use generic way of checking BLAS linkage for recent numpy versions#762
use generic way of checking BLAS linkage for recent numpy versions#762boegel merged 8 commits intoeasybuilders:developfrom
Conversation
|
@wpoely86: please review? |
|
looks good |
|
Easyblocks unit test suite PASSed (see https://jenkins1.ugent.be/job/easybuild-easyblocks-pr-builder/1434/console for more details). This pull request is now ready for review/testing. Please try and find someone who can tackle this; contact @boegel if you're not sure what to do. |
|
Easyblocks unit test suite PASSed (see https://jenkins1.ugent.be/job/easybuild-easyblocks-pr-builder/1435/console for more details). This pull request is now ready for review/testing. Please try and find someone who can tackle this; contact @boegel if you're not sure what to do. |
|
Easyblocks unit test suite PASSed (see https://jenkins1.ugent.be/job/easybuild-easyblocks-pr-builder/1440/console for more details). This pull request is now ready for review/testing. Please try and find someone who can tackle this; contact @boegel if you're not sure what to do. |
|
Easyblocks unit test suite PASSed (see https://jenkins1.ugent.be/job/easybuild-easyblocks-pr-builder/1441/console for more details). This pull request is now ready for review/testing. Please try and find someone who can tackle this; contact @boegel if you're not sure what to do. |
|
Easyblocks unit test suite PASSed (see https://jenkins1.ugent.be/job/easybuild-easyblocks-pr-builder/1443/console for more details). This pull request is now ready for review/testing. Please try and find someone who can tackle this; contact @boegel if you're not sure what to do. |
|
Tested by rebuilding with numpy 1.9 and 1.10 + Python easyconfigs, good to go, thanks for the review @wpoely86! |
use generic way of checking BLAS linkage for recent numpy versions
implement generic way of checking BLAS linkage in numpy
check for
blas_dotsymbol (cfr. #757) doesn't always work, apparently (seems like it's not there when optimising more aggressively (-O3 vs -O2), or whne-fPICis used)cfr. numpy/numpy#6675 (comment)
cc @gppezzi