Adding some relevant folks on cc. I don't expect any action on this in the short term though.

There are probably opportunities to reduce the duration of time that this overlap happens (e.g., with PlzNavigate), but it's likely quite hard to eliminate entirely.  In cross-process navigations, we don't show the new process until it sends the DidCommitProvisionalLoad IPC, which is after it has been running for a short time.  The old process then runs its unload handler in the background after the new process is showing.  This means the overlap lasts at least from the creation of the new process, to creating the Document and issuing the commit IPC, to the end of the unload handler in the old process.

Even if we ran unload first (i.e., not in the background) and then loaded the new page after quitting the old process, we'd likely end up with much longer durations of a blank white page showing.  I'm not sure it will be worth the complexity of supporting both possible orderings in the code.  (For example, I was able to remove a substantial amount of code after we changed unload handlers to run in the background.)

Could still be worth collecting data on how bad the problem is, and whether reducing the amount of time that it happens would have a meaningful benefit.

This issue has been Available for over a year. If it's no longer important or seems unlikely to be fixed, please consider closing it out. If it is important, please re-triage the issue.

Sorry for the inconvenience if the bug really should have been left as Available.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot