Closed
Bug 1280518
Opened 8 years ago
Closed 4 months ago
Displayport shown on APZ minimap is inconsistent with what is rendered
Categories
(Core :: Panning and Zooming, defect, P3)
Tracking
()
RESOLVED
DUPLICATE
of bug 1604280
Tracking | Status | |
---|---|---|
firefox47 | --- | unaffected |
firefox48 | --- | wontfix |
firefox49 | --- | wontfix |
firefox50 | --- | wontfix |
firefox51 | --- | fix-optional |
firefox52 | --- | fix-optional |
People
(Reporter: botond, Unassigned)
References
(Blocks 1 open bug)
Details
(Keywords: regression, Whiteboard: [gfx-noted])
STR: 1. Turn on apz.minimap.enabled in about:config 2. Navigate to https://bugzilla.mozilla.org/show_bug.cgi?id=1234158 3. Press "End" to scroll to the bottom 4. Press "Up" a few times to scroll up a bit, by less than a page. 5. Wheel-scroll back down to the bottom. At the end of the wheel-scroll, the displayport rect shown on the minimap does not extend down to the bottom of the page, while the viewport does, as if we were checkerboarding, but we are not in fact checkerboarding - the bottom of the page is rendered in full.
Comment 1•8 years ago
|
||
I can sort of repro on OS X. It's most likely a regression from bug 1257641, because the empty transaction updates the scroll offset in the frame metrics, which changes where the minimap draws the displayport. However the displayport hasn't actually moved. Presumably this means anywhere we're computing the page-relative displayport in the compositor it might be a little off, so we should fix this.
Blocks: 1257641
status-firefox47:
--- → unaffected
status-firefox48:
--- → affected
status-firefox49:
--- → affected
status-firefox50:
--- → affected
Keywords: regression
Priority: -- → P3
Whiteboard: [gfx-noted]
Comment 2•8 years ago
|
||
I took a look through the users of FrameMetrics::GetDisplayPort and other than the minimap the only other call site that looks like it would be affected is the one that calls SyncFrameMetrics to sync the displayport to java-land on Fennec. So it probably doesn't impact user-visible correctness as much as I thought it might.
Doesn't sound like we'd uplift a fix to beta, so wontfix for 48.
Updated•8 years ago
|
Version: unspecified → 48 Branch
Updated•8 years ago
|
Updated•8 years ago
|
Updated•8 years ago
|
Updated•2 years ago
|
Severity: normal → S3
Reporter | ||
Comment 4•4 months ago
|
||
This issue no longer occurs with the WebRender implementation of the minimap added in bug 1604280.
You need to log in
before you can comment on or make changes to this bug.
Description
•