From: Roland Scheidegger <rscheidegger_lists@hispeed.ch>
To: Linux and Kernel Video <video4linux-list@redhat.com>
Cc: linux-kernel@vger.kernel.org
Subject: Re: PROBLEM: four bttv tuners in one PC crashed
Date: Tue, 07 Mar 2006 20:08:54 +0100 [thread overview]
Message-ID: <440DDA46.2010503@hispeed.ch> (raw)
In-Reply-To: <440D7384.5030307@cern.ch>
Jiri Tyr wrote:
> Michel Bardiaux wrote:
>
> >I dont see anything suspicious conflict. But 'overlay' has been
> >mentionned, I'm not familiar with that but I assume it means you're
> >doing video grabs directly to the X display, right?
>
> Yes, I'm using xawtv + overlay directly to the X display :0.0. Is better
> way to use it with some WM (KDE, Gnome, ...)?
>
> >Then it could be a problem with the graphics card or its driver.
>
> On the mainboard is graphics card Intel 915. I have tryed it with i810
> (with / without DRI) and vesa driver. I have also tryed PCIE graphics
> card from ATI. In all of those cases the PC freeze and as well as if is
> in PCI slot only ONE tuner!
>
> >Just my 2 cents, because I am totally unfamiliar with such uses, sounds
> >like a problem for the MythTV crowd.
>
> I'm totally lost! I have here 25 PCs (100 tuners) what always freeze ;o(
FWIW, I have a box (with only one bttv card) which does not work stable
neither due to the capture card. It DID run stable at one point, but I
think that was kernel 2.2-ish... I tried quite a few things, IIRC when
it did run stable I had to tweak pci latency timer in the bios, but this
no longer works. Tried kernels with/without apic, acpi, smp,...
I also get the same OCERR error messages, and very similar oops. Though
it usually locks up only once every few hours or even days. Sometimes
hard and sometimes soft lockups.
The bttv driver/chip seems to cause random memory corruption sometimes,
processes will just start dying... My guess is that somehow dma
transfers to random / wrong addresses happens when there are lost irqs /
pci bus contention, either due to a driver bug or a chip bug (something
like losing parts of the risc program) though if there is a chip bug,
that behaviour never happened with some other OS so a workaround should
be possible), but I could be ways off.
This was with a i440bx chipset btw, it runs just fine with a newer box
(though overlay had problems there too, but since the box is faster I'm
just using grabdisplay, which didn't do anything for stability on the
old box). Tried both a bt848 and bt878 card which didn't change anything.
Roland
next prev parent reply other threads:[~2006-03-07 19:09 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-03-06 15:34 PROBLEM: four bttv tuners in one PC crashed Jiri Tyr
2006-03-06 15:56 ` Duncan Sands
2006-03-06 15:58 ` Jiri Tyr
2006-03-06 18:39 ` Jiri Tyr
2006-03-07 8:22 ` Jiri Tyr
2006-03-07 11:50 ` Jiri Tyr
2006-03-07 12:32 ` Duncan Sands
2006-03-15 8:48 ` Jiri Tyr
2006-03-15 14:21 ` Sami Farin
2006-03-15 14:30 ` Jiri Tyr
2006-03-15 15:25 ` Jiri Tyr
2006-03-15 15:41 ` Sami Farin
2006-03-07 19:08 ` Roland Scheidegger [this message]
2006-03-07 19:43 ` Duncan Sands
[not found] <5NpZk-7wW-13@gated-at.bofh.it>
[not found] ` <5NJ1x-1OE-15@gated-at.bofh.it>
[not found] ` <5NPJk-3GD-3@gated-at.bofh.it>
[not found] ` <5NRBv-6ze-9@gated-at.bofh.it>
2006-03-11 10:08 ` Bodo Eggert
2006-03-12 11:21 ` Sami Farin
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=440DDA46.2010503@hispeed.ch \
--to=rscheidegger_lists@hispeed.ch \
--cc=linux-kernel@vger.kernel.org \
--cc=video4linux-list@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox