From owner-freebsd-multimedia@FreeBSD.ORG Mon Oct 16 17:21:20 2006 Return-Path: X-Original-To: freebsd-multimedia@freebsd.org Delivered-To: freebsd-multimedia@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 2292A16A407 for ; Mon, 16 Oct 2006 17:21:20 +0000 (UTC) (envelope-from rick@kiwi-computer.com) Received: from kiwi-computer.com (megan.kiwi-computer.com [63.224.10.3]) by mx1.FreeBSD.org (Postfix) with SMTP id 785A843D46 for ; Mon, 16 Oct 2006 17:21:19 +0000 (GMT) (envelope-from rick@kiwi-computer.com) Received: (qmail 77884 invoked by uid 2001); 16 Oct 2006 17:21:18 -0000 Date: Mon, 16 Oct 2006 12:21:18 -0500 From: "Rick C. Petty" To: usleepless@gmail.com Message-ID: <20061016172118.GA77730@keira.kiwi-computer.com> References: <45317970.5000508@bellsouth.net> <20061015064102.8780.qmail@web30310.mail.mud.yahoo.com> <20061016015911.GC57865@keira.kiwi-computer.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.4.2.1i Cc: freebsd-multimedia@freebsd.org Subject: Re: New port: pvrxxx for Hauppauge PVR150/500 X-BeenThere: freebsd-multimedia@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list Reply-To: rick-freebsd@kiwi-computer.com List-Id: Multimedia discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 16 Oct 2006 17:21:20 -0000 On Mon, Oct 16, 2006 at 08:40:04AM +0200, usleepless@gmail.com wrote: > > correct. however, ivtv seems to be able to download it much faster. i > have experimented with speeding up the i2c stuff, but failed ( ie gave > up ). Hmm, I could take a look at it. In my infinite spare time... > >Feel free (anyone) to look into the iic code and pull it out from under > >GIANT. > > will that help? reason i ask is because with the i2c, the cpu is > responsible for every bit-switch/line-switch in the protocol. so > through the pci-interface, it tells the card to pull the line up, you > wait a very short time, and tell it to pull the i2c-line down. etc... I'm assuming there is little (if any) delay between driving the data line and then toggling the clock line. I think as long as your timing loop guarantees that at most you're transferring 400 kbit/s, you should be fine. I think this can be done by using some type of timer in the kernel, so you can go in & out of the kernel and allow other interrupts to process, etc. > this weekend, i removed Giant from the pvrxxx-driver itself. i am on > UP myself, so i don't know what it is good for. Hrm, I thought GIANT was more useful than just for SMP. I guess it depends on whether the cxm driver sits in a timing loop, and whether other interrupts can happen while you're programming the DSP chip. My guess is that you'll notice some improvement even in UP. -- Rick C. Petty