Don't require `alloca`s for consuming simple enums #138582

scottmcm · 2025-03-16T23:53:51Z

Well, 4 months later I'm finally back to this.

For example, if you pass an Option<u32> to a function, don't immediately write it to an alloca then read it again.

rustbot · 2025-03-16T23:53:55Z

rustbot has assigned @saethlin.
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

scottmcm · 2025-03-17T00:13:27Z

@bors try @rust-timer queue

bors · 2025-03-17T00:14:38Z

⌛ Trying commit 019c41a with merge a5a9937...

Don't require `alloca`s for creating or consuming simple enums This lets, for example, `Option<u32>` stay a `ScalarPair` the whole time, never needing to get written to stack.

bors · 2025-03-17T02:20:35Z

☀️ Try build successful - checks-actions
Build commit: a5a9937 (a5a99374d77fd499d01fb91c55bcb4ef8dad712a)

rust-timer · 2025-03-17T03:39:46Z

Finished benchmarking commit (a5a9937): comparison URL.

Overall result: ❌✅ regressions and improvements - please read the text below

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 the most reliable metric that we have; it was used to determine the overall result at the top of this comment. However, even this metric can sometimes exhibit noise.

	mean	range	count
Regressions ❌ (primary)	0.4%	[0.2%, 1.6%]	187
Regressions ❌ (secondary)	0.6%	[0.2%, 6.7%]	129
Improvements ✅ (primary)	-0.5%	[-1.0%, -0.3%]	9
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	0.3%	[-1.0%, 1.6%]	196

Max RSS (memory usage)

Results (primary -2.0%, secondary 1.9%)

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)	1.9%	[1.9%, 1.9%]	1
Improvements ✅ (primary)	-2.0%	[-2.0%, -2.0%]	1
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-2.0%	[-2.0%, -2.0%]	1

Cycles

Results (primary 1.1%, secondary 2.1%)

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)	1.1%	[0.8%, 1.5%]	11
Regressions ❌ (secondary)	2.5%	[1.1%, 3.5%]	10
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-2.7%	[-2.7%, -2.7%]	1
All ❌✅ (primary)	1.1%	[0.8%, 1.5%]	11

Binary size

Results (primary -0.2%, secondary -0.0%)

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)	0.0%	[0.0%, 0.0%]	20
Improvements ✅ (primary)	-0.2%	[-0.6%, -0.0%]	65
Improvements ✅ (secondary)	-0.1%	[-0.3%, -0.0%]	22
All ❌✅ (primary)	-0.2%	[-0.6%, -0.0%]	65

Bootstrap: 775.741s -> 776.411s (0.09%)
Artifact size: 365.10 MiB -> 364.26 MiB (-0.23%)

scottmcm · 2025-03-17T07:47:45Z

Wow, those numbers are horrible :/

(Well, other than the size improvements)

scottmcm · 2025-03-17T17:33:48Z

@bors try @rust-timer queue

Don't require `alloca`s for consuming simple enums For example, if you pass an `Option<u32>` to a function, don't immediately write it to an `alloca` then read it again.

bors · 2025-03-17T17:35:00Z

⌛ Trying commit 3477a13 with merge 132433d...

bors · 2025-03-17T19:44:20Z

☀️ Try build successful - checks-actions
Build commit: 132433d (132433d7556f612643abac6a48c4b80911f5e9f4)

rust-timer · 2025-03-17T21:40:10Z

Finished benchmarking commit (132433d): comparison URL.

Overall result: ❌✅ regressions and improvements - please read the text below

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 the most reliable metric that we have; it was used to determine the overall result at the top of this comment. However, even this metric can sometimes exhibit noise.

	mean	range	count
Regressions ❌ (primary)	0.2%	[0.1%, 0.5%]	9
Regressions ❌ (secondary)	0.5%	[0.2%, 1.4%]	45
Improvements ✅ (primary)	-0.4%	[-0.8%, -0.2%]	14
Improvements ✅ (secondary)	-0.3%	[-0.3%, -0.3%]	3
All ❌✅ (primary)	-0.1%	[-0.8%, 0.5%]	23

Max RSS (memory usage)

Results (primary -2.9%)

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)	-2.9%	[-2.9%, -2.9%]	1
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-2.9%	[-2.9%, -2.9%]	1

Cycles

Results (secondary 3.5%)

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)	3.5%	[2.4%, 5.4%]	9
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-	-	0

Binary size

Results (primary -0.1%, secondary 0.0%)

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)	0.0%	[0.0%, 0.0%]	4
Regressions ❌ (secondary)	0.0%	[0.0%, 0.0%]	38
Improvements ✅ (primary)	-0.1%	[-0.4%, -0.0%]	54
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-0.1%	[-0.4%, 0.0%]	58

Bootstrap: 774.895s -> 775.425s (0.07%)
Artifact size: 365.10 MiB -> 365.16 MiB (0.02%)

scottmcm · 2025-03-18T00:59:11Z

Wow, that one's almost entirely green in primary and almost entirely red in secondary O_o

scottmcm · 2025-07-06T05:46:53Z

tests/codegen/enum/enum-match.rs

-// CHECK-NEXT: %[[IS_B:.+]] = icmp eq i8 %0, 2
-// CHECK-NEXT: %[[TRUNC:.+]] = and i8 %0, 1
-// CHECK-NEXT: %[[R:.+]] = select i1 %[[IS_B]], i8 13, i8 %[[TRUNC]]
+// CHECK-NEXT: %[[IS_B:.+]] = icmp eq i8 %e, 2
+// CHECK-NEXT: %[[R:.+]] = select i1 %[[IS_B]], i8 13, i8 %e


