UNSTABLE pull / linux-focal-cuda12.1-py3.10-gcc9-experimental-split-build / test (default)

[huydhn]

This is a new update to CI and folks will need to include the merge base https://hud.pytorch.org/hud/pytorch/pytorch/236fbcbdf44ea429b11ac44c1814af56ca243605/1 to get it pass on their PR. So, I keep the job as unstable in the next few days to soften the CI impact until most devs move pass that commit.

cc @seemethere @malfet @pytorch/pytorch-dev-infra @PaliC

[pytorch-bot]

Hello there! From the UNSTABLE prefix in this issue title, it looks like you are attempting to unstable a job in PyTorch CI. The information I have parsed is below:

• Job name: pull / linux-focal-cuda12.1-py3.10-gcc9-experimental-split-build / test (default)
• Credential: huydhn

Within ~15 minutes, pull / linux-focal-cuda12.1-py3.10-gcc9-experimental-split-build / test (default) and all of its dependants will be unstable in PyTorch CI. Please verify that the job name looks correct. With great power comes great responsibility.

[huydhn]

I'm seeing the new failure showing on my PR at #128873 (comment)

[huydhn]

This has been unstable for 2 weeks while green in trunk, so I'm closing this now.

[WeiChunyu-star]

is this the same issue？#130545
@huydhn
Please help to check it out

[huydhn]

For the context, the issue has been fixed, but you need to do a rebase on your PR for the test to pass. It's not ideal, but it's a good practice to base your PR on a recent commit like a few days old at most. Just FYI, you can rebase easily with @pytorchbot rebase (to PyTorch stable branch) or @pytorch rebase -b main (to latest main commit) on your PR.