Tests seem ok with latest trunk.

Looks a lot betterm the rendering issues I remember seeing are gone. There are still two issues when I test locally: On both tests: the canvas content position seems to have an off by one error: they seem to be 1 pixel too low and 1 pixel too much to the right The image as a whole is significantly darker than the reference On fast/canvas/webgl/webgl-composite-modes-repaint.html: the whole image is very dark, much darker than the reference or fast/canvas/webgl/webgl-composite-modes.html result

(In reply to comment #2) > Looks a lot betterm the rendering issues I remember seeing are gone. > > There are still two issues when I test locally: > > On both tests: the canvas content position seems to have an off by one error: they seem to be 1 pixel too low and 1 pixel too much to the right > The image as a whole is significantly darker than the reference > > On fast/canvas/webgl/webgl-composite-modes-repaint.html: the whole image is very dark, much darker than the reference or fast/canvas/webgl/webgl-composite-modes.html result I think originally those to tests have different problems. webgl-composite-modes didn't show right because 1st frame display was broken. This was fixed by r139033 . webgl-composite-modes-repaint was broken because of issue that was fixed by r138265 . Now we have completely different issues. And still different for both tests. I think webgl-composite-modes is affected by https://bugs.webkit.org/show_bug.cgi?id=102374 . webgl-composite-modes-repaint might have wrong opacity on layer, but it just a guess. I'd suggest to separate this to different bugs.

Closing this bug because the EFL port has been removed from trunk. If you feel this bug applies to a different upstream WebKit port and was closed in error, please either update the title and reopen the bug, or leave a comment to request this.