Commit graph

7398 commits

Author SHA1 Message Date
Todd Gamblin
e7ab8d476d Merge pull request #219 from eschnett/eschnett/gmp
Update gmp to 6.1.0
2015-12-07 16:46:44 -08:00
Todd Gamblin
6bc86a6baa Merge pull request #220 from eschnett/eschnett/mpc
Update mpc to 1.0.3
2015-12-07 16:46:32 -08:00
Todd Gamblin
b0434fa486 Merge pull request #221 from davidbeckingsale/features/cleverleaf-boost
Update CleverLeaf package to use Boost and fix SAMRAI MPI usage
2015-12-07 16:46:15 -08:00
Todd Gamblin
fcabf0ef0d Merge pull request #222 from mplegendre/bugfix/github-203-gcc5-builds
Fix issue with gcc 5.x not building on RHEL6
2015-12-07 15:55:21 -08:00
Matthew LeGendre
f8e046bb33 Fix issue with gcc 5.x not building on RHEL6 2015-12-07 15:49:47 -08:00
Gregory L. Lee
1a22a507d6 mesa requires +python variant of libxml2 2015-12-07 15:36:36 -08:00
David Beckingsale
36272c8136 Fix MPI selection in SAMRAI and remove narrow dependency specification 2015-12-07 14:58:16 -08:00
David Beckingsale
77bf133cd2 Update CleverLeaf package to use Boost and fix SAMRAI MPI usage 2015-12-07 14:33:31 -08:00
Erik Schnetter
aff95ee587 Update mpc to 1.0.3 2015-12-05 11:23:40 -05:00
Erik Schnetter
29537b9991 Update gmp to 6.1.0 2015-12-05 11:21:47 -05:00
Peter Scheibel
530e049d4a Added nose 1.3.7 source along with with the lgpl license (renamed from lgpl.txt
to LICENSE)
2015-12-04 17:55:20 -08:00
Gregory L. Lee
5891847ae4 added py-pillow and updated py-basemap to depend on it 2015-12-03 15:35:28 -08:00
Gregory L. Lee
a56c7e5f1a Merge branch 'develop' of https://github.com/scalability-llnl/spack into develop 2015-12-03 15:02:49 -08:00
Gregory L. Lee
0f4efc018d updated package versions 2015-12-03 15:01:44 -08:00
Todd Gamblin
5a5990be09 Merge pull request #214 from epfl-scitas/packages/boost
Boost: mpi, python and compression for iostream variants
2015-12-03 09:24:21 -06:00
Massimiliano Culpo
50bd4d2e4e mirror : deals correctly with variants that optionally enable resources (if they are archive URLs) 2015-12-03 11:21:11 +01:00
Peter Scheibel
b98b612204 Try to be more careful to only suppress the warning about nose already being
imported. Other warnings should be output.
2015-12-02 18:10:28 -08:00
Mario Melara
028cca16e6 Got rid of the unusued default strategy=None 2015-12-02 10:15:33 -08:00
Nicolas Richart
1a132e4fdb Missing declaration of the debug variant 2015-12-02 16:18:29 +01:00
Simon Thompson
aa137d8be0 fix build to work against depends for modules created by setup.py and add python3 support 2015-12-02 14:20:11 +00:00
Nicolas Richart
4bb6f23ae1 Adding variant for mpi and python + compression for iostream 2015-12-02 13:26:00 +01:00
Massimiliano Culpo
39a3cfd4d9 reource directive accepts 'basename' keyword
llvm : libc++ variant
2015-12-02 12:24:37 +01:00
Massimiliano Culpo
4b2168ab8e resources : added missing check after fetch 2015-12-02 10:13:39 +01:00
Mario Melara
0544b164fc Updated script of mini-xml for adios package 2015-12-01 11:59:49 -08:00
Mario Melara
4e0d47f1dd building with GCC needs special variables added. Intel compilers build just fine 2015-12-01 11:59:19 -08:00
Mario Melara
d7fdb8e015 Got rid of hard coded strings for adding default 2015-12-01 11:57:54 -08:00
Mario Melara
58f2b39bc8 Changed default to CPU_TARGET env var. Helps deal with target differences between cori and edison 2015-12-01 11:57:17 -08:00
Mario Melara
89fbe4fdfa Changed add_target to include variables rather than hard-coded strings 2015-12-01 11:55:38 -08:00
Massimiliano Culpo
b85dccca92 resources : added error handling for destination keyword on resource directive 2015-12-01 15:12:26 +01:00
Massimiliano Culpo
d3d9b5401b resources : changed name of stage folder 2015-12-01 13:26:44 +01:00
Massimiliano Culpo
a075d581ef resource : fetch strategy constructed from kwargs instead or hardcoded URLFetchStrategy 2015-12-01 12:56:46 +01:00
Massimiliano Culpo
a173ab1e31 clang : default OpenMP is libomp 2015-12-01 12:16:08 +01:00
Todd Gamblin
114e4d3fe6 Merge pull request #213 from khuck/develop
Had to force a patch to the OMPT support, so the hash for v0.1 has changed.
2015-11-30 21:42:49 -08:00
Kevin Huck
cd1207329b Updating the OMPT hash again... 2015-11-30 10:32:07 -08:00
Kevin Huck
d5c7cd5e50 Had to force a patch to the OMPT support, so the hash for v0.1 has
changed.
2015-11-30 09:48:15 -08:00
Todd Gamblin
55237b094a Use new travis insfrastructure (sudo:false) 2015-11-29 22:09:11 -08:00
Todd Gamblin
45ffcdee7b Add very basic Travis CI support. 2015-11-29 22:02:13 -08:00
Tom Scogland
b253308eee returning flux to a building state for web release
Flux package reworked to include all new dependencies, fixed issues with a
clean-build of hwloc, lua and czmq as well that prevented flux from building
cold on a minimal system.
2015-11-29 19:22:31 -08:00
Todd Gamblin
05b791a621 Add citation information to README 2015-11-28 19:45:36 -08:00
Todd Gamblin
cc349e9a32 Add default repos.yaml 2015-11-28 19:11:51 -08:00
Todd Gamblin
dfcf567de0 Add a cleaned up repo command. 2015-11-28 19:00:24 -08:00
Todd Gamblin
73ef06018e Integrate namespace attribute into spec, spec DAG, spec YAML. 2015-11-28 16:26:23 -08:00
Todd Gamblin
a338e0efd5 Fix spack.db -> spack.repo in bin/spack 2015-11-28 16:21:31 -08:00
Todd Gamblin
7383bd393e Fixed bug #42: problem with satisfies() for virtual dependencies.
- _cross_provider_maps() had suffered some bit rot (map returned was
  ill-formed but still worked for cases with one vdep)

