Missed optimization when using split_first

Impl 1:

pub fn f(input: &mut &[u64]) -> Option<u64> {
    match input {
        [] => None,
        [first, rest @ ..] => {
            *input = rest;
            Some(*first)
        }
    }
}

Impl 2:

pub fn f(input: &mut &[u64]) -> Option<u64> {
    let (first, rest) = input.split_first()?;
    *input = rest;
    Some(*first)
}

Godbolt: https://godbolt.org/z/aasq15z7a

These functions should be equivalent, but the second function generates one more if than the first one. This also happens when using .get(0) and then doing the slicing manually. I expected the impls to generate the same code.

Quoting myself on Zulip:

I've seen this before too. The extra comparison is a null pointer check and is going to be the result of something creating a Option<&T>

This is a metadata preservation failure during SimplifyCFG speculation: https://llvm.godbolt.org/z/frs8o853f

Preserving !nonnull here used to be illegal, but thanks to https://reviews.llvm.org/D141386 we are allowed to preserve it now. Only need to drop !noundef.

Upstream patch: https://reviews.llvm.org/D146629

Fixed by #114048, needs codegen test.

add codegen test for rust-lang#109328

Auto merge of rust-lang#125347 - tesuji:needtests, r=nikic

Rollup merge of rust-lang#126906 - GrigorenkoPV:fixme-split_at_first,…

Unrolled build for rust-lang#126906

Navigation Menu

Repository navigation

Missed optimization when using `split_first` #109328

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Missed optimization when using split_first #109328

Description

Activity

JakobDegen commented on Mar 18, 2023

nikic commented on Mar 19, 2023

nikic commented on Mar 22, 2023

nikic commented on Aug 14, 2023

Metadata

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Issue actions

Missed optimization when using `split_first` #109328