Subject: Re: support for unconfigured PCI devices, esp. SD card reader
To: Anne Bennett <anne@porcupine.montreal.qc.ca>
From: Quentin Garnier <cube@cubidou.net>
List: port-i386
Date: 07/30/2007 01:50:43
--NJ5+aVN4Egd/eJfU
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
On Sun, Jul 29, 2007 at 06:10:35PM -0400, Anne Bennett wrote:
>=20
> Hi, all.
>=20
> I have an HP Pavilion a1620n (Intel-based), and I have not yet managed
> to get all of its PCI devices completely recognized. I tried updating my
> kernel sources to the latest NetBSD 3.1 with:
>=20
> cd /usr/src/sys && cvs -q update -dP -r netbsd-3-1
>=20
> (and of course re-creating the kernel), but the command didn't patch
> /usr/src/sys/dev/pci/{pcidevs,pcidevs.h,pcidevs_data.h}, nor any
> /usr/src/sys/dev/pci/devname.c files, which I would expect to be needed to
> recognize new devices. The ones still giving me a bit of a hard time are:
>=20
> (1) unconfigured at pci1 ... Conexant product 0x2f20 (card reader?)
> (2) vga1 at pci0 ... Intel product 0x2772 [graphics card]
> (3) pcib0 at pci0 ... Intel product 0x27b1 (has sub-devices, anyway)
> (4) unconfigured at pci0 ... Intel SMBus Controller 8201GB
> (5) fw at fwohci0 not configured
>=20
> "2" and "3" seem to work despite the lack of a "name", and I don't
> even know what an SMBus Controller does, so that one is probably not
-current as of yesterday has support for the SMBus controller. Not
worth spending time on it, though.
> urgent either. "5" probably needs an entry in the kernel config file
> (maybe "fw* at fwohci?") but I don't see a handy example in the sample
> kernel configs.
As Pavel said, the Conexant chip is a modem (I'd guess you have a modem
PCI card in that box).
=46rom the dmesg output, I really don't see where the card reader could
be; even if umass(4) was missing, we'd have a line for an unconfigured
USB device.
Could you run acpidump from pkgsrc (in sysutils)? And make available
the file from acpidump -o file? I wonder if the card reader could be
an ACPI device; not that it would help us supporting it, but at least
locate it.
--=20
Quentin Garnier - cube@cubidou.net - cube@NetBSD.org
"You could have made it, spitting out benchmarks
Owe it to yourself not to fail"
Amplifico, Spitting Out Benchmarks, Hometakes Vol. 2, 2005.
--NJ5+aVN4Egd/eJfU
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (NetBSD)
iQEVAwUBRq0n09goQloHrPnoAQI19ggAsh96qofATink2e9QXg5JE/M8DYRNlmpZ
vNW6RwJaPl1Y/0XqeYWtl7pciRf8S+HUKOAPM3dYVzs7C5EfIyPoNHCCh9GL8yvo
u+bHVj3WPe9M+7QnlsgYJ6JoirDa+mz+jugpy4rZDFvATjE+iPX7hdaAbwcJYUyU
cEIuHUDWW6yR21yd2cQpiP7npdVLOJc2Wr5pHt00/UL7L/dQyAvT2gl3jpEaQYFz
qTeZD+TsXPvtKRLqnpjd9st10VxZW0h3Rh/gucysC+ichALtv865n5+i7kQX9itA
V4ZwOWhfQxtPisdZFaquFtv3a+mU+U96snBWYsJOXAb/7biN4Qi/FA==
=V+Dk
-----END PGP SIGNATURE-----
--NJ5+aVN4Egd/eJfU--