Advice on "new" hardware.... (long post)

Kevin Kinsey kdk at daleco.biz
Wed Oct 3 15:23:23 PDT 2007


Hello,

Out of necessity, I recently replaced my workstation motherboard
(old one was apparently suffering a breakdown in the IDE controller
chip).  Funds being rather low, I opted for a fairly low priced
"combo" deal from a well-known online retailer.  I may be 
regretting that now.

I'm having a few ACPI related error messages, some problems with
USB, Xorg doesn't look too pretty, and sound isn't running.
Booting with ACPI off doesn't seem to help, and makes USB
worse.

BIOS date is July of this year, and so far I'm not actually
sure were to look for any update (appears to be a PcChips
board [yeah, I know]) and I wonder if there'd be an update
anyway, so soon.

The main ACPI problem that is evident is simply the error
message:  "acpi_tz0: _TMP value is absurd, ignored (-269.2C)".
That can be ignored, but I'm guessing that most of the other
problems are probably related to FBSD and this board's ACPI
tables not "playing nice" with each other.  Any chance at
all I'm right?

Sound problem is the biggie, and it seems to be here:

# cat /dev/sndstat
FreeBSD Audio Driver (newpcm)
Installed devices:
pcm0: <SiS 7012> at io 0xd800, 0xd400 irq 18 bufsz 16384 kld snd_ich (1p/1r/1v channels duplex default)

# dmesg | grep pcm
pcm0: <SiS 7012> port 0xd800-0xd8ff,0xd400-0xd47f irq 18 at device 2.7 on pci0
pcm0: AC97 reset timed out.
pcm0: <Analog Devices AD1888 AC97 Codec>
pcm0: ac97 codec reports dac not ready
pcm0: ac97 link rate calibration timed out after 1868021 us

And this is logged by the kernel after "play foo.wav":

   kernel: pcm0:virtual:0:dsp0.2: play interrupt timeout, channel dead

The above are "symptoms" (along with the fact that there's
no sound); is it likely that the problem is related to this?:

# pciconf -lv | grep -A4 pcm
pcm0 at pci0:2:7:  class=0x040100 card=0x18801019 chip=0x70121039 rev=0xa0 hdr=0x00
    vendor     = 'Silicon Integrated Systems (SiS)'
    device     = 'SiS7013 PCI Audio Accelerator'
    class      = multimedia
    subclass   = audio

It's not really a 7012, but a 7013 chip.  /sys/dev/sound/pci/ich.c indicates
that the 7012 "needs special handling" and that the "7012 register isn't
like the standard ICH".  I'm guessing that either the 7013 *is* more like
the standard ICH, or else it's different even than the 7012, because it doesn't
seem to wanna play right with pcm(4) when it thinks it's a 7012....

The USB problem so far isn't a biggie; it's worst when I boot
without ACPI; then I can't get USB to function at all (use it
for the mouse, mainly, and can work around that with a PS/2
mouse, I guess).

So, is there any hope for this hardware?  I could try HEAD, I guess,
but it doesn't appear that any work's been done in ICH on HEAD for
2 months or so (I'm currently running STABLE from yesterday).  What
about the ACPI?  Should I beg over on hackers@?  Anybody wanna donate
a nice modern mobo set?  [</evilgrin>, just had to ask].

Thanks for any insight!

Kevin Kinsey
-- 
Senate, n.:
	A body of elderly gentlemen charged with high duties and
misdemeanors.
		-- Ambrose Bierce


More information about the freebsd-questions mailing list