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

[SOF-v2.7] backport fixes for CI builds #8387

Closed

Conversation

kv2019i
Copy link
Collaborator

@kv2019i kv2019i commented Oct 25, 2023

Fix the Zephyr Windows build and Intel QB TGL build with following backports:

ec0667a xtensa-build-zephyr: remove ipc option
4e74d2b .github/zephyr: de-hardcode the remote name in the Windows build too

No changes to functionality.

Fixes commit 4bc6488 (".github/zephyr: de-hardcode the name of the
zephyr remote")

(Yay for duplication)

Signed-off-by: Marc Herbert <[email protected]>
(cherry picked from commit d661616)
Signed-off-by: Kai Vehmanen <[email protected]>
For legacy CAVS platforms (TGL/ADL/EHL), the default
build config is IPC4 now, and the overlay file is
already emptied. Remove the option in this build
script.

Signed-off-by: Chao Song <[email protected]>
(cherry picked from commit a17c282)
Signed-off-by: Kai Vehmanen <[email protected]>
@kv2019i
Copy link
Collaborator Author

kv2019i commented Oct 25, 2023

@marc-hb Uh, this is hopeless, it would seem we would need to backport a lot of github workflow stuff to get the Windows builds working. Now we have issues with removing the IPC4 option, and the problem with ninja tool missing. If we get the QB build to pass, I think I'll stop here.

@marc-hb
Copy link
Collaborator

marc-hb commented Oct 25, 2023

I don't think Github Actions are hopeless on this branch: in fact some Linux configurations and some Windows configurations did compile in https://github.com/thesofproject/sof/actions/runs/6639325877/job/18037461117?pr=8387! I'll take a look at the failures today. We'll most likely need my choco install workaround.

I don't understand why you're cherry-picking the IPC4 removal though?

If we get the QB build to pass, I think I'll stop here.

On the contrary, I think that is hopeless. QB never new anything about branches, see internal issue PTS-45980

Granted, the failure in https://sof-ci.01.org/sof-pr-viewer/#/build/PR8387/build13001947 looks like a temporary glitch, not like a real incompatibility and problem. But it's only a matter of time. That's because Github Actions are branched at the same time and QB code (or Jenkins code for that matter) is not.

EDIT: detailed results missing in (failed) https://sof-ci.01.org/sofpr/PR8387/build14140/devicetest/index.html and (passed) ACE https://sof-ci.01.org/sofpr/PR8387/build14139/devicetest/index.html, likely due to the "new build service".

@marc-hb
Copy link
Collaborator

marc-hb commented Oct 25, 2023

@marc-hb
Copy link
Collaborator

marc-hb commented Oct 25, 2023

I don't think Github Actions are hopeless on this branch: in fact some Linux configurations and some Windows configurations did compile in https://github.com/thesofproject/sof/actions/runs/6639325877/job/18037461117?pr=8387! I'll take a look at the failures today. We'll most likely need my choco install workaround.

I looked at these logs. If you drop the IPC4 removal and cherry-pick the choco install (#8254) then everything should Just Work (TM)

@kv2019i
Copy link
Collaborator Author

kv2019i commented Oct 26, 2023

@marc-hb wrote:

I don't understand why you're cherry-picking the IPC4 removal though?

It has magic code to use the IPC4 rimage toml file if you build for "tgl" and do not pass any IPC version on the command-line.

Without this, QB will build a IPC4 SOF binary, but use IPC3 toml file to prepare the ri file and resulting binary will not boot.

If we get the QB build to pass, I think I'll stop here.

On the contrary, I think that is hopeless. QB never new anything about branches, see internal issue PTS-45980

Maybe so. I was checking the QB build log and to my eyes, it picked all the correct stuff (with west).

Now the build failed to some other problem (west init), so maybe this is a hopeless road after all.

@kv2019i
Copy link
Collaborator Author

kv2019i commented Oct 26, 2023

Closing, this seems to lead to just more problems..

@marc-hb
Copy link
Collaborator

marc-hb commented Oct 26, 2023

Works fine in #8406

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.

3 participants