<html> <head> <base href="https://bugs.freedesktop.org/"> </head> <body> <div> <a class="bz_bug_link bz_status_NEW " title="NEW - [i915] Feature request: Add support for fencing for PRIME setups" href="https://bugs.freedesktop.org/show_bug.cgi?id=95472#c64">Comment # 64</a> on <a class="bz_bug_link bz_status_NEW " title="NEW - [i915] Feature request: Add support for fencing for PRIME setups" href="https://bugs.freedesktop.org/show_bug.cgi?id=95472">bug 95472</a> from <a class="email" href="mailto:mario.kleiner@tuebingen.mpg.de" title="Mario Kleiner <mario.kleiner@tuebingen.mpg.de>"> Mario Kleiner</a> <pre>So Michel was right, and with the attached patch, page-flipped renderoffload works nicely on Intel Haswell + AMD Tonga under amdgpu-kms, flips properly waiting on the dmabuf fence for longer running rendering. I also retested Intel Ivybridge + AMD Cedar under radeon-kms and Intel Ironlake + GeForce 330M under nouveau-kms to make sure it doesn't cause other problems. Maybe we can still get this into Linux 4.8-rc6? That would be splendid. I'll also attach a hacky patch on top of Chris eb-fences branch which seems to remove the Z-Tearing for windowed copy-swaps - and replaces it be the regular horizontal tearing we always have with DRI3/Present, even for single gpu, because we are too slow to get blitting started ahead of scanout.</pre> </div> <hr> You are receiving this mail because: <ul> <li>You are the assignee for the bug.</li> <li>You are the QA Contact for the bug.</li> <li>You are on the CC list for the bug.</li> </ul> </body> </html>