<html> <head> <base href="https://bugs.freedesktop.org/"> </head> <body> <div> <a class="bz_bug_link bz_status_NEW " title="NEW - record-paint-alpha-clip-mask.xcb.rgb24 random failure" href="https://bugs.freedesktop.org/show_bug.cgi?id=84330#c5">Comment # 5</a> on <a class="bz_bug_link bz_status_NEW " title="NEW - record-paint-alpha-clip-mask.xcb.rgb24 random failure" href="https://bugs.freedesktop.org/show_bug.cgi?id=84330">bug 84330</a> from <a class="email" href="mailto:psychon@znc.in" title="Uli Schlachter <psychon@znc.in>"> Uli Schlachter</a> <pre>Ok, I was wrong in that this is a bug in the server. The part that I forgot is: Each render composite operation is only applied to the extents of the affected region. This means that "no trapezoids" means "no extents" means "no effect" and this is indeed a no-op. I don't know if the RENDER spec says that anywhere, but it is how the implementation behaves. The comment in the code indicates this with the comment saying "assert(trap extents == extents)". Obviously (as this bug shows), this assert is wrong. It might even still be wrong when we have a non-empty list of traps. So my proposed fix is to just drop this optimisation completely. I sent the following fix to the cairo mailing list: <a href="https://lists.cairographics.org/archives/cairo/2016-July/027619.html">https://lists.cairographics.org/archives/cairo/2016-July/027619.html</a></pre> </div> <hr> You are receiving this mail because: <ul> <li>You are the QA Contact for the bug.</li> </ul> </body> </html>