Switch rtems target to panic unwind

[thesummer]

Switch the RTEMS target to panic_unwind.

Relates to rust-lang/backtrace-rs#682

r? @workingjubilee

rustbot has assigned @workingjubilee.
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

These commits modify compiler targets.
(See the Target Tier Policy.)

☔ The latest upstream changes (presumably #133533) made this pull request unmergeable. Please resolve the merge conflicts.

@workingjubilee I hope the release stress from the last release is over. Could you please have a look at this PR?

Integrating it even without the update to the backtrace submodule would already help me to experiment with some unit test setups.

The changes here look reasonable to me but the backtrace changes need to merge before this right?

@tgross35 I think it is save to merge this one here first as well. Currently the port also requires that backtrace is updated locally, so merging this won't change that.

The backtrace change is so small that maybe we should just try to get that in around the same time. @workingjubilee is backtrace stuck from CI failures or could rust-lang/backtrace-rs#682 get into a release soonish?

@tgross35 Updating the backtrace submodule was blocked by some performance metrics (see this PR #130417).

I am not sure if the CI failures of rust-lang/backtrace-rs#682 are related to that.

confused. If this was all that was needed, why wasn't it just done from the get-go...?

Most of the CI failures in backtrace have nothing to do with the ruzstd merge and can simply be fixed by fixing them.

CI will be fixed enough when rust-lang/backtrace-rs#687 rust-lang/backtrace-rs#679 merges

@workingjubilee Sorry, for the confusion. I am new to rust compiler development and was working myself upwards from a bare-metal/panic abort system to RTEMS support.

Happy New Year btw.

As of #135053 we now have a backtrace version that includes the libunwind change for the RTEMS target and builds without zstd support, thus without immediately requiring us to bring in ruzstd, thus without immediately addressing any performance questions.

This is a tier 3 target so I am entitled to ship a change recklessly to it without heed to whether it still builds¹. MUAHAHAHA!

@bors r+ rollup

Footnotes

1. ...also it's at the maintainer's request. ↩

📌 Commit 3f94047 has been approved by workingjubilee

It is now in the queue for this repository.

@workingjubilee thanks a lot. Also for your efforts to punch through the CI problems in the backtrace repo.