Mailing List Archive

On 3/21/2011 10:56 PM, Dan Serban wrote:
> While I wrote an original email detailing absolutely everything I could
> about my PVR-150's locking up for the past few months. I think I should
> just take this one step at a time.
>
> It seems that the mailing list contains some posts from frustrated users
> out there that have the same problem. In some cases it seems like a patch
> will be applied to handle the issue.
>
> Is there a patch against any current kernel release?
>
> I'll start there, but I do have a lot of time invested in gathering
> possible information on this, my backend has been unusable since November
> due to this bug. I've tried many different things to alleviate the issue,
> though I feel that maybe I'm overlooking something simple. Like a bug in
> the IVTV driver.
>
> Let me know if that's the case :).

I had what may be the same problem on a system with an AMD Athlon X2 4850e. I fixed it by disabling AMD Cool n Quiet in the BIOS. If you see messages like this in your dmesg output:
ivtv0: DMA TIMEOUT 00000001 0
and if you've got a similar CPU, then maybe the same BIOS change will work for you.

HTH,
Helen

_______________________________________________
ivtv-users mailing list
ivtv-users@ivtvdriver.org
http://ivtvdriver.org/mailman/listinfo/ivtv-users

On 03/21/11 20:31, faginbagin wrote:
>
>
> On 3/21/2011 10:56 PM, Dan Serban wrote:
>> While I wrote an original email detailing absolutely everything I could
>> about my PVR-150's locking up for the past few months. I think I
>> should
>> just take this one step at a time.
>>
>> It seems that the mailing list contains some posts from frustrated users
>> out there that have the same problem. In some cases it seems like a
>> patch
>> will be applied to handle the issue.
>>
>> Is there a patch against any current kernel release?
>>
>> I'll start there, but I do have a lot of time invested in gathering
>> possible information on this, my backend has been unusable since
>> November
>> due to this bug. I've tried many different things to alleviate the
>> issue,
>> though I feel that maybe I'm overlooking something simple. Like a
>> bug in
>> the IVTV driver.
>>
>> Let me know if that's the case :).
>
> I had what may be the same problem on a system with an AMD Athlon X2
> 4850e. I fixed it by disabling AMD Cool n Quiet in the BIOS. If you
> see messages like this in your dmesg output:
> ivtv0: DMA TIMEOUT 00000001 0
> and if you've got a similar CPU, then maybe the same BIOS change will
> work for you.
>

No, in fact my dmesg output is eerily quiet. I do believe it has
something to do with my PCI bus being saturated, as, once in a while I
do get a green recording. Though again, no dmesg output telling me
there were any DMA errors.

Cool 'n quiet is turned off for me, this is a 939 socket opteron 165
dual core.

Sometimes I do get to see things like:

[164049.321003] ivtv1: All encoder MPG stream buffers are full. Dropping
data.
[164049.325004] ivtv1: Cause: the application is not reading fast enough.

Upping the buffers via the module command line sends the system into a
non-bootable death spiral, I think the problem stems from just too many
PCI cards in the system, though I'd hate to remove what I've got :/

> HTH,
> Helen
>
> _______________________________________________
> ivtv-users mailing list
> ivtv-users@ivtvdriver.org
> http://ivtvdriver.org/mailman/listinfo/ivtv-users


_______________________________________________
ivtv-users mailing list
ivtv-users@ivtvdriver.org
http://ivtvdriver.org/mailman/listinfo/ivtv-users

On 03/21/11 19:56, Dan Serban wrote:
> While I wrote an original email detailing absolutely everything I could
> about my PVR-150's locking up for the past few months. I think I should
> just take this one step at a time.
>
> It seems that the mailing list contains some posts from frustrated users
> out there that have the same problem. In some cases it seems like a patch
> will be applied to handle the issue.
>
> Is there a patch against any current kernel release?
>
> I'll start there, but I do have a lot of time invested in gathering
> possible information on this, my backend has been unusable since November
> due to this bug. I've tried many different things to alleviate the issue,
> though I feel that maybe I'm overlooking something simple. Like a bug in
> the IVTV driver.
>
> Let me know if that's the case :).
>
> _______________________________________________
> ivtv-users mailing list
> ivtv-users@ivtvdriver.org
> http://ivtvdriver.org/mailman/listinfo/ivtv-users
>

Thought I'd add a little more hardware information as my email was just
_too_ sparse :).

Motherboard: MSI 7125 Socket 939 era
CPU: Opteron 165 (dual core, not overclocked)
1x Hauppauge PVR 500
2x Hauppauge PVR 150
2x Sil 3114 PCI 4x SATA controller (1 on-board, 1 in a pci slot)

All 4 DRAM banks are populated with 1GB sticks.

The curious thing about this whole setup is the PCI addressable memory
space, it's a little odd, for an older machine as this (with no further
BIOS updates), I'd expect lesser addressable space via the BIOS, but I
get almost all of the 4GB, free reports 3996116 total memory.

I've fiddled with the BIOS memory mappings, one resulted in all of the
memory being available and the PCI cards didn't get much of anything,
the other way disabled 1 GB (3gigs available to the system) and the PCI
cards seemed to starve for addressable memory as well.

To conclude, I do have my suspicions that somehow my PCI bus beings
overburdened would cause these lockups, though they are extremely
difficult to diagnose and/or reproduce.

_______________________________________________
ivtv-users mailing list
ivtv-users@ivtvdriver.org
http://ivtvdriver.org/mailman/listinfo/ivtv-users

On 03/21/11 19:56, Dan Serban wrote:
> While I wrote an original email detailing absolutely everything I could
> about my PVR-150's locking up for the past few months. I think I should
> just take this one step at a time.
>
> It seems that the mailing list contains some posts from frustrated users
> out there that have the same problem. In some cases it seems like a patch
> will be applied to handle the issue.
>
> Is there a patch against any current kernel release?
>
> I'll start there, but I do have a lot of time invested in gathering
> possible information on this, my backend has been unusable since November
> due to this bug. I've tried many different things to alleviate the issue,
> though I feel that maybe I'm overlooking something simple. Like a bug in
> the IVTV driver.
>
> Let me know if that's the case :).
>
> _______________________________________________
> ivtv-users mailing list
> ivtv-users@ivtvdriver.org
> http://ivtvdriver.org/mailman/listinfo/ivtv-users
>

Is this a bug with IVTV? Do I have too many PCI devices? Is IO a
bottleneck and possibly causing one of the PVR150's to hang?

What can I look at to resolve my problem?

_______________________________________________
ivtv-users mailing list
ivtv-users@ivtvdriver.org
http://ivtvdriver.org/mailman/listinfo/ivtv-users

----- Original Message -----
> From: "Dan Serban" <dserban@lodgingcompany.com>

> On 03/21/11 19:56, Dan Serban wrote:

> > Is there a patch against any current kernel release?
> >
> > I'll start there, but I do have a lot of time invested in gathering
> > possible information on this, my backend has been unusable since
> > November due to this bug. I've tried many different things to alleviate the
> > issue, though I feel that maybe I'm overlooking something simple. Like a
> > bug in