AFAIK the reason this and disappeared is because by staying in SSA the whole time it avoids the range information on the load getting dropped (llvm/llvm-project#134403) that @dtcxzyw previously identified in llvm/llvm-project#134093 (comment)

scottmcm · 2025-07-06T07:09:28Z

@bors try @rust-timer queue

Don't require `alloca`s for consuming simple enums Well, 4 months later I'm finally back to this. Will stay as Draft until #143502 lands, since this builds on that. For example, if you pass an `Option<u32>` to a function, don't immediately write it to an `alloca` then read it again.

bors · 2025-07-06T07:10:39Z

⌛ Trying commit ae2daed with merge 81bafbf...

scottmcm · 2025-07-06T17:34:37Z

@bors try @rust-timer queue

Don't require `alloca`s for consuming simple enums Well, 4 months later I'm finally back to this. For example, if you pass an `Option<u32>` to a function, don't immediately write it to an `alloca` then read it again.

bors · 2025-07-06T17:35:48Z

⌛ Trying commit f452f6d with merge 7ec415a...

bors · 2025-07-06T19:48:19Z

☀️ Try build successful - checks-actions
Build commit: 7ec415a (7ec415a2dd00f671ff064d2ad88c2a73f6f6a4a5)

rust-timer · 2025-07-07T01:22:33Z

Finished benchmarking commit (7ec415a): comparison URL.

Overall result: ❌✅ regressions and improvements - please read the text below

Benchmarking this pull request means it may be perf-sensitive – we'll automatically label it not fit for rolling up. You can override this, but we strongly advise not to, due to possible changes in compiler perf.

Next Steps: If you can justify the regressions found in this try perf run, please do so in sufficient writing along with @rustbot label: +perf-regression-triaged. If not, please fix the regressions and do another perf run. If its results are neutral or positive, the label will be automatically removed.

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

Instruction count

Our most reliable metric. Used to determine the overall result above. However, even this metric can be noisy.

	mean	range	count
Regressions ❌ (primary)	0.4%	[0.1%, 0.9%]	172
Regressions ❌ (secondary)	0.5%	[0.0%, 6.8%]	177
Improvements ✅ (primary)	-0.5%	[-0.8%, -0.4%]	11
Improvements ✅ (secondary)	-0.2%	[-0.8%, -0.0%]	9
All ❌✅ (primary)	0.4%	[-0.8%, 0.9%]	183

Max RSS (memory usage)

Results (secondary 2.8%)

A less reliable metric. May be of interest, but not used to determine the overall result above.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	4.3%	[1.6%, 5.7%]	9
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-1.8%	[-2.4%, -0.8%]	3
All ❌✅ (primary)	-	-	0

Cycles

Results (primary 2.8%, secondary 0.6%)

A less reliable metric. May be of interest, but not used to determine the overall result above.

	mean	range	count
Regressions ❌ (primary)	2.8%	[2.4%, 3.3%]	2
Regressions ❌ (secondary)	4.9%	[0.8%, 8.5%]	5
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-6.6%	[-11.7%, -1.3%]	3
All ❌✅ (primary)	2.8%	[2.4%, 3.3%]	2

Binary size

Results (primary -0.1%, secondary -0.0%)

A less reliable metric. May be of interest, but not used to determine the overall result above.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-0.1%	[-0.3%, -0.0%]	66
Improvements ✅ (secondary)	-0.0%	[-0.2%, -0.0%]	28
All ❌✅ (primary)	-0.1%	[-0.3%, -0.0%]	66

Bootstrap: 461.042s -> 461.941s (0.19%)
Artifact size: 372.12 MiB -> 370.56 MiB (-0.42%)

scottmcm · 2025-07-07T02:03:20Z

@bors try @rust-timer queue

rust-timer · 2025-07-07T02:03:22Z

Awaiting bors try build completion.

@rustbot label: +S-waiting-on-perf

bors · 2025-07-07T02:04:30Z

⌛ Trying commit bc074f1 with merge abd7739...

Don't require `alloca`s for consuming simple enums Well, 4 months later I'm finally back to this. For example, if you pass an `Option<u32>` to a function, don't immediately write it to an `alloca` then read it again.

bors · 2025-07-07T04:09:52Z

☀️ Try build successful - checks-actions
Build commit: abd7739 (abd773984d3b89a23c04f8cf58c43e835b4a970f)

rust-timer · 2025-07-07T04:09:54Z

Queued abd7739 with parent ca98d4d, future comparison URL.
There are currently 4 preceding artifacts in the queue.
It will probably take at least ~5.2 hours until the benchmark run finishes.

rustbot assigned saethlin Mar 16, 2025

rustbot added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. T-compiler Relevant to the compiler team, which will review and decide on the PR/issue. labels Mar 16, 2025