Miri: handling of SNaN inputs in `f*::pow` operations by LorrensP-2158466 · Pull Request #142514 · rust-lang/rust

LorrensP-2158466 · 2025-06-14T16:14:35Z

fixes miri/#4286 and related to #138062 and miri/#4208.

For the following cases of the powf or powi operations, Miri returns either 1.0 or an arbitrary NaN:

powf(SNaN, 0.0)
powf(1.0, SNaN)
powi(SNaN, 0)

Also added a macro in miri/tests/pass/float.rs which conveniently checks if both are indeed returned from such an operation.

Made these changes in the rust repo so I could test against stdlib, since these were impacted some time ago and were fixed in #138062. Tested with:

env MIRIFLAGS=-Zmiri-many-seeds ./x miri --no-fail-fast std core coretests -- f32 f64

This was successful. This does take a while, so I recommend using --no-doc and separate use of f32 or f64

The pr is somewhat split up into 3 main commits, which implement the cases described above. The first commit also introduces the macro, and the last commit is just a global refactor of some things.

r? @RalfJung

rustbot · 2025-06-14T16:14:40Z

The Miri subtree was changed

cc @rust-lang/miri

RalfJung · 2025-06-27T14:00:04Z

src/tools/miri/src/intrinsics/mod.rs

+fn random_nan<S: Semantics>(rng: &mut StdRng) -> IeeeFloat<S> {
+    if rng.random() {
+        IeeeFloat::<S>::snan(Some(rng.random()))
+    } else {
+        IeeeFloat::<S>::qnan(Some(rng.random()))
+    }
+}


We already have extensive logic for random NaNs (generate_nan), please use that instead.

oh yeah 🤦

RalfJung · 2025-06-27T14:02:54Z

src/tools/miri/src/intrinsics/mod.rs

+        ("powf32" | "powf64", [base, exp]) if *base == one => {
+            let rng = ecx.machine.rng.get_mut();
+            // Handle both the musl and glibc cases non-deterministically.
+            if !exp.is_signaling() || rng.random() { one } else { random_nan(rng) }


This is too compact to be readable. Also, this needs to honor the (new) float_nondet flag.

let return_nan = exp.is_signaling() && ecx.machine.float_nondet && rng.random(); if return_nan { ...

src/tools/miri/src/intrinsics/mod.rs

RalfJung · 2025-06-27T14:06:20Z

src/tools/miri/tests/pass/float.rs

+    }
+
+    // x^(SNaN) = (1 | NaN)
+    test_snan_nondet!(f32::powf(SNAN_F32, 0.0));


Please don't invent new logic, but use existing logic. We already have ensure_nondet.

RalfJung · 2025-06-27T14:06:56Z

Sorry for the slow review, I was traveling. And due to an upcoming deadline I'll be quite busy for another 2 weeks, unfortunately.

RalfJung · 2025-06-27T14:18:11Z

@rustbot author

rustbot · 2025-06-27T14:18:15Z

Reminder, once the PR becomes ready for a review, use @rustbot ready.

RalfJung · 2025-06-27T20:19:22Z

@bors2 try jobs=x86_64-gnu-aux

rust-bors · 2025-06-27T20:19:27Z

⌛ Trying commit 550a13b with merge 1dc8c00…

To cancel the try build, run the command @bors2 try cancel.

Miri: handling of SNaN inputs in `f*::pow` operations    fixes [miri/#4286](rust-lang/miri#4286) and related to #138062 and [miri/#4208](rust-lang/miri#4208 (comment)). For the following cases of the powf or powi operations, Miri returns either `1.0` or an arbitrary `NaN`: - `powf(SNaN, 0.0)` - `powf(1.0, SNaN)` - `powi(SNaN, 0)` Also added a macro in `miri/tests/pass/float.rs` which conveniently checks if both are indeed returned from such an operation. Made these changes in the rust repo so I could test against stdlib, since these were impacted some time ago and were fixed in #138062. Tested with: ```fish env MIRIFLAGS=-Zmiri-many-seeds ./x miri --no-fail-fast std core coretests -- f32 f64 ``` This was successful. This does take a while, so I recommend using `--no-doc` and separate use of `f32` or `f64` The pr is somewhat split up into 3 main commits, which implement the cases described above. The first commit also introduces the macro, and the last commit is just a global refactor of some things. r? `@RalfJung` try-job: x86_64-gnu-aux

rust-bors · 2025-06-27T22:09:03Z

☀️ Try build successful (CI)
Build commit: 1dc8c00 (1dc8c007860b211c121063d9186c775b10a289d8, parent: fe5f3dedf7b4d6bea2cadb17343f747d70b4c66b)

LorrensP-2158466 · 2025-06-29T14:27:14Z

Sorry for the slow review, I was traveling. And due to an upcoming deadline I'll be quite busy for another 2 weeks, unfortunately.

No worries! I had exams, so I couldn't respond to the review if it had been done sooner.

LorrensP-2158466 · 2025-06-29T15:55:31Z

@rustbot ready

src/tools/miri/src/intrinsics/mod.rs

src/tools/miri/tests/pass/float.rs

RalfJung · 2025-06-29T20:27:43Z

Thanks! I added a commit with some more tests. Should be good to go now :)
@bors r+ rollup

bors · 2025-06-29T20:27:45Z

📌 Commit 06dd8bc has been approved by RalfJung

It is now in the queue for this repository.

plus various minor tweaks

RalfJung · 2025-06-29T20:32:48Z

@bors r+

bors · 2025-06-29T20:32:51Z

📌 Commit 67ab61e has been approved by RalfJung

It is now in the queue for this repository.

Rollup of 14 pull requests Successful merges: - #142429 (`tests/ui`: A New Order [13/N]) - #142514 (Miri: handling of SNaN inputs in `f*::pow` operations) - #143066 (Use let chains in the new solver) - #143090 (Workaround for memory unsafety in third party DLLs) - #143118 (`tests/ui`: A New Order [15/N]) - #143159 (Do not freshen `ReError`) - #143168 (`tests/ui`: A New Order [16/N]) - #143176 (fix typos and improve clarity in documentation) - #143187 (Add my work email to mailmap) - #143190 (Use the `new` method for `BasicBlockData` and `Statement`) - #143195 (`tests/ui`: A New Order [17/N]) - #143196 (Port #[link_section] to the new attribute parsing infrastructure) - #143199 (Re-disable `tests/run-make/short-ice` on Windows MSVC again) - #143219 (Show auto trait and blanket impls for `!`) r? `@ghost` `@rustbot` modify labels: rollup

LorrensP-2158466 · 2025-06-30T15:57:00Z

Thanks! I added a commit with some more tests. Should be good to go now :)

Oh wow, thanks!

Rollup merge of #142514 - LorrensP-2158466:miri-float-nondet-pow, r=RalfJung Miri: handling of SNaN inputs in `f*::pow` operations fixes [miri/#4286](rust-lang/miri#4286) and related to #138062 and [miri/#4208](rust-lang/miri#4208 (comment)). For the following cases of the powf or powi operations, Miri returns either `1.0` or an arbitrary `NaN`: - `powf(SNaN, 0.0)` - `powf(1.0, SNaN)` - `powi(SNaN, 0)` Also added a macro in `miri/tests/pass/float.rs` which conveniently checks if both are indeed returned from such an operation. Made these changes in the rust repo so I could test against stdlib, since these were impacted some time ago and were fixed in #138062. Tested with: ```fish env MIRIFLAGS=-Zmiri-many-seeds ./x miri --no-fail-fast std core coretests -- f32 f64 ``` This was successful. This does take a while, so I recommend using `--no-doc` and separate use of `f32` or `f64` The pr is somewhat split up into 3 main commits, which implement the cases described above. The first commit also introduces the macro, and the last commit is just a global refactor of some things. r? `@RalfJung`

