Skip to content
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.

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

Parallel layout is slower than sequential on my Linux system #17127

Open
mbrubeck opened this issue Jun 1, 2017 · 1 comment
Open

Parallel layout is slower than sequential on my Linux system #17127

mbrubeck opened this issue Jun 1, 2017 · 1 comment

Comments

@mbrubeck
Copy link
Contributor

@mbrubeck mbrubeck commented Jun 1, 2017

Using release builds of Servo on my 4-core, 8-thread Linux laptop, time spent in layout is 25% to 100% longer in parallel mode than in sequential mode (-y 1), across a few different real-world pages (Wikipedia article, Google search results, MySpace page from tp5n). This doesn't appear to be a recent regression; Servo builds from up to a year ago have approximately the same performance in both parallel and sequential mode. The recent update to rayon 0.7.1 did not fix this problem.

One of the biggest deltas is "LayoutMain", which frequently takes 10x to 30x longer in parallel mode than in sequential mode.

Running the same benchmarks on my 4-core iMac running macOS, parallel layout is significantly faster than sequential layout, as expected.

System details:

  • Hardware: Lenovo Thinkpad P50
  • CPU: Intel(R) Xeon(R) CPU E3-1505M v5 @ 2.80GHz
  • OS: Debian 9.0
  • CPU governor: "performance"
  • no other processes using significant CPU during benchmark runs
@heycam
Copy link
Member

@heycam heycam commented Jun 27, 2017

This reproduces on my P50 (i7-6820HQ @ 2.70GHz) on Linux Mint 18 with the performance governor.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Linked pull requests

Successfully merging a pull request may close this issue.

None yet
3 participants
You can’t perform that action at this time.