Skip to content

Conversation

@vkraleti
Copy link
Contributor

@vkraleti vkraleti commented Dec 5, 2025

Add kas fragment and extend build-yocto.yml to also cover qcom-distro-selinux as another variant for prop image.

Add `qcom-distro-selinux-prop-image.yml` to enable SELinux support in
proprietary image builds. This configuration sets the distro to
qcom-distro-selinux and builds qcom-multimedia-proprietary-image.

Signed-off-by: Viswanath Kraleti <viswanath.kraleti@oss.qualcomm.com>
Add jobs for building qcom-multimedia-proprietary-image with
qcom-distro-selinux distro to generate images with SELinux enabled.

Signed-off-by: Viswanath Kraleti <viswanath.kraleti@oss.qualcomm.com>
- name: qcom-distro-kvm
yamlfile: ':ci/qcom-distro-kvm.yml'
- name: qcom-distro-selinux-prop-image
yamlfile: ':ci/qcom-distro-selinux-prop-image.yml'
Copy link
Contributor

Choose a reason for hiding this comment

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

No, this should not be done in the 'cache warm up' stage. I'll send corresponding changes to drop other jobs from this part.

Choose a reason for hiding this comment

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

Thanks for clarifying that this shouldn’t be part of the cache warm-up stage. Hope you are working on this and will share a change for this.

Copy link
Contributor

Choose a reason for hiding this comment

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

@lumag , i think it makes sense to kvm and sota in the warmup section since they are generic distro built for several machines, so the warmup seems 'useful' in this case (though we might benefit to split into 2 warmup sections for 'base' and slightly modified distro.. but it's pushing too much).

However we should not have any of the 'prop' distro or images in the warmup, since there is very little benefit since the prop content is typically machine specific or proprietary blob.

do you agree?

Copy link
Contributor

Choose a reason for hiding this comment

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

I think, the main purpose of the warmup is to push armv8a / armv8-2a builds into the sstate-cache. Please correct me if I'm wrong, but neither kvm nor sota will result in a big number of additions to the sstate-cache (on the other hand, building them at this stage will take more time as the packages are not a part of the cache).

Performing separate prop build at this stage doesn't make sense. However I still hope to land #1106, removing a separate build step.

Copy link
Contributor

Choose a reason for hiding this comment

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

I think, the main purpose of the warmup is to push armv8a / armv8-2a builds into the sstate-cache. Please correct me if I'm wrong, but neither kvm nor sota will result in a big number of additions to the sstate-cache (on the other hand, building them at this stage will take more time as the packages are not a part of the cache).

there are a few changes for kvm and sota, but maybe not much, so perhaps it's indeed easier to keep the warmup simple, so let's move them in the compile section.

Performing separate prop build at this stage doesn't make sense. However I still hope to land #1106, removing a separate build step.

It had been a while since I checked #1106, it does look really good now. we could split compile into compile_armv8 and compile_armv7 to simplify a bit the matrix definition, but it can be done separately.

@github-actions
Copy link

github-actions bot commented Dec 9, 2025

Test run workflow

Test jobs for commit 355e480

@test-reporting-app
Copy link

Test Results

 14 files  ±0   28 suites  ±0   36m 3s ⏱️ +6s
 50 tests ±0   50 ✅ ±0  0 💤 ±0  0 ❌ ±0 
304 runs  ±0  304 ✅ ±0  0 💤 ±0  0 ❌ ±0 

Results for commit 355e480. ± Comparison against base commit 376521a.

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.

5 participants