fix OpenBLAS 0.3.15 patch to correctly set the CPU core type for Tiger Lake

[micaeljtoliveira]

(created using eb --new-pr)

[akesandgren]

LGTM, patch matches changes from 0.3.15 to 0.3.18

[akesandgren]

@boegelbot Please test @ generoso

[boegelbot]

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

PR test command 'EB_PR=15845 EB_ARGS= /opt/software/slurm/bin/sbatch --job-name test_PR_15845 --ntasks=4 ~/boegelbot/eb_from_pr_upload_generoso.sh' executed!

• exit code: 0
• output:

Submitted batch job 9092

Test results coming soon (I hope)...

- notification for comment with ID 1232778622 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
FAILED
Build succeeded for 1 out of 2 (2 easyconfigs in total)
cns2 - 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/89d26aff7ca2dcc5596ba64568a108c0 for a full test report.

[akesandgren]

The updated patch fails to apply to 0.3.12.
Please provide a working, separate, one for 0.3.12

[micaeljtoliveira]

@akesandgren Sorry for the long delay in addressing the problem with 0.3.12. I tried to fix the patch, but it seems that for it to actually work one would need to backport a lot more changes from later versions of OpenBLAS. So instead I just kept the current patch as it is for that version. Hope that's okay.

[akesandgren]

Test report by @akesandgren
SUCCESS
Build succeeded for 2 out of 2 (2 easyconfigs in total)
b-an02.hpc2n.umu.se - Linux Ubuntu 20.04, x86_64, Intel(R) Xeon(R) CPU E5-2690 v4 @ 2.60GHz, Python 3.8.10
See https://gist.github.com/482c201aa6e3227e04338df8ec8a4061 for a full test report.

[akesandgren]

@micaeljtoliveira Can you take a look at this, https://gist.github.com/sassy-crick/cc750a24bd1404c1c9a6f8a2c09c10aa ?

Built on a "11th Gen Intel(R) Core(TM) i7-1185G7 @ 3.00GHz"

[micaeljtoliveira]

@akesandgren Not quite sure what to do about this. The issue with 0.3.12 is to be expected. As mentioned above, to fix it one would need to backport more fixes made in later versions of OpenBLAS. I tried to do that, but without much success: compilation what still broken even after including several more changes to the patch. Regarding the Lapack timing, that's a strange error, as I can't see any changes in the PR that could do that.

[akesandgren]

Just wanted you to verify that it's what you'd expect for the 0.3.12 build.
The timing error can be ignored, we don't use that part as far as I know.

[akesandgren]

LGTM

[akesandgren]

Going in, thanks @micaeljtoliveira!