Shrink `FnAbi` #100999

nnethercote · 2022-08-25T12:52:21Z

Because they can take up a lot of memory in debug and release builds.

r? @bjorn3

rustbot · 2022-08-25T12:52:26Z

Some changes occurred in compiler/rustc_codegen_gcc

cc @antoyo

Some changes occurred in compiler/rustc_codegen_cranelift

cc @bjorn3

Some changes occurred to the CTFE / Miri engine

cc @rust-lang/miri

compiler/rustc_codegen_cranelift/src/abi/pass_mode.rs

nnethercote · 2022-08-25T13:12:10Z

@bors try @rust-timer queue

rust-timer · 2022-08-25T13:12:12Z

Awaiting bors try build completion.

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

bors · 2022-08-25T13:12:19Z

⌛ Trying commit 71c4b14a9585e2595cd28fdff1975ebf558d1686 with merge 0866a29e0084cb0441c0a82a9aa5e1853d404e2a...

compiler/rustc_const_eval/src/interpret/terminator.rs

bors · 2022-08-25T14:32:02Z

☀️ Try build successful - checks-actions
Build commit: 0866a29e0084cb0441c0a82a9aa5e1853d404e2a (0866a29e0084cb0441c0a82a9aa5e1853d404e2a)

rust-timer · 2022-08-25T14:32:04Z

Queued 0866a29e0084cb0441c0a82a9aa5e1853d404e2a with parent 4d45b07, future comparison URL.

rust-timer · 2022-08-25T15:46:03Z

Finished benchmarking commit (0866a29e0084cb0441c0a82a9aa5e1853d404e2a): comparison URL.

Overall result: ❌ regressions - no 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.

@bors rollup=never
@rustbot label: +S-waiting-on-review -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.

	mean¹	range	count²
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	0.2%	[0.2%, 0.2%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-	-	0

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.

	mean¹	range	count²
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	2.1%	[2.1%, 2.1%]	1
Improvements ✅ (primary)	-3.2%	[-7.1%, -1.8%]	8
Improvements ✅ (secondary)	-2.6%	[-2.6%, -2.6%]	1
All ❌✅ (primary)	-3.2%	[-7.1%, -1.8%]	8

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.

	mean¹	range	count²
Regressions ❌ (primary)	2.2%	[2.2%, 2.2%]	1
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-2.8%	[-3.9%, -2.1%]	5
All ❌✅ (primary)	2.2%	[2.2%, 2.2%]	1

the arithmetic mean of the percent change ↩ ↩² ↩³
number of relevant changes ↩ ↩² ↩³

nnethercote · 2022-08-25T23:26:19Z

A few very nice wins for max-rss, all in the opt and debug builds, which are the only ones affected by this change. Particularly for some of the larger benchmarks. And if you check the "Show non-relevant results" box, you'll see more green, for ones that don't meet the significance threshold (because max-rss is noisy) but are probably real (because there are so many of them, and it lines up with the significant ones). Also some hints of small speedups in the cycles and wall-time results.

Because `PassMode::Cast` is by far the largest variant, but is relatively rare. This requires making `PassMode` not impl `Copy`, and `Clone` is no longer necessary. This causes lots of sigil adjusting, but nothing very notable.

Currently they try to be very precise. But they are wrong, i.e. they don't match what's happening in the loop below. This code isn't hot enough for it to matter that much.

Because it's only ever set to `None` or `Some(Reg::i32())`.

Because it's only needed for that variant. This shrinks the types and clarifies the logic.

nnethercote · 2022-08-26T01:51:41Z

r? @bjorn3

bjorn3 · 2022-08-26T07:04:11Z

@bors r+ rollup=never

bors · 2022-08-26T07:04:13Z

📌 Commit f974617 has been approved by bjorn3

It is now in the queue for this repository.

bors · 2022-08-27T14:00:56Z

⌛ Testing commit f974617 with merge 332cc8f...

bors · 2022-08-27T16:41:52Z

☀️ Test successful - checks-actions
Approved by: bjorn3
Pushing 332cc8f to master...

rust-timer · 2022-08-28T16:46:19Z

Finished benchmarking commit (332cc8f): comparison URL.

Overall result: ❌ regressions - no action needed

@rustbot label: -perf-regression

Instruction count

This is a highly reliable metric that was used to determine the overall result at the top of this comment.

	mean¹	range	count²
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	1.3%	[1.3%, 1.3%]	2
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-	-	0

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.

	mean¹	range	count²
Regressions ❌ (primary)	3.8%	[3.8%, 3.8%]	1
Regressions ❌ (secondary)	3.3%	[3.3%, 3.3%]	1
Improvements ✅ (primary)	-3.9%	[-7.9%, -2.6%]	5
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-2.6%	[-7.9%, 3.8%]	6

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.

	mean¹	range	count²
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	3.7%	[3.7%, 3.7%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-3.9%	[-3.9%, -3.9%]	1
All ❌✅ (primary)	-	-	0

the arithmetic mean of the percent change ↩ ↩² ↩³
number of relevant changes ↩ ↩² ↩³

Shrink `FnAbi` Because they can take up a lot of memory in debug and release builds. r? `@bjorn3`

rustbot added the T-compiler Relevant to the compiler team, which will review and decide on the PR/issue. label Aug 25, 2022

nnethercote changed the title ~~Shrink fn abi~~ Shrink FnAbi Aug 25, 2022

bjorn3 reviewed Aug 25, 2022

View reviewed changes

compiler/rustc_codegen_cranelift/src/abi/pass_mode.rs Outdated Show resolved Hide resolved

rustbot added the S-waiting-on-perf Status: Waiting on a perf run to be completed. label Aug 25, 2022

RalfJung reviewed Aug 25, 2022

View reviewed changes

compiler/rustc_const_eval/src/interpret/terminator.rs Show resolved Hide resolved

rustbot added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. and removed S-waiting-on-perf Status: Waiting on a perf run to be completed. labels Aug 25, 2022

nnethercote added 7 commits August 26, 2022 09:30

Add size assertions for FnAbi and ArgAbi.

263c426

Box CastTarget within PassMode.

e4bf113

Because `PassMode::Cast` is by far the largest variant, but is relatively rare. This requires making `PassMode` not impl `Copy`, and `Clone` is no longer necessary. This causes lots of sigil adjusting, but nothing very notable.

Change FnAbi::fixed_count to a u32.

4df7bff

Change FnAbi::args to a boxed slice.

b75b3b3

Simplify arg capacity calculations.

feeaa4d

Currently they try to be very precise. But they are wrong, i.e. they don't match what's happening in the loop below. This code isn't hot enough for it to matter that much.

Turn ArgAbi::pad into a bool.

b853e8a

Because it's only ever set to `None` or `Some(Reg::i32())`.

Move ArgAbi::pad_i32 into PassMode::Cast.

f974617

Because it's only needed for that variant. This shrinks the types and clarifies the logic.

nnethercote force-pushed the shrink-FnAbi branch from 71c4b14 to f974617 Compare August 26, 2022 01:49

rust-highfive assigned bjorn3 Aug 26, 2022

bors 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 Aug 26, 2022

bors added the merged-by-bors This PR was explicitly merged by bors. label Aug 27, 2022

bors merged commit 332cc8f into rust-lang:master Aug 27, 2022

rustbot added this to the 1.65.0 milestone Aug 27, 2022

nnethercote deleted the shrink-FnAbi branch August 28, 2022 22:40

bjorn3 pushed a commit to bjorn3/rust that referenced this pull request Oct 23, 2022

Auto merge of rust-lang#100999 - nnethercote:shrink-FnAbi, r=bjorn3

1239a02

Shrink `FnAbi` Because they can take up a lot of memory in debug and release builds. r? `@bjorn3`

antoyo pushed a commit to antoyo/rust that referenced this pull request Jun 19, 2023

Auto merge of rust-lang#100999 - nnethercote:shrink-FnAbi, r=bjorn3

1a61b6e

Shrink `FnAbi` Because they can take up a lot of memory in debug and release builds. r? `@bjorn3`

andjo403 mentioned this pull request Aug 3, 2024

Add range attribute to scalar function results and arguments #128371

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Shrink `FnAbi` #100999

Shrink `FnAbi` #100999

nnethercote commented Aug 25, 2022 •

edited

Loading

rustbot commented Aug 25, 2022

nnethercote commented Aug 25, 2022

rust-timer commented Aug 25, 2022

bors commented Aug 25, 2022

bors commented Aug 25, 2022

rust-timer commented Aug 25, 2022

rust-timer commented Aug 25, 2022

nnethercote commented Aug 25, 2022

nnethercote commented Aug 26, 2022

bjorn3 commented Aug 26, 2022

bors commented Aug 26, 2022

bors commented Aug 27, 2022

bors commented Aug 27, 2022

rust-timer commented Aug 28, 2022

Shrink FnAbi #100999

Shrink FnAbi #100999

Conversation

nnethercote commented Aug 25, 2022 • edited Loading

rustbot commented Aug 25, 2022

nnethercote commented Aug 25, 2022

rust-timer commented Aug 25, 2022

bors commented Aug 25, 2022

bors commented Aug 25, 2022

rust-timer commented Aug 25, 2022

rust-timer commented Aug 25, 2022

Overall result: ❌ regressions - no action needed

Instruction count

Max RSS (memory usage)

Cycles

Footnotes

nnethercote commented Aug 25, 2022

nnethercote commented Aug 26, 2022

bjorn3 commented Aug 26, 2022

bors commented Aug 26, 2022

bors commented Aug 27, 2022

bors commented Aug 27, 2022

rust-timer commented Aug 28, 2022

Overall result: ❌ regressions - no action needed

Instruction count

Max RSS (memory usage)

Cycles

Footnotes

Shrink `FnAbi` #100999

Shrink `FnAbi` #100999

nnethercote commented Aug 25, 2022 •

edited

Loading