578134 - REFTEST TEST-UNEXPECTED-FAIL | file:///c:/talos-slave/mozilla-central-win7-opt-u-reftest-d2d/build/reftest/tests/layout/reftests/box-shadow/boxshadow-rounded-spread.html

Status: RESOLVED WORKSFORME

(Core :: Web Painting, defect)

Description

[Armen [:armenzg]]

Attachment: log

After enabling reftests for Direct 2D we are hitting this failure in this test suite.

Comment 1

[Zack Weinberg (:zwol)]

Attachment: analyzed log

There are a very small number of divergent pixels on the outer edge of the circle.  I have no idea whether it is practical to fix that.

Comment 2

[John O'Duinn [:joduinn] (please use "needinfo?" flag)]

(In reply to comment #1)
> Created attachment 456909 [details]
> analyzed log
> 
> There are a very small number of divergent pixels on the outer edge of the
> circle.  I have no idea whether it is practical to fix that.

If that becomes the formal decision from Dev, can the test be modified to return green for that case?

Comment 3

[Jeff Muizelaar [:jrmuizel]]

(In reply to comment #2)
> (In reply to comment #1)
> > Created attachment 456909 [details] [details]
> > analyzed log
> > 
> > There are a very small number of divergent pixels on the outer edge of the
> > circle.  I have no idea whether it is practical to fix that.
> 
> If that becomes the formal decision from Dev, can the test be modified to
> return green for that case?

I expect so yes.

Comment 4

[Bas Schouten (:bas.schouten)]

(In reply to comment #2)
> (In reply to comment #1)
> > Created attachment 456909 [details] [details]
> > analyzed log
> > 
> > There are a very small number of divergent pixels on the outer edge of the
> > circle.  I have no idea whether it is practical to fix that.
> 
> If that becomes the formal decision from Dev, can the test be modified to
> return green for that case?

We've talked fuzzy matching before, but it's a tricky area. And additionally it's probably only a good idea if we understand why there are these tiny discrepancies.

Comment 5

[Robert O'Callahan (:roc) (email my personal email if necessary)]

I actually think that what we should do here is change nsCSSRendering::PaintBackgroundWithSC to paint the background color by filling the background shape instead of clipping to the background shape and then filling the dirty area. This should actually be faster as well as making these test failures go away.

Comment 6

[Ed Morley [:emorley]]

Mass marking whiteboard:[orange] bugs WFM (to clean up TBPL bug suggestions) that:
* Haven't changed in > 6months
* Whose whiteboard contains none of the strings: {disabled,marked,random,fuzzy,todo,fails,failing,annotated,leave open,time-bomb}
* Passed a (quick) manual inspection of bug summary/whiteboard to ensure they weren't a false positive.

I've also gone through and searched for cases where the whiteboard wasn't labelled correctly after test disabling, by using attachment description & basic comment searches. However if the test for which this bug was about has in fact been disabled/annotated/..., please accept my apologies & reopen/mark the whiteboard appropriately so this doesn't get re-closed in the future (and please ping me via IRC or email so I can try to tweak the saved searches to avoid more edge cases).

Sorry for the spam! Filter on: #FFA500