Use wrapping arithmetic in `from_str_radix` by maxdexh · Pull Request #163099 · rust-lang/rust

maxdexh · 2026-09-21T09:22:57Z

Uses wrapping arithmetic in the fast/unchecked loop of from_str_radix.

Originally this PR was written using unchecked arithmetic:
This generates identical assembly when compiling with -O: https://godbolt.org/z/Gvb7G9YEM
However, it significantly simplifies the asm of -O -Coverflow-checks (e.g. RfL): https://godbolt.org/z/Pah9fMfrd
More precisely, the -O -Coverflow-checks output becomes identical to the -O one: https://godbolt.org/z/WoYMThP8e
In conclusion, it makes more sense to use wrapping operations here.

Darksonn · 2026-09-21T09:25:11Z

@rust-timer queue

maxdexh · 2026-09-21T09:25:27Z

You need to run bors (@bors try is enough)

Darksonn · 2026-09-21T09:25:43Z

@bors try

[EXPERIMENT] Use unchecked arithmetic in `from_str_radix`

maxdexh · 2026-09-21T09:57:19Z

Is rustbot broken? @rustbot label T-libs

rustbot · 2026-09-21T09:57:22Z

Error: Label waiting-on-author can only be set by Rust team members

Please file an issue on GitHub at triagebot if there's a problem with this bot, or reach out on #triagebot on Zulip.

Darksonn · 2026-09-21T10:15:14Z

The waiting-on-author label has an S- prefix

maxdexh · 2026-09-21T10:20:23Z

Alright so godbolt reveals that they generate the same optimized IR with overflow checks (for u32). This PR causes the overflow checked version to become identical to that as well: https://godbolt.org/z/1xTf3Y3j8

i.e. this should not change runtime perf when running without overflow checks, but with overflow checks it should cause a decent amount of improvement (and get rid of the panics)

rust-bors · 2026-09-21T10:25:11Z

☀️ Try build successful (CI)
Build commit: 1d6776a (1d6776a3cda0d0b6962859a966a20cfe82558cab)
Base parent: 220b36c (220b36c420c49c59923f54cd4a76634fac98a067)

maxdexh · 2026-09-21T10:37:51Z

Since this makes the correctness of can_not_overflow soundness critical (and that function is underdocumented), I added a correctness proof as a comment. Shouldn't affect the perf run

rust-timer · 2026-09-21T11:09:11Z

Finished benchmarking commit (1d6776a): comparison URL.

Overall result: ✅ improvements - no action needed

Benchmarking means the PR may be perf-sensitive. Consider adding rollup=never if this change is not fit for rolling up.

@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
Improvements ✅ (secondary)	-0.4%	[-0.4%, -0.4%]	1
All ❌✅ (primary)	-	-	0

Max RSS (memory usage)

Results (primary -2.4%, 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)	2.8%	[2.8%, 2.8%]	1
Improvements ✅ (primary)	-2.4%	[-2.9%, -1.9%]	2
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	-2.4%	[-2.9%, -1.9%]	2

Cycles

Results (primary -2.0%, secondary -4.4%)

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)	-2.0%	[-2.0%, -2.0%]	1
Improvements ✅ (secondary)	-4.4%	[-4.4%, -4.4%]	1
All ❌✅ (primary)	-2.0%	[-2.0%, -2.0%]	1

Binary size

Results (primary 0.0%, 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)	0.1%	[0.0%, 0.3%]	5
Regressions ❌ (secondary)	0.1%	[0.0%, 0.1%]	4
Improvements ✅ (primary)	-0.2%	[-0.2%, -0.2%]	1
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	0.0%	[-0.2%, 0.3%]	6

Bootstrap: 499.144s -> 497.776s (-0.27%)
Artifact size: 408.93 MiB -> 409.71 MiB (0.19%)

maxdexh · 2026-09-21T11:11:14Z

r? libs

edit: lol

Darksonn · 2026-09-21T11:15:45Z

@rustbot label I-libs-nominated

clarfonthey · 2026-09-22T17:06:07Z

If this produces the same assembly for wrapping we should use wrapping instead of unchecked, since just adding unsafe code is a higher bar than changing an operation to always wrap.

maxdexh · 2026-09-22T17:08:24Z

Agreed. I'll add a comment mentioning this though

clarfonthey · 2026-09-22T18:31:12Z

@rustbot ready

Actually handing review back to Alice

Darksonn

LGTM

@bors r+

View changes since this review

rust-bors · 2026-09-22T18:47:30Z

📌 Commit 84a341b has been approved by Darksonn

It is now in the queue for this repository.

