On Sun, Jun 21, 2009 at 01:31:45AM +0530, Manish Jain wrote:
> 
> Hi Roland,
> 
> Looks like nobody at freebsd-questions has any answers for me. Maybe 
> 7.2-x86 doesn't like newish amd hardware. Anyway, here are the 
> contents of /dev/sndstat :
> 
> > FreeBSD Audio Driver (newpcm: 32bit 2007061600/i386)
> > Installed devices:
> > pcm0: <HDA ATI RS690/780 HDMI PCM #0 Digital> at cad 0 nid 1 on hdac0 kld 
> > snd_hda [MPSAFE] (1p:1v/0r:0v channels default)
> > pcm1: <HDA Realtek ALC885 PCM #0 Analog> at cad 0 nid 1 on hdac1 kld 
> > snd_hda [MPSAFE] (1p:1v/1r:1v channels duplex)
> > pcm2: <HDA Realtek ALC885 PCM #1 Analog> at cad 0 nid 1 on hdac1 kld 
> > snd_hda [MPSAFE] (1p:1v/1r:1v channels duplex)
> > pcm3: <HDA Realtek ALC885 PCM #2 Digital> at cad 0 nid 1 on hdac1 kld 
> > snd_hda [MPSAFE] (1p:1v/1r:1v channels duplex)

> I tried snd_hda, even though it's listed as the driver for Intel
> HDA, not Realtek HDA on AMD hardware. It doesn't work. The volume
> control applet still complains no sound device has been configured.

The fact that /dev/sndstat exists means that there is a sound driver
that works. From the contents you can see that it is indeed the snd_hda
driver ("HDA Realtek ALC885 "). You can also see that there are four
sound outputs. Currently the digital HDMI is the default.

You can configure the sound hardware through sysctls. If you run "sysctl
hw.snd" you'll see a list of things you can change. With 'sysctl -d' you
can see what a particular sysctl does. In this case one that is
interesting is 'hw.snd.default_unit'. The different outputs in
/dev/sndstat will be connected to different sockets on the machine
(e.g. front/back panel). Try setting hw.snd.default_unit to 1, 2 or 3
and see if that helps.

It could be that you have to unload and reload the snd_hda module after
setting the default channel. You might also have to restart the volume
control applet.

> /dev/ums0 does get created I if plug in the mouse after the login 
> prompt has appeared.

Ok. I think it should be a bug that the mouse isn't properly detected
when connected at boot, but since the USB stack has been replaced in
8-CURRENT, I don't know if anyone is willing to go bughunting in the 7.x
USB stack.

> But the mouse would behave crazy till I changed the Xorg generated
> xorg.conf's setting of ZAxisMapping from "4 5 6 7" to "4 5".

That is not abnormal. A lot of mice need special zaxis mappings.

> I have still more problems. After all ports finished installing, 
> some X applications just don't start (eg Acroread). I can't any 
> error messages from the terminal because gnome-terminal itself 
> doesn't start.

Then use an xterm instead of a gnome-terminal. Acroread depends on Linux
emulation. Maybe you're missing some bits there?

> Given time, maybe I could ultimately figure this problem out, but if 
> you have any ready ideas, please be kind enough to share them me.
> 
> Here are some messages.
 
> #/var/log/console.log :
> > Jun 20 01:17:01 pbbi kernel: Jun 20 01:17:01 pbbi pulseaudio[1276]: main.c: 
> > Daemon startup without any loaded modules, refusing to work.

I've never used pulseaudio. I prefer esound.

> > Jun 20 01:18:04 pbbi kernel: Jun 20 01:18:04 pbbi 
> > gnome-keyring-daemon[1251]: dbus failure unregistering from session: 
> > Connection is closed

This looks pretty harmless.

> #/var/log/messages :
> > Jun 20 02:00:07 pbbi kernel: info: [drm] Loading RS780 CP Microcode
> > Jun 20 02:00:07 pbbi kernel: info: [drm] Loading RS780 PFP Microcode
> > Jun 20 02:00:07 pbbi kernel: info: [drm] Resetting GPU
> > Jun 20 02:00:07 pbbi kernel: info: [drm] writeback test succeeded in 1 usecs
> > Jun 20 02:00:07 pbbi kernel: drm0: [ITHREAD]
> > Jun 20 02:00:07 pbbi console-kit-daemon[1135]: WARNING: kvm_getenvv failed

Look at kvm_getenvv(3), in the BUGS section: 

     "In order for kvm_getenvv(3) to function correctly, procfs(5) must be
      mounted on /proc"

> loader.conf looks like following :
> > kern.maxdsiz="734003200"
> > ums_load="YES"
> > linprocfs_load="YES"
> > linsysfs_load="YES"
> > sound_load="YES"
> > snd_hda_load="YES"

Looks OK. 

Roland
-- 
R.F.Smith                                   http://www.xs4all.nl/~rsmith/
[plain text _non-HTML_ PGP/GnuPG encrypted/signed email much appreciated]
pgp: 1A2B 477F 9970 BA3C 2914  B7CE 1277 EFB0 C321 A725 (KeyID: C321A725)

Attachment: pgpfFPTsBATGd.pgp
Description: PGP signature

Reply via email to