Closed Bug 939191 Opened 11 years ago Closed 11 years ago

FirefoxOS scrolling FPS test appears to be broken

Categories

(Testing Graveyard :: Eideticker, defect, P1)

x86_64
Linux
defect

Tracking

(Not tracked)

RESOLVED FIXED

People

(Reporter: wlach, Assigned: wlach)

Details

(Keywords: perf, Whiteboard: [c=automation p= s=2013.11.22 u=])

Attachments

(1 file)

So we're not measuring fps very well on FirefoxOS, that's a known problem.

However, with the settings app, we have the additional problem that it's not working as intended. It's supposed to be scrolling continually, but what we're seeing instead is a bit of movement followed by nothing. We should fix it.

Example of bad behaviour:

http://people.mozilla.org/~dhunt/eideticker-ci-dashboard/b2g.inari.mozilla-central.master.tests/framediff-view.html?title=Frame%20Difference%20FirefoxOS%20Settings%20Scrolling%20Performance%20%282013-11-12%29&video=videos/video-1384268771.71.webm&framediff=framediffs/framediff-1384268781.05.json&actionlog=actionlogs/action-log-1384268781.05.json
Keywords: perf
Priority: -- → P1
Whiteboard: [c=automation p= s= u=]
Working on this. The problem is that we're trying to scroll around the area of the header, which doesn't work at all. I'm probably going to use a similar strategy as we did with the callog test: get actual dimensions of settings content area, then scroll that.
Assignee: nobody → wlachance
Using a similar method to what we used in the callog scrolling test. Seems to work pretty ok.
Attachment #8334935 - Flags: review?(dave.hunt)
Comment on attachment 8334935 [details] [diff] [review]
0001-Bug-939191-Fix-settings-scrolling-test-r-davehunt.patch

Review of attachment 8334935 [details] [diff] [review]:
-----------------------------------------------------------------

Works well. With the debugging removed r=me.

::: src/tests/b2g/appscrolling/settings.py
@@ +7,5 @@
> +        settings_container = self.device.marionette.find_element(
> +            By.CSS_SELECTOR, '#root > div')
> +        settings_location = settings_container.location
> +        settings_swipe_padding = 16
> +        print settings_container.location

I'm guessing these are left over from debugging?

@@ +14,5 @@
> +        scroll_x1 = settings_location['x'] + settings_container.size['width'] / 2
> +        scroll_y1 = settings_location['y'] + (settings_container.size['height'] -
> +                                              settings_swipe_padding)
> +        scroll_y2 = settings_location['y'] + settings_swipe_padding
> +        print (scroll_x1, scroll_y1, scroll_y2)

Also debugging?
Attachment #8334935 - Flags: review?(dave.hunt) → review+
Status: NEW → RESOLVED
Closed: 11 years ago
Resolution: --- → FIXED
Whiteboard: [c=automation p= s= u=] → [c=automation p= s=2013.11.22 u=]
Product: Testing → Testing Graveyard
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: