Revert "Auto merge of #127537 - veluca93:struct_tf, r=BoxyUwU"

[Kobzol]

This reverts #127537 (commit acb4e8b), reversing changes made to 100fde5.

Opening to see if this can help resolve the recent perf. results instability.

[Kobzol]

@bors try @rust-timer queue

[rustbot]

r? @lcnr

rustbot has assigned @lcnr.
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

[bors]

⌛ Trying commit 47e6b5d with merge dfc9c99...

[bors]

☀️ Try build successful - checks-actions
Build commit: dfc9c99 (dfc9c9968aad89eaa962ab1d90ce46db9b683cb9)

[rust-timer]

Finished benchmarking commit (dfc9c99): 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)	1.4%	[0.4%, 3.0%]	3
Regressions ❌ (secondary)	0.5%	[0.5%, 0.5%]	1
Improvements ✅ (primary)	-0.5%	[-1.0%, -0.2%]	63
Improvements ✅ (secondary)	-0.7%	[-2.2%, -0.3%]	18
All ❌✅ (primary)	-0.4%	[-1.0%, 3.0%]	66

Max RSS (memory usage)

Results (primary -0.6%, secondary -0.8%)

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)	0.8%	[0.4%, 1.8%]	19
Regressions ❌ (secondary)	0.6%	[0.4%, 0.9%]	13
Improvements ✅ (primary)	-0.9%	[-3.8%, -0.4%]	87
Improvements ✅ (secondary)	-0.9%	[-3.6%, -0.4%]	124
All ❌✅ (primary)	-0.6%	[-3.8%, 1.8%]	106

Cycles

Results (primary -0.5%, secondary -0.1%)

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)	1.2%	[0.5%, 2.8%]	13
Regressions ❌ (secondary)	1.2%	[0.4%, 3.4%]	47
Improvements ✅ (primary)	-0.8%	[-2.9%, -0.4%]	73
Improvements ✅ (secondary)	-1.1%	[-3.6%, -0.4%]	67
All ❌✅ (primary)	-0.5%	[-2.9%, 2.8%]	86

Binary size

Results (primary -0.7%, secondary -1.4%)

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)	-0.7%	[-1.6%, -0.0%]	81
Improvements ✅ (secondary)	-1.4%	[-2.3%, -0.6%]	24
All ❌✅ (primary)	-0.7%	[-1.6%, -0.0%]	81

Bootstrap: 790.965s -> 750.306s (-5.14%)
Artifact size: 338.44 MiB -> 338.34 MiB (-0.03%)

[Kobzol]

We could do a bunch of perf. runs, but the desired result (removal of instability) would be a bit hard to detect (we'd need to check if the results don't change much, but even the baseline results after reverting are quite large).

So I'd suggest reverting, then checking if it helped and then try to land again with some improvements.

[Kobzol]

r? lqd

@bors rollup=never

[lqd]

Ah I had also added this to #129851 — but do we want to try just this revert and see what happens for a few merges, before trying with #129714 if needed?

[lqd]

This one will at least fix a few regressions including bootstrap, while the collector behavior on the other was pretty weird and we may need to try both anyways.

@bors r+

[bors]

📌 Commit 47e6b5d has been approved by lqd

It is now in the queue for this repository.

[Kobzol]

I checked the collector logs of #129714 and everything seemed to have happened correctly. So I think that it might just have been some database fluke. Let's try reverting just #127537 first, and see what happens. The other PR didn't seem that suspicious.

[Kobzol]

@bors p=10

All current perf. results seem really weird and misleading, which is confusing developers on merged PRs. We hope that by reverting this PR, it should be fixed, so ideally the revert should be merged ASAP.

[bors]

⌛ Testing commit 47e6b5d with merge 94885bc...

[bors]

☀️ Test successful - checks-actions
Approved by: lqd
Pushing 94885bc to master...

[rust-timer]

Finished benchmarking commit (94885bc): comparison URL.

Overall result: ✅ improvements - no action needed

@rustbot label: -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
Regressions ❌ (secondary)	0.5%	[0.3%, 0.6%]	2
Improvements ✅ (primary)	-0.5%	[-1.1%, -0.2%]	61
Improvements ✅ (secondary)	-0.7%	[-2.2%, -0.3%]	26
All ❌✅ (primary)	-0.5%	[-1.1%, -0.2%]	61

Max RSS (memory usage)

Results (primary -0.6%, secondary -0.7%)

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)	0.8%	[0.5%, 1.9%]	14
Regressions ❌ (secondary)	0.7%	[0.4%, 1.2%]	16
Improvements ✅ (primary)	-0.8%	[-2.3%, -0.4%]	101
Improvements ✅ (secondary)	-1.0%	[-3.5%, -0.4%]	82
All ❌✅ (primary)	-0.6%	[-2.3%, 1.9%]	115

Cycles

Results (primary -0.6%, secondary -0.1%)

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)	1.9%	[0.4%, 6.2%]	6
Regressions ❌ (secondary)	1.1%	[0.4%, 4.7%]	43
Improvements ✅ (primary)	-0.8%	[-3.1%, -0.4%]	63
Improvements ✅ (secondary)	-0.9%	[-2.4%, -0.4%]	58
All ❌✅ (primary)	-0.6%	[-3.1%, 6.2%]	69

Binary size

Results (primary -0.7%, secondary -1.4%)

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)	-0.7%	[-1.6%, -0.0%]	82
Improvements ✅ (secondary)	-1.4%	[-2.3%, -0.5%]	25
All ❌✅ (primary)	-0.7%	[-1.6%, -0.0%]	82

Bootstrap: missing data
Artifact size: 338.38 MiB -> 338.33 MiB (-0.02%)