Use adaptive sparse encoding for metadata tables by TaKO8Ki · Pull Request #164484 · rust-lang/rust

TaKO8Ki · 2026-10-17T18:30:12Z

I have added this implementation partially following the approach suggested in the existing FIXME comment.

Each 64-element block stores a cumulative count of preceding non-default values, enabling constant-time lookup by combining that count with a popcount of the bitmap bits before the requested position.

In local measurements, .rmeta file sizes decreased for all four crates in both Check and Release builds:

Crate	Check reduction	Release reduction
tokio	33.68%	27.25%
tokio-util	29.18%	21.93%
serde_core	34.45%	18.37%
serde_json	27.07%	16.96%

LLM disclosure: I used an LLM for some of the analysis in this PR.

TaKO8Ki · 2026-10-17T18:30:35Z

@bors try @rust-timer queue

Use adaptive sparse encoding for metadata tables

rust-bors · 2026-10-17T19:29:50Z

☀️ Try build successful (CI)
Build commit: cef35eb (cef35eba824ffd8f61ba555d23a22c5938662a2a)
Base parent: 2834d57 (2834d57549e77dbd7e9fa42f51aa349f8a00368b)

rust-timer · 2026-10-17T20:15:46Z

Finished benchmarking commit (cef35eb): comparison URL.

Overall result: ❌ regressions - please read:

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.

Next, please: If you can, justify the regressions found in this try perf run in writing along with @rustbot label: +perf-regression-triaged. If not, fix the regressions and do another perf run. Neutral or positive results will clear the label automatically.

@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)	0.4%	[0.1%, 2.6%]	97
Regressions ❌ (secondary)	1.0%	[0.1%, 2.6%]	87
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	0.4%	[0.1%, 2.6%]	97

Max RSS (memory usage)

Results (primary -3.8%, secondary -4.2%)

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)	-3.8%	[-7.2%, -0.7%]	230
Improvements ✅ (secondary)	-4.2%	[-7.1%, -1.0%]	225
All ❌✅ (primary)	-3.8%	[-7.2%, -0.7%]	230

Cycles

Results (primary 1.8%, secondary -6.4%)

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

	mean	range	count
Regressions ❌ (primary)	1.8%	[1.8%, 1.8%]	1
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-6.4%	[-9.1%, -2.8%]	6
All ❌✅ (primary)	1.8%	[1.8%, 1.8%]	1

Binary size

Results (primary -0.6%)

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)	-0.6%	[-0.6%, -0.5%]	4
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-0.6%	[-0.6%, -0.5%]	4

Bootstrap: 489.187s -> 489.99s (0.16%)
Artifact size: 406.93 MiB -> 409.04 MiB (0.52%)

TaKO8Ki · 2026-10-18T06:41:03Z

I think the instruction-count regressions, averaging 0.66% across the affected benchmark cases, are an acceptable trade-off for the 20.15% average reduction in crate metadata size. https://perf.rust-lang.org/compare.html?start=2834d57549e77dbd7e9fa42f51aa349f8a00368b&end=cef35eba824ffd8f61ba555d23a22c5938662a2a&stat=size%3Acrate_metadata

Kobzol · 2026-10-18T07:52:43Z

CC @jyn514 those metadata size reduction results will probably be interesting :)

nnethercote

This needs some detailed comments explaining how the new scheme works.

The metadata wins are very nice. It might be possible to reduce the instruction count regressions.

View changes since this review

nnethercote · 2026-10-22T07:47:44Z

+            len
+        };
+        let sparse =
+            u32::try_from(non_defaults).is_ok() && index_bytes + non_defaults * width < len * width;


The sparse form is used if it saves any amount of space. If a table is almost full the space saving will be small but we will pay the full speed cost of the sparse layout. It would be worth trying a few different policies, e.g. only use the sparse form if the table is less than 90% full or 75%, or 50%. The best fraction will hopefully give most of the metadata size reductions while incurring a smaller instruction count cost.

nnethercote · 2026-10-22T07:52:46Z

+                return Default::default();
+            }
+            let base =
+                u32::from_le_bytes(metadata.blob()[start + 8..start + 12].try_into().unwrap());


You could read metadata.blob()[start + 12] and then read from offsets 0 and 8, which would avoid one bounds check.

rustbot · 2026-10-22T20:02:20Z

This PR was rebased onto a different main commit. Here's a range-diff highlighting what actually changed.

Rebasing is a normal part of keeping PRs up to date, so no action is needed—this note is just to help reviewers.

Use adaptive sparse encoding for metadata tables

rust-bors · 2026-10-22T20:08:41Z

Try build cancelled. Cancelled workflows:

https://github.com/rust-lang/rust/actions/runs/39845404589

Hint: if you want to run another try build, you do not need to manually cancel the previous one. Just run @bors try and bors will cancel the previous build automatically.

