simd intrinsics: add simd_shuffle_generic and other missing intrinsics #119213

RalfJung · 2023-12-22T11:32:55Z

Also tweak the simd_shuffle docs a bit.

rustbot · 2023-12-22T11:33:02Z

Failed to set assignee to calebzulawski: invalid assignee

Note: Only org members with at least the repository "read" role, users with write permissions, or people who have commented on the PR may be assigned.

RalfJung · 2023-12-22T11:34:17Z

r? @workingjubilee

library/core/src/intrinsics/simd.rs

calebzulawski · 2023-12-22T13:56:06Z

library/core/src/intrinsics/simd.rs

+    /// Returns a new vector such that element `i` is selected from `xy[IDX[i]]`, where `xy`
+    /// is the concatenation of `x` and `y`. It is a compile-time error if `IDX[i]` is out-of-bounds
+    /// of `xy`.
+    pub fn simd_shuffle_generic<T, U, const IDX: &'static [u32]>(x: T, y: T) -> U;


I haven't seen this intrinsic before 🤔 this is what we need in std::simd

It's an experiment to see how far we can get with the current const-generic support. I don't think std::simd can use it yet, that would need generic_const_exprs which is still a highly experimental feature.

it would be nice if we could stabilize a small enough subset of GCE that this sort of thing becomes feasible.

From talking with @lcnr that seems pretty far out currently. And in particular reference types in const generics have pretty thorny unsolved theoretical questions as well.

rustbot · 2023-12-28T12:22:41Z

Failed to set assignee to calebzulawski: invalid assignee

Note: Only org members with at least the repository "read" role, users with write permissions, or people who have commented on the PR may be assigned.

RalfJung · 2023-12-28T12:22:52Z

I realized there are more intrinsics missing, so I added those as well.

library/core/src/intrinsics/simd.rs

RalfJung · 2024-01-27T13:58:45Z

@workingjubilee friendly review ping :)

workingjubilee · 2024-01-30T09:32:10Z

'sup, sorry about the delay. I had the usual holidays-induced madness and then work on shipping soundness fixes to Rust crates and then a blackout in winter! will take a look this week/after a nap.

RalfJung · 2024-01-30T09:37:05Z

No worries, I was just making sure this doesn't fall through the cracks. :)

workingjubilee · 2024-02-09T22:34:05Z

it seems sometimes I have an interesting interpretation of "this week".

library/core/src/intrinsics/simd.rs

workingjubilee · 2024-02-10T08:00:44Z

library/core/src/intrinsics/simd.rs

+    /// The stores happen in left-to-right order.
+    /// (This is relevant in case two of the stores overlap.)


( probably not relevant, but fun facts to know and tell: if the machine instruction gets executed, this guarantee also affects the state of the CPU during exception handling. )

Wait, exception handles can observe these instructions as non-atomic? Wow.

Yeah, on Intel CPUs at least, for most variants of these instructions, the operands (esp. the mask operand) get updated so that continuing with the same operands will complete the operation, completing a single store for each index.

What do you mean "get updated"? Does this change the contents of some other register, where the mask operand is stored? I hope it restored the original value when the instruction is done?

What do you mean "get updated"? Does this change the contents of some other register, where the mask operand is stored?

Yes, the mask register.

I hope it restored the original value when the instruction is done?

uhhh probably! 😁

I hope it restored the original value when the instruction is done?

uhhh probably! 😁

nope, it's defined to clear the mask register: https://www.felixcloutier.com/x86/vscatterdps:vscatterdpd:vscatterqps:vscatterqpd

Okay... well as long as LLVM takes that into account it's all good.

library/core/src/intrinsics/simd.rs

rustbot · 2024-02-10T09:20:05Z

Some changes occurred to the platform-builtins intrinsics. Make sure the
LLVM backend as well as portable-simd gets adapted for the changes.

cc @antoyo, @GuillaumeGomez, @bjorn3, @calebzulawski, @programmerjake

workingjubilee · 2024-02-10T09:26:22Z

cool!

@bors r+

workingjubilee · 2024-02-10T09:27:05Z

library/core/src/intrinsics/simd.rs

@@ -204,7 +204,7 @@ extern "platform-intrinsic" {
    ///
    /// `T` must be a vector.
    ///
-    /// `V` must be a vector with the same element type as `T` and the same length as `IDX`.
+    /// `U` must be a vector with the same element type as `T` and the same length as `IDX`.


ah, good catch. ^^;

bors · 2024-02-10T09:27:18Z

📌 Commit 3bc490d has been approved by workingjubilee

It is now in the queue for this repository.

workingjubilee · 2024-02-10T09:29:51Z

@bors rollup

…iaskrgr Rollup of 9 pull requests Successful merges: - rust-lang#119213 (simd intrinsics: add simd_shuffle_generic and other missing intrinsics) - rust-lang#120272 (Suppress suggestions in derive macro) - rust-lang#120773 (large_assignments: Allow moves into functions) - rust-lang#120874 (Take empty `where` bounds into account when suggesting predicates) - rust-lang#120882 (interpret/write_discriminant: when encoding niched variant, ensure the stored value matches) - rust-lang#120883 (interpret: rename ReadExternStatic → ExternStatic) - rust-lang#120890 (Adapt `llvm-has-rust-patches` validation to take `llvm-config` into account.) - rust-lang#120895 (don't skip coercions for types with errors) - rust-lang#120896 (Print kind of coroutine closure) r? `@ghost` `@rustbot` modify labels: rollup

Rollup merge of rust-lang#119213 - RalfJung:simd_shuffle, r=workingjubilee simd intrinsics: add simd_shuffle_generic and other missing intrinsics Also tweak the simd_shuffle docs a bit. r? `@calebzulawski`