…d, r=Darksonn Use wrapping arithmetic in `from_str_radix` Uses wrapping arithmetic in the fast/unchecked loop of `from_str_radix`. Originally this PR was written using unchecked arithmetic: This generates identical assembly when compiling with `-O`: https://godbolt.org/z/Gvb7G9YEM However, it significantly simplifies the asm of `-O -Coverflow-checks` (e.g. RfL): https://godbolt.org/z/Pah9fMfrd More precisely, the `-O -Coverflow-checks` output becomes identical to the `-O` one: https://godbolt.org/z/WoYMThP8e In conclusion, it makes more sense to use wrapping operations here.

tgross35 · 2026-09-22T19:40:55Z

If the goal is to actually remove panics and keep them gone then why no codegen test?

View changes since the review

The function can already panic from the radix check. This change just brings the loop in line with the no-overflow-check version, which is the default. Adding a test feels more like an obligation that we will keep it this way. Idk if that is what we want, but I'm open to discussion on zulip ^^

…uwer Rollup of 13 pull requests Successful merges: - #156949 (Detect missing else in let statement) - #160436 (stabilize `Box::take`) - #160570 (macro_metavar_expr_concat: support concatenating into string literals) - #162837 (Dedicated Display type for CStr::display) - #163099 (Use wrapping arithmetic in `from_str_radix`) - #163166 (Tiny cleanups to deferred liveness) - #161667 (Add `f16` inline ASM support for `nvptx64-nvidia-cuda`) - #163063 (Restore `Send` and `Sync` for `BorrowedCursor`) - #163097 (OpenBSD/sparc64 has switched from GCC to Clang) - #163126 (Skip redundant storage-conflict updates during coroutine layout) - #163135 (librustdoc: remove stale dep on base64) - #163146 (tests: Update `f16b` codegen test for LoongArch and RISC-V) - #163159 (treat inductive cycles as ambig)

Rollup merge of #163099 - maxdexh:int-from-str-loop-unchecked, r=Darksonn Use wrapping arithmetic in `from_str_radix` Uses wrapping arithmetic in the fast/unchecked loop of `from_str_radix`. Originally this PR was written using unchecked arithmetic: This generates identical assembly when compiling with `-O`: https://godbolt.org/z/Gvb7G9YEM However, it significantly simplifies the asm of `-O -Coverflow-checks` (e.g. RfL): https://godbolt.org/z/Pah9fMfrd More precisely, the `-O -Coverflow-checks` output becomes identical to the `-O` one: https://godbolt.org/z/WoYMThP8e In conclusion, it makes more sense to use wrapping operations here.

…d, r=Darksonn Use wrapping arithmetic in `from_str_radix` Uses wrapping arithmetic in the fast/unchecked loop of `from_str_radix`. Originally this PR was written using unchecked arithmetic: This generates identical assembly when compiling with `-O`: https://godbolt.org/z/Gvb7G9YEM However, it significantly simplifies the asm of `-O -Coverflow-checks` (e.g. RfL): https://godbolt.org/z/Pah9fMfrd More precisely, the `-O -Coverflow-checks` output becomes identical to the `-O` one: https://godbolt.org/z/WoYMThP8e In conclusion, it makes more sense to use wrapping operations here.

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 Sep 21, 2026

This comment has been minimized.

Sign in to view

rust-bors Bot pushed a commit that referenced this pull request Sep 21, 2026

Auto merge of #163099 - maxdexh:int-from-str-loop-unchecked, r=<try>

1d6776a

[EXPERIMENT] Use unchecked arithmetic in `from_str_radix`

rustbot added the T-libs Relevant to the library team, which will review and decide on the PR/issue. label Sep 21, 2026

Darksonn closed this Sep 21, 2026

Darksonn reopened this Sep 21, 2026

rustbot added the S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. label Sep 21, 2026

This comment has been minimized.

