Do not inline integer literals which are out of range in format_args! #116633

agluszak · 2023-10-11T13:16:18Z

Inlining integers (even those out of range) was introduced in #106824.

Closes #116631.

Closes #115423.

rustbot · 2023-10-11T13:16:28Z

r? @compiler-errors

(rustbot has picked a reviewer for you, use r? to override)

agluszak · 2023-10-11T13:20:45Z

Technically that's a breaking change...

compiler-errors · 2023-10-11T13:22:02Z

Hm yeah. Nominating for T-lang, impl looks ok but I didn't look at it too long -- will do after T-lang confirms we should re-break this.

In the mean time, I'll do a crater run.

@bors try

bors · 2023-10-11T13:23:13Z

⌛ Trying commit b0ae218 with merge b67bddc...

Do not inline integer linterals which are out of range in format_args! Inlining integers (even those out of range) was introduced in rust-lang#106824. Closes rust-lang#116631.

bors · 2023-10-11T14:36:32Z

☀️ Try build successful - checks-actions
Build commit: b67bddc (b67bddc71c8ce18bca43818e26d6290f72650a92)

compiler-errors · 2023-10-11T14:37:05Z

@craterbot check

craterbot · 2023-10-11T14:37:18Z

👌 Experiment pr-116633 created and queued.
🤖 Automatically detected try build b67bddc
🔍 You can check out the queue and this experiment's details.

ℹ️ Crater is a tool to run experiments across parts of the Rust ecosystem. Learn more

craterbot · 2023-10-11T14:37:32Z

🚧 Experiment pr-116633 is now running

ℹ️ Crater is a tool to run experiments across parts of the Rust ecosystem. Learn more

lukas-code · 2023-10-11T15:11:49Z

compiler/rustc_ast/src/ast.rs

+            LitIntType::Signed(IntTy::I32) => i32::MAX as u128,
+            LitIntType::Signed(IntTy::I64) => i64::MAX as u128,
+            LitIntType::Signed(IntTy::I128) => i128::MAX as u128,
+            LitIntType::Signed(IntTy::Isize) => isize::MAX as u128,


This uses the host's isize, but should be using use the target's isize. Same for usize below.

Fair. What should I do to obtain these values for the target?

You can get size of isize and usize from tcx.data_layout.pointer_size. That will return a Size, which has the signed_int_max and unsigned_int_max methods that seem useful. It looks like the nearest tcx is the one in the LoweringContext, which you get as self from here:

rust/compiler/rustc_ast_lowering/src/format.rs

Line 15 in b0ae218

pub(crate) fn lower_format_args(&mut self, sp: Span, fmt: &FormatArgs) -> hir::ExprKind<'hir> {

lukas-code · 2023-10-11T15:18:24Z

compiler/rustc_ast/src/ast.rs

+            LitIntType::Unsigned(UintTy::U64) => u64::MAX as u128,
+            LitIntType::Unsigned(UintTy::U128) => u128::MAX,
+            LitIntType::Unsigned(UintTy::Usize) => usize::MAX as u128,
+            LitIntType::Unsuffixed => u128::MAX,


Suggested change

LitIntType::Unsuffixed => u128::MAX,

LitIntType::Unsuffixed => i32::MAX,

Unsuffixed and unconstrained integer literals should have type i32, even if they overflow.

See the old behavior before inlining: https://rust.godbolt.org/z/PMG7ac4d4

asquared31415 · 2023-10-11T17:53:38Z

tests/ui/fmt/issue-116631.rs

@@ -0,0 +1,3 @@
+fn main() {
+    format_args!("{}\n", 0xffff_ffff_u8); //~ ERROR literal out of range for `u8`


Should this have a test for each of the different integer types?

At the very least it should have a test case that covers #115423 (which I don't interpet is about being out of range, but rather about not respecting the type of the literal).

asquared31415 · 2023-10-11T17:54:23Z

According to the linked PR, it's meant to be gated behind -Zflatten-format-args=yes, is that gating broken somehow?

Edit: ah that flag seems to have been stabilized to default to on, I didn't see that

craterbot · 2023-10-13T09:44:33Z

🎉 Experiment pr-116633 is completed!
📊 13 regressed and 0 fixed (376984 total)
📰 Open the full report.

⚠️ If you notice any spurious failure please add them to the blacklist!
ℹ️ Crater is a tool to run experiments across parts of the Rust ecosystem. Learn more

scottmcm · 2023-10-18T14:36:14Z

We discussed this in the @rust-lang/lang triage meeting today.

Everyone agreed that this was "just" a bug, and should be fixed. Printing a token that's explicitly marked an i8 and getting a result outside the range of an i8 is clearly wrong.

Ideally this would trigger the overflowing_literals lint same as using a token like this as an expression in other places.

traviscross · 2023-10-24T17:47:40Z

@rustbot labels -I-lang-nominated

compiler-errors · 2023-10-25T15:21:43Z

Awaiting feedback above

@rustbot author

agluszak · 2023-10-26T22:03:57Z

I'll fix it on Saturday or Sunday

JohnCSimon · 2024-02-11T20:26:02Z

@agluszak

ping from triage - can you post your status on this PR? This PR has not received an update in a few months. Thank you!

