Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Sat, 30 Nov 2002 21:25:08 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Sat, 30 Nov 2002 21:25:08 -0500 Received: from ns.indranet.co.nz ([210.54.239.210]:60891 "EHLO mail.acheron.indranet.co.nz") by vger.kernel.org with ESMTP id ; Sat, 30 Nov 2002 21:25:07 -0500 Date: Sun, 01 Dec 2002 15:27:52 +1300 From: Andrew McGregor To: Chris Ison , linux-kernel@vger.kernel.org Subject: Re: kernel space access to user space functions Message-ID: <23630000.1038709672@localhost.localdomain> In-Reply-To: <3DE91CBF.295C1491@bigpond.net.au> References: <3DE91CBF.295C1491@bigpond.net.au> X-Mailer: Mulberry/3.0.0b9 (Linux/x86) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1895 Lines: 48 --On Sunday, December 01, 2002 06:17:03 +1000 Chris Ison wrote: > I realize I asked this previously, but the answer given was not to the > question I asked. > > How can I get a kernel module to call a function within a program? As others have said, you can't. Unix in general doesn't work this way. However, all is not lost. There are other ways to accomplish what you want. > The reason being I am creating a software midi driver and already have a > small program that does what I want the driver to do, problem is all the > math in the program is floating point. > > What I would like to do, is be able to run the program, and have the > kernel software midi driver call a function within the program to que up > midi events, and have the program do all the hard work of the wavetable > synth. Actually, you want to have the program open a MIDI device, and read MIDI from there, then output to an audio device. ALSA already has the facilities to do this, and it should not be hard to write a MIDI cross-queue device for OSS if you need that. For low latency, you're going to have to be very careful about the structure of the main loop of this thing, and probably use mmap-ed audio under ALSA, or the low-latency DMA tricks with OSS. > This way, any improvements to the software don't have to be translated > to the driver, and visa versa. > > How can I make this happen. And please give an example. Since there are several softsynths done already, why not look at how they did it. http://www.xdt.com/ar/linux-snd/ is a very big list of linux sound apps. There should be examples amongst them. Andrew - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/