Put inline functions in only one CGU

[Zoxc]

Revival of #70459 for the purpose of up-to-date performance tests.

cc @nnethercote

[rustbot]

r? @wesleywiser

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

[nnethercote]

@bors try @rust-timer queue

[bors]

⌛ Trying commit 0424a16 with merge b521c7bd21ae87a8124b52029719e9598c747b83...

[bors]

☀️ Try build successful - checks-actions
Build commit: b521c7bd21ae87a8124b52029719e9598c747b83 (b521c7bd21ae87a8124b52029719e9598c747b83)

[rust-timer]

Finished benchmarking commit (b521c7bd21ae87a8124b52029719e9598c747b83): 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)	19.3%	[0.3%, 1248.3%]	152
Regressions ❌ (secondary)	3.5%	[0.5%, 35.9%]	157
Improvements ✅ (primary)	-6.8%	[-42.8%, -0.2%]	32
Improvements ✅ (secondary)	-2.8%	[-5.9%, -0.4%]	28
All ❌✅ (primary)	14.7%	[-42.8%, 1248.3%]	184

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)	5.9%	[0.6%, 52.7%]	81
Regressions ❌ (secondary)	2.9%	[0.7%, 5.8%]	109
Improvements ✅ (primary)	-10.4%	[-20.8%, -2.8%]	9
Improvements ✅ (secondary)	-4.0%	[-4.0%, -4.0%]	1
All ❌✅ (primary)	4.2%	[-20.8%, 52.7%]	90

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)	23.2%	[1.0%, 1144.8%]	135
Regressions ❌ (secondary)	5.6%	[1.3%, 34.0%]	121
Improvements ✅ (primary)	-10.1%	[-40.1%, -1.0%]	18
Improvements ✅ (secondary)	-7.4%	[-14.2%, -3.6%]	16
All ❌✅ (primary)	19.3%	[-40.1%, 1144.8%]	153

Binary size

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)	11.0%	[0.0%, 60.1%]	114
Regressions ❌ (secondary)	4.5%	[0.0%, 35.2%]	101
Improvements ✅ (primary)	-3.4%	[-9.2%, -0.2%]	27
Improvements ✅ (secondary)	-0.3%	[-0.6%, -0.1%]	5
All ❌✅ (primary)	8.2%	[-9.2%, 60.1%]	141

Bootstrap: 653.837s -> 722.986s (10.58%)

[Zoxc]

I pushed another variation which keeps inline(always in multiple CGUs. That could use a perf run too.

[lqd]

@bors try @rust-timer queue

[bors]

⌛ Trying commit e2789cb with merge 57e40c96b520dd84471d5c2cb4878d0571bbcb6d...

[bors]

☀️ Try build successful - checks-actions
Build commit: 57e40c96b520dd84471d5c2cb4878d0571bbcb6d (57e40c96b520dd84471d5c2cb4878d0571bbcb6d)

[rust-timer]

Finished benchmarking commit (57e40c96b520dd84471d5c2cb4878d0571bbcb6d): 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)	10.3%	[0.3%, 548.4%]	201
Regressions ❌ (secondary)	3.1%	[0.4%, 32.6%]	172
Improvements ✅ (primary)	-12.7%	[-34.9%, -0.7%]	14
Improvements ✅ (secondary)	-3.0%	[-5.8%, -1.0%]	13
All ❌✅ (primary)	8.8%	[-34.9%, 548.4%]	215

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)	8.6%	[0.7%, 40.4%]	21
Regressions ❌ (secondary)	2.6%	[1.8%, 3.4%]	2
Improvements ✅ (primary)	-5.8%	[-13.9%, -2.2%]	10
Improvements ✅ (secondary)	-3.4%	[-5.8%, -2.1%]	9
All ❌✅ (primary)	4.0%	[-13.9%, 40.4%]	31

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)	15.5%	[1.1%, 537.7%]	147
Regressions ❌ (secondary)	4.4%	[1.0%, 29.7%]	133
Improvements ✅ (primary)	-17.6%	[-33.8%, -8.4%]	9
Improvements ✅ (secondary)	-2.5%	[-2.7%, -2.2%]	5
All ❌✅ (primary)	13.6%	[-33.8%, 537.7%]	156

Binary size

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)	15.9%	[0.1%, 33.1%]	57
Regressions ❌ (secondary)	15.0%	[0.1%, 30.7%]	20
Improvements ✅ (primary)	-1.3%	[-3.7%, -0.4%]	7
Improvements ✅ (secondary)	-0.4%	[-0.6%, -0.3%]	2
All ❌✅ (primary)	14.0%	[-3.7%, 33.1%]	64

Bootstrap: 654.49s -> 700.78s (7.07%)

[lqd]

@bors try @rust-timer queue

[bors]

⌛ Trying commit f62342a with merge 3d2052a0df0ea3414f2385c3befd14d093c00e10...

[bors]

☀️ Try build successful - checks-actions
Build commit: 3d2052a0df0ea3414f2385c3befd14d093c00e10 (3d2052a0df0ea3414f2385c3befd14d093c00e10)

[rust-timer]

Finished benchmarking commit (3d2052a0df0ea3414f2385c3befd14d093c00e10): 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)	12.1%	[0.3%, 549.2%]	166
Regressions ❌ (secondary)	3.0%	[0.5%, 32.6%]	146
Improvements ✅ (primary)	-10.0%	[-34.8%, -0.2%]	17
Improvements ✅ (secondary)	-2.6%	[-6.0%, -0.9%]	19
All ❌✅ (primary)	10.0%	[-34.8%, 549.2%]	183

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)	10.1%	[1.2%, 42.1%]	18
Regressions ❌ (secondary)	2.4%	[1.5%, 2.9%]	5
Improvements ✅ (primary)	-2.8%	[-14.1%, -0.4%]	32
Improvements ✅ (secondary)	-2.7%	[-6.7%, -0.9%]	43
All ❌✅ (primary)	1.8%	[-14.1%, 42.1%]	50

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)	17.3%	[1.0%, 543.5%]	126
Regressions ❌ (secondary)	4.8%	[0.9%, 30.0%]	95
Improvements ✅ (primary)	-15.8%	[-33.5%, -2.3%]	10
Improvements ✅ (secondary)	-6.4%	[-12.8%, -2.6%]	20
All ❌✅ (primary)	14.9%	[-33.5%, 543.5%]	136

Binary size

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)	12.6%	[0.1%, 33.3%]	72
Regressions ❌ (secondary)	3.6%	[0.1%, 30.6%]	91
Improvements ✅ (primary)	-1.2%	[-3.4%, -0.4%]	7
Improvements ✅ (secondary)	-0.5%	[-0.6%, -0.3%]	2
All ❌✅ (primary)	11.4%	[-3.4%, 33.3%]	79

Bootstrap: 650.027s -> 735.767s (13.19%)

[nnethercote]

Bootstrap and artifact size results are also bad, in addition to the compile-time results :(