
Original version added --no-gcc to CFLAGS when compiling with intel
compilers. This does not appear to be needed and indeed causes problems
(see #18894) with newer intel compilers; I have modified so it is not
added for intel@19: (I confirmed it is needed/works for intel@20, based
on comments in #18854 looks like holds for intel@19 as well).
(Also fix old formatting issue flake8 was complaining about)
* Update of py-redis for merlin-1.7.5
* Add hiredis variant and python versions for 3.5.x versions.
* Update var/spack/repos/builtin/packages/py-redis/package.py
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
* Added slurm version 20-02-4-1 and support to build slurmrestd
* cleanup formatting
* cleanup libjwt to pass unittests
* missed one boilerplate
* hacking a pass in unittests
* defer to default install routine in libjwt
This commit refactors the computation of the search path
for aclocal in its own method, so that it's easier to reuse
for packages that need to have a custom autoreconf phase.
Co-authored-by: Toyohisa Kameyama <kameyama@riken.jp>
The r-devtools package was not installable due to a few issues.
- The rstudioapi spec was for 0.11.0 but the rstudioapi version is
actually 0.11. This caused an error during concretization.
- Set r-usethis to depend on rlang@0.4.3: rather than r-lang@0.4.3.
- Set r-usethis to depend on r-gh@1.1.0: rather than r-gh@1.1.0.
- Added version r-gh-1.1.0 as it is not currently present in spack.
* Added CUDAHOSTCXX variable needed to compile with cuda and mpi.
* Added guard for setting CUDAHOSTCXX with MPI.
* Acceptable working version of dealii+cuda+mpi.
* cpio: Fix issue compiling with newer intel compilers (#18854)
Do not add --no-gcc for recent intel compilers (e.g. 20.x)
* cpio: Remove --no-gcc flag for intel@19 as well as intel@20
Based on comments from @nrichart, removing --no-gcc option for intel@19
as well as intel@20
* Provide draco-7_8_0.
+ Also provide a patchfile for draco-7_6_0 to support CrayPE builds.
+ Version 7.8.0 has a new variant `+caliper`.
+ Sort dependencies alphabetically after grouping by required and optional.
* Remove patchfile that is no longer needed.
+ Newer versions of draco do not require this patch.
+ Older versions of draco are not supported for spectrum-mpi.
* Change new variant +caliper to default to False.
* pandoc: add variant for texlive
Modifies the pandoc package by adding a variant for texlive, which is only needed for PDF output. Enables this variant by default.
* Fix whitespace
Fix for #19095
When given +openmp, add the correct compiler openmp flag to the link
stage. This seems to be required for %intel compilers.
I do this for all compilers, not just %intel, because it does not seem
to harm anything and might be beneficial for others (and just seems
'correct').
* py-scikit-image: bump version
* address reviewer comments
* address reviewer comments
* address reviewer comments
* py-scikit-image : update dependencies : part 2
* cloudpicke is a docs only dependency, enable it with a variant if necessary
* address reviewer comments
* cleanup build vs run deps
* address reviewer comments
* Initial cut at FLCL spackage. Works with GCC so far.
* Update spackage to list release which supports spack. Add @agaspar as a maintainer. Default unit tests to disabled when building with spack.
* Change url to 0.2.
* Nope, 0.3.
* add package py-lmodule version 0.1.0
Lmodule is tested with lmod >= 7.x. Lmod 6 has different json
structure in spider which is not supported by lmodule
* py-charm4py: new package
Charm++ for python
Installation notes:
1) charm4py ships with its own charm++ tarball. It really wants
to use the version it ships with. It also builds charm++ in a special way to
produce libcharm.so (but not charmc, etc), so it does not seem
worthwhile to try to hack to build using a spack installed charmpp.
