perf: Avoid a heap allocation per basic block in MoveData's location maps - #160245

xmakro · 2026-07-31T05:14:09Z

MoveData holds two LocationMaps, and each was an IndexVec<BasicBlock, Vec<T>> that heap-allocated one Vec per basic block. Building the move data for a body allocated a small vector for every block, twice over. Store the entries in one flat buffer indexed by a per-block start offset, so each map is two allocations regardless of block count.

Kobzol · 2026-07-31T07:11:32Z

@bors try @rust-timer queue

perf: Avoid a heap allocation per basic block in MoveData's location maps

rust-bors · 2026-07-31T09:17:30Z

☀️ Try build successful (CI)
Build commit: a4ce053 (a4ce0538412cb5e1c770e8ba940d2922b1af928e)
Base parent: 8ab9fdf (8ab9fdff5a91b9f2b5ed57fb0275452d9a0d0280)

rust-timer · 2026-07-31T10:41:47Z

Finished benchmarking commit (a4ce053): comparison URL.

Overall result: ✅ improvements - no action needed

Benchmarking means the PR may be perf-sensitive. It's automatically marked not fit for rolling up. Overriding is possible but disadvised: it risks changing compiler perf.

@bors rollup=never rustc-perf
@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
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-0.3%	[-0.5%, -0.2%]	28
Improvements ✅ (secondary)	-0.3%	[-0.5%, -0.1%]	22
All ❌✅ (primary)	-0.3%	[-0.5%, -0.2%]	28

Max RSS (memory usage)

Results (primary -0.4%, secondary -2.7%)

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

	mean	range	count
Regressions ❌ (primary)	0.5%	[0.5%, 0.5%]	2
Regressions ❌ (secondary)	5.7%	[5.7%, 5.7%]	1
Improvements ✅ (primary)	-2.1%	[-2.1%, -2.1%]	1
Improvements ✅ (secondary)	-5.5%	[-8.3%, -1.5%]	3
All ❌✅ (primary)	-0.4%	[-2.1%, 0.5%]	3

Cycles

Results (primary 0.4%, secondary 0.2%)

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

	mean	range	count
Regressions ❌ (primary)	0.8%	[0.5%, 2.1%]	11
Regressions ❌ (secondary)	1.1%	[0.5%, 3.1%]	10
Improvements ✅ (primary)	-2.0%	[-2.6%, -1.4%]	2
Improvements ✅ (secondary)	-2.1%	[-3.8%, -0.7%]	4
All ❌✅ (primary)	0.4%	[-2.6%, 2.1%]	13

Binary size

This perf run didn't have relevant results for this metric.

Bootstrap: 492.09s -> 496.702s (0.94%)
Artifact size: 390.63 MiB -> 390.63 MiB (-0.00%)

xmakro · 2026-07-31T10:47:03Z

r? @nnethercote

nnethercote

Nice change! Just a few nits to fix.

View changes since this review

nnethercote · 2026-07-31T11:35:39Z

 impl<T> IndexMut<Location> for LocationMap<T> {
    fn index_mut(&mut self, index: Location) -> &mut Self::Output {
-        &mut self.map[index.block][index.statement_index]
+        let offset = self.offset(index);


unfortunately &mut self.data[self.offset(index)] fails borrowck:

error[E0502]: cannot borrow `*self` as immutable because it is also borrowed as mutable | | fn index_mut(&mut self, index: Location) -> &mut Self::Output { | - let's call the lifetime of this reference `'1` | &mut self.data[self.offset(index)] | ---------------^^^^--------------- | | | | | | | immutable borrow occurs here | | mutable borrow occurs here | returning this value requires that `self.data` is borrowed for `'1`

…maps

xmakro · 2026-07-31T11:47:46Z

Thanks for the quick review!

nnethercote · 2026-07-31T12:05:18Z

@bors r+

rust-bors · 2026-07-31T12:05:21Z

📌 Commit b146323 has been approved by nnethercote

It is now in the queue for this repository.

cjgillot · 2026-07-31T14:34:24Z

+    /// `block_starts[bb]` gives the index in `data` where block `bb`'s entries
+    /// start; each block has one entry per statement plus one for its terminator.
+    data: Vec<T>,
+    block_starts: IndexVec<BasicBlock, usize>,


Should we eventually move to use an IndexVec<PointIndex> with a shared DenseLocationMap?

View changes since the review

rust-bors · 2026-08-01T11:13:43Z

☀️ Test successful - CI
Approved by: nnethercote
Duration: 3h 11m 2s
Pushing c47bb2f to main...

github-actions · 2026-08-01T11:16:52Z

What is this?

This is an experimental post-merge analysis report that shows differences in test outcomes between the merged PR and its parent PR.

Comparing b6a3d79 (parent) -> c47bb2f (this PR)

Test differences

Show 2 test diffs

2 doctest diffs were found. These are ignored, as they are noisy.

Test dashboard

Run

cargo run --manifest-path src/ci/citool/Cargo.toml -- \
    test-dashboard c47bb2f39658c00b109693df9611d7238bedf6fa --output-dir test-dashboard

And then open test-dashboard/index.html in your browser to see an overview of all executed tests.

