Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 03 Jan 2012 13:48:35 +0100
From:      "O. Hartmann" <ohartman@mail.zedat.fu-berlin.de>
To:        Christian Gusenbauer <c47g@gmx.at>
Cc:        freebsd-stable@freebsd.org
Subject:   Re: x11/nvidia-driver: crash with FreeBSD 9.0-STABLE/amd64 when module loaded via /boot/loader.conf
Message-ID:  <4F02F923.2030903@mail.zedat.fu-berlin.de>
In-Reply-To: <201201031311.32751.c47g@gmx.at>
References:  <4F02D538.1020109@mail.zedat.fu-berlin.de> <201201031311.32751.c47g@gmx.at>

next in thread | previous in thread | raw e-mail | index | archive | help
This is an OpenPGP/MIME signed message (RFC 2440 and 3156)
--------------enig970125229F4BE0F05D30C28F
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

On 01/03/12 13:11, Christian Gusenbauer wrote:
> Hi Oliver!
>=20
> I'm using stable/9 and amd64 setup, too, but I do not experience such p=
age=20
> faults as you do. I've an older GeForce 7600 GT card and I'm using driv=
er=20
> 285.05.09 - neither OS nor driver built with clang. Perhaps it's a clan=
g=20
> issue?
>=20
> Ciao,
> Christian.
>=20
> On Tuesday 03 January 2012 11:15:20 O. Hartmann wrote:
>> On one of my FreeBSD 9.0 boxes running FreeBSD 9.0-STABLE/amd64 (CLANG=

>> built), equipted with a nVidia GT760Ti graphics board, loading the
>> nvidia module via /boot/loader.conf results in a page fault with rando=
m
>> error messages before rebooting.
>>
>> Booting the box without the nvidia.ko module loaded in he preboot phas=
e,
>> results in a stable system. I can then load the module via kldload and=

>> the system jumps then into X11-screen as usual.
>>
>> I recompile x11/nvidia-driver on a regular basis when recompiling the
>> kernel, so the module is always up to date and en par with the kernel
>> sources.
>>
>> The problem occurs idendependently of having kernel module drm.ko load=
ed
>> or not. I also tried falling back to the older nvidia driver 285.05.09=

>> (I use at the moment successfully 290.10).
>>
>> Before issuing a PR and doing a full debug reconfiguration of my box,
>> I'd like to ask whether this is a commong, well known problem since wi=
th
>> FreeBSD 9.0, I'd like to stay with STABLE rather than with RELEASE.
>>
>> Regards,
>> Oliver

Hello Christian,
I just checked with a laptop which has a dedicated nVidia Quadro 3100M
(Dell Latidude E6510), which has also world and kernel build with CLANG.
I do not experience this pagefault on that laptop either. I also have a
box running nearly the same hardware and the same GTX560Ti, but FreeBSD
10.0-CURRENT, and this box' OS is also built via CLANG. There is also no
issue. Maybe it's an BIOS issue triggered by the BIOS.
I guess, I can not circumvent any debugging ...

Thanks,
regards,
Oliver


--------------enig970125229F4BE0F05D30C28F
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: OpenPGP digital signature
Content-Disposition: attachment; filename="signature.asc"

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.18 (FreeBSD)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iF4EAREIAAYFAk8C+S8ACgkQU6Ni+wtCKv+YXwEAmYVMnbHZnmGFp5lAuhojTw5F
B038paPFzVkDM2J2Mi8A+gN3T6mQIIkWXU83NPW4ZdQ12kgxqASQnq4wdoKGP3tq
=SguW
-----END PGP SIGNATURE-----

--------------enig970125229F4BE0F05D30C28F--



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4F02F923.2030903>