…s, r=workingjubilee Stop using `unsized_const_parameters` in core/std `feature(unsized_const_parameters)` is an incomplete feature and should not be used by core/std as it makes it can make it significantly harder to evolve the feature. It also just generally opens the possibility of introducing bugs on stable through std's backdoor. The only usage of this feature in std is the `simd_shuffle_intrinsic` added in rust-lang#119213. It doesn't seem to be used anywhere as far as I can tell so it is removed in this PR. All tests and codegen logic etc have been kept however. r? `@workingjubilee`

Rollup merge of rust-lang#128150 - BoxyUwU:std_only_sized_const_params, r=workingjubilee Stop using `unsized_const_parameters` in core/std `feature(unsized_const_parameters)` is an incomplete feature and should not be used by core/std as it makes it can make it significantly harder to evolve the feature. It also just generally opens the possibility of introducing bugs on stable through std's backdoor. The only usage of this feature in std is the `simd_shuffle_intrinsic` added in rust-lang#119213. It doesn't seem to be used anywhere as far as I can tell so it is removed in this PR. All tests and codegen logic etc have been kept however. r? `@workingjubilee`

rustbot added S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. T-libs Relevant to the library team, which will review and decide on the PR/issue. labels Dec 22, 2023

RalfJung force-pushed the simd_shuffle branch from 911c637 to 8a126eb Compare December 22, 2023 11:34

rustbot assigned workingjubilee Dec 22, 2023

calebzulawski reviewed Dec 22, 2023

View reviewed changes

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

calebzulawski reviewed Dec 22, 2023

View reviewed changes

RalfJung force-pushed the simd_shuffle branch from 8a126eb to a1b5879 Compare December 22, 2023 14:03

calebzulawski approved these changes Dec 22, 2023

View reviewed changes

RalfJung changed the title ~~simd intrinsics: add simd_shuffle_generic~~ simd intrinsics: add simd_shuffle_generic and other missing intrinsics Dec 28, 2023

calebzulawski approved these changes Dec 28, 2023

View reviewed changes

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

RalfJung force-pushed the simd_shuffle branch from 8e4c416 to 4201c0a Compare December 28, 2023 12:36

calebzulawski approved these changes Dec 28, 2023

View reviewed changes

calebzulawski mentioned this pull request Dec 29, 2023

Use core::intrinsics rust-lang/portable-simd#386

Merged

workingjubilee reviewed Feb 10, 2024

View reviewed changes

RalfJung added 4 commits February 10, 2024 10:13

simd intrinsics: add simd_shuffle_generic

d96f0c3

add more missing simd intrinsics

5219af6

simd_scatter: mention left-to-right order

aa64c73

various docs tweaks

3bc490d

RalfJung force-pushed the simd_shuffle branch from 4c3c2fb to 3bc490d Compare February 10, 2024 09:20

workingjubilee reviewed Feb 10, 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 Feb 10, 2024

matthiaskrgr mentioned this pull request Feb 11, 2024

Rollup of 9 pull requests #120903

Merged

bors merged commit 5f9457c into rust-lang:master Feb 11, 2024
11 checks passed

rustbot added this to the 1.78.0 milestone Feb 11, 2024

RalfJung deleted the simd_shuffle branch February 11, 2024 20:51

BoxyUwU mentioned this pull request Jul 24, 2024

Stop using unsized_const_parameters in core/std #128150

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

simd intrinsics: add simd_shuffle_generic and other missing intrinsics #119213

simd intrinsics: add simd_shuffle_generic and other missing intrinsics #119213

RalfJung commented Dec 22, 2023

rustbot commented Dec 22, 2023

RalfJung commented Dec 22, 2023

calebzulawski Dec 22, 2023

RalfJung Dec 22, 2023 •

edited

Loading

workingjubilee Feb 9, 2024

RalfJung Feb 10, 2024

rustbot commented Dec 28, 2023

RalfJung commented Dec 28, 2023

RalfJung commented Jan 27, 2024

workingjubilee commented Jan 30, 2024

RalfJung commented Jan 30, 2024

workingjubilee commented Feb 9, 2024

workingjubilee Feb 10, 2024

RalfJung Feb 10, 2024

workingjubilee Feb 10, 2024

RalfJung Feb 10, 2024 •

edited

Loading

workingjubilee Feb 10, 2024

programmerjake Feb 11, 2024

RalfJung Feb 11, 2024

rustbot commented Feb 10, 2024

workingjubilee commented Feb 10, 2024

workingjubilee Feb 10, 2024

bors commented Feb 10, 2024

workingjubilee commented Feb 10, 2024

		/// The stores happen in left-to-right order.
		/// (This is relevant in case two of the stores overlap.)

simd intrinsics: add simd_shuffle_generic and other missing intrinsics #119213

simd intrinsics: add simd_shuffle_generic and other missing intrinsics #119213

Conversation

RalfJung commented Dec 22, 2023

rustbot commented Dec 22, 2023

RalfJung commented Dec 22, 2023

Choose a reason for hiding this comment

RalfJung Dec 22, 2023 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

rustbot commented Dec 28, 2023

RalfJung commented Dec 28, 2023

RalfJung commented Jan 27, 2024

workingjubilee commented Jan 30, 2024

RalfJung commented Jan 30, 2024

workingjubilee commented Feb 9, 2024

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

RalfJung Feb 10, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

rustbot commented Feb 10, 2024

workingjubilee commented Feb 10, 2024

Choose a reason for hiding this comment

bors commented Feb 10, 2024

workingjubilee commented Feb 10, 2024

RalfJung Dec 22, 2023 •

edited

Loading

RalfJung Feb 10, 2024 •

edited

Loading