bjoernager/rust
bjoernager/rust
1
Fork 0
Code Issues Pull requests Activity

Optimized vec::IntoIter::next_chunk impl

Browse source 
```
test vec::bench_next_chunk                               ... bench:         696 ns/iter (+/- 22)
x86_64v1, pr
test vec::bench_next_chunk                               ... bench:         309 ns/iter (+/- 4)

znver2, default
test vec::bench_next_chunk                               ... bench:      17,272 ns/iter (+/- 117)
znver2, pr
test vec::bench_next_chunk                               ... bench:         211 ns/iter (+/- 3)
```

The znver2 default impl seems to be slow due to inlining decisions. It goes through `core::array::iter_next_chunk`
which has a deeper call tree.
This commit is contained in:
The 8472 2022-06-26 22:53:25 +02:00
parent 7425fb293f
commit 2f9f2e507e
4 changed files with 64 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

4
library/alloc/src/lib.rs
View file

@ -90,6 +90,7 @@
#![feature(alloc_layout_extra)]
#![feature(allocator_api)]
#![feature(array_chunks)]
#![feature(array_into_iter_constructors)]
#![feature(array_methods)]
#![feature(array_windows)]
#![feature(assert_matches)]
@ -120,8 +121,11 @@
#![feature(hasher_prefixfree_extras)]
#![feature(inplace_iteration)]
#![feature(iter_advance_by)]
#![feature(iter_next_chunk)]
#![feature(layout_for_ptr)]
#![feature(maybe_uninit_array_assume_init)]
#![feature(maybe_uninit_slice)]
#![feature(maybe_uninit_uninit_array)]
#![cfg_attr(test, feature(new_uninit))]
#![feature(nonnull_slice_from_raw_parts)]
#![feature(pattern)]