[TOY] Extend GVN to perform local value numbering. #143333

cjgillot · 2025-07-02T15:56:21Z

This PR is more a toy than anything else, but I still think the implementation is sound.

Current GVN MIR opt can be easily extended to track values that change inside a basic block. This PR attempts that.

r? @ghost for perf

cjgillot · 2025-07-05T14:13:23Z

[TOY] Extend GVN to perform local value numbering. This PR is more a toy than anything else, but I still think the implementation is sound. Current GVN MIR opt can be easily extended to track values that change inside a basic block. This PR attempts that. r? `@ghost` for perf

bors · 2025-07-05T14:14:39Z

⌛ Trying commit c3be10b with merge faceb52...

bors · 2025-07-05T17:10:23Z

☀️ Try build successful - checks-actions
Build commit: faceb52 (faceb52e72c06f94883ba351a73c83544ab4154b)

rust-timer · 2025-07-05T18:38:46Z

Finished benchmarking commit (faceb52): 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)	15.4%	[0.2%, 937.4%]	227
Regressions ❌ (secondary)	10.3%	[0.2%, 569.6%]	237
Improvements ✅ (primary)	-1.1%	[-1.8%, -0.4%]	2
Improvements ✅ (secondary)	-1.3%	[-4.1%, -0.3%]	16
All ❌✅ (primary)	15.3%	[-1.8%, 937.4%]	229

Max RSS (memory usage)

Results (primary 9.1%, secondary 2.4%)

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

	mean	range	count
Regressions ❌ (primary)	71.9%	[1.6%, 377.4%]	14
Regressions ❌ (secondary)	28.4%	[2.2%, 134.5%]	15
Improvements ✅ (primary)	-2.0%	[-7.5%, -0.7%]	79
Improvements ✅ (secondary)	-2.5%	[-3.5%, -0.8%]	80
All ❌✅ (primary)	9.1%	[-7.5%, 377.4%]	93

Cycles

Results (primary 44.8%, secondary 29.7%)

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

	mean	range	count
Regressions ❌ (primary)	46.5%	[2.3%, 299.4%]	27
Regressions ❌ (secondary)	40.6%	[2.4%, 263.2%]	28
Improvements ✅ (primary)	-2.1%	[-2.1%, -2.1%]	1
Improvements ✅ (secondary)	-4.2%	[-6.7%, -2.4%]	9
All ❌✅ (primary)	44.8%	[-2.1%, 299.4%]	28

Binary size

Results (primary -0.1%, secondary 0.1%)

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

	mean	range	count
Regressions ❌ (primary)	0.2%	[0.0%, 0.6%]	37
Regressions ❌ (secondary)	0.5%	[0.0%, 1.7%]	70
Improvements ✅ (primary)	-0.3%	[-1.3%, -0.0%]	54
Improvements ✅ (secondary)	-1.0%	[-12.8%, -0.0%]	24
All ❌✅ (primary)	-0.1%	[-1.3%, 0.6%]	91

Bootstrap: 458.797s -> 582.037s (26.86%)
Artifact size: 372.11 MiB -> 361.76 MiB (-2.78%)

cjgillot · 2025-07-06T08:54:20Z

@bors try @rust-timer queue

[TOY] Extend GVN to perform local value numbering. This PR is more a toy than anything else, but I still think the implementation is sound. Current GVN MIR opt can be easily extended to track values that change inside a basic block. This PR attempts that. r? `@ghost` for perf

bors · 2025-07-06T08:55:31Z

⌛ Trying commit e521268 with merge dc4aa90...

cjgillot · 2025-07-06T08:56:01Z

@bors try @rust-timer queue

[TOY] Extend GVN to perform local value numbering. This PR is more a toy than anything else, but I still think the implementation is sound. Current GVN MIR opt can be easily extended to track values that change inside a basic block. This PR attempts that. r? `@ghost` for perf

bors · 2025-07-06T08:57:11Z

⌛ Trying commit 0293254 with merge 4d16c54...

bors · 2025-07-06T11:13:52Z

☀️ Try build successful - checks-actions
Build commit: 4d16c54 (4d16c544e50b0f84cc5347467c108e93106aa8f9)

bors · 2025-09-19T16:10:43Z

☔ The latest upstream changes (presumably #145737) made this pull request unmergeable. Please resolve the merge conflicts.

cjgillot · 2025-09-21T15:08:51Z

Now that #145737 is merged.
@bors try @rust-timer queue

[TOY] Extend GVN to perform local value numbering.

rust-bors · 2025-09-21T17:20:38Z

☀️ Try build successful (CI)
Build commit: 695bd40 (695bd40e2c808675c28012225875aeb7508ec032, parent: 1d23da6b7304d9e2a2c3dcb1b0aaa709cb9bc4ad)

rust-timer · 2025-09-21T18:46:30Z

Finished benchmarking commit (695bd40): 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)	0.5%	[0.1%, 4.9%]	109
Regressions ❌ (secondary)	0.7%	[0.1%, 6.5%]	55
Improvements ✅ (primary)	-0.6%	[-0.9%, -0.4%]	4
Improvements ✅ (secondary)	-0.3%	[-1.2%, -0.2%]	26
All ❌✅ (primary)	0.5%	[-0.9%, 4.9%]	113

Max RSS (memory usage)

Results (primary 1.9%, secondary 4.6%)

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

	mean	range	count
Regressions ❌ (primary)	1.9%	[1.0%, 2.8%]	2
Regressions ❌ (secondary)	4.6%	[3.1%, 5.0%]	8
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	1.9%	[1.0%, 2.8%]	2

Cycles

Results (primary 4.3%, secondary -0.7%)

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

	mean	range	count
Regressions ❌ (primary)	4.3%	[3.5%, 5.1%]	2
Regressions ❌ (secondary)	4.1%	[3.2%, 5.7%]	3
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-3.1%	[-4.3%, -2.0%]	6
All ❌✅ (primary)	4.3%	[3.5%, 5.1%]	2

Binary size

Results (primary -0.1%, secondary -0.3%)

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

	mean	range	count
Regressions ❌ (primary)	0.1%	[0.0%, 0.9%]	13
Regressions ❌ (secondary)	0.0%	[0.0%, 0.3%]	39
Improvements ✅ (primary)	-0.1%	[-2.1%, -0.0%]	46
Improvements ✅ (secondary)	-1.4%	[-13.0%, -0.0%]	10
All ❌✅ (primary)	-0.1%	[-2.1%, 0.9%]	59

Bootstrap: 470.725s -> 472.748s (0.43%)
Artifact size: 387.92 MiB -> 390.07 MiB (0.55%)

cjgillot · 2025-09-21T20:28:24Z

@bors try @rust-timer queue

[TOY] Extend GVN to perform local value numbering.

rust-bors · 2025-09-21T22:39:10Z

☀️ Try build successful (CI)
Build commit: ed91d44 (ed91d44e9b64826073d03e70308259977dc35b52, parent: 7e4b8d702fedccc9c7803773a22c2e053ac3b004)

rust-timer · 2025-09-21T23:48:58Z

Finished benchmarking commit (ed91d44): 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)	0.3%	[0.1%, 2.1%]	125
Regressions ❌ (secondary)	0.3%	[0.1%, 0.7%]	62
Improvements ✅ (primary)	-0.6%	[-0.9%, -0.3%]	5
Improvements ✅ (secondary)	-0.3%	[-1.4%, -0.1%]	22
All ❌✅ (primary)	0.3%	[-0.9%, 2.1%]	130

Max RSS (memory usage)

Results (primary -0.5%, secondary 3.4%)

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

	mean	range	count
Regressions ❌ (primary)	0.8%	[0.7%, 0.9%]	2
Regressions ❌ (secondary)	4.1%	[1.0%, 5.0%]	9
Improvements ✅ (primary)	-1.7%	[-2.7%, -0.6%]	2
Improvements ✅ (secondary)	-2.9%	[-2.9%, -2.9%]	1
All ❌✅ (primary)	-0.5%	[-2.7%, 0.9%]	4

Cycles

Results (primary 2.0%, secondary -2.9%)

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

	mean	range	count
Regressions ❌ (primary)	2.0%	[1.7%, 2.3%]	2
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-2.9%	[-5.3%, -2.0%]	8
All ❌✅ (primary)	2.0%	[1.7%, 2.3%]	2

Binary size

Results (primary -0.1%, secondary -0.3%)

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

	mean	range	count
Regressions ❌ (primary)	0.1%	[0.0%, 0.8%]	13
Regressions ❌ (secondary)	0.0%	[0.0%, 0.3%]	36
Improvements ✅ (primary)	-0.1%	[-2.2%, -0.0%]	46
Improvements ✅ (secondary)	-1.4%	[-13.0%, -0.0%]	10
All ❌✅ (primary)	-0.1%	[-2.2%, 0.8%]	59

Bootstrap: 471.863s -> 473.827s (0.42%)
Artifact size: 387.93 MiB -> 390.21 MiB (0.59%)

GVN: Ensure indirect is first projection in try_as_place. I haven't found any report for this bug on existing code, but managed to trigger it with rust-lang/rust#143333

rustbot added the T-compiler Relevant to the compiler team, which will review and decide on the PR/issue. label Jul 2, 2025

This comment has been minimized.

