Add `noundef` metadata for fits-in-target-pointer-size array immediate arguments by jieyouxu · Pull Request #123425 · rust-lang/rust

jieyouxu · 2024-04-03T18:33:44Z

noundef is only added if the small array immediate fits in the target pointer size and if optimizations are enabled.

rustbot · 2024-04-03T18:33:52Z

rustbot has assigned @TaKO8Ki.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

tests/codegen/union-abi.rs

clubby789 · 2024-04-03T22:15:38Z

@bors try @rust-timer queue

bors · 2024-04-03T22:16:49Z

⌛ Trying commit 31abdb2 with merge 6be5114...

…<try> Add `noundef` metadata for fits-in-target-pointer-size array immediate arguments `noundef` is only added if the small array immediate fits in the target pointer size and if optimizations are enabled. Closes rust-lang#123183.

bors · 2024-04-03T23:48:02Z

☀️ Try build successful - checks-actions
Build commit: 6be5114 (6be511458bb70e6b7db8ec27a49f5a0067274688)

rust-timer · 2024-04-04T01:18:54Z

Finished benchmarking commit (6be5114): comparison URL.

Overall result: no relevant changes - no action needed

Benchmarking this pull request likely means that it is perf-sensitive, so we're automatically marking it as not fit for rolling up. While you can manually mark this PR as fit for rollup, we strongly recommend not doing so since this PR may lead to changes in compiler perf.

@bors rollup=never
@rustbot label: -S-waiting-on-perf -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)	0.1%	[0.1%, 0.1%]	1
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-2.4%	[-2.4%, -2.4%]	1
All ❌✅ (primary)	0.1%	[0.1%, 0.1%]	1

Cycles

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)	-	-	0
Regressions ❌ (secondary)	3.9%	[3.9%, 3.9%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-2.4%	[-2.4%, -2.4%]	1
All ❌✅ (primary)	-	-	0

Binary size

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

Bootstrap: 666.914s -> 667.807s (0.13%)
Artifact size: 318.16 MiB -> 318.17 MiB (0.01%)

tests/codegen/array-immediate-param-noundef.rs

dianqk · 2024-04-07T01:45:16Z

Is it working on the following code?

#[repr(transparent)]
pub struct Foo([u8; 4]);

#[repr(transparent)]
pub struct Baz(Foo);

#[no_mangle]
pub fn replace_repr_transparent_struct_short_array_2(r: &mut Baz, v: Baz) -> Baz {
    std::mem::replace(r, v)
}

clubby789 · 2024-04-07T13:14:34Z

define i32 @replace_repr_transparent_struct_short_array_2(ptr noalias nocapture noundef align 1 dereferenceable(4) %r, i32 %0) unnamed_addr #0 {
start:
  %result.sroa.0.0.copyload = load i32, ptr %r, align 1
  store i32 %0, ptr %r, align 1
  ret i32 %result.sroa.0.0.copyload
}

jieyouxu · 2024-04-07T13:18:42Z

aaaaa right, recursive #[repr(transparent)], I have not considered that....

jieyouxu · 2024-04-07T15:02:14Z

@rustbot author

jieyouxu · 2024-04-07T15:18:54Z

I made the is_transparent_array check recursive now, so it should see through nested #[repr(transparent)] cases (added a test case for that).

@rustbot ready

tests/codegen/array-immediate-param-noundef.rs

… immediate arguments

dianqk

LGTM. But I believe this still requires someone with r+ privileges to take a look.

r? compiler

davidtwco · 2024-04-16T10:29:40Z

@bors r=davidtwco,DianQK

bors · 2024-04-16T10:29:44Z

📌 Commit 6bf21cf has been approved by davidtwco,DianQK

It is now in the queue for this repository.

…davidtwco,DianQK Add `noundef` metadata for fits-in-target-pointer-size array immediate arguments `noundef` is only added if the small array immediate fits in the target pointer size and if optimizations are enabled. Closes rust-lang#123183.

bors · 2024-04-16T15:29:42Z

⌛ Testing commit 6bf21cf with merge 57a2597...

RalfJung · 2024-04-16T16:53:08Z

compiler/rustc_ty_utils/src/abi.rs

+                // `#[repr(transparent)] enum` containing array. Note that `#[repr(transparent)]`
+                // can contain other `#[repr(transparent)]` structs or enums, which can eventually
+                // contain an array!
+                if arg.layout.ty.is_array() || is_transparent_array(cx, arg.layout) {


I don't understand why arrays would be special here. What if this is an array of unions, or something like that?

