Jeff, is it expected that gradient rendering is dependent on the destination surface size in some way?
Yes, bug 966996 caused some of that. It might be worth reverting the clip extents part of that patch to see if it helps. It's also very believable that gradient drawing could not be start position invariant. Skia's implementation currently has this problem.
Created attachment 8480935 [details] [diff] [review] Add test fuzz Unfortunately that didn't help :(
Attachment #8480935 - Flags: review?(jmuizelaar)
Attachment #8480935 - Flags: review?(jmuizelaar) → review+
Status: NEW → RESOLVED
Last Resolved: 4 years ago
Resolution: --- → FIXED
Target Milestone: --- → mozilla34
You need to log in before you can comment on or make changes to this bug.