From owner-freebsd-multimedia Tue Sep 2 16:00:13 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id QAA03080 for multimedia-outgoing; Tue, 2 Sep 1997 16:00:13 -0700 (PDT) Received: from ic4.ic.dk (qmailr@globip70.image.dk [194.234.57.70]) by hub.freebsd.org (8.8.7/8.8.7) with SMTP id PAA03026 for ; Tue, 2 Sep 1997 15:59:58 -0700 (PDT) Received: (qmail 11284 invoked from network); 2 Sep 1997 22:56:01 -0000 Received: from ic1.ic.dk (193.88.65.12) by ic4.ic.dk with SMTP; 2 Sep 1997 22:56:01 -0000 Received: from jblhome by ic1.ic.dk with UUCP id AA03778 (5.65c8/IDA-1.4.4j); Wed, 3 Sep 1997 00:57:01 +0200 Received: (from jacob@localhost) by pippin.jblhome.ping.dk (8.8.7/8.7.3) id AAA02221; Wed, 3 Sep 1997 00:27:01 +0200 (CEST) To: Luigi Rizzo Cc: multimedia@FreeBSD.ORG Subject: Re: Unsuccesful attempt at SB16PNP and no PNP bios; what am I doing wrong? References: <199709022020.WAA06643@labinfo.iet.unipi.it> From: Jacob Bohn Lorensen Date: 03 Sep 1997 00:27:01 +0200 In-Reply-To: Luigi Rizzo's message of Tue, 2 Sep 1997 22:20:30 +0200 (MET DST) Message-Id: <87sovne4i2.fsf@pippin.jblhome.ping.dk> Lines: 63 X-Mailer: Gnus v5.3/Emacs 19.34 X-Charset: ISO_8859-1 X-Char-Esc: 29 Sender: owner-freebsd-multimedia@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Luigi Rizzo writes: > > So, I went ahead with sb_dsp.c. in the function sb16pnp_attach, I > > inserted the following lines (marked with +) > > > > read_pnp_parms ( &d , 0 ) ; > > + d.irq[0] = 7; > > + d.drq[0] = 3; d.drq[1] = 5; > > + d.port[0] = 0x220; d.port[1] = 0x330; d.port[2] = 0x388; > > dev->id_iobase = d.port[0]; > > write_pnp_parms ( &d , 0 , 1); > correct. The only doubt is if irq7 might conflict with the parallel > port. It doesn't (I am running the lpt port in polled mode...) > > It seems that both isa_dmastatus() and isa_dmastop() have been checked > > into src-cur-3027, so I did not need to patch isa.c, only autoconf.c, > > files.i386 and of course, copying pnp.c and pnp.h. > correct, there is only one minor difference in isa_dmastatus, where my > code returns 0 if the channel is idle, whereas -current returns -2. > I'll fix this in the next snap (sometime later this week). Ok, it seems everything works - sort-of-ok - after all! The only reason I did not discover this was because the cables were not correctly mounted (*blush*), and that the sound is extremely low. I am not able to control the volume with xmix - there seems to be only two volumes: no sound and almost no sound (I have the SOunblaster connected to my stereo and have turned the volume control on the stereo on max). Furthermore, the volume level alternates between on/off when I try to raise or lower it. /usr/sbin/mixer seems to work ok, but I still can't get enough output. > > default ioctl snd1 subdev 4 fn 0x402c7413 fail > > default ioctl snd1 subdev 4 fn 0x402c7413 fail > yes I forgot some debugging messages enabled. I'll remove them in the > next snap. Still they tell you that you are actually getting interrupts > for the device, meaning that the card is working. If you use a real .au > file you should be able to hear it (some mixer calls are still > unimplemented for the SB16 so it might be that you cannot control the > volume, but the basic mechanisms seem to work on your card. The > problem is, I don't have the card yet so I cannot test it. That's right, I seem to be able to play .au files just fine (except for the volume problem mentioned above). I seem, however, to be unable to record (cat /dev/audio >/tmp/testsample); it just hangs until I press Ctrl-C, then I get this message in the syslog: timeout flushing dma2, cnt 0x0 flags 0x00000451 is this related to the -2 vs. 0 return value from dmastatus, you mentioned? Thanks, Jacob. -- Jacob Lorensen; Mosebuen 33, 1.; DK-2820 Gentofte, Denmark; +45-31560401 PGP ID = E596F0B5; PGP Fingerprint = 1E8726467436DC4A 723B6678C5AD9E71