powerpc64-ibm-aix: fix cfg(target_abi) value

[RalfJung]

Our powerpc64-ibm-aix target currently sets cfg_abi to "vec-extabi", which means that programs compiled for this target think they use that ABI. Even our inline asm logic trusts this field. But that's a lie, we're actually using the default ABI since we are never setting EnableAIXExtendedAltivecABI on the LLVM side. We should fix that discrepancy.

Between changing how we generate the code, and changing the label we put into cfg_abi, the latter is the less risky change. So let's do that.

This has been tried before in #153830. We then decided to wait a bit while the target maintainers investigate whether they want to change the ABI or not. I think we have waited long enough. The last comment from them that I found is this one which says "I'd conclude the extend vector ABI should be disabled for now"; there have been further questions but no further communication. We can always still change the ABI in the future, but for now let's fix the obvious bug where the ABI we report in cfg_abi does not match the actual ABI we are compiling for.

Cc @Gelbpunkt @daltenty @gilamn5tr @amy-kwan @taiki-e

[rustbot]

r? @TaKO8Ki

rustbot has assigned @TaKO8Ki.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

Why was this reviewer chosen?

The reviewer was selected based on:

• Owners of files modified in this PR: compiler
• compiler expanded to 75 candidates
• Random selection from 20 candidates

[RalfJung]

Given this, IMO we should just always error here, and never allow these registers to be used on AIX. This would remove the last cfg_abi check in the asm logic, which is good -- hopefully we'll never accidentally add any new such checks in the future. :)
@taiki-e @Amanieu would that be okay?

View changes since the review

[taiki-e]

Yeah, I think it's fine to always emit error here. If you decide to use vec-default as the default, enabling vec-extabi will likely require adding a target modifier, so there's no need to refer to the cfg_abi.

[tgross35]

Cc @xingxue-ibm who actively works on libc for AIX. xingxue, consider adding yourself to https://github.com/rust-lang/rust/blob/31c76a10a742d10a749ab4e19cec324864314b70/src/doc/rustc/src/platform-support/aix.md so people other than me know that :)

[taiki-e]

My impression is that the target maintainer at the time wanted to enable the vec-extabi but failed to actually do so because they misunderstood the role of the abi field. (#153830 (comment))

If the current target maintainer wants to leave it disabled, I think this should be fine.

[RalfJung]

Cc @ecnelises