Have 2 browser windows (A and B) showing different contents. (Better to be on a slow machine with low memory.) Say window A is in the front and B is in the back. Load a new page in A and switch to window B while A is loading a new page. Make sure a part of A is visible. You can see that the window A clears and then shows the content of B before displaying the content of new page. Wouldn't it make it much with less memory, if Mozilla did not do this?
over to layout for further investigation. This is probably easily duped against some other bug but I didn't see it.
Assignee: asa → clayton
Status: UNCONFIRMED → NEW
Component: Browser-General → Layout
Ever confirmed: true
QA Contact: doronr → petersen
Mark, can you please triage this from Clayton's list.
Assignee: clayton → attinasi
The problem is that the offscreen buffer used to render into is shared across all of the windows. When the new window is being updated, if the connection is too slow, there is nothing new to paint the first time through the update so it ends up painting whatever was left in the offscreen buffer. This is a know problem. I'll search for the duplicate now...
OK - I couldn't find the duplicate, so this will become the reference bug against which future incarnations will be dupe'd. A related bug is bug 31981, however it is much more specific than this general description. It does point to the most likely solution to the problem, however.
Severity: major → normal
Status: NEW → ASSIGNED
OS: Mac System 8.6 → All
Hardware: Macintosh → All
Summary: contents of one window showing up in another while loading a new page → contents of other regions or windows showing up while loading a new page (offscreen buffer residual)
Target Milestone: --- → M20
Marking beta3 - if not then, when? Also, CC'd Waterson since he has a broad view of the system. It looks like there are potentially many ways this could be happening... From discussion I've had (with kmccluck) the problem stems from (synchronous ?) painting occuring before there are any frames to paint, and the back-buffer is merely dumping what it has in it already (from the last page presumably).
Approving for beta3: this has to be fixed because it is so ugly. McClusky is working on it too so We'll work on it together.
*** Bug 48273 has been marked as a duplicate of this bug. ***
This is actually a dup of bug 34283, owned by kmcclusk. The issues are outlined in that bug. *** This bug has been marked as a duplicate of 34283 ***
Status: ASSIGNED → RESOLVED
Last Resolved: 18 years ago
Resolution: --- → DUPLICATE
Marking verified dup of 34283.
Status: RESOLVED → VERIFIED
You need to log in before you can comment on or make changes to this bug.