Rollup of 14 pull requests Successful merges: - rust-lang/rust#142429 (`tests/ui`: A New Order [13/N]) - rust-lang/rust#142514 (Miri: handling of SNaN inputs in `f*::pow` operations) - rust-lang/rust#143066 (Use let chains in the new solver) - rust-lang/rust#143090 (Workaround for memory unsafety in third party DLLs) - rust-lang/rust#143118 (`tests/ui`: A New Order [15/N]) - rust-lang/rust#143159 (Do not freshen `ReError`) - rust-lang/rust#143168 (`tests/ui`: A New Order [16/N]) - rust-lang/rust#143176 (fix typos and improve clarity in documentation) - rust-lang/rust#143187 (Add my work email to mailmap) - rust-lang/rust#143190 (Use the `new` method for `BasicBlockData` and `Statement`) - rust-lang/rust#143195 (`tests/ui`: A New Order [17/N]) - rust-lang/rust#143196 (Port #[link_section] to the new attribute parsing infrastructure) - rust-lang/rust#143199 (Re-disable `tests/run-make/short-ice` on Windows MSVC again) - rust-lang/rust#143219 (Show auto trait and blanket impls for `!`) r? `@ghost` `@rustbot` modify labels: rollup

Rollup of 14 pull requests Successful merges: - rust-lang#142429 (`tests/ui`: A New Order [13/N]) - rust-lang#142514 (Miri: handling of SNaN inputs in `f*::pow` operations) - rust-lang#143066 (Use let chains in the new solver) - rust-lang#143090 (Workaround for memory unsafety in third party DLLs) - rust-lang#143118 (`tests/ui`: A New Order [15/N]) - rust-lang#143159 (Do not freshen `ReError`) - rust-lang#143168 (`tests/ui`: A New Order [16/N]) - rust-lang#143176 (fix typos and improve clarity in documentation) - rust-lang#143187 (Add my work email to mailmap) - rust-lang#143190 (Use the `new` method for `BasicBlockData` and `Statement`) - rust-lang#143195 (`tests/ui`: A New Order [17/N]) - rust-lang#143196 (Port #[link_section] to the new attribute parsing infrastructure) - rust-lang#143199 (Re-disable `tests/run-make/short-ice` on Windows MSVC again) - rust-lang#143219 (Show auto trait and blanket impls for `!`) r? `@ghost` `@rustbot` modify labels: rollup

…-foreign-items, r=RalfJung Miri: non-deterministic floating point operations in `foreign_items` Part of [rust-lang/miri/rust-lang#3555](rust-lang/miri#3555 (comment)), this pr does the `foreign_items` work. Some things have changed since rust-lang#138062 and rust-lang#142514. I moved the "helpers" used for creating fixed outputs and clamping operations to their defined ranges to `helpers.rs`. These are now also extended to handle the floating-point operations in `foreign_items`. Tests in `miri/tests/float.rs` were changed/added. Failing tests in `std` were extracted, run under miri with `-Zmiri-many-seeds=0..1000` and changed accordingly. Double checked with `-Zmiri-many-seeds`. I noticed that the C standard doesn't specify the output ranges for all of its mathematical operations; it just specifies them as: ``` Returns The sinh functions return sinh x. ``` So I used [Wolfram|Alpha](https://www.wolframalpha.com/).

…-foreign-items, r=RalfJung Miri: non-deterministic floating point operations in `foreign_items` Part of [rust-lang/miri/rust-lang#3555](rust-lang/miri#3555 (comment)), this pr does the `foreign_items` work. Some things have changed since rust-lang#138062 and rust-lang#142514. I moved the "helpers" used for creating fixed outputs and clamping operations to their defined ranges to `math.rs`. These are now also extended to handle the floating-point operations in `foreign_items`. Tests in `miri/tests/float.rs` were changed/added. Failing tests in `std` were extracted, run under miri with `-Zmiri-many-seeds=0..1000` and changed accordingly. Double checked with `-Zmiri-many-seeds`. I noticed that the C standard doesn't specify the output ranges for all of its mathematical operations; it just specifies them as: ``` Returns The sinh functions return sinh x. ``` So I used [Wolfram|Alpha](https://www.wolframalpha.com/).

