For comment #5, I know why it's not failing in the previous cycle.
We clone the kernel source tree with master branch, not the master-next branch with -proposed kernel.
Therefore in 4.18.0-12, the branch we use for testing is 4.18.0-11, and it does not include this patch.
But now with 4.18.0-13, we're using 4.18.0-12 for the test, that's how it gets this confusing result.
For comment #5,
I know why it's not failing in the previous cycle.
We clone the kernel source tree with master branch, not the master-next branch with -proposed kernel.
Therefore in 4.18.0-12, the branch we use for testing is 4.18.0-11, and it does not include this patch.
But now with 4.18.0-13, we're using 4.18.0-12 for the test, that's how it gets this confusing result.