RalfJung · 2024-04-16T16:59:29Z

compiler/rustc_ty_utils/src/abi.rs

+    let mut adt_layout = outermost_layout;
+    // Recursively walk a layout, seeing through all `#[repr(transparent)]` layers.
+    while adt_layout.is_transparent::<LayoutCx<'tcx, TyCtxt<'tcx>>>()
+        && let Some((_, layout)) = adt_layout.non_1zst_field(cx)


Note that unions can be transparent, so wouldn't this do the wrong thing then?

At least for structs, there's not really any need to check for is_transparent. This is an optimization so we can just make use of the fact that a struct is a wrapper (i.e., non_1zst_field returns Some) even without a repr.

RalfJung · 2024-04-16T17:01:11Z

@bors r-
because of missing tests and since at first glance it seems some of these tests may actually fail -- better safe than sorry

RalfJung · 2024-04-16T17:00:11Z

tests/codegen/array-immediate-param-noundef.rs

+
+// # Negative examples
+
+// This inner struct is *not* `#[repr(transparent)]`, so we must not emit `noundef` for the outer


Why "must not"? With the layout we pick this would still be correct.

RalfJung · 2024-04-16T17:00:38Z

tests/codegen/array-immediate-param-noundef.rs

This testcase is missing negative tests with transparent unions, and arrays of unions, to ensure they do not get the attribute.

RalfJung · 2024-04-16T17:08:08Z

How do I get bors to process the next PR?
@bors force

RalfJung · 2024-04-16T17:10:34Z

compiler/rustc_ty_utils/src/abi.rs

                // so we pick an appropriately sized integer type instead.
                arg.cast_to(Reg { kind: RegKind::Integer, size });
+
+                // Let's see if we can add a `noundef`. This is only legal for arrays, definitely


I think the elephant in the room here is padding, isn't it? Arrays are fine as they do not have any padding. But the word "padding" does not appear anywhere in the PR diff or description, so I am confused.

bors · 2024-04-16T17:35:55Z

☀️ Try build successful - checks-actions
Build commit: 57a2597 (57a25976fe3c471108640effc63529372c2b61ca)

scottmcm · 2024-04-16T17:45:30Z

I agree that there's nothing particularly special about arrays here -- it might be possible for other aggregates too. I just opened the issue saying arrays because that's the place I noticed it and I was confident about. (Since [i32; 1] definitely can't be undef.)

But for example we don't put noundef on { i32, i32 } returns either right now, it looks like, and we probably could. That's probably better as a "not this PR", though.

jieyouxu · 2024-04-16T18:28:21Z

Thanks for the catch, I'll revisit this PR in a few days when less busy

jieyouxu · 2024-05-04T15:03:15Z

(Still busy, will have more time towards the end of this month)

jieyouxu · 2024-06-09T12:22:14Z

Unfortunately I don't have the bandwidth to work on this, but other people are welcomed to pick this up.

…<try> Annotate eligible small immediate arguments with `noundef` Retake of rust-lang#123425. We try to annotate small (fits within target pointer width) aggregate arguments passed as immediates (specifically casted as "appropriately sized integer type") with `noundef`. Example: ```rs #[no_mangle] pub fn short_array_u64x1(v: [u64; 1]) -> [u64; 1] { v } ``` currently produces ```llvm define i64 `@short_array_u64x1(i64` %0) ... ``` This PR changes that to ```llvm define noundef i64 `@short_array_u64x1(i64` noundef %0) ... ``` The `noundef` attribute is added only if the immediate value has no padding. Specifically, the conservative heuristic we use is to: - Peel away layers of `#[repr(Rust)]` or `#[repr(transparent)]` wrappers if present - Check for innermost simple arrays (whose element type is a primitive type) that can fit within target pointer width Union immediates or otherwise anything that contains unions will not have `noundef` attribute applied. Closes rust-lang#123183. cc `@/RalfJung` who pointed out various problems with the previous take, hopefully I addressed most of them in this take. r? `@ghost` (perf)

rustbot assigned TaKO8Ki Apr 3, 2024

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 Apr 3, 2024

jieyouxu commented Apr 3, 2024

View reviewed changes

tests/codegen/union-abi.rs Outdated Show resolved Hide resolved

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 Apr 3, 2024

This comment has been minimized.

