* hpx5: Convert to AutotoolsPackage, several updates
- convert to AutotoolsPackage
- add several variants
- add several dependencies
- add new version 4.0.0
- don’t set compiler flags explicitly
* hpx5: Don’t import os
* hpx5: Clean up configure arguments
* hpx5: Correct libffi configure declaration
Also add new variant “+instrumentation”.
* hpx5: Correct CUDA, MPI, OpenCL configuration
* hpx5: Use built-in libffi
Spack’s libffi installs its headers in a strange place, and hpx5 can’t pick them up.
* libpciaccess: Convert to AutotoolsPackage, remove Darwin work-around
Other packages that depend on libpciaccess can do so via a conditional dependency (via a variant); we don’t need a dummy package for Darwin any more.
* Disable libpciaccess dependency on non-Darwin platforms
* hwloc: Remove direct dependency on libpciaccess
* Added support for building Theano with GPU support. Added a package
for the libgpuarry libary needed for Theano.
* Added dependency on py-pexpect package which ipython requires.
Fixes#1476
Concretization uses compilers defined in config files and if those
are not available defaults to searching typical paths where the
detected operating system would have a compiler. If there is an OS
update, the detected OS can change; in this case all compilers
defined in the config files would no longer match (because they would
be associated with the previous OS version). The error message in
this case was too vague. This commit adds logic for detecting when it
is likely that the OS has been updated (in particular when that
affects compiler concretization) and improves the information provided
to the user in the error message.
* Dont propagate flags between different compilers
Fixes#2786
Previously when a spec had no parents with an equivalent compiler,
Spack would default to adding the compiler flags associated with the
root of the DAG. This eliminates that default.
* added test for compiler flag propagation
* simplify compiler flag propagation logic
Fixes#3428
Users can run 'spack compiler find' to automatically initialize their
compilers.yaml configuration file. It also turns out that Spack will
implicitly initialize the compilers configuration file as part of
detecting compilers if none are found (so if a user were to attempt to
concretize a spec without running 'spack compiler find' it would not
fail). However, in this case Spack was overlooking its own implicit
initialization of the config files and would report that no new
compilers were found. This commit removes implicit initialization when
the user calls 'spack compiler find'.
This did not surface until #2999 because the 'spack compiler' command
defaulted to using a scope 'user/platform' that was not accounted for
in get_compiler_config (where the implicit initialization logic
predates the addition of this new scope); #2999 removed the scope
specification when checking through config files, leading to the
implicit initialization.
* hwloc: Add missing dependencies
- add several missing dependencies
- add variants for various features
- add new version 1.11.6
* hwloc: Add descriptions to variants
* New package pocl
* pocl: Update dependencies, add self-test
* pocl: Don't require LLVM shared libraries
LLVM build fails with shared libraries.
* Add patch
* Update
* Update
* Make build and install work; install test still fails
* Split pocl into pocl proper and pocl-test
* Add debug output
* pocl: Update to 0.14-rc
* pocl: Correct flake8 error
* pocl: Heed code review recommendations
* pocl: Add newline at end of file
* pocl: Correct flake8 error
I don’t want to use an even longer line by putting the whole variant declaration onto a single line, nor do I think that having an overlong line and adding `# noqa` at the end is more readable than splitting a string over three lines.
* pocl: Correct dependency type for libtool
