[Webkit-unassigned] [Bug 39123] Unneeded fillRect call in RenderBoxModelObject::paintFillLayerExtended

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=39123


Darin Adler <darin at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |darin at apple.com




--- Comment #8 from Darin Adler <darin at apple.com>  2010-05-20 10:39:43 PST ---
The issue is not whether fillRect can be slow. It's whether the cost of checking for this special case pays for itself or not. If the special case is rare enough, the check for it can be more costly than the optimization.

Further, some graphics systems already have early exits for the case of filling objects with a fully-transparent color. So adding a second check of that might help performance on some platforms and hurt performance slightly on others. The better fix if this is the case might be to make sure the lower level graphics code checks for this consistently on all platforms. It's not clear to me that putting this optimization into the rendering layer is the best way to make things faster.

The reason to check at a higher level is that you can save even more computation and overhead if you check at as high a level as possible.

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.