Rollup merge of #143906 - LorrensP-2158466:miri-float-nondet-foreign-items, r=RalfJung Miri: non-deterministic floating point operations in `foreign_items` Part of [rust-lang/miri/#3555](rust-lang/miri#3555 (comment)), this pr does the `foreign_items` work. Some things have changed since #138062 and #142514. I moved the "helpers" used for creating fixed outputs and clamping operations to their defined ranges to `math.rs`. These are now also extended to handle the floating-point operations in `foreign_items`. Tests in `miri/tests/float.rs` were changed/added. Failing tests in `std` were extracted, run under miri with `-Zmiri-many-seeds=0..1000` and changed accordingly. Double checked with `-Zmiri-many-seeds`. I noticed that the C standard doesn't specify the output ranges for all of its mathematical operations; it just specifies them as: ``` Returns The sinh functions return sinh x. ``` So I used [Wolfram|Alpha](https://www.wolframalpha.com/).

…items, r=RalfJung Miri: non-deterministic floating point operations in `foreign_items` Part of [rust-lang/miri/#3555](#3555 (comment)), this pr does the `foreign_items` work. Some things have changed since rust-lang/rust#138062 and rust-lang/rust#142514. I moved the "helpers" used for creating fixed outputs and clamping operations to their defined ranges to `math.rs`. These are now also extended to handle the floating-point operations in `foreign_items`. Tests in `miri/tests/float.rs` were changed/added. Failing tests in `std` were extracted, run under miri with `-Zmiri-many-seeds=0..1000` and changed accordingly. Double checked with `-Zmiri-many-seeds`. I noticed that the C standard doesn't specify the output ranges for all of its mathematical operations; it just specifies them as: ``` Returns The sinh functions return sinh x. ``` So I used [Wolfram|Alpha](https://www.wolframalpha.com/).

…items, r=RalfJung Miri: non-deterministic floating point operations in `foreign_items` Part of [rust-lang/miri/#3555](rust-lang/miri#3555 (comment)), this pr does the `foreign_items` work. Some things have changed since rust-lang/rust#138062 and rust-lang/rust#142514. I moved the "helpers" used for creating fixed outputs and clamping operations to their defined ranges to `math.rs`. These are now also extended to handle the floating-point operations in `foreign_items`. Tests in `miri/tests/float.rs` were changed/added. Failing tests in `std` were extracted, run under miri with `-Zmiri-many-seeds=0..1000` and changed accordingly. Double checked with `-Zmiri-many-seeds`. I noticed that the C standard doesn't specify the output ranges for all of its mathematical operations; it just specifies them as: ``` Returns The sinh functions return sinh x. ``` So I used [Wolfram|Alpha](https://www.wolframalpha.com/).

…-foreign-items, r=RalfJung Miri: non-deterministic floating point operations in `foreign_items` Part of [rust-lang/miri/rust-lang#3555](rust-lang/miri#3555 (comment)), this pr does the `foreign_items` work. Some things have changed since rust-lang#138062 and rust-lang#142514. I moved the "helpers" used for creating fixed outputs and clamping operations to their defined ranges to `math.rs`. These are now also extended to handle the floating-point operations in `foreign_items`. Tests in `miri/tests/float.rs` were changed/added. Failing tests in `std` were extracted, run under miri with `-Zmiri-many-seeds=0..1000` and changed accordingly. Double checked with `-Zmiri-many-seeds`. I noticed that the C standard doesn't specify the output ranges for all of its mathematical operations; it just specifies them as: ``` Returns The sinh functions return sinh x. ``` So I used [Wolfram|Alpha](https://www.wolframalpha.com/).

…items, r=RalfJung Miri: non-deterministic floating point operations in `foreign_items` Part of [rust-lang/miri/#3555](rust-lang/miri#3555 (comment)), this pr does the `foreign_items` work. Some things have changed since rust-lang/rust#138062 and rust-lang/rust#142514. I moved the "helpers" used for creating fixed outputs and clamping operations to their defined ranges to `math.rs`. These are now also extended to handle the floating-point operations in `foreign_items`. Tests in `miri/tests/float.rs` were changed/added. Failing tests in `std` were extracted, run under miri with `-Zmiri-many-seeds=0..1000` and changed accordingly. Double checked with `-Zmiri-many-seeds`. I noticed that the C standard doesn't specify the output ranges for all of its mathematical operations; it just specifies them as: ``` Returns The sinh functions return sinh x. ``` So I used [Wolfram|Alpha](https://www.wolframalpha.com/).

