Date: Mon, 25 Sep 2006 05:56:38 -0700 From: David Wolfskill <david@catwhisker.org> To: Andrew Kolchoogin <andrew@rinet.ru> Cc: freebsd-stable@freebsd.org Subject: Re: Direct Rendering Manager problems Message-ID: <20060925125638.GW698@bunrab.catwhisker.org> In-Reply-To: <45179684.10401@rinet.ru> References: <45179684.10401@rinet.ru>
next in thread | previous in thread | raw e-mail | index | archive | help
--VXH143WYhfIPisUc
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
On Mon, Sep 25, 2006 at 12:42:44PM +0400, Andrew Kolchoogin wrote:
> Dear colleagues,
>=20
> I've recently discovered some oddity with DRM on FreeBSD.
>=20
> Maybe it's a FAQ, but it seems to me that both on-demand kernel=20
> module loading and loading them in /boot/loader.conf is broken on recent=
=20
> RELENG_6.
That does not appear to be my experience; more details below.
> I've an ATI Radeon Mobility 9700 and self-built RELENG_6. When I=20
> started X, X.org server reports that no DRI available. Manual looking=20
> into /dev shows that no /dev/dri directory presents at all. Loading=20
> radeon.ko in /boot/loader.conf doesn't fix the problem. Also no=20
> DRM-related lines appears in dmesg output.
>=20
> But problem goes away after adding line "device radeondrm" to=20
> kernel configuration file, rebuilding kernel and rebooting the machine.=
=20
> Direct Rendering Manager initialises correctly, and node /dev/dri/card0=
=20
> exists.
>=20
> Are there any suggestions for additional testing?
Here is some relevant information from my current laptop. Note that its
disk drive was moved from the previous laptop (same model, but used an
nVidia video card), so I had merely commented out loading the nvidia
module while I was working on switching to the radeon module (and hadn't
yet got around to deleting the line).
g1-18(6.2-P)[1] uname -a
FreeBSD g1-18.catwhisker.org. 6.2-PRERELEASE FreeBSD 6.2-PRERELEASE #185: S=
un Sep 24 06:36:42 PDT 2006 root@g1-18.catwhisker.org.:/common/S1/obj/u=
sr/src/sys/LAPTOP_30W i386
g1-18(6.2-P)[2] ls -la /dev/dri/
total 1
dr-xr-xr-x 2 root wheel 512 Sep 24 10:19 .
dr-xr-xr-x 5 root wheel 512 Dec 31 1969 ..
crw-rw---- 1 root wheel 0, 135 Sep 24 10:18 card0
g1-18(6.2-P)[3] kldstat
Id Refs Address Size Name
1 16 0xc0400000 6a00b8 kernel
2 3 0xc0aa1000 1adb8 linux.ko
3 1 0xc0abc000 59f00 acpi.ko
4 1 0xc487f000 6000 linprocfs.ko
5 1 0xc4a13000 2000 rtc.ko
6 1 0xc4c15000 1e000 radeon.ko
7 1 0xc4c33000 e000 drm.ko
g1-18(6.2-P)[4] cat /boot/loader.conf
boot_verbose=3D"YES"
debug.mpsafenet=3D"1"
hw.ata.atapi_dma=3D"1"
# nvidia_load=3D"YES"
linux_load=3D"YES"
g1-18(6.2-P)[5] grep -i driver /etc/X11/XF86Config
Driver "keyboard"
Driver "mouse"
### Available Driver options are:-
Driver "radeon"
g1-18(6.2-P)[6] grep 'ATI Technologies Inc Radeon' /var/log/Xorg.0.log
(--) PCI:*(1:0:0) ATI Technologies Inc Radeon R250 Lf [FireGL 9000] rev 1, =
Mem @ 0xe0000000/27, 0xfcff0000/16, I/O @ 0xc000/8, BIOS @ 0x80000000/17
g1-18(6.2-P)[7]=20
So it looks to me as if X handles loading the radeon.ko module and
that (in turn) creates the /dev/dri/* entries -- as expected.
Peace,
david
--=20
David H. Wolfskill david@catwhisker.org
Believe SORBS at your own risk: 63.193.123.122 has been static since Aug 19=
99.
See http://www.catwhisker.org/~david/publickey.gpg for my public key.
--VXH143WYhfIPisUc
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (FreeBSD)
iEYEARECAAYFAkUX0gUACgkQmprOCmdXAD3utACfRq8sao5Ibb2YW4IqHZdjHxTI
4ZQAn3ZoIIR5KECxg4uskeahvAHLt5Qd
=jjay
-----END PGP SIGNATURE-----
--VXH143WYhfIPisUc--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060925125638.GW698>