consistently enable usempi toolchain option in OSU-Micro-Benchmarks easyconfigs

[ocaisa]

No description provided.

[branfosj]

Looking at the OSU easyconfigs, we have two separate setups:

1. Using configopts = 'CC="$MPICC" CXX="$MPICC"'
2. Using toolchainopts = {'usempi': True}

If we, instead, switch them all to use method 2 then this automatically sets CC and CXX as required?

[ocaisa]

I'm on board, I also noticed that when creating #15018
I'll update this and there as well.

[ocaisa]

@boegelbot please test @ generoso

[boegelbot]

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

PR test command 'EB_PR=15039 EB_ARGS= /opt/software/slurm/bin/sbatch --job-name test_PR_15039 --ntasks=4 ~/boegelbot/eb_from_pr_upload_generoso.sh' executed!

• exit code: 0
• output:

Submitted batch job 8178

Test results coming soon (I hope)...

- notification for comment with ID 1048858355 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 16 out of 16 (16 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/0f6d5225944ca69fcb9cecec8aa38f99 for a full test report.

[smoors]

lgtm

[smoors]

Test report by @smoors
FAILED
Build succeeded for 21 out of 24 (16 easyconfigs in total)
node250.hydra.os - Linux CentOS Linux 7.9.2009, x86_64, Intel(R) Xeon(R) CPU E5-2650 v4 @ 2.20GHz (broadwell), 1 x NVIDIA Tesla P100-PCIE-16GB, 470.82.01, Python 2.7.5
See https://gist.github.com/5ddc192a2705065ab689a7b8931bad99 for a full test report.

[smoors]

the failures are because our cluster does not currently support the iimpic toolchain.

[smoors]

Going in, thanks @ocaisa!