Open Bug 1368421 Opened 8 years ago Updated 1 year ago

every other touchpad scroll ignored on page

Categories

(Core :: Panning and Zooming, defect, P3)

55 Branch
All
macOS
defect

Tracking

()

Tracking Status
firefox53 --- wontfix
firefox54 --- affected
firefox55 --- affected

People

(Reporter: tnikkel, Unassigned)

References

()

Details

(Whiteboard: [gfx-noted])

Try to (two finger) touchpad scroll on the page https://www.lavelocita.cc/opinion-page/zwifters Every other touchpad scroll seems to be ignored.
I'm not able to reproduce on OS X or Windows. On Windows the two-finger touchpad scroll is flaky for me but that's true on all sites, so I think it's a hardware problem on my end. More details needed - what OS/version/hardware are you using? Is it consistently reproducible, exactly every other touchpad scroll?
Flags: needinfo?(tnikkel)
I'm testing on OS X. After more testing it seems that it is touchpad scrolls that you start while momentum from a previous touchpad scroll is still scrolling the page.
Flags: needinfo?(tnikkel)
On Linux, *touchpad* scrolling on the site works fine, but during two-finger *touchscreen* scrolling I see the symptoms that Timothy described.
Oh, I can repro now. For some reason it doesn't happen on my main profile but I can see it in other profiles and in a clean profile. (on OS X). Seems to happen on release, beta, and nightly, so not a recent regression.
OS: Unspecified → Mac OS X
Priority: -- → P3
Hardware: Unspecified → All
Whiteboard: [gfx-noted]
Version: Trunk → 55 Branch
(In reply to Botond Ballo [:botond] from comment #3) > On Linux, *touchpad* scrolling on the site works fine, but during two-finger > *touchscreen* scrolling I see the symptoms that Timothy described. Actually, it doesn't seem to be specific to two-finger touchscreen scrolling - one-finger scrolling (i.e. regular touch-panning) has the same issue. It looks like, when you touch-scroll the page, the page changes in some way that causes the event regions of the layer bearing the root content APZC to vanish. As a result, said APZC doesn't catch any events, and they just go to the root APZC (which doesn't do anything useful with them). Eventually, the event regions come back, and APZ event handling goes back to normal. I haven't investigated further, and I don't plan to in the immediate future. I mainly wanted to make sure this isn't a regression from bug 1180799, and I'm now convinced it's not.
Severity: normal → S3
You need to log in before you can comment on or make changes to this bug.