revert for benchmark

[bvanjoi]

This PR reverts #118311 and #118188, with the intent to assess the performance impact brought about by #118311

[rustbot]

r? @WaffleLapkin

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

[rustbot]

Some changes occurred to MIR optimizations

cc @rust-lang/wg-mir-opt

[lqd]

@bors try @rust-timer queue

[bors]

⌛ Trying commit 6caa433 with merge b28572e...

[bors]

☀️ Try build successful - checks-actions
Build commit: b28572e (b28572ef1c60b5dde65c77fb50c1a726ecec8491)

[pnkfelix]

(As I mentioned on #118311 (comment), I don't think that measuring the effect of reverting both PR #118311 and PR #118188 will not tell us much about the performance impact of reverting PR #118311, because PR #118188 itself introduced a number of performance improvements, and so reverting that will presumably induce regressions that could well mask the effect we are trying to determine.)

[rust-timer]

Finished benchmarking commit (b28572e): comparison URL.

Overall result: ❌✅ regressions and improvements - ACTION NEEDED

Benchmarking this pull request likely means that it is perf-sensitive, so we're automatically marking it as not fit for rolling up. While you can manually mark this PR as fit for rollup, we strongly recommend not doing so since this PR may lead to changes in compiler perf.

Next Steps: If you can justify the regressions found in this try perf run, please indicate this with @rustbot label: +perf-regression-triaged along with sufficient written justification. If you cannot justify the regressions please fix the regressions and do another perf run. If the next run shows neutral or positive results, the label will be automatically removed.

@bors rollup=never
@rustbot label: -S-waiting-on-perf +perf-regression

Instruction count

This is a highly reliable metric that was used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	0.3%	[0.1%, 0.4%]	22
Regressions ❌ (secondary)	0.6%	[0.1%, 1.1%]	33
Improvements ✅ (primary)	-0.3%	[-0.4%, -0.3%]	7
Improvements ✅ (secondary)	-0.4%	[-0.5%, -0.3%]	6
All ❌✅ (primary)	0.1%	[-0.4%, 0.4%]	29

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)	2.2%	[2.2%, 2.2%]	1
Regressions ❌ (secondary)	1.2%	[1.2%, 1.2%]	1
Improvements ✅ (primary)	-5.2%	[-7.6%, -0.8%]	3
Improvements ✅ (secondary)	-3.0%	[-4.6%, -1.6%]	13
All ❌✅ (primary)	-3.4%	[-7.6%, 2.2%]	4

Cycles

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
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-1.3%	[-1.3%, -1.3%]	1
Improvements ✅ (secondary)	-2.5%	[-2.7%, -2.2%]	5
All ❌✅ (primary)	-1.3%	[-1.3%, -1.3%]	1

Binary size

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

Bootstrap: 674.352s -> 672.281s (-0.31%)
Artifact size: 313.40 MiB -> 313.37 MiB (-0.01%)

[bvanjoi]

We can draw two conclusions from this PR and #118411:

• As indicated in the above comments, the combination of merge DefKind::Coroutine into Defkind::Closure #118311 and resolve: Feed the def_kind query immediately on DefId creation #118188 resulted in an optimization
• The comment found at aligns the behavior to that prior to #118311 #118411 (comment) underlines that there's no equivalence between ExprKind::Closure(Closure { movability: Some(_), .. }) => DefKind::Coroutine and tcx.coroutine_kind(def_id).is_some(). Therefore, we can obtain some performance optimization through certain modification as demonstrated in aligns the behavior to that prior to #118311 #118411.

[bvanjoi]

So, should we consider merging #118411?

cc @petrochenkov @pnkfelix

[petrochenkov]

I wasn't assigned to this PR, so I forgot about it entirely.
This revert served its purpose, so it can be closed now.
#118411 should be reviewed and merged in some form.