You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Oct 3, 2018. It is now read-only.
We had a leak recently (bug 1264161) where every time a browser window was opened and closed it would leak a bunch of stuff. Surprisingly (to me) none of our automation tests for this, since leak checking is done at shutdown which was too late for this particular scenario. AWSY however would have caught this sort of a leak if it opened and closed windows in addition to tabs. Something to consider adding, and I imagine it shouldn't be too hard. It could be as simple as an extra phase at the end of the test (after "After TP5, tabs closed [+30s, forced GC]") where we cycle through the TP5 set opening and closing each page in a new window, then do a force GC and grab another snapshot. I strongly suspect that such a test would have showed the leak.
The text was updated successfully, but these errors were encountered:
We had a leak recently (bug 1264161) where every time a browser window was opened and closed it would leak a bunch of stuff. Surprisingly (to me) none of our automation tests for this, since leak checking is done at shutdown which was too late for this particular scenario. AWSY however would have caught this sort of a leak if it opened and closed windows in addition to tabs. Something to consider adding, and I imagine it shouldn't be too hard. It could be as simple as an extra phase at the end of the test (after "After TP5, tabs closed [+30s, forced GC]") where we cycle through the TP5 set opening and closing each page in a new window, then do a force GC and grab another snapshot. I strongly suspect that such a test would have showed the leak.
The text was updated successfully, but these errors were encountered: