Closed
Description
The built in unit tester presently contains some very crude one-off automated benchmarking stuff. It was initially a clone of the Go testing.B
benchmark system (http://golang.org/pkg/testing/#B) but started to drift towards doing statistical measurement in order to get a more stable signal.
I believe Bryan O'Sullivan's criterion
library (https://github.com/bos/criterion) is the state of the art here. We should port / emulate whatever it does well.
Metadata
Metadata
Assignees
Type
Projects
Milestone
Relationships
Development
No branches or pull requests
Activity
catamorphism commentedon Aug 26, 2013
(bug triage) Good idea; not a blocker IMO.
Aatch commentedon Mar 4, 2014
Visiting for triage.
This would be nice, but still isn't a high-priority.
schuster commentedon Apr 5, 2014
I thought I might try to write this to help me learn Rust. I've found the testing guide and the various benchmark tests in the code. Are there any other resources I should be aware of, or design goals to keep in mind (other than emulating criterion)?
I'll try to get something simple working, post it, and request feedback as I go.
japaric commentedon May 26, 2014
I'm tackling this issue here, I'll love to get feedback!
kud1ing commentedon Oct 16, 2014
Is this fixed by https://github.com/japaric/criterion.rs ?
Is it already usable for supporting #18085 ?
steveklabnik commentedon Jan 21, 2015
I'm pulling a massive triage effort to get us ready for 1.0. As part of this, I'm moving stuff that's wishlist-like to the RFCs repo, as that's where major new things should get discussed/prioritized.
This issue has been moved to the RFCs repo: rust-lang/rfcs#651
criterion
for benchmarking rust-lang/rfcs#651Auto merge of rust-lang#6812 - Y-Nak:fix-6792, r=matthiaskrgr