* Additional versions of py-jsonschema.
* Tweak to force Maestro to use jsonschema@3.2.0:
* Correction of whitespace (flake8 error).
* Merges importlib's Python version conditons
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
* Add new versions of spfft
* Extend CudaPackage and use virtual fftw package
Co-authored-by: Simon Pintarelli <simon.pintarelli@cscs.ch>
* Add CUDA 11 compatibility note
* Depend on older cuda <= 10 for spfft <= 0.9.11
Co-authored-by: Simon Pintarelli <simon.pintarelli@cscs.ch>
* introduce logic for boost+context dependency and generic_context variant
* fix OTF2 instrumentation minor problem
* default coroutine impl depends on platform
* fix flake8
* add reference to ~generic_coroutines conflict info
* Update var/spack/repos/builtin/packages/hpx/package.py
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
* Octopus: Add support for version 10.0.
Fix compilation when using the MKL as a provider for BLAS/LAPACK. Octopus will now detect that the MKL also provides the FFTW API and will refuse to compile when both the FFTW library and the MKL are given to the configure script.
* Octopus: Add supported version range for libxc.
* berkeley-db: add version 18.1.40, update build options in package
* combine adamjstewart's changes
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
* [kassiopeia] New package
* [kassiopeia] Remove master branch, update dependencies
* Update var/spack/repos/builtin/packages/kassiopeia/package.py
Unable to test since I do not have a license to intel-parallel-studio, but I see no reason why it would not work if.
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
* [kassiopeia] depends_on mpi
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
* [kassiopeia] cmake_args with self.spec.satisfies and elses
* [kassiopeia] args.extend -> args.append
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
* h5py: explicitly specify version
hdf5@1.10.5 on Cray is wrongly detected as 1.8.4.
* Update var/spack/repos/builtin/packages/py-h5py/package.py
Thanks. Also had this first, then CI was complaining about line length ...
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
Before:
```console
$ licensee diff --license mit LICENSE-MIT
Comparing to MIT License:
Input Length: 1092
License length: 1020
Similarity: 92.46%
diff --git a/LICENSE b/LICENSE
index 0ce42af..be0ff1c 100644
--- a/LICENSE
+++ b/LICENSE
@@ -1,3 +1,4 @@
{+spack project developers. see the top-level copyright file for details.+}
permission is hereby granted, free of charge, to any person obtaining a copy of
this software and associated documentation files (the "software"), to deal in
the software without restriction, including without limitation the rights to
```
After:
```console
$ licensee diff --license mit LICENSE-MIT
Comparing to MIT License:
Input Length: 1020
License length: 1020
Similarity: 100.00%
Exact match!
```
This gets us a 100% license match from GitHub's `licensee` tool.
`spack buildcache list` was trying to construct an `Arch` object and
compare it to `arch_for_spec(<spec>)`. for each spec in the buildcache.
`Arch` objects are only intended to be constructed for the machine they
describe. The `ArchSpec` object (part of the `Spec`) is the descriptor
that lets us talk about architectures anywhere.
- [x] Modify `spack buildcache list` and `spack buildcache install` to
filter with `Spec` matching instead of using `Arch`.
- [x] Make it easier to get a `Spec` with a proper `ArchSpec` from an
`Arch` object via new `Arch.to_spec()` method.
- [x] Pull `spack.architecture.default_arch()` out of
`spack.architecture.sys_type()` so we can get an `Arch` instead of
a string.
Ci is currently failing on brew update with the error:
```
Error: Cannot install bazelisk because conflicting formulae are installed.
bazel: because Bazelisk replaces the bazel binary
Please `brew unlink bazel` before continuing.
Unlinking removes a formula's symlinks from /usr/local. You can
link the formula again after the install finishes. You can --force this
install, but the build may fail or cause obscure side effects in the
resulting software.
```
Avoiding:
```
$ brew update
$ brew upgrade
```
solves the issue by preventing the risk of conflicting formulae
* Run Python2.6 unit tests on Github Actions
* Skip url tests on Python 2.6 to reduce waiting times
* Skip foreground background tests on Python 2.6 to reduce waiting times
* Removed references to Travis in the documentation
* Deleted install_patchelf.sh (can be installed from repo on CentOS 6)
* Update LBANN, Hydrogen, Aluminum to inherit CudaPackage
* Update CMake constraints: LBANN, Hydrogen, and Aluminum now require
cmake@3.16.0: (better support for pthreads with nvcc)
* Aluminum: add variants for host-enabled MPI and RMA features in a
MPI-GPU RDMA-enabled library
* NCCL: add versions 2.7.5-1, 2.7.6-1, and 2.7.8-1
* Hydrogen: add version 1.4.0
* LBANN: add versions 0.99 and 0.100
* Aluminum: add versions 0.4.0 and 0.5.0
* new package(s): py-gql
and related dependencies:
py-aiohttp
py-async-timeout
py-graphql-core
py-idna-ssl
py-multidict
py-websockets
py-yarl
new versions:
py-requests
* fixes
Co-authored-by: Andrew W Elble <aweits@skl-a-00.rc.rit.edu>
* NWChem 7.0.0
* add python2 for 6.8.1. removed 6.8 https://github.com/spack/spack/pull/17779#discussion_r462700413
* nwchem 6.8.1 breaks with gcc 10 and later
* restored extra python bits for version 6.8.1. add env. definition of basis libraries
* changes for flake8
* url fixed
* prevent 6.8.1 being compiled with gcc 10
* Ferret: Add missing dependency with curl.
* Ferret: Don't force using the static version of libgfortran.
* Ferret: Ensure Spack's compiler wrappers are used.
This allows properly setting the rpaths.
* Ferret: Add support for versions 7.3 to 7.6.
* Ferret: Add a variant to install Ferret standard datasets.
* Ferret: Define some useful runtime environnement variables.
* Ferret: Fix flake8.
Also add myself as a maintainer as suggested by @alalazo.
As discussed in issue #17638, wherein kahip fails to build when
scons is dependent on python@3.
This converts the print statements in various SConstruct files
into python3 friendly print functions.
I found most of the affected SConstruct files in both @2.00 and
the later versions I found on web, but some files were only in @2.00.
I split the patches into two files for that reason, but have not
tried the later versions.
