update custom easyblock for PETSc to consider include/suitesparse subdirectory for SuiteSparse headers

[tanmoy1989]

(created using eb --new-pr)
SuiteSparse can install its headers into a subdirectory of the include directory instead. To be safe, both flavours of include directory locations were considered for SuiteSparse.

[zao]

Test report by @zao

Overview of tested easyconfigs (in order)

• FAIL (build issue) PETSc-3.19.2-foss-2022b.eb (partial log available at https://gist.github.com/zao/b035ccb550e7c11195a42bb78e706a87)

Build succeeded for 0 out of 1 (1 easyconfigs in total)
eb-mix.zao.se - Linux Rocky Linux 9.4 (Blue Onyx), x86_64, AMD Ryzen 9 3900X 12-Core Processor (zen2), Python 3.9.18
See https://gist.github.com/zao/4fa75f6450d7a51a0a2a1c51522b2df3 for a full test report.

[zao]

It seems like the PETSc build gets angry if any of the SuiteSparse include directory candidates given on the command line doesn't exist, see the test report in #3391 (comment) above.

This logic probably needs to guard the additional entry behind whether the directory actually exists on the filesystem.

[tanmoy1989]

Thanks @zao: I would really appreciate if you could provide suggestions on how to proceed further. I am running out of options and a bit confused too.

[boegel]

@tanmoy1989 While I'm not opposing the proposed change, as long as it doesn't cause trouble with builds that were working fine before, I do wonder what's triggering the different subdirectory to be used instead?

[tanmoy1989]

@boegel: I am not sure if I understood your question completely, however, I am fully Ok if you suggest any changes to be made before merging the PR.

[boegel]

@tanmoy1989 I guess my question is what triggered you to look into implemented these changes.

I'm not sure how come we did not run into this before, but maybe I'm missing something...

[tanmoy1989]

@boegel Well, the main triggering point was the stage where PETSc was failing to build. Then @zao suggested some changes (here and over EasyBuild slack) and I followed that which ultimately brings down to this custom easyblock. The .eb file for PETSc can be found here where a detailed discussion with @zao took place.

[boegel]

@tanmoy1989 Thanks for clarifying, I'll see if I can get this merged ASAP to be included in the upcoming EasyBuild v4.9.3...

There's a small code style issue to fix:

./easybuild/easyblocks/p/petsc.py:269:23: E111 indentation is not a multiple of 4

We can deal with this ourselves though if that's easier for you

[tanmoy1989]

@boegel Many thanks! I would really appreciate if the indentation part could be kindly taken care of.

[boegel]

lgtm

[boegel]

@boegelbot please test @ generoso
EB_ARGS="PETSc-3.19.2-foss-2022b.eb PETSc-3.20.3-foss-2023a.eb --installpath /tmp/$USER/pr3391"

[boegelbot]

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

PR test command 'EB_PR=3391 EB_ARGS="PETSc-3.19.2-foss-2022b.eb PETSc-3.20.3-foss-2023a.eb --installpath /tmp/$USER/pr3391" EB_CONTAINER= EB_REPO=easybuild-easyblocks /opt/software/slurm/bin/sbatch --job-name test_PR_3391 --ntasks=4 ~/boegelbot/eb_from_pr_upload_generoso.sh' executed!

• exit code: 0
• output:

Submitted batch job 14185

Test results coming soon (I hope)...

- notification for comment with ID 2329225545 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

Overview of tested easyconfigs (in order)

• SUCCESS PETSc-3.19.2-foss-2022b.eb
• SUCCESS PETSc-3.20.3-foss-2023a.eb

Build succeeded for 2 out of 2 (2 easyconfigs in total)
cns3 - Linux Rocky Linux 8.9, x86_64, Intel(R) Xeon(R) CPU E5-2667 v3 @ 3.20GHz (haswell), Python 3.6.8
See https://gist.github.com/boegelbot/20ed88bcf6dfc9f9ef015172e3380c8f for a full test report.