Sign in to view

rustbot removed the S-waiting-on-perf Status: Waiting on a perf run to be completed. label Apr 4, 2024

dianqk reviewed Apr 7, 2024

View reviewed changes

tests/codegen/array-immediate-param-noundef.rs Outdated Show resolved Hide resolved

rust-cloud-vms bot force-pushed the array-imm-noundef-param branch from 31abdb2 to e1aa348 Compare April 7, 2024 14:55

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 Apr 7, 2024

rust-cloud-vms bot force-pushed the array-imm-noundef-param branch from e1aa348 to 7a67c9d Compare April 7, 2024 15:16

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 Apr 7, 2024

dianqk reviewed Apr 10, 2024

View reviewed changes

tests/codegen/array-immediate-param-noundef.rs Outdated Show resolved Hide resolved

tests/codegen/array-immediate-param-noundef.rs Outdated Show resolved Hide resolved

tests/codegen/array-immediate-param-noundef.rs Outdated Show resolved Hide resolved

jieyouxu added 3 commits April 10, 2024 14:24

codegen: add noundef metadata for fits-in-target-pointer-size array…

eb00c05

… immediate arguments

codegen: revert union-abi test

b9f3e3d

codegen: only emit noundef for arrays, not unions or other types

b7b78bb

codegen: update array-immediate-param-noundef test

6bf21cf

rust-cloud-vms bot force-pushed the array-imm-noundef-param branch from 7a67c9d to 6bf21cf Compare April 10, 2024 14:41

dianqk approved these changes Apr 11, 2024

View reviewed changes

rustbot assigned davidtwco and unassigned TaKO8Ki Apr 11, 2024

davidtwco approved these changes Apr 16, 2024

View reviewed changes

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 Apr 16, 2024

RalfJung reviewed Apr 16, 2024

View reviewed changes

bors 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-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. labels Apr 16, 2024

RalfJung reviewed Apr 16, 2024

View reviewed changes

jieyouxu marked this pull request as draft June 8, 2024 18:25

jieyouxu closed this Jun 9, 2024

jieyouxu deleted the array-imm-noundef-param branch July 1, 2024 11:39

jieyouxu mentioned this pull request Jul 1, 2024

Annotate eligible small immediate arguments with noundef #127210

Closed


		// # Negative examples

		// This inner struct is not `#[repr(transparent)]`, so we must not emit `noundef` for the outer

Uh oh!

Conversation

jieyouxu commented Apr 3, 2024

Uh oh!

rustbot commented Apr 3, 2024

Uh oh!

Uh oh!

clubby789 commented Apr 3, 2024

Uh oh!

This comment has been minimized.

bors commented Apr 3, 2024

Uh oh!

bors commented Apr 3, 2024

Uh oh!

This comment has been minimized.

rust-timer commented Apr 4, 2024

Overall result: no relevant changes - no action needed

Uh oh!

Uh oh!

dianqk commented Apr 7, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

clubby789 commented Apr 7, 2024

Uh oh!

jieyouxu commented Apr 7, 2024

Uh oh!

jieyouxu commented Apr 7, 2024

Uh oh!

jieyouxu commented Apr 7, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

dianqk left a comment

Choose a reason for hiding this comment

Uh oh!

davidtwco commented Apr 16, 2024

Uh oh!

bors commented Apr 16, 2024

Uh oh!

bors commented Apr 16, 2024

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

RalfJung commented Apr 16, 2024

Uh oh!

Choose a reason for hiding this comment

Uh oh!

RalfJung Apr 16, 2024 • 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!

RalfJung commented Apr 16, 2024

Uh oh!

Choose a reason for hiding this comment

Uh oh!

bors commented Apr 16, 2024

Uh oh!

scottmcm commented Apr 16, 2024

Uh oh!

jieyouxu commented Apr 16, 2024

Uh oh!

jieyouxu commented May 4, 2024

Uh oh!

jieyouxu commented Jun 9, 2024

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

10 participants

dianqk commented Apr 7, 2024 •

edited

Loading

jieyouxu commented Apr 7, 2024 •

edited

Loading

RalfJung Apr 16, 2024 •

edited

Loading