Skip to content

Add benchmarks for push, insert, extend, and pushpop #31

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Sep 11, 2016
Merged

Add benchmarks for push, insert, extend, and pushpop #31

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
abf82ff
Add benchmarks for push, insert, extend, and pushpop
nipunn1313 Aug 27, 2016
951f1d0
Add .travis.yml. Bench closures return values. Feature gate.
nipunn1313 Aug 28, 2016
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 5 additions & 0 deletions .travis.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,5 +3,10 @@ rust:
- nightly
- beta
- stable
script: |
cargo build --verbose &&
cargo test --verbose &&
([ $TRAVIS_RUST_VERSION != nightly ] || cargo test --verbose --features benchmarks) &&
([ $TRAVIS_RUST_VERSION != nightly ] || cargo bench --verbose --features benchmarks bench)
notifications:
webhooks: http://build.servo.org:54856/travis
3 changes: 3 additions & 0 deletions Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,3 +13,6 @@ documentation = "http://doc.servo.org/smallvec/"
name = "smallvec"
path = "lib.rs"
doctest = false

[features]
benchmarks = []
52 changes: 52 additions & 0 deletions lib.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,6 +5,8 @@
//! Small vectors in various sizes. These store a certain number of elements inline and fall back
//! to the heap for larger allocations.

#![cfg_attr(feature = "benchmarks", feature(test))]

use std::borrow::{Borrow, BorrowMut};
use std::cmp;
use std::fmt;
Expand Down Expand Up @@ -1001,3 +1003,53 @@ pub mod tests {
assert_eq!(vec.drain().len(), 3);
}
}

#[cfg(all(feature = "benchmarks", test))]
mod bench {
extern crate test;
use SmallVec;
use self::test::Bencher;

#[bench]
fn bench_push(b: &mut Bencher) {
b.iter(|| {
let mut vec: SmallVec<[u64; 16]> = SmallVec::new();
for x in 0..100 {
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Also, I think the compiler will be pretty much able to optimize these loops out. IIRC there was some builtin support for black box testing for this kind of things. I think we should be able to enforce that using it, or at least confirm it's not being optimized out.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I believe if the closure returns a value then the optimizer will not remove it.

I was empirically able to determine that the optimizer was not removing the code either way (for now), but I will include the vec return value to ensure it.

vec.push(x);
}
vec
});
}

#[bench]
fn bench_insert(b: &mut Bencher) {
b.iter(|| {
let mut vec: SmallVec<[u64; 16]> = SmallVec::new();
for x in 0..100 {
vec.insert(0, x);
}
vec
});
}

#[bench]
fn bench_extend(b: &mut Bencher) {
b.iter(|| {
let mut vec: SmallVec<[u64; 16]> = SmallVec::new();
vec.extend(0..100);
vec
});
}

#[bench]
fn bench_pushpop(b: &mut Bencher) {
b.iter(|| {
let mut vec: SmallVec<[u64; 16]> = SmallVec::new();
for x in 0..100 {
vec.push(x);
vec.pop();
}
vec
});
}
}