Rename `assert_uninit_valid` intrinsic #105613

Noratrieb · 2022-12-12T14:52:59Z

It's not about "uninit" anymore but about "filling with 0x01 bytes" so the name should at least try to reflect that.

This is actually not fully correct though, as it does still panic for all uninit with -Zstrict-init-checks. I'm not sure what the best way is to deal with that not causing confusion. I guess we could just remove the flag? I don't think having it makes a lot of sense anymore with the direction that we have chose to go. It could be relevant again if #100423 lands so removing it may be a bit over eager.

r? @RalfJung

rustbot · 2022-12-12T14:53:07Z

Some changes occurred in compiler/rustc_codegen_cranelift

cc @bjorn3

Hey! It looks like you've submitted a new PR for the library teams!

If this PR contains changes to any rust-lang/rust public library APIs then please comment with @rustbot label +T-libs-api -T-libs to tag it appropriately. If this PR contains changes to any unstable APIs please edit the PR description to add a link to the relevant API Change Proposal or create one if you haven't already. If you're unsure where your change falls no worries, just leave it as is and the reviewer will take a look and make a decision to forward on if necessary.

Examples of T-libs-api changes:

Stabilizing library features
Introducing insta-stable changes such as new implementations of existing stable traits on existing stable types
Introducing new or changing existing unstable library APIs (excluding permanently unstable features / features without a tracking issue)
Changing public documentation in ways that create new stability guarantees
Changing observable runtime behavior of library APIs

Some changes occurred to the CTFE / Miri engine

cc @rust-lang/miri

bjorn3 · 2022-12-12T17:09:28Z

library/core/src/intrinsics.rs

    ///
    /// This intrinsic does not have a stable counterpart.
    #[rustc_const_unstable(feature = "const_assert_type2", issue = "none")]
    #[rustc_safe_intrinsic]
-    pub fn assert_uninit_valid<T>();
+    #[cfg(not(bootstrap))]
+    pub fn assert_ones_valid<T>();


This reads to me like is about if all bytes being 0b11111111 is valid.

Additionally it still seems to check uninit, not 0x01.

Yeah, I think the better approach here is probably to document and rename to match what this intrinsic is for: Checking if it is valid to call mem::uninitialized::<T>(). This whole thing is a wart on a wart, and we should probably just own that this is so special-cased.

Calling it ones_valid still leads to confusion, because the behavior changes to something else with -Zstrict-init-checks.

Yeah, I think the better approach here is probably to document and rename to match what this intrinsic is for: Checking if it is valid to call mem::uninitialized::().

Agreed. So what would be a good name? assert_mem_uninitialized_valid?

Yup, that's what I would go for.

RalfJung · 2022-12-13T10:42:41Z

This is actually not fully correct though, as it does still panic for all uninit with -Zstrict-init-checks. I'm not sure what the best way is to deal with that not causing confusion. I guess we could just remove the flag?

I think that flag is still quite useful, e.g. for cargo careful.

It's not about "uninit" anymore but about "filling with 0x01 bytes" so the name should at least try to reflect that.

saethlin · 2022-12-13T17:58:56Z

library/core/src/intrinsics.rs

@@ -959,13 +959,14 @@ extern "rust-intrinsic" {
    #[rustc_safe_intrinsic]
    pub fn assert_zero_valid<T>();

-    /// A guard for unsafe functions that cannot ever be executed if `T` has invalid
-    /// bit patterns: This will statically either panic, or do nothing.
+    /// A guard for `std::mem::uninitialized`. Checks whether a repeated bit pattern `0x01`


Checks whether a repeated bit pattern 0x01 is legal for T

Except if we decide to change the bit pattern (unlikely) or -Zstrict-init-checks is passed. Yes it's redundant, but it might be better to just say "Checks whether mem::uninitialized should panic for T" 🤷

library/core/src/intrinsics.rs

RalfJung · 2022-12-21T10:06:56Z

@bors r+

bors · 2022-12-21T10:06:58Z

📌 Commit 6f21ba4 has been approved by RalfJung

It is now in the queue for this repository.

…id, r=RalfJung Rename `assert_uninit_valid` intrinsic It's not about "uninit" anymore but about "filling with 0x01 bytes" so the name should at least try to reflect that. This is actually not fully correct though, as it does still panic for all uninit with `-Zstrict-init-checks`. I'm not sure what the best way is to deal with that not causing confusion. I guess we could just remove the flag? I don't think having it makes a lot of sense anymore with the direction that we have chose to go. It could be relevant again if rust-lang#100423 lands so removing it may be a bit over eager. r? `@RalfJung`

…id, r=RalfJung Rename `assert_uninit_valid` intrinsic It's not about "uninit" anymore but about "filling with 0x01 bytes" so the name should at least try to reflect that. This is actually not fully correct though, as it does still panic for all uninit with `-Zstrict-init-checks`. I'm not sure what the best way is to deal with that not causing confusion. I guess we could just remove the flag? I don't think having it makes a lot of sense anymore with the direction that we have chose to go. It could be relevant again if rust-lang#100423 lands so removing it may be a bit over eager. r? ``@RalfJung``

bors · 2022-12-21T23:20:07Z

⌛ Testing commit 6f21ba4 with merge bdbe392...

bors · 2022-12-22T02:16:52Z

☀️ Test successful - checks-actions
Approved by: RalfJung
Pushing bdbe392 to master...

rust-timer · 2022-12-22T03:36:40Z

Finished benchmarking commit (bdbe392): comparison URL.

Overall result: no relevant changes - no action needed

@rustbot label: -perf-regression

Instruction count

This benchmark run did not return any relevant results for this metric.

