Revert "Make PROC_MACRO_DERIVE_RESOLUTION_FALLBACK a hard error"

[pnkfelix]

This reverts commit 7d82cad aka PR #84022

I am doing this to buy us some time with respect to issue #106337 w.r.t. the 1.67 release.

[rustbot]

r? @cjgillot

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

[pnkfelix]

beta-nominating, checking with @wesleywiser or @Mark-Simulacrum to see about trying to get this in before the 1.67-beta is uplifted to stable.

@rustbot label: beta-nominated

[Mark-Simulacrum]

@bors r+ p=1

[bors]

📌 Commit 5fa1347 has been approved by Mark-Simulacrum

It is now in the queue for this repository.

[petrochenkov]

I don't think that we should revert this, or at least it should be re-landed on the next or next+1 release.
Whatever the solution for #64079 will be, it certainly won't look like what PROC_MACRO_DERIVE_RESOLUTION_FALLBACK does.

[bors]

⌛ Testing commit 5fa1347 with merge 85da15c...

[bors]

☀️ Test successful - checks-actions
Approved by: Mark-Simulacrum
Pushing 85da15c to master...

[rust-timer]

Finished benchmarking commit (85da15c): comparison URL.

Overall result: no relevant changes - no action needed

@rustbot label: -perf-regression

Instruction count

This benchmark run did not return any relevant results for this metric.

Max RSS (memory usage)

Results

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	0.7%	[0.7%, 0.7%]	1
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-2.4%	[-2.4%, -2.4%]	1
All ❌✅ (primary)	0.7%	[0.7%, 0.7%]	1

Cycles

This benchmark run did not return any relevant results for this metric.

[pnkfelix]

I don't think that we should revert this, or at least it should be re-landed on the next or next+1 release. Whatever the solution for #64079 will be, it certainly won't look like what PROC_MACRO_DERIVE_RESOLUTION_FALLBACK does.

Obviously I'm in favor of the revert, but I mostly agree with the sentiment of @petrochenkov 's statements. I don't think the lang team has spent very much time at all thinking about #64079, and that there are other points in the solution space that should be considered. I am thinking/hoping we will re-land the hard-error-update sometime in the next 12 to 18 weeks, which I suppose is a slightly longer schedule than what @petrochenkov had proposed.