Job duration changes

x86_64-msvc-ext1: 2h 33m -> 1h 24m (-45.1%)
arm-android: 1h 13m -> 1h 40m (+37.1%)
x86_64-gnu-gcc-core-tests: 11m 38s -> 7m 20s (-36.9%)
dist-loongarch64-linux: 1h 55m -> 1h 15m (-34.9%)
dist-ohos-x86_64: 59m 4s -> 1h 17m (+31.0%)
x86_64-mingw-1: 2h 19m -> 3h 1m (+29.6%)
i686-gnu-nopt-2: 2h 21m -> 1h 40m (-29.1%)
dist-riscv64-linux-gnu: 1h 7m -> 1h 26m (+29.0%)
x86_64-gnu-parallel-frontend: 1h 32m -> 1h 52m (+22.4%)
dist-android: 27m 20s -> 21m 25s (-21.7%)

How to interpret the job duration changes?

Job durations can vary a lot, based on the actual runner instance
that executed the job, system noise, invalidated caches, etc. The table above is provided
mostly for t-infra members, for simpler debugging of potential CI slow-downs.

rust-timer · 2026-08-01T11:57:57Z

Finished benchmarking commit (c47bb2f): comparison URL.

Overall result: ✅ improvements - no action needed

@rustbot label: -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
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-0.3%	[-0.5%, -0.2%]	23
Improvements ✅ (secondary)	-0.3%	[-0.5%, -0.1%]	21
All ❌✅ (primary)	-0.3%	[-0.5%, -0.2%]	23

Max RSS (memory usage)

Results (primary -4.2%, secondary -1.5%)

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)	-4.2%	[-8.0%, -0.4%]	2
Improvements ✅ (secondary)	-1.5%	[-4.6%, -0.4%]	7
All ❌✅ (primary)	-4.2%	[-8.0%, -0.4%]	2

Cycles

Results (primary -0.8%, secondary -0.5%)

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)	0.7%	[0.6%, 1.0%]	3
Improvements ✅ (primary)	-0.8%	[-1.8%, -0.4%]	7
Improvements ✅ (secondary)	-1.0%	[-1.3%, -0.5%]	7
All ❌✅ (primary)	-0.8%	[-1.8%, -0.4%]	7

Binary size

This perf run didn't have relevant results for this metric.

Bootstrap: 491.498s -> 490.834s (-0.14%)
Artifact size: 390.44 MiB -> 390.52 MiB (0.02%)

rustbot added S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. T-compiler Relevant to the compiler team, which will review and decide on the PR/issue. labels Jul 31, 2026

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 31, 2026

This comment has been minimized.

Sign in to view

rust-bors Bot pushed a commit that referenced this pull request Jul 31, 2026

Auto merge of #160245 - xmakro:perf/location-map-single-alloc, r=<try>

a4ce053

perf: Avoid a heap allocation per basic block in MoveData's location maps

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 Jul 31, 2026

rustbot assigned nnethercote Jul 31, 2026

xmakro marked this pull request as ready for review July 31, 2026 10:47

rustbot added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. and removed S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. labels Jul 31, 2026

nnethercote reviewed Jul 31, 2026

View reviewed changes

perf: Avoid a heap allocation per basic block in MoveData's location …

b146323

…maps

xmakro force-pushed the perf/location-map-single-alloc branch from 57527a3 to b146323 Compare July 31, 2026 11:45

nnethercote approved these changes Jul 31, 2026

View reviewed changes

rust-bors Bot added S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels Jul 31, 2026

cjgillot reviewed Jul 31, 2026

View reviewed changes

This comment has been minimized.

Sign in to view

rust-bors Bot added merged-by-bors This PR was explicitly merged by bors. and removed S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. labels Aug 1, 2026

rust-bors Bot merged commit c47bb2f into rust-lang:main Aug 1, 2026
13 of 14 checks passed

rustbot added this to the 1.99.0 milestone Aug 1, 2026

xmakro mentioned this pull request Aug 5, 2026

Back MoveData's location maps with a shared DenseLocationMap #160572

Draft

Uh oh!

Uh oh!

Conversation

xmakro commented Jul 31, 2026

Uh oh!

Kobzol commented Jul 31, 2026

Uh oh!

This comment has been minimized.

This comment has been minimized.

rust-bors Bot commented Jul 31, 2026

Uh oh!

This comment has been minimized.

rust-timer commented Jul 31, 2026

Overall result: ✅ improvements - no action needed

Uh oh!

xmakro commented Jul 31, 2026

Uh oh!

nnethercote left a comment • edited by rustbot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

xmakro commented Jul 31, 2026

Uh oh!

nnethercote commented Jul 31, 2026

Uh oh!

rust-bors Bot commented Jul 31, 2026

Uh oh!

cjgillot Jul 31, 2026 • edited by rustbot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

This comment has been minimized.

rust-bors Bot commented Aug 1, 2026

Uh oh!

Uh oh!

github-actions Bot commented Aug 1, 2026

Test differences

Job duration changes

Uh oh!

rust-timer commented Aug 1, 2026

Overall result: ✅ improvements - no action needed

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants

nnethercote left a comment •

edited by rustbot

Loading

cjgillot Jul 31, 2026 •

edited by rustbot

Loading