Again, this issue is still relevant. Updating did again break it.

Show audit status?

Here's the current hovercard:

After D20877, we'll forcefully override this option when generating diffs internally.

I deployed the --sshd-key %k stuff to secure and it looks like that shaved ~1,000-2000ms off the total cost. The ssh-auth cost has dropped to about 200ms:

I saw things hang during deploy and the OpenSSH bug should have been fixed years ago.

I configured ControlMaster on secure.

Under the --merge workflow (which does a --no-ff merge and always generates a merge commit; as opposed to the --squash strategy, which does a rebase-like squash merge) the git p4 submit --commit ... fails:

I'm passing these additional flags to git p4 submit:

If you haven't logged in to Perforce recently (I think?), the sync can fail like this:

Agreed that supporting YubiKey OTP is pointless - it's impractical and basically a dead legacy feature at this point. WebAuthn has emerged as the de-facto standard for hardware tokens.

I am running through GPG for the first time:

This makes a lot of sense, thanks or the time to explain.

I'm going to merge this into T11535 since it's basically the same issue.

This hasn't come up again, so it looks like it probably isn't a feature that would see a lot of general use.

Why not just implement the basic path for arc land and fail with an error if the user has some esoteric setup?

It's possible I'm misreading/misunderstanding this, so maybe the right algorithm is actually to select the "p4" remote ahead of the "origin" remote, i.e. swap steps (2) and (3).

the config with the dépôt path to “fetch” latest changelists from is saved inside .git directory

You don’t need to pass any arguments to git p4 sync: the config with the dépôt path to “fetch” latest changelists from is saved inside .git directory. Did you follow the exact instructions I provided?

I got Git/Perforce working after a large amount of bashing my head into errors like these:

One change we can make here is to explicitly specify -c diff.suppressBlankEmpty false in the arc diff command, as we currently use --src-prefix and --dst-prefix to override diff.mnemonicprefix. This is likely desirable, but it would be good to also parse these diffs properly if they arrive through --raw or copy/pasting, etc.