Sign in to view

cjgillot force-pushed the local-value-numbering branch from c7dac45 to 62766e3 Compare July 2, 2025 22:11

This comment has been minimized.

Sign in to view

cjgillot added the A-mir-opt-GVN Area: MIR opt Global Value Numbering (GVN) label Jul 5, 2025

This comment has been minimized.

Sign in to view

rustbot added the S-waiting-on-perf Status: Waiting on a perf run to be completed. label Jul 5, 2025

This comment has been minimized.

Sign in to view

rustbot added perf-regression Performance regression. and removed S-waiting-on-perf Status: Waiting on a perf run to be completed. labels Jul 5, 2025

cjgillot force-pushed the local-value-numbering branch from c3be10b to e521268 Compare July 6, 2025 08:54

This comment has been minimized.

Sign in to view

rustbot added the S-waiting-on-perf Status: Waiting on a perf run to be completed. label Jul 6, 2025

cjgillot force-pushed the local-value-numbering branch from e521268 to 0293254 Compare July 6, 2025 08:55

This comment has been minimized.

Sign in to view

cjgillot added 2 commits September 19, 2025 19:46

Make Value Copy by arena-allocating vectors.

50c72b7

Use a VnIndex in Address projection.

f092b9b

cjgillot force-pushed the local-value-numbering branch from 477da90 to 2705998 Compare September 19, 2025 21:59

This comment has been minimized.

Sign in to view

rust-bors bot added a commit that referenced this pull request Sep 21, 2025

Auto merge of #143333 - cjgillot:local-value-numbering, r=<try>

695bd40

[TOY] Extend GVN to perform local value numbering.

rustbot added the S-waiting-on-perf Status: Waiting on a perf run to be completed. label Sep 21, 2025

cjgillot added 3 commits September 21, 2025 15:30

Only decide once whether to use a complex projection.

87b464b

Simplify project.

02551b9

Extend GVN to perform local value numbering.

99438d1

This comment has been minimized.

Sign in to view

rustbot removed the S-waiting-on-perf Status: Waiting on a perf run to be completed. label Sep 21, 2025

Separate SSA vs non-SSA locals map.

aa12f6a

cjgillot force-pushed the local-value-numbering branch from 2705998 to aa12f6a Compare September 21, 2025 20:28

This comment has been minimized.

Sign in to view

rust-bors bot added a commit that referenced this pull request Sep 21, 2025

Auto merge of #143333 - cjgillot:local-value-numbering, r=<try>

ed91d44

[TOY] Extend GVN to perform local value numbering.

rustbot added the S-waiting-on-perf Status: Waiting on a perf run to be completed. label Sep 21, 2025

This comment has been minimized.

Sign in to view

rustbot removed the S-waiting-on-perf Status: Waiting on a perf run to be completed. label Sep 21, 2025

Dylan-DPC added S-experimental Status: Ongoing experiment that does not require reviewing and won't be merged in its current state. and removed S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. labels Sep 28, 2025

[TOY] Extend GVN to perform local value numbering. #143333

Are you sure you want to change the base?

[TOY] Extend GVN to perform local value numbering. #143333

Uh oh!

Conversation

cjgillot commented Jul 2, 2025

Uh oh!

This comment has been minimized.

This comment has been minimized.

This comment has been minimized.

cjgillot commented Jul 5, 2025

Uh oh!

This comment has been minimized.

bors commented Jul 5, 2025

Uh oh!

bors commented Jul 5, 2025

Uh oh!

This comment has been minimized.

rust-timer commented Jul 5, 2025

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

Uh oh!

cjgillot commented Jul 6, 2025

Uh oh!

This comment has been minimized.

bors commented Jul 6, 2025

Uh oh!

cjgillot commented Jul 6, 2025

Uh oh!

This comment has been minimized.

bors commented Jul 6, 2025

Uh oh!

This comment has been minimized.

bors commented Jul 6, 2025

Uh oh!

This comment has been minimized.

bors commented Sep 19, 2025

Uh oh!

cjgillot commented Sep 21, 2025

Uh oh!

This comment has been minimized.

This comment has been minimized.

rust-bors bot commented Sep 21, 2025

Uh oh!

This comment has been minimized.

rust-timer commented Sep 21, 2025

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

Uh oh!

cjgillot commented Sep 21, 2025

Uh oh!

This comment has been minimized.

This comment has been minimized.

rust-bors bot commented Sep 21, 2025

Uh oh!

This comment has been minimized.

rust-timer commented Sep 21, 2025

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

Uh oh!

Uh oh!