Rollup of 14 pull requests Successful merges: - rust-lang/rust#142429 (`tests/ui`: A New Order [13/N]) - rust-lang/rust#142514 (Miri: handling of SNaN inputs in `f*::pow` operations) - rust-lang/rust#143066 (Use let chains in the new solver) - rust-lang/rust#143090 (Workaround for memory unsafety in third party DLLs) - rust-lang/rust#143118 (`tests/ui`: A New Order [15/N]) - rust-lang/rust#143159 (Do not freshen `ReError`) - rust-lang/rust#143168 (`tests/ui`: A New Order [16/N]) - rust-lang/rust#143176 (fix typos and improve clarity in documentation) - rust-lang/rust#143187 (Add my work email to mailmap) - rust-lang/rust#143190 (Use the `new` method for `BasicBlockData` and `Statement`) - rust-lang/rust#143195 (`tests/ui`: A New Order [17/N]) - rust-lang/rust#143196 (Port #[link_section] to the new attribute parsing infrastructure) - rust-lang/rust#143199 (Re-disable `tests/run-make/short-ice` on Windows MSVC again) - rust-lang/rust#143219 (Show auto trait and blanket impls for `!`) r? `@ghost` `@rustbot` modify labels: rollup

rustbot assigned RalfJung Jun 14, 2025

rustbot added the S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. label Jun 14, 2025

RalfJung reviewed Jun 27, 2025

View reviewed changes

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 Jun 27, 2025

LorrensP-2158466 force-pushed the miri-float-nondet-pow branch from 550a13b to f49980c Compare June 29, 2025 15:55

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 Jun 29, 2025

RalfJung reviewed Jun 29, 2025

View reviewed changes

make some powf and powi cases involving SNaN non-deterministic

60a4828

RalfJung force-pushed the miri-float-nondet-pow branch from 6718b82 to 06dd8bc Compare June 29, 2025 20:26

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 Jun 29, 2025

RalfJung force-pushed the miri-float-nondet-pow branch from 06dd8bc to e059457 Compare June 29, 2025 20:31

add float_nan test for powf

67ab61e

plus various minor tweaks

RalfJung force-pushed the miri-float-nondet-pow branch from e059457 to 67ab61e Compare June 29, 2025 20:32

dianqk mentioned this pull request Jun 30, 2025

Rollup of 14 pull requests #143233

Merged

bors merged commit d2dc99c into rust-lang:master Jun 30, 2025
10 checks passed

rustbot added this to the 1.90.0 milestone Jun 30, 2025

LorrensP-2158466 deleted the miri-float-nondet-pow branch July 13, 2025 10:49

LorrensP-2158466 mentioned this pull request Jul 13, 2025

Miri: non-deterministic floating point operations in foreign_items #143906

Merged

Uh oh!

Conversation

LorrensP-2158466 commented Jun 14, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

rustbot commented Jun 14, 2025

Uh oh!

RalfJung Jun 27, 2025

Choose a reason for hiding this comment

Uh oh!

LorrensP-2158466 Jun 29, 2025

Choose a reason for hiding this comment

Uh oh!

RalfJung Jun 27, 2025

Choose a reason for hiding this comment

Uh oh!

Uh oh!

RalfJung Jun 27, 2025

Choose a reason for hiding this comment

Uh oh!

RalfJung commented Jun 27, 2025

Uh oh!

RalfJung commented Jun 27, 2025

Uh oh!

rustbot commented Jun 27, 2025

Uh oh!

RalfJung commented Jun 27, 2025

Uh oh!

rust-bors bot commented Jun 27, 2025

Uh oh!

rust-bors bot commented Jun 27, 2025

Uh oh!

LorrensP-2158466 commented Jun 29, 2025

Uh oh!

LorrensP-2158466 commented Jun 29, 2025

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

RalfJung commented Jun 29, 2025

Uh oh!

bors commented Jun 29, 2025

Uh oh!

RalfJung commented Jun 29, 2025

Uh oh!

bors commented Jun 29, 2025

Uh oh!

LorrensP-2158466 commented Jun 30, 2025

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

LorrensP-2158466 commented Jun 14, 2025 •

edited

Loading