2) Originally, the installation was failing due to unresolved cuda
symbols when setup.py was doing a ctypes.CDLL of libcharm.so (in order
to verify version?). This appears to be due to the fact that
libcharm.so had undefined cuda symbols, but did not show libcudart.so as
a dependency (in e.g. ldd output). To fix this, I had to add
libcudart.so explicitly when linking libcharm.so, but since setup.py
untars a tarball to build libcharm, the solution was a tad convoluted:
2a) Add a patch in spack to py-charm4py which creates a patchfile
"spack-charm4py-setup.py.patch" which will modify a Makefile file (after it
is untarred) to add the flags in env var SPACK_CHARM4PY_EXTRALIBS to
the link command for libcharm.so
2b) The spack patch file also patches setup.py to run patch using the
aforementioned patchfile to patch the Makefile after it is untarred, and
sets the SPACK_CHARM4PY_EXTRALIBS variable appropriately in the setup
environment.
* Update var/spack/repos/builtin/packages/py-charm4py/package.py
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
* Update var/spack/repos/builtin/packages/py-charm4py/package.py
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
* py-charm4py: flake8 fixes
remove useless import
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
* Update Package, Pymol 2.4
* Fixed flake8 stuff
* more style fixes
* missing ( at EOF
* added py-pymol 2.3 back this
* extra line removal
* white space in empty line removal
* added libpng and py-pyqt5 to prefix_path
* Fix 'unexpected product version' error for macOS 11.0
* Adjustment: add the minimum version that this macOS patch is necessary.
* Adding a keyword to prevent the patch being applied to systems other than darwin (macOS)
* Deleting quotation marks
* AMD ROCm 3.8.0 - roctracer-dev
* Update var/spack/repos/builtin/packages/roctracer-dev/package.py
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
* added py-ply dependency
* remove py-ply
* Update var/spack/repos/builtin/packages/roctracer-dev/package.py
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
* aomp 3.7.0 and rccl 3.8.0 update
* Bump up to ROCm 3.8.0 support on AOMP
* Create 0001-Add-amdgcn-to-devicelibs-bitcode-names-3.8.patch
* Create 0001-Add-amdgcn-to-devicelibs-bitcode-names.patch
* Update var/spack/repos/builtin/packages/aomp/package.py
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
It looks like intel compilers generate warnings for omp pragmas when
openmp flag is not given, which due to other flags set get promoted to
errors.
This adds a flag to ignore the pragma omp warnings (icc diagnostic
number 3180 on %intel@14:).
* Add rocblas 3.8.0 and add all Tensile deps
* Deploy rocm_smi to the bin/ folder so that it is in $PATH
* BUILD_WITH_TENSILE_HOST=ON on 3.7.0+ and fix flake8
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
[py-particle] format
[py-particle] switch to pypi downloads
[py-particle] specify dependencies in more details
[py-particle] format
Update var/spack/repos/builtin/packages/py-particle/package.py
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
Update var/spack/repos/builtin/packages/py-particle/package.py
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
Update var/spack/repos/builtin/packages/py-particle/package.py
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
* Created +python_deps variant
- the timemory python bindings can still be imported without these runtime packages and forcing a dependence by default significantly increases the spack install time
* Added conflict
- added conflicts('+python_deps', when='~python')
* rocm-3.8.0 updates for hipblas,rocsolver,rocm-opencl
* rocm-3.8.0 updates to rocalution and rename and change rocmvalidationsuite
* rocm-3.8.0 update to miopen-hip
* Revert "rocm-3.8.0 updates for hipblas,rocsolver,rocm-opencl"
This reverts commit 2542e8b1bede3d4e9e0ec1c3efd789f052826c21.
* rocm-3.8.0 changes for rocsolver and hipblas
* new package: py-gitpython
* Update var/spack/repos/builtin/packages/py-gitpython/package.py
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
Co-authored-by: sbulut <sbulut@3vgeomatics.com>
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
Co-authored-by: Sinan81 <Sinan81@earth>
* Add more updates for kokkos 3.2 release, particularly nvcc-wrapper
* Use an ordinary Package
Co-authored-by: Jeremiah J Wilke <jjwilke@kokkos-dev-2.sandia.gov>
* Flang master branch is now the preferred version.