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.

	mean	range	count
Regressions ❌ (primary)	3.1%	[3.1%, 3.1%]	1
Regressions ❌ (secondary)	2.3%	[2.0%, 3.3%]	14
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	3.1%	[3.1%, 3.1%]	1

Cycles

This benchmark run did not return any relevant results for this metric.

…, r=RalfJung Rename `assert_uninit_valid` intrinsic It's not about "uninit" anymore but about "filling with 0x01 bytes" so the name should at least try to reflect that. This is actually not fully correct though, as it does still panic for all uninit with `-Zstrict-init-checks`. I'm not sure what the best way is to deal with that not causing confusion. I guess we could just remove the flag? I don't think having it makes a lot of sense anymore with the direction that we have chose to go. It could be relevant again if rust-lang#100423 lands so removing it may be a bit over eager. r? `@RalfJung`

This commit fixes compilation errors but not runtime ones. Related changes: rust-lang/rust#104986 rust-lang/rust#105657 rust-lang/rust#105603 rust-lang/rust#105613

…, r=RalfJung Rename `assert_uninit_valid` intrinsic It's not about "uninit" anymore but about "filling with 0x01 bytes" so the name should at least try to reflect that. This is actually not fully correct though, as it does still panic for all uninit with `-Zstrict-init-checks`. I'm not sure what the best way is to deal with that not causing confusion. I guess we could just remove the flag? I don't think having it makes a lot of sense anymore with the direction that we have chose to go. It could be relevant again if rust-lang#100423 lands so removing it may be a bit over eager. r? `@RalfJung`

Upgrade our toolchain to `nightly-2023-01-23`. The changes here are related to the following changes: - rust-lang/rust#104986 - rust-lang/rust#105657 - rust-lang/rust#105603 - rust-lang/rust#105613 - rust-lang/rust#105977 - rust-lang/rust#104645

Update the toolchain to use nightly-2023-04-16. Changes were related to the following changes to the toolchain: - rust-lang/rust#108944 - rust-lang/rust#108148 - rust-lang/rust#108471 - rust-lang/rust#109358 - rust-lang/rust#109849 - rust-lang/rust#109819 - rust-lang/rust#109718 - rust-lang/rust#109092 - rust-lang/rust#108856 - rust-lang/rust#105613 - rust-lang/rust#103042 - rust-lang/rust#109716 - rust-lang/rust#108340 - rust-lang/rust#102906 - rust-lang/rust#98112 - rust-lang/rust#108080

rustbot assigned RalfJung Dec 12, 2022

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. T-libs Relevant to the library team, which will review and decide on the PR/issue. labels Dec 12, 2022

Noratrieb force-pushed the rename-assert_uninit_valid branch from d2220ff to e3537b8 Compare December 12, 2022 14:54

bjorn3 reviewed Dec 12, 2022

View reviewed changes

This comment has been minimized.

Sign in to view

Rename assert_uninit_valid intrinsic

8b2a7da

It's not about "uninit" anymore but about "filling with 0x01 bytes" so the name should at least try to reflect that.

Noratrieb force-pushed the rename-assert_uninit_valid branch from e3537b8 to 8b2a7da Compare December 13, 2022 17:08

saethlin reviewed Dec 13, 2022

View reviewed changes

RalfJung reviewed Dec 21, 2022

View reviewed changes

library/core/src/intrinsics.rs Outdated Show resolved Hide resolved

less specific wording

6f21ba4

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 Dec 21, 2022

fee1-dead mentioned this pull request Dec 21, 2022

Rollup of 17 pull requests #106006

Closed

fee1-dead mentioned this pull request Dec 21, 2022

Rollup of 12 pull requests #106007

Closed

bors added the merged-by-bors This PR was explicitly merged by bors. label Dec 22, 2022

bors merged commit bdbe392 into rust-lang:master Dec 22, 2022

rustbot added this to the 1.68.0 milestone Dec 22, 2022

Noratrieb deleted the rename-assert_uninit_valid branch December 22, 2022 05:24

celinval mentioned this pull request Jan 24, 2023

Upgrade toolchain to nightly-2023-01-23 model-checking/kani#2149

Merged

4 tasks

celinval mentioned this pull request Apr 26, 2023

Upgrade the toolchain to nightly-2023-04-16 model-checking/kani#2406

Merged

4 tasks

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Rename `assert_uninit_valid` intrinsic #105613

Rename `assert_uninit_valid` intrinsic #105613

Noratrieb commented Dec 12, 2022

rustbot commented Dec 12, 2022

bjorn3 Dec 12, 2022

bjorn3 Dec 12, 2022

saethlin Dec 12, 2022 •

edited

Loading

RalfJung Dec 13, 2022

saethlin Dec 13, 2022

This comment has been minimized.

RalfJung commented Dec 13, 2022

saethlin Dec 13, 2022

RalfJung commented Dec 21, 2022

bors commented Dec 21, 2022

bors commented Dec 21, 2022

bors commented Dec 22, 2022

rust-timer commented Dec 22, 2022

Rename assert_uninit_valid intrinsic #105613

Rename assert_uninit_valid intrinsic #105613

Conversation

Noratrieb commented Dec 12, 2022

rustbot commented Dec 12, 2022

Choose a reason for hiding this comment

Choose a reason for hiding this comment

saethlin Dec 12, 2022 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

This comment has been minimized.

RalfJung commented Dec 13, 2022

Choose a reason for hiding this comment

RalfJung commented Dec 21, 2022

bors commented Dec 21, 2022

bors commented Dec 21, 2022

bors commented Dec 22, 2022

rust-timer commented Dec 22, 2022

Overall result: no relevant changes - no action needed

Rename `assert_uninit_valid` intrinsic #105613

Rename `assert_uninit_valid` intrinsic #105613

saethlin Dec 12, 2022 •

edited

Loading