TaKO8Ki · 2026-10-22T20:09:17Z

@bors try @rust-timer queue

rust-timer · 2026-10-22T20:09:20Z

This pull request is already queued and waiting for a try build to finish.

rust-bors · 2026-10-22T20:09:25Z

⌛ Trying commit de21c98 with merge a7d52aa…

To cancel the try build, run the command @bors try cancel.

Workflow: https://github.com/rust-lang/rust/actions/runs/39846373166

Use adaptive sparse encoding for metadata tables

TaKO8Ki · 2026-10-23T04:53:19Z

@bors try @rust-timer queue

rust-timer · 2026-10-23T04:53:21Z

This pull request is already queued and waiting for a try build to finish.

rust-bors · 2026-10-23T04:53:27Z

⌛ Trying commit de21c98 with merge 6ea9563…

To cancel the try build, run the command @bors try cancel.

Workflow: https://github.com/rust-lang/rust/actions/runs/39902737994

Use adaptive sparse encoding for metadata tables

TaKO8Ki · 2026-10-23T05:01:02Z

@rust-timer build a7d52aa

rust-timer · 2026-10-23T05:01:05Z

Queued a7d52aa with parent c303e41, future comparison URL.
There are currently 0 preceding artifacts in the queue.
It will probably take at least ~0.8 hours until the benchmark run finishes.

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 Oct 17, 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 Oct 17, 2026

This comment has been minimized.

Sign in to view

rust-bors Bot pushed a commit that referenced this pull request Oct 17, 2026

Auto merge of #164484 - TaKO8Ki:sparse-metadata-tables, r=<try>

cef35eb

Use adaptive sparse encoding for metadata tables

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 Oct 17, 2026

rustbot assigned nnethercote Oct 18, 2026

TaKO8Ki marked this pull request as ready for review October 18, 2026 09:39

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 Oct 18, 2026

TaKO8Ki added the perf-regression-triaged The performance regression has been triaged. label Oct 20, 2026

nnethercote reviewed Oct 22, 2026

View reviewed changes

nnethercote added S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels Oct 22, 2026

use adaptive sparse encoding for metadata tables

72ffc10

TaKO8Ki force-pushed the sparse-metadata-tables branch from fa18cf1 to 0a7e283 Compare October 22, 2026 20:02

rustbot added the S-waiting-on-perf Status: Waiting on a perf run to be completed. label Oct 22, 2026

rust-bors Bot pushed a commit that referenced this pull request Oct 22, 2026

Auto merge of #164484 - TaKO8Ki:sparse-metadata-tables, r=<try>

eb11157

Use adaptive sparse encoding for metadata tables

reduce sparse metadata table lookup overhead

de21c98

TaKO8Ki force-pushed the sparse-metadata-tables branch from 0a7e283 to de21c98 Compare October 22, 2026 20:07

rust-lang deleted a comment from rust-bors Bot Oct 22, 2026

rust-bors Bot pushed a commit that referenced this pull request Oct 22, 2026

Auto merge of #164484 - TaKO8Ki:sparse-metadata-tables, r=<try>

a7d52aa

Use adaptive sparse encoding for metadata tables

rust-lang deleted a comment from rust-timer Oct 22, 2026

rust-bors Bot pushed a commit that referenced this pull request Oct 23, 2026

Auto merge of #164484 - TaKO8Ki:sparse-metadata-tables, r=<try>

6ea9563

Use adaptive sparse encoding for metadata tables

Conversation

TaKO8Ki commented Oct 17, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

TaKO8Ki commented Oct 17, 2026

Uh oh!

This comment has been minimized.

This comment has been minimized.

rust-bors Bot commented Oct 17, 2026

Uh oh!

This comment has been minimized.

rust-timer commented Oct 17, 2026

Overall result: ❌ regressions - please read:

Instruction count

Max RSS (memory usage)

Cycles

Binary size

Uh oh!

TaKO8Ki commented Oct 18, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Kobzol commented Oct 18, 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!

nnethercote Oct 22, 2026

Choose a reason for hiding this comment

Uh oh!

nnethercote Oct 22, 2026

Choose a reason for hiding this comment

Uh oh!

rustbot commented Oct 22, 2026

Uh oh!

rust-bors Bot commented Oct 22, 2026

Uh oh!

TaKO8Ki commented Oct 22, 2026

Uh oh!

rust-timer commented Oct 22, 2026

Uh oh!

rust-bors Bot commented Oct 22, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

TaKO8Ki commented Oct 23, 2026

Uh oh!

rust-timer commented Oct 23, 2026

Uh oh!

rust-bors Bot commented Oct 23, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

TaKO8Ki commented Oct 23, 2026

Uh oh!

rust-timer commented Oct 23, 2026

Uh oh!

This branch has not been deployed

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

TaKO8Ki commented Oct 17, 2026 •

edited

Loading

TaKO8Ki commented Oct 18, 2026 •

edited

Loading

nnethercote left a comment •

edited by rustbot

Loading

rust-bors Bot commented Oct 22, 2026 •

edited

Loading

rust-bors Bot commented Oct 23, 2026 •

edited

Loading