From mboxrd@z Thu Jan 1 00:00:00 1970 From: bugzilla-daemon@freedesktop.org Subject: [Bug 38800] glXSwapBuffersMscOML is slow on AMD Fusion but not on Intel 945 w/Atom Date: Thu, 7 Jul 2011 10:46:36 -0700 (PDT) Message-ID: <20110707174636.AEFA213004F@annarchy.freedesktop.org> References: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Received: from annarchy.freedesktop.org (annarchy.freedesktop.org [131.252.210.176]) by gabe.freedesktop.org (Postfix) with ESMTP id CC6DF9E758 for ; Thu, 7 Jul 2011 10:46:36 -0700 (PDT) In-Reply-To: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: dri-devel-bounces+sf-dri-devel=m.gmane.org@lists.freedesktop.org Errors-To: dri-devel-bounces+sf-dri-devel=m.gmane.org@lists.freedesktop.org To: dri-devel@lists.freedesktop.org List-Id: dri-devel@lists.freedesktop.org https://bugs.freedesktop.org/show_bug.cgi?id=38800 --- Comment #33 from Simon Farnsworth 2011-07-07 10:46:36 PDT --- (In reply to comment #31) > (In reply to comment #30) > > So wait for bo to be idle before waiting to be out of vblank > > Bad idea I'm afraid: Consider the case where the ioctl is called outside of > vblank, but by the time the BO is idle we're in vblank. We'd need to (probably > needlessly) wait for a full frame before the flip takes effect. Against that, and arguing in favour of Jerome's method: if userspace is submitting enough rendering that the BO goes idle just as vblank starts, it's already on a performance knife-edge. We're just arguing about which way to push you if you're already on that knife-edge. As Jerome's method is guaranteed to at least accurately report the timestamps of page flips, it's probably the right way to push you; I've been caught up in other issues today, but I'll try to code up an RFC patch for Jerome's idea tomorrow. -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug.