also build BLIS backend for FlexiBLAS v3.0.4 with GCC/10.3.0

[smoors]

(created using eb --new-pr)

this is needed to avoid failing test reports with

== 2022-04-26 11:57:35,990 easyblock.py:3049 INFO Checking for banned/required linked shared libraries...
== 2022-04-26 11:57:35,990 run.py:179 INFO Auto-enabling streaming output of 'flexiblas list' command because logging to stdout is enabled
== 2022-04-26 11:57:35,990 run.py:214 DEBUG run_cmd: running cmd flexiblas list (in /tmp/vsc10009/ebinstall/15334/software/Roary/3.13.0-foss-2021a)
== 2022-04-26 11:57:35,991 run.py:233 INFO running cmd: flexiblas list
== 2022-04-26 11:57:37,130 build_log.py:169 ERROR EasyBuild crashed with an error (at easybuild/base/exceptions.py:124 in __init__): cmd "flexiblas list" exited with exit code -6 and output:
[1;2;31mflexiblas Failed to get the BLAS backend (BLIS) from the configuration.

see also #15334

[smoors]

@boegelbot: please test @ generoso

[boegelbot]

@smoors: Request for testing this PR well received on login1

PR test command 'EB_PR=15347 EB_ARGS= /opt/software/slurm/bin/sbatch --job-name test_PR_15347 --ntasks=4 ~/boegelbot/eb_from_pr_upload_generoso.sh' executed!

• exit code: 0
• output:

Submitted batch job 8425

Test results coming soon (I hope)...

- notification for comment with ID 1109867512 processed

Message to humans: this is just bookkeeping information for me,
it is of no use to you (unless you think I have a bug, which I don't).

[boegel]

Hold on, how can we have missed this in earlier testing?!

cc @bartoldeman

[boegelbot]

Test report by @boegelbot
SUCCESS
Build succeeded for 1 out of 1 (1 easyconfigs in total)
cns1 - Linux Rocky Linux 8.5, x86_64, Intel(R) Xeon(R) CPU E5-2667 v3 @ 3.20GHz (haswell), Python 3.6.8
See https://gist.github.com/ca4a1d6d3068315072942dffe477740a for a full test report.

[smoors]

Hold on, how can we have missed this in earlier testing?!

it's also weird that in generoso the test did not fail without this PR.

[boegel]

@smoors It makes no sense to me that flexiblas list is complaining with

flexiblas Failed to get the BLAS backend (BLIS) from the configuration.

if it wasn't built with a BLIS backend at all... How would it even know about BLIS in that case?!

Are you sure that the problem with flexiblas list producing that error isn't self-inflicted somehow, for example by having $FLEXIBLAS set to BLIS in your environment for some reason?

Note: flexiblas list is being run by the toolchain mechanism in EasyBuild framework, to determine for which FlexiBLAS backend libraries a sanity check should be performed whether there's any direct linking being done that library, as opposed to correctly linking to libflexiblas.so instead (as a part of sanity_check_linked_shared_libs).

@bartoldeman Do you have any ideas here?

I'll rebuild FlexiBLAS/3.0.4-GCC-10.3.0 on generoso without that change for now, to avoid trouble...

[smoors]

@boegel mystery solved :)

$  cat ~/.flexiblasrc
default=BLIS

it could still be useful to add support for BLIS here, but I understand that you don't want to introduce unnecessary changes to a toolchain module.

[boegel]

@smoors Now that the mystery is solved, this is just an enhancement, which indeed makes sense, so let's get it applied...

[boegel]

@boegelbot please test @ generoso

[boegelbot]

@boegel: Request for testing this PR well received on login1

PR test command 'EB_PR=15347 EB_ARGS= /opt/software/slurm/bin/sbatch --job-name test_PR_15347 --ntasks=4 ~/boegelbot/eb_from_pr_upload_generoso.sh' executed!

• exit code: 0
• output:

Submitted batch job 8433

Test results coming soon (I hope)...

- notification for comment with ID 1111094335 processed

Message to humans: this is just bookkeeping information for me,
it is of no use to you (unless you think I have a bug, which I don't).

[boegel]

Test report by @boegel
SUCCESS
Build succeeded for 1 out of 1 (1 easyconfigs in total)
node2607.swalot.os - Linux CentOS Linux 7.9.2009, x86_64, Intel(R) Xeon(R) CPU E5-2660 v3 @ 2.60GHz (haswell), Python 3.6.8
See https://gist.github.com/9f49252e465887e180e1a1534a20d714 for a full test report.

[boegelbot]

Test report by @boegelbot
SUCCESS
Build succeeded for 1 out of 1 (1 easyconfigs in total)
cns1 - Linux Rocky Linux 8.5, x86_64, Intel(R) Xeon(R) CPU E5-2667 v3 @ 3.20GHz (haswell), Python 3.6.8
See https://gist.github.com/731b5da7921ca4637539027d6f14f0cf for a full test report.

[boegel]

Test report by @boegel
SUCCESS
Build succeeded for 1 out of 1 (1 easyconfigs in total)
node3519.doduo.os - Linux RHEL 8.4, x86_64, AMD EPYC 7552 48-Core Processor (zen2), Python 3.6.8
See https://gist.github.com/9bedc482b8fd48e5623a97f79b9bf90d for a full test report.

[boegel]

@boegelbot please test @ jsc-zen2

[boegelbot]

@boegel: Request for testing this PR well received on jsczen2l1.int.jsc-zen2.easybuild-test.cluster

PR test command 'EB_PR=15347 EB_ARGS= /opt/software/slurm/bin/sbatch --job-name test_PR_15347 --ntasks=8 ~/boegelbot/eb_from_pr_upload_jsc-zen2.sh' executed!

• exit code: 0
• output:

Submitted batch job 1146

Test results coming soon (I hope)...

- notification for comment with ID 1111193805 processed

Message to humans: this is just bookkeeping information for me,
it is of no use to you (unless you think I have a bug, which I don't).

[boegelbot]

Test report by @boegelbot
SUCCESS
Build succeeded for 1 out of 1 (1 easyconfigs in total)
jsczen2c1.int.jsc-zen2.easybuild-test.cluster - Linux Rocky Linux 8.5, x86_64, AMD EPYC 7742 64-Core Processor (zen2), Python 3.6.8
See https://gist.github.com/5683fcb5fc68cb36c9dd062bad8debd6 for a full test report.

[boegel]

lgtm

[boegel]

Going in, thanks @smoors!