{lang}[iimkl/2025b] SciPy-bundle v2025.07#23826
Conversation
…numpy-2.3.2_fix_selected_kind_for_ifort.patch
Updated software
|
|
Test report by @branfosj |
|
@boegelbot please test @ jsc-zen3 |
|
@branfosj: Request for testing this PR well received on jsczen3l1.int.jsc-zen3.fz-juelich.de PR test command '
Test results coming soon (I hope)... Details- notification for comment with ID 3269927284 processed Message to humans: this is just bookkeeping information for me, |
|
Test report by @boegelbot |
|
Test report by @Thyre |
|
Reported to Intel via support ticket 06682863 |
|
Test report by @branfosj |
|
Test report by @branfosj |
|
Test report by @branfosj |
|
@boegelbot please test @ jsc-zen3 |
|
@branfosj: Request for testing this PR well received on jsczen3l1.int.jsc-zen3.fz-juelich.de PR test command '
Test results coming soon (I hope)... Details- notification for comment with ID 3270643177 processed Message to humans: this is just bookkeeping information for me, |
|
Test report by @boegelbot |
|
Test report by @branfosj |
|
Test report by @branfosj |
|
Test report by @branfosj |
|
Test report by @branfosj |
|
Test report by @Thyre |
|
Hmm, is my test failure and the one of boegelbot maybe caused by still using |
|
@branfosj From the support ticket I filed:
|
I've added that and removed the patch. We might want to target it more than this, but it will be good to check that it works. |
|
@boegelbot please test @ jsc-zen3 |
|
@branfosj: Request for testing this PR well received on jsczen3l1.int.jsc-zen3.fz-juelich.de PR test command '
Test results coming soon (I hope)... Details- notification for comment with ID 3273812990 processed Message to humans: this is just bookkeeping information for me, |
|
Test report by @branfosj |
|
Test report by @boegelbot |
|
Test report by @Thyre |
| toolchain = {'name': 'iimkl', 'version': '2025b'} | ||
| toolchainopts = {'pic': True, 'lowopt': True, 'strict': True, | ||
| # '-fveclib=none' to avoid 'undefined symbol: pow8_h' in numpy | ||
| # see https://github.com/numpy/numpy/issues/27111 |
There was a problem hiding this comment.
-fveclib=none may cost significant performance. Have you tried linking with -lsvml instead (small vector math library)?
(untested, it may not work...)
There was a problem hiding this comment.
I've not found where we can set this that prevents the build failure.
There was a problem hiding this comment.
Getting it to work is far more important than element-wise exponentials being a bit faster!
|
Test report by @branfosj |
|
Test report by @branfosj |
|
Test report by @branfosj |
|
Test report by @boegel |
We require the framework fixes for the rpath wrapper, otherwise Meson fails to detect lots of features unfortunately. |
|
Test report by @boegel |
|
Going in, thanks @branfosj! |
(created using
eb --new-pr)-cflag specified easybuild-framework#4981