Bugzilla

(In reply to Emilio Cobos Álvarez (:emilio) from comment #7)
> Do you know why we haven't unsuppressed by the time the test runs? The test runs on load, so we should have unsuppressed then.
> 
> Is the issue we don't hit [this code](https://searchfox.org/mozilla-central/rev/1cbb3aa1ce39581528eb75b08848fe85dd481f20/layout/base/PresShell.cpp#9849) on time? Or something else?

Yeah the code actually hit, but it's too late. In the content process after calling UnsuppressAndInvalidate we build the scroll data in question and it's sent to APZ. In the meantime, the UnsuppressAndInvalidate reaches to the parent process via an IPC call, then in the parent process we do display list building with the proper LayersId and it's sent to APZ. But unfortunately we got the proper LayersId in APZ the scroll data from the content process has been already processed.

(In reply to Emilio Cobos Álvarez (:emilio) from comment #7)
> Do you know why we haven't unsuppressed by the time the test runs? The test runs on load, so we should have unsuppressed then.
> 
> Is the issue we don't hit [this code](https://searchfox.org/mozilla-central/rev/1cbb3aa1ce39581528eb75b08848fe85dd481f20/layout/base/PresShell.cpp#9849) on time? Or something else?

Yeah the code actually hit, but it's too late. In the content process after calling UnsuppressAndInvalidate we build the scroll data in question and it's sent to APZ. In the meantime, the UnsuppressAndInvalidate reaches to the parent process via an IPC call, then in the parent process we do display list building with the proper LayersId and it's sent to APZ. But unfortunately when we got the proper LayersId in APZ the scroll data from the content process has been already processed.