-
Notifications
You must be signed in to change notification settings - Fork 504
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
test on debian12 #8928
base: master
Are you sure you want to change the base?
test on debian12 #8928
Conversation
@ysiraichi , this is the error message I see for Debian12 CUDA 12.8 build:
|
This is odd. Are we using some kind of bazel cache for building it? |
Yes, remote_cache is enabled for the bazel build. Will this impact the outcome? |
I'm not sure. But can we try doing it without using the remote cache? |
I disabled the remote cache and rebuilt again. PyTorch can be build with CUDA 12.8 successfully. However, for PyTorch/XLA, it will fail with xla/infra/ansible/config/apt.yaml Line 38 in cbff76f
|
Maybe we can replace that with |
Great! 12.8 build can pass now. The compilation takes ~2hour30mins without remote cache, which takes ~1hour to complete. I will see if we can enable it again. |
I think that the problem there was that the cache stored some |
Since cuda 12.8 requires Debian12, this PR tests if we can use Debian12 for the base image.