<html> <head> <base href="https://bugs.freedesktop.org/" /> </head> <body> <div> <a class="bz_bug_link bz_status_NEW " title="NEW - Add FP64 support to the i965 shader backends" href="https://bugs.freedesktop.org/show_bug.cgi?id=92760#c70">Comment # 70</a> on <a class="bz_bug_link bz_status_NEW " title="NEW - Add FP64 support to the i965 shader backends" href="https://bugs.freedesktop.org/show_bug.cgi?id=92760">bug 92760</a> from <a class="email" href="mailto:itoral@igalia.com" title="Iago Toral <itoral@igalia.com>"> Iago Toral</a> <pre>Created <a href="attachment.cgi?id=122690" name="attach_122690" title="Sample test showcasing non-uniform control flow problems in align16">attachment 122690</a> <a href="attachment.cgi?id=122690&action=edit" title="Sample test showcasing non-uniform control flow problems in align16">[details]</a> Sample test showcasing non-uniform control flow problems in align16 I forgot to attach this. The test executes non-uniform control flow with DF operations in a vertex shader. If we run this in gen7, it will execute in align16 mode. The test renders a quad and is expected to produce 3 red vertices and 1 green vertex, but the result is 2 red and 2 green, which means that one vertex is taking the incorrect branch of the conditional, confirming Curro's point that we have an execmask problem. The test produces the expected result in gen8 because it runs VS in scalar mode.</pre> </div> <hr> You are receiving this mail because: <ul> <li>You are the QA Contact for the bug.</li> </ul> </body> </html>