FYI: when a PR is ready for review, send a message containing
@rustbot ready to switch to S-waiting-on-review so the PR is in the reviewer's backlog.

Dylan-DPC · 2024-04-10T13:08:57Z

Closing this as inactive. Feel free to reöpen this pr or create a new pr if you get the time to work on this. Thanks

Don't inline integer literals when they overflow - new attempt Basically rust-lang#116633 but I implemented the suggested changes. Fixes rust-lang#115423. Fixes rust-lang#116631. This is my first contribution to this repo so please let me know if I'm supposed to change something :)

Rollup merge of rust-lang#123935 - tstsrt:fix-115423, r=oli-obk Don't inline integer literals when they overflow - new attempt Basically rust-lang#116633 but I implemented the suggested changes. Fixes rust-lang#115423. Fixes rust-lang#116631. This is my first contribution to this repo so please let me know if I'm supposed to change something :)

Fix rust-lang#116631

b0ae218

rustbot assigned compiler-errors Oct 11, 2023

rustbot added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. T-compiler Relevant to the compiler team, which will review and decide on the PR/issue. labels Oct 11, 2023

compiler-errors added the I-lang-nominated Nominated for discussion during a lang team meeting. label Oct 11, 2023

craterbot added S-waiting-on-crater Status: Waiting on a crater run to be completed. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels Oct 11, 2023

lukas-code reviewed Oct 11, 2023

View reviewed changes

apiraino mentioned this pull request Oct 11, 2023

format_args! does not respect integer literals' type #115423

Closed

asquared31415 reviewed Oct 11, 2023

View reviewed changes

agluszak changed the title ~~Do not inline integer linterals which are out of range in format_args!~~ Do not inline integer literals which are out of range in format_args! Oct 11, 2023

craterbot added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. and removed S-waiting-on-crater Status: Waiting on a crater run to be completed. labels Oct 13, 2023

rustbot removed the I-lang-nominated Nominated for discussion during a lang team meeting. label Oct 24, 2023

rustbot 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 25, 2023

WaffleLapkin mentioned this pull request Dec 6, 2023

Don't throw away unused arguments of format_args #118659

Closed

Dylan-DPC closed this Apr 10, 2024

tstsrt mentioned this pull request Apr 14, 2024

Don't inline integer literals when they overflow - new attempt #123935

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Do not inline integer literals which are out of range in format_args! #116633

Do not inline integer literals which are out of range in format_args! #116633

agluszak commented Oct 11, 2023 •

edited by apiraino

Loading

rustbot commented Oct 11, 2023

agluszak commented Oct 11, 2023

compiler-errors commented Oct 11, 2023

bors commented Oct 11, 2023

bors commented Oct 11, 2023

compiler-errors commented Oct 11, 2023

craterbot commented Oct 11, 2023

craterbot commented Oct 11, 2023

lukas-code Oct 11, 2023

agluszak Oct 11, 2023

lukas-code Oct 11, 2023 •

edited

Loading

lukas-code Oct 11, 2023

asquared31415 Oct 11, 2023

pnkfelix Oct 18, 2023

asquared31415 commented Oct 11, 2023 •

edited

Loading

craterbot commented Oct 13, 2023

scottmcm commented Oct 18, 2023

traviscross commented Oct 24, 2023

compiler-errors commented Oct 25, 2023

agluszak commented Oct 26, 2023

JohnCSimon commented Feb 11, 2024

Dylan-DPC commented Apr 10, 2024

	LitIntType::Unsuffixed => u128::MAX,
	LitIntType::Unsuffixed => i32::MAX,

		@@ -0,0 +1,3 @@
		fn main() {
		format_args!("{}\n", 0xffff_ffff_u8); //~ ERROR literal out of range for `u8`

Do not inline integer literals which are out of range in format_args! #116633

Do not inline integer literals which are out of range in format_args! #116633

Conversation

agluszak commented Oct 11, 2023 • edited by apiraino Loading

rustbot commented Oct 11, 2023

agluszak commented Oct 11, 2023

compiler-errors commented Oct 11, 2023

bors commented Oct 11, 2023

bors commented Oct 11, 2023

compiler-errors commented Oct 11, 2023

craterbot commented Oct 11, 2023

craterbot commented Oct 11, 2023

lukas-code Oct 11, 2023

Choose a reason for hiding this comment

agluszak Oct 11, 2023

Choose a reason for hiding this comment

lukas-code Oct 11, 2023 • edited Loading

Choose a reason for hiding this comment

lukas-code Oct 11, 2023

Choose a reason for hiding this comment

asquared31415 Oct 11, 2023

Choose a reason for hiding this comment

pnkfelix Oct 18, 2023

Choose a reason for hiding this comment

asquared31415 commented Oct 11, 2023 • edited Loading

craterbot commented Oct 13, 2023

scottmcm commented Oct 18, 2023

traviscross commented Oct 24, 2023

compiler-errors commented Oct 25, 2023

agluszak commented Oct 26, 2023

JohnCSimon commented Feb 11, 2024

Dylan-DPC commented Apr 10, 2024

agluszak commented Oct 11, 2023 •

edited by apiraino

Loading

lukas-code Oct 11, 2023 •

edited

Loading

asquared31415 commented Oct 11, 2023 •

edited

Loading