- ProviderIndex.satisfies() was only checking whether the result map
  was non-empty.  It should check whether all common vdeps are *in*
  the result map, as that indicates there is *some* way to satisfy
  *all* of them.  We were checking whether there was some way to
  satisfy *any one* of them, which is wrong.

- Above would cause a problem when there is more than one vdep provider.

- Added test that covers this case.

- Added `constrained()` method to Spec. Analogous to `normalized()`:
  `constrain():constrained() :: normalize():normalized()`
2015-11-27 23:06:18 -08:00
Todd Gamblin
763d850a0e Add tests for importing packages directly.
Added test for, e.g.:

    import spack.pkg.builtin.mock.mpich
    import spack.pkg.builtin.mock.mpich as mpich
    from spack.pkg.builtin.mock.mpich import Mpich

Among others.  These ensure that direct package imports work so that
packages can be extended.
2015-11-26 14:23:12 -08:00
Todd Gamblin
89d5127900 New, cleaner package repository structure.
Package repositories now look like this:

    top-level-dir/
        repo.yaml
        packages/
            libelf/
                package.py
            mpich/
                package.py
            ...

This leaves room at the top level for additional metadata, source,
per-repo configs, indexes, etc., and it makes it easy to see that
something is a spack repo (just look for repo.yaml and packages).
2015-11-26 14:19:27 -08:00
Massimiliano Culpo
1fe626ec7c resource directive : sketch of implementation + clang / llvm use case 2015-11-26 17:53:33 +01:00
Peter Scheibel
c5ed7e2a15 On systems which have already installed nose (perhaps an older version)
importing nose (which is also in Spack's external/ directory) outputs a warning
(since there is more than one nose package). This avoids printing the warning
to the user.
2015-11-25 18:33:52 -08:00
Peter Scheibel
fc96f62794 Updated documentation/naming in tally_plugin to reflect that the count of the
total number of tests run excludes skipped tests
2015-11-25 18:08:17 -08:00
Peter Scheibel
d50a18d9eb XML output for unit tests is now enabled with an option (disabled by default) 2015-11-25 17:10:45 -08:00