zdb: dump ZAP_FLAG_UINT64_KEY ZAPs properly

[robn]

[Sponsors: Klara, Inc., Wasabi Technology, Inc.]

Motivation and Context

I was using zdb to look around my local pool, and it kept crashing on this very strange ZAP that made no sense to me. Turns out it was the BRT ZAP, which has binary keys rather than text keys, and zdb had no idea what to make of that.

Description

Detect ZAP_FLAG_UINT64_KEY ZAPs in dump_zap(). If seen, synthesize a text "key" from the first uint64, and show the value using the same "array of integers" code used elsewhere.

This is not especially useful for keys longer than one uint64 (eg DDT keys, which are 40 bytes), but that's sorta difficult to make a generic display for and zdb already has structured formatters for dedup and block reference tables, so not super useful anyway. Really, I just wanted something useful and not-crashy.

How Has This Been Tested?

Hand tested on Linux and FreeBSD.

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Performance enhancement (non-breaking change which improves efficiency)
• Code cleanup (non-breaking change which makes code smaller or more readable)
• Breaking change (fix or feature that would cause existing functionality to change)
• Library ABI change (libzfs, libzfs_core, libnvpair, libuutil and libzfsbootenv)
• Documentation (a change to man pages or other documentation)

Checklist:

• My code follows the OpenZFS code style requirements.
• I have updated the documentation accordingly.
• I have read the contributing document.
• I have added tests to cover my changes.
• I have run the ZFS Test Suite with this change applied.
• All commit messages are properly formatted and contain Signed-off-by.

[amotin]

I saw this also, but was thinking towards extending ZAP API to allow key length reporting, after which we could to it properly and universally, but also haven't got to it yet.

[robn]

Yeah, I'd like to do that. Seems like the _uint64 variants might not even be necessary if it can automatically decide what to do from the the flag. That's be really nice for zap_lookup inside a cursor especially.

Alas, not really on my radar at the moment, but I've chucked it on the list.