Closed Bug 1321422 Opened 8 years ago Closed 7 years ago

tps test seems to rely on invalidations of hidden frames

Categories

(Testing :: Talos, defect)

Product:
Testing
Component:
Talos
Version:
Version 3
Type:
defect
Priority:
Not set
Severity:
normal

Tracking

(Not tracked)

Status:
RESOLVED WONTFIX

People

(Reporter: neil, Unassigned)

References

(Blocks 1 open bug)

Details

For some reason, hidden frames get invalidated. This seemed to be unnecessary and could even cause session-windows-restored to fire early in some cases, so I tried to remove these invalidations in bug 1290158, however this "caused" a tps regression.

After playing around with the tps test locally for a bit, I noticed that its code to check for paint transactions was getting confused when I removed hidden invalidations - there are now fewer transactions, which means that the test has to wait longer to get the number of transactions that it wants.

I was hoping you could shed some light on why you test these transaction ids - they are confusing in as much as the last transaction id on the window utils and the transaction id in the paint events don't even seem to match up.
Flags: needinfo?(mconley)
(In reply to neil@parkwaycc.co.uk from comment #0)
> I was hoping you could shed some light on why you test these transaction ids
> - they are confusing in as much as the last transaction id on the window
> utils and the transaction id in the paint events don't even seem to match up.

Sure - I wrote about this in a mailing list post a few months back: https://groups.google.com/forum/#!searchin/mozilla.dev.platform/MozAfterPaint/mozilla.dev.platform/pCLwWdYc_GY/j9A-vWm3AgAJ

Hopefully that clears it up?
Flags: needinfo?(mconley)
is ther any work to do here?
Flags: needinfo?(neil)
Status: NEW → RESOLVED
Closed: 7 years ago
Resolution: --- → WONTFIX
Flags: needinfo?(neil)
You need to log in before you can comment on or make changes to this bug.