add note about remove_duplicate_unreachable_blocks in source comments

[miguelraz]

gotta write down negative results, for science!

@Nilstrieb

#110524 (comment)

Please signal for rollup.

[rustbot]

r? @wesleywiser

(rustbot has picked a reviewer for you, use r? to override)

[rustbot]

Some changes occurred to MIR optimizations

cc @rust-lang/wg-mir-opt

[compiler-errors]

I don't think we usually comment about failed perf attempts, since I would assume very hot code (e.g. trait selection) would be riddled with these comments -- though perhaps someone like @nnethercote knows about best practice here.

[compiler-errors]

Even if we do want to leave a note about perf sensitivity here, it should probably left as a comment on the field instead?

[nnethercote]

I agree.

[miguelraz]

That sounds reasonable - I'm open to whatever maintainers feel is best.

[nnethercote]

Comments about perf are good when the code has unexpected things in it. E.g. often code that is optimized is written in a way that is different to the obvious way, and a comment about that is helpful: "this code is hot, hence...".

This PR's comment is effectively saying "this code is written in the obvious way, not an optimized way". And @compiler-errors is right that we don't normally write comments like that. If there were many of them I think they'd get annoying. If there was a reason to think that this code would be especially hot, then maybe, but it doesn't strike me that way.

[Noratrieb]

I just thought that it'd be quite likely that someone in the future might see this and see an opportunity to optimize. Only to waste their time. But if you think it's not worth it, fair.

[apiraino]

Switching to waiting on author, IIUC indications were left for some changes? thanks

@rustbot author

[compiler-errors]

I'm gonna close this since it doesn't seem like adding this note is useful. Thanks for the PR!