fix unsoundness in Step::forward_unchecked for signed integers #122461
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
rustbot
added
S-waiting-on-review
This comment has been minimized.
This comment has been minimized.
the8472
force-pushed
the
fix-step-forward-unchecked
branch
from
ed3e39f to
f0487c0
Compare
Amanieu
reviewed
Mar 13, 2024
This comment has been minimized.
This comment has been minimized.
the8472
force-pushed
the
fix-step-forward-unchecked
branch
from
f0487c0 to
d3cab9f
Compare
|
The Miri subtree was changed cc @rust-lang/miri |
Amanieu
reviewed
Mar 14, 2024
| @@ -1 +1 @@ | |||
| The loop took around 7s | |||
| The loop took around 12s | |||
There was a problem hiding this comment.
The test contains a loop over a range of unconstrained integer types which defaults to i32.
Aiui in miri "time" ticks based on the number of mir statements executed or something like that. Since the Range impl changed this also affects rate of time.
There was a problem hiding this comment.
Argh this test again. We should probably do something a bit better here.
When isolation is enabled (the default), Miri doesn't have access to the system clock, so we use the end of a basic block as an extremely crude approximation of clock ticks, so that we can have a clock when isolation is enabled. This test asserts that the magic number for NANOSECONDS_PER_BASIC_BLOCK gets us an isolation clock that's remotely close to the real passage of time. Because when we first did this I think it was off by a factor of 100.
This diff is fine for this PR.
There was a problem hiding this comment.
I'm still a bit surprised that the number of basic blocks is almost doubling?
|
@bors r+ |
bors
added
S-waiting-on-bors
20 tasks
matthiaskrgr
added a commit
to matthiaskrgr/rust
that referenced
this pull request
Mar 14, 2024
…, r=Amanieu fix unsoundness in Step::forward_unchecked for signed integers Fixes rust-lang#122420 ```rust pub fn foo(a: i8, b: u8) -> i8 { unsafe { a.checked_add_unsigned(b).unwrap_unchecked() } } ``` still compiles down to a single arithmetic instruction ([godbolt](https://rust.godbolt.org/z/qsd3xYWfE)). But we may be losing some loop optimizations if llvm can no longer easily derive that it's a finite counted loop from the no-wrapping flags.
bors
added a commit
to rust-lang-ci/rust
that referenced
this pull request
Mar 14, 2024
…iaskrgr Rollup of 11 pull requests Successful merges: - rust-lang#104353 (Add CStr::bytes iterator) - rust-lang#114038 (unix time module now return result) - rust-lang#119676 (rustdoc-search: search types by higher-order functions) - rust-lang#120699 (Document `TRACK_DIAGNOSTIC` calls.) - rust-lang#121899 (Document how removing a type's field can be bad and what to do instead) - rust-lang#121940 (Mention Register Size in `#[warn(asm_sub_register)]`) - rust-lang#122397 (Various cleanups around the const eval query providers) - rust-lang#122405 (Add methods to create StableMIR constant) - rust-lang#122416 (Various style improvements to `rustc_lint::levels`) - rust-lang#122440 (const-eval: organize and extend tests for required-consts) - rust-lang#122461 (fix unsoundness in Step::forward_unchecked for signed integers) r? `@ghost` `@rustbot` modify labels: rollup
jhpratt
added a commit
to jhpratt/rust
that referenced
this pull request
Mar 14, 2024
…, r=Amanieu fix unsoundness in Step::forward_unchecked for signed integers Fixes rust-lang#122420 ```rust pub fn foo(a: i8, b: u8) -> i8 { unsafe { a.checked_add_unsigned(b).unwrap_unchecked() } } ``` still compiles down to a single arithmetic instruction ([godbolt](https://rust.godbolt.org/z/qsd3xYWfE)). But we may be losing some loop optimizations if llvm can no longer easily derive that it's a finite counted loop from the no-wrapping flags.
matthiaskrgr
added a commit
to matthiaskrgr/rust
that referenced
this pull request
Mar 14, 2024
…, r=Amanieu fix unsoundness in Step::forward_unchecked for signed integers Fixes rust-lang#122420 ```rust pub fn foo(a: i8, b: u8) -> i8 { unsafe { a.checked_add_unsigned(b).unwrap_unchecked() } } ``` still compiles down to a single arithmetic instruction ([godbolt](https://rust.godbolt.org/z/qsd3xYWfE)). But we may be losing some loop optimizations if llvm can no longer easily derive that it's a finite counted loop from the no-wrapping flags.
This was referenced Mar 14, 2024
bors
added a commit
to rust-lang-ci/rust
that referenced
this pull request
Mar 14, 2024
…iaskrgr Rollup of 9 pull requests Successful merges: - rust-lang#104353 (Add CStr::bytes iterator) - rust-lang#119676 (rustdoc-search: search types by higher-order functions) - rust-lang#120699 (Document `TRACK_DIAGNOSTIC` calls.) - rust-lang#121899 (Document how removing a type's field can be bad and what to do instead) - rust-lang#122405 (Add methods to create StableMIR constant) - rust-lang#122416 (Various style improvements to `rustc_lint::levels`) - rust-lang#122421 (Improve `Step` docs) - rust-lang#122440 (const-eval: organize and extend tests for required-consts) - rust-lang#122461 (fix unsoundness in Step::forward_unchecked for signed integers) Failed merges: - rust-lang#122397 (Various cleanups around the const eval query providers) r? `@ghost` `@rustbot` modify labels: rollup
rust-timer
added a commit
to rust-lang-ci/rust
that referenced
this pull request
Mar 14, 2024
Rollup merge of rust-lang#122461 - the8472:fix-step-forward-unchecked, r=Amanieu fix unsoundness in Step::forward_unchecked for signed integers Fixes rust-lang#122420 ```rust pub fn foo(a: i8, b: u8) -> i8 { unsafe { a.checked_add_unsigned(b).unwrap_unchecked() } } ``` still compiles down to a single arithmetic instruction ([godbolt](https://rust.godbolt.org/z/qsd3xYWfE)). But we may be losing some loop optimizations if llvm can no longer easily derive that it's a finite counted loop from the no-wrapping flags.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fixes #122420
still compiles down to a single arithmetic instruction (godbolt).
But we may be losing some loop optimizations if llvm can no longer easily derive that it's a finite counted loop from the no-wrapping flags.