From owner-freebsd-x11@FreeBSD.ORG Tue Jul 3 08:26:48 2007 Return-Path: X-Original-To: freebsd-x11@freebsd.org Delivered-To: freebsd-x11@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 6E14F16A400 for ; Tue, 3 Jul 2007 08:26:48 +0000 (UTC) (envelope-from adamk@voicenet.com) Received: from b.mx.visualtech.com (a.imap.visualtech.com [208.16.19.9]) by mx1.freebsd.org (Postfix) with ESMTP id 4951D13C4AE for ; Tue, 3 Jul 2007 08:26:48 +0000 (UTC) (envelope-from adamk@voicenet.com) Received: from [192.168.0.101] (c-68-45-180-45.hsd1.nj.comcast.net [68.45.180.45]) by b.mx.visualtech.com (Postfix) with ESMTP id 38C9B483F2; Tue, 3 Jul 2007 04:27:06 -0400 (EDT) From: Adam K Kirchhoff To: vehemens In-Reply-To: <200707021802.28460.vehemens@verizon.net> References: <200707021802.28460.vehemens@verizon.net> Content-Type: text/plain Date: Tue, 03 Jul 2007 04:26:44 -0400 Message-Id: <1183451204.9886.4.camel@sorrow> Mime-Version: 1.0 X-Mailer: Evolution 2.10.1 Content-Transfer-Encoding: 7bit Cc: freebsd-x11@freebsd.org Subject: Re: BERYL Response Times X-BeenThere: freebsd-x11@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: X11 on FreeBSD -- maintaining and support List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 03 Jul 2007 08:26:48 -0000 On Mon, 2007-07-02 at 18:02 -0700, vehemens wrote: > I seeing BERYL window moves and cube rotation hesitation as much as several > seconds after running for several minutes. Also after running for 30 minutes > or so, I see a significant increase in response delays. > > I've read that BERYL has some problems, but none mention this type of runtime > problem. > > The system is somewhat low end (XP 2100, 512MB, RADEON RV250), but it doesn't > appear to be a resource problem. > > Is anyone else seeing this? If not, what's your configuration? > _______________________________________________ Unless you applied some patches floating around this mailing list, the radeon driver does not currently support AIGLX, which means that you are using the copy render path. This is *much* slower than the normal texture_from_pixmap render path. The one time I tried this, beryl performed exactly as you described. Adam