From owner-freebsd-multimedia Fri Jan 26 3: 2:55 2001 Delivered-To: freebsd-multimedia@freebsd.org Received: from relay2.agava.net.ru (2.oivt.mipt.ru [193.125.142.2]) by hub.freebsd.org (Postfix) with ESMTP id 6A34037B402 for ; Fri, 26 Jan 2001 03:02:35 -0800 (PST) Received: from juil.domain (juil.domain [192.168.1.50]) by relay2.agava.net.ru (Postfix) with ESMTP id F3FC1437D7; Fri, 26 Jan 2001 14:02:31 +0300 (MSK) Date: Fri, 26 Jan 2001 14:02:41 +0300 (MSK) From: Ilya Martynov X-X-Sender: To: Alexander Matey Cc: Subject: Re: Aureal card sound driver stops working In-Reply-To: <20010124150003.A3049@cis.ohio-state.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-multimedia@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org AM> [..skip..] AM> AM> With as much details as you gave me, it looks like these nfs timeouts AM> cause the application that has opened the device to die silently without AM> closing the device properly. XMMS haven't die. If this happens (lock of /dev/dsp) than I can see dialog box telling me something about that it can't longer output sound (sorry I don't remember this message exactly). However XMMS doesn't opens /dev/pcm directly - it works via esd. I've not noticed core dumps of esd (in log messages). I can be mistaken now since some time have passed but I'm almost sure I've not seen it. Any anyway esd doesn't know anything about NFS. I'm going to try to reproduce it with playing via XMMS mp3 from my local filesystem. I belive that this problem is not caused by NFS timeouts. AM> newpcm still thinks that the device is open and doesn't let you AM> reopen it. AM> AM> It is not a driver that checks whether the device is busy, it is newpcm AM> that tracks open/close requests and keeps the record on a higher level. Another questions. Aureal Vortex 8810 is supposed to have four independent chanels. Right? I've created additional pcm devices: ls -l /dev/dsp /dev/dsp0* lrwxrwxrwx 1 root wheel 4 Dec 8 02:10 /dev/dsp -> dsp0 crw-rw-rw- 1 root wheel 30, 3 Jan 24 21:29 /dev/dsp0 crw-rw-rw- 1 root wheel 30, 0x00010003 Dec 8 02:11 /dev/dsp0.1 crw-rw-rw- 1 root wheel 30, 0x00020003 Dec 8 02:11 /dev/dsp0.2 crw-rw-rw- 1 root wheel 30, 0x00030003 Dec 8 02:11 /dev/dsp0.3 crw-rw-rw- 1 root wheel 30, 0x00040003 Dec 8 02:11 /dev/dsp0.4 Before this problem with locking /dev/dsp is happen I do able to create four applications that use these four chanels - I've checked it. If it is a problem in newpcm because it haven't registered close request for /dev/dsp, which is actually is same as /dev/dsp0.1 (is it right?) why I can't use /dev/dsp0.2, /dev/dsp0.3 or /dev/dsp0.4? Once /dev/dsp becomes locked they also become locked. -- Ilya Martynov AGAVA Software Company, http://www.agava.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-multimedia" in the body of the message