* Flang master branch can now use LLVM 9
* Remove master as this was never used by Flang.
* Add LLVM-Flang release_90 and release_90.
Magma is not currently compatible with CUDA-11. While this is reflected
in the package, it is done with a comment in a `depends_on` directive,
which has the effect of trying to install a version of CUDA that may be
different from the one in the current environment, without any message
to the end user. A `conflicts` is a better way to handle this.
* Disable bash completion by default.
* flake8
* Adding explicit dependence on libuuid
* Adding explicit dependence on cryptsetup
This way we don't pick up host crypto packages by mistake.
* Fixing the completion directory.
* Update var/spack/repos/builtin/packages/util-linux/package.py
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
* flake8
* Removing libuuid linkage according to @michaelkuhn on #18696
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
This PR adds the current release version of mumax and tweaks the install
of the previous beta version.
- Set the url parameter to reflect the release version over the beta
version. Hopefully, this will be consistent going forward.
- Set an explicit url for the previous beta version.
- Accept values for `cuda_arch`. The previous version had its own list
but the release version does not.
- Replace the built in cuda compute capabilities list with the one
provided by Spack for the 3.10beta version.
This PR fixes a couple of things with the libbeagle package.
- libbeagle can only be built for one GPU type. Add a test for that.
- version 2 had the arch statement in
libhmsbeagle/GPU/kernels/Makefile.am but version 3 has it in
configure.ac. Put the variant specified value in configure.ac for
consistency.
Due to recent changes in the `netcdf-c` package, it is now necessary to explicitly request a non-mpi-enabled hdf5 build if building a non-mpi-enabled seacas.
* libvdwxc: unbreak concretization, request fftw-api
mixing both fftw and fftw-api in a dependency tree can trigger the
following:
```
$ spack spec cp2k@master +sirius
==> [2020-09-16-12:36:06.552981] sirius applying constraint gsl
==> [2020-09-16-12:36:06.554270] sirius applying constraint openblas@0.3.10%gcc@7.5.0~consistent_fpcsr~ilp64+pic+shared threads=none arch=linux-opensuse_leap15-sandybridge
Traceback (most recent call last):
File "./bin/spack", line 64, in <module>
sys.exit(spack.main.main())
File "/data/tiziano/debug-spack/spack2/lib/spack/spack/main.py", line 762, in main
return _invoke_command(command, parser, args, unknown)
File "/data/tiziano/debug-spack/spack2/lib/spack/spack/main.py", line 490, in _invoke_command
return_val = command(parser, args)
File "/data/tiziano/debug-spack/spack2/lib/spack/spack/cmd/spec.py", line 103, in spec
spec.concretize()
File "/data/tiziano/debug-spack/spack2/lib/spack/spack/spec.py", line 2228, in concretize
user_spec_deps=user_spec_deps),
File "/data/tiziano/debug-spack/spack2/lib/spack/spack/spec.py", line 2716, in normalize
visited, all_spec_deps, provider_index, tests)
File "/data/tiziano/debug-spack/spack2/lib/spack/spack/spec.py", line 2654, in _normalize_helper
dep, visited, spec_deps, provider_index, tests)
File "/data/tiziano/debug-spack/spack2/lib/spack/spack/spec.py", line 2613, in _merge_dependency
visited, spec_deps, provider_index, tests)
File "/data/tiziano/debug-spack/spack2/lib/spack/spack/spec.py", line 2654, in _normalize_helper
dep, visited, spec_deps, provider_index, tests)
File "/data/tiziano/debug-spack/spack2/lib/spack/spack/spec.py", line 2554, in _merge_dependency
provider = self._find_provider(dep, provider_index)
File "/data/tiziano/debug-spack/spack2/lib/spack/spack/spec.py", line 2489, in _find_provider
providers = provider_index.providers_for(vdep)
File "/data/tiziano/debug-spack/spack2/lib/spack/spack/provider_index.py", line 80, in providers_for
return sorted(s.copy() for s in result)
File "/data/tiziano/debug-spack/spack2/lib/spack/llnl/util/lang.py", line 249, in <lambda>
lambda s, o: o is not None and s._cmp_key() < o._cmp_key())
TypeError: '<' not supported between instances of 'str' and 'NoneType'
```
while at the same point disallowing MKL as a fftw provider.
Solving both issues by depending on `fftw-api@3` instead and adding a
conflict on `^fftw~mpi` when using `+mpi` (thanks to alalazo).
* cp2k: use conflicts instead of runtime checks for fftw/openblas variants
* Initial Draft of Cosmoflow Spackage
Need to add in logic to streamline cpu/gpu builds
* Added ~cuda logic to cosmoflow spackage
Added logic to support a ~cuda build for cosmoflow
* Requested Changes to Cosmoflow Spackage
Made requested changes to cosmoflow spackage
* Likwid versions >= 5.0.0 depend on Lua 5.2.
According to https://github.com/RRZE-HPC/likwid/issues/324 recent
versions of Likwid require Lua 5.2.
* Update var/spack/repos/builtin/packages/likwid/package.py
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
MatIO development has switched to github from sourceforge. Updated the `git` and `url` variables and added the four new versions (1.5.14 -- 1.5.17) that have been released since the last update of this package.
* qbox: install to correct directory structure
* qbox: Have qb executable put in bin rather than src subdir
* qbox: Fix python script shebangs to use python from path
* qbox: Add dependencies on gnuplot, python2 for utilities
* qbox: fix flake8 issue
* qbox: Add $prefix/util to PATH
* Initial CRADL Spackage Work
Currently resolving ```--single-version-externally-managed``` error
* Fixed GPUtil Issues
Thanks to Vinay Ramakrishnaiah for overwriting install
* Finished CRADL Install Function
Finished CRADL install function which is basically copying the scripts
to the install directory. Also resolved flake8 issues for PR purposes
* Add new package: webbench
* Update var/spack/repos/builtin/packages/webbench/package.py
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
* cpio: add --rtlib=compiler-rt for %fj
* cpio: simplify if
* Update var/spack/repos/builtin/packages/cpio/package.py
This seems better.
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
* Add new package: hping
* Update var/spack/repos/builtin/packages/hping/package.py
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
* Update var/spack/repos/builtin/packages/hping/package.py
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
* Update package.py
Fix flake8 errors
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
* ncurses: adding external support.
* Update var/spack/repos/builtin/packages/ncurses/package.py
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
* Update var/spack/repos/builtin/packages/ncurses/package.py
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
* Update var/spack/repos/builtin/packages/ncurses/package.py
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
* Fixing includes.
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
* Checksummed New Flux Versions
Checksummed new flux versions to let spack detect them
* Added CXXFlags to build Flux-sched
Added missing cxxflags to build flux-sched
* Adding Cuda Variant to SW4Lite
Added cuda variant of sw4lite as per guidance in README
* Updated SW4Lite+cuda to Current Header Conventions
Updated sw4lite+cuda to use current conventions for spackage include
dirs
* Fixing FLake8 Issue with Sw4lite+cuda Fix
Fixed overly long line and further underlined sticky note reminding me
to run flake8 BEFORE pushing
* Switching to Spack Compiler Wrapper
Switching to spack compiler wrapper for consistency
* Orca: Add new versions.
* Orca: Support OpenMPI without the legacy wrappers.
By default, Spack builds OpenMPI without the legacy wrappers when using the Slurm scheduler. This breaks Orca since its binaries are hardcoded to call "mpirun". To workaround this issue, add a "mpirun" wrapper which calls "srun" when required.
* cp2k: do not support ~openmp for v8+
* sirius: version bump
* cp2k: fix overlapping deps for elpa
fixes#18029
* cp2k: update SIRIUS dependency for v8+
* spfft: requires CMake 3.11+
* cp2k: fix build with +sirius
* darshan-util: remove return(-1) from void function
* Update var/spack/repos/builtin/packages/darshan-util/package.py
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
* Update var/spack/repos/builtin/packages/darshan-util/package.py
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
