[clad] Bump clad to version v1.9

[vgvassilev]

No description provided.

[vgvassilev]

@guitargeek, could you do some performance benchmarks to see if we have not regressed?

[github-actions]

Test Results

    18 files      18 suites   3d 19h 49m 39s ⏱️
 2 719 tests  2 719 ✅ 0 💤 0 ❌
47 253 runs  47 253 ✅ 0 💤 0 ❌

Results for commit aa18e0b.

♻️ This comment has been updated with latest results.

[guitargeek]

I have checked, and there are no changes in jitting time and runtime for ATLAS and CMS Higgs combination fits.

[dpiparo]

@guitargeek can this PR then be approved?

[guitargeek]

Well, not as it is. We can't use Clad master as a dependency. There first needs to be a new release and then this PR needs to be updated to use the tag of that release.

[vgvassilev]

Also there is a regression on windows that I am bisecting.

[dpiparo]

Thanks, good that this is being followed up.

[vgvassilev]

@guitargeek ping.

[guitargeek]

Hi @vgvassilev, thanks for the ping, I forgot to report on my findings here. Clad 1.9 works for both the ATLAS and CMS Higgs combination examples that we have and presented at ICHEP 2024.

[vgvassilev]

Hi @vgvassilev, thanks for the ping, I forgot to report on my findings here. Clad 1.9 works for both the ATLAS and CMS Higgs combination examples that we have and presented at ICHEP 2024.

Great to hear. Let's move forward then.

[guitargeek]

Thanks for updating Clad in ROOT to v1.9!