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

slang: verible: Use gcc-9 #103

Merged
merged 1 commit into from
Apr 13, 2021
Merged

slang: verible: Use gcc-9 #103

merged 1 commit into from
Apr 13, 2021

Conversation

ajelinski
Copy link
Contributor

GitHub Actions virtual environments doesn't have gcc-8 preinstalled
anymore: actions/runner-images#2950

GitHub Actions virtual environments doesn't have gcc-8 preinstalled
anymore: actions/runner-images#2950
Copy link
Contributor

@PiotrZierhoffer PiotrZierhoffer left a comment

Choose a reason for hiding this comment

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

Lgtm, merge on green

@tgorochowik
Copy link
Member

@mithro
Copy link
Member

mithro commented Apr 9, 2021

You should really be using the Compiler provided in the conda environment. Using the system provided compiler is generally bad idea.

@ajelinski
Copy link
Contributor Author

@tgorochowik This would also require updating the image we build on. Currently used Ubuntu 16.04 for all Linux builds doesn't have gcc-10: https://github.com/actions/virtual-environments/blob/main/images/linux/Ubuntu1604-README.md

@mithro AFAIR these two were impossible to build with a compiler from Conda and at least slang is a static binary. Still, this fix is just an adjustment to the GitHub Actions changes. These two packages have been built with OS-based gcc already.

@mithro
Copy link
Member

mithro commented Apr 9, 2021

@ajelinski - Can we make sure that is a GitHub issue which is tracking that this problem exists and needs to be fixed in the future?

@ajelinski
Copy link
Contributor Author

@mithro This seems to be a conscious change in the OS image. I can see they are constantly making software changes in the images: https://github.com/actions/virtual-environments/commits/main/images/linux/Ubuntu1604-README.md

@umarcor
Copy link
Member

umarcor commented Apr 9, 2021

My proposal would be to use one single docker image for having an stable build environment, regardless of the host. The purpose of such image is not for users to consume it, but just to use it in CI as a homogeneization layer on top of whichever VM (virtual environment) the CI provider offers. I think that the dockerfile that builds and the CI that updates the image would fit in hdl/conda-ci. We might then consider providing a ready-to-use hdlc/conda container image in hdl/containers. But that is a different discussion.

However, I agree with Tim that, in the mid term, it'd be better to have it done regardless of the compiler on the system.

@mithro
Copy link
Member

mithro commented Apr 9, 2021

@mithro This seems to be a conscious change in the OS image. I can see they are constantly making software changes in the images: https://github.com/actions/virtual-environments/commits/main/images/linux/Ubuntu1604-README.md

I meant a GitHub issue about the fact the conda package is using the system compiler package.

@ajelinski
Copy link
Contributor Author

ajelinski commented Apr 13, 2021

@mithro Sorry, clearly I misunderstood you. I've added such an issue: #104

@umarcor I think it's a good idea, maybe you could add a GH issue?

I'm merging the PR since it fixes failures and the issues aren't really about the changes made.

@ajelinski ajelinski merged commit 81fc92f into master Apr 13, 2021
@ajelinski ajelinski deleted the fix-verible-slang branch April 13, 2021 13:47
@umarcor
Copy link
Member

umarcor commented Apr 13, 2021

@ajelinski I have this pending in my notifications/ToDo: hdl/conda-ci#13 (comment). I just exposed the idea here for others to be aware.

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

Successfully merging this pull request may close these issues.

None yet

5 participants