Run builds with 4 CPU cores
What does this MR do?
We have two Mac Pros reserved for CI/development with 24 cores each. Right now that means we can only run 6 builds or development VMs in parallel.
With 4 cores VMs this goes to 12 so it should speed up the CI considerably. Job timings are below:
name | 8 cores (pipeline) | 4 cores (pipeline) |
---|---|---|
create_gitlab_base_img | 32min | 32min |
create_toolchain_base_img | 59min | 65min |
create_runner_10.3_img | 21min | 26min |
create_runner_11.2.1_img | 31min | 39min |
create_runner_12_img | 47min | 57min |
So it is around 25% slower overall for xcode, 10% for toolchain, identical for base. However, with twice as much capacity we can finish the stage in half the time, so it is still a net benefit.
Why was this MR needed?
What's the best way to test this MR?
What are the relevant issue numbers?
Edited by Adrien Kohlbecker