about summary refs log tree commit diff
path: root/tests/rustdoc-js-std/parser-errors.js
diff options
context:
space:
mode:
authorMatthias Krüger <matthias.krueger@famsik.de>2023-03-18 12:04:21 +0100
committerGitHub <noreply@github.com>2023-03-18 12:04:21 +0100
commit7ebf2cd2b8f1f8c11f9339ef337d336a91d6f7dc (patch)
tree599c1fff28765ed61a3acb84485935c9c2417d04 /tests/rustdoc-js-std/parser-errors.js
parent9599f3cc5430039b615bc5171be41a6733083dc0 (diff)
parent675c4aa2c1c142415d4e95bf550ec0b1de2493d0 (diff)
downloadrust-7ebf2cd2b8f1f8c11f9339ef337d336a91d6f7dc.tar.gz
rust-7ebf2cd2b8f1f8c11f9339ef337d336a91d6f7dc.zip
Rollup merge of #108772 - jyn514:faster-tidy, r=the8472
Speed up tidy quite a lot

I highly recommend reviewing this commit-by-commit. Based on #106440 for convenience.

## Timings

These were collected by running `x test tidy -v` to copy paste the command, then using [`samply record`](https://github.com/mstange/samply).

before (8 threads)
![image](https://user-images.githubusercontent.com/23638587/222965319-352ad2c8-367c-4d74-960a-e4bb161a6aff.png)

after (8 threads) ![image](https://user-images.githubusercontent.com/23638587/222965323-fa846f4e-727a-4bf8-8e3b-1b7b40505cc3.png)

before (64 threads) ![image](https://user-images.githubusercontent.com/23638587/222965302-dc88020c-19e9-49d9-a87d-cad054d717f3.png)
after (64 threads) ![image](https://user-images.githubusercontent.com/23638587/222965335-e73d7622-59de-41d2-9cc4-1bd67042a349.png)

The last commit makes tidy use more threads, so comparing "before (8 threads)" to "after (64 threads)" is IMO the most realistic comparison. Locally, that brings the time for me to run tidy down from 4 to .9 seconds, i.e. the majority of the time for `x test tidy` is now spend running `fmt --check`.

r? `@the8472`
Diffstat (limited to 'tests/rustdoc-js-std/parser-errors.js')
0 files changed, 0 insertions, 0 deletions