Sign in to view

maxdexh force-pushed the int-from-str-loop-unchecked branch from eca2552 to 2e0c4ed Compare September 21, 2026 10:36

rustbot removed the S-waiting-on-perf Status: Waiting on a perf run to be completed. label Sep 21, 2026

maxdexh changed the title ~~[EXPERIMENT] Use unchecked arithmetic in from_str_radix~~ Use unchecked arithmetic in from_str_radix Sep 21, 2026

rustbot assigned Darksonn Sep 21, 2026

maxdexh marked this pull request as ready for review September 21, 2026 11:11

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 Sep 21, 2026

rustbot added the I-libs-nominated Nominated for discussion during a libs team meeting. label Sep 21, 2026

Darksonn removed the S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. label Sep 21, 2026

Darksonn reviewed Sep 21, 2026

View reviewed changes

Comment thread library/core/src/num/mod.rs Outdated

maxdexh force-pushed the int-from-str-loop-unchecked branch from 2e0c4ed to 75f50bf Compare September 21, 2026 11:25

maxdexh changed the title ~~Use unchecked arithmetic in from_str_radix~~ Use wrapping arithmetic in from_str_radix Sep 22, 2026

maxdexh force-pushed the int-from-str-loop-unchecked branch from 75f50bf to 3e43dc9 Compare September 22, 2026 17:16

clarfonthey added S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. and removed I-libs-nominated Nominated for discussion during a libs team meeting. S-waiting-on-t-libs Status: Awaiting decision from T-libs labels Sep 22, 2026

clarfonthey reviewed Sep 22, 2026

View reviewed changes

Comment thread library/core/src/num/mod.rs Outdated

Use wrapping arithmetic in from_str_radix

84a341b

maxdexh force-pushed the int-from-str-loop-unchecked branch from 3e43dc9 to 84a341b Compare September 22, 2026 18:25

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 Sep 22, 2026

Darksonn approved these changes Sep 22, 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 Sep 22, 2026

JonathanBrouwer mentioned this pull request Sep 22, 2026

Rollup of 15 pull requests #163175

Closed

tgross35 reviewed Sep 22, 2026

View reviewed changes

JonathanBrouwer mentioned this pull request Sep 22, 2026

Rollup of 13 pull requests #163178

Merged

rust-bors Bot merged commit fa0d183 into rust-lang:main Sep 23, 2026
13 checks passed

rustbot added this to the 1.100.0 milestone Sep 23, 2026

maxdexh deleted the int-from-str-loop-unchecked branch September 23, 2026 06:35

github-actions Bot mentioned this pull request Oct 2, 2026

Toolchain upgrade to nightly-2026-09-25 flux-rs/flux#1803

Merged

Uh oh!

Conversation

maxdexh commented Sep 21, 2026 • edited by rustbot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Darksonn commented Sep 21, 2026

Uh oh!

This comment has been minimized.

maxdexh commented Sep 21, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Darksonn commented Sep 21, 2026

Uh oh!

This comment has been minimized.

maxdexh commented Sep 21, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

rustbot commented Sep 21, 2026

Uh oh!

Darksonn commented Sep 21, 2026

Uh oh!

maxdexh commented Sep 21, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

rust-bors Bot commented Sep 21, 2026

Uh oh!

This comment has been minimized.

maxdexh commented Sep 21, 2026

Uh oh!

rust-timer commented Sep 21, 2026

Overall result: ✅ improvements - no action needed

Instruction count

Max RSS (memory usage)

Cycles

Binary size

Uh oh!

maxdexh commented Sep 21, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Darksonn commented Sep 21, 2026

Uh oh!

Uh oh!

clarfonthey commented Sep 22, 2026

Uh oh!

maxdexh commented Sep 22, 2026

Uh oh!

Uh oh!

clarfonthey commented Sep 22, 2026

Uh oh!

Darksonn 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!

rust-bors Bot commented Sep 22, 2026

Uh oh!

tgross35 Sep 22, 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!

maxdexh Sep 22, 2026 • edited 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!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants

maxdexh commented Sep 21, 2026 •

edited by rustbot

Loading

maxdexh commented Sep 21, 2026 •

edited

Loading

maxdexh commented Sep 21, 2026 •

edited

Loading

maxdexh commented Sep 21, 2026 •

edited

Loading

maxdexh commented Sep 21, 2026 •

edited

Loading

Darksonn left a comment •

edited by rustbot

Loading

tgross35 Sep 22, 2026 •

edited by rustbot

Loading

maxdexh Sep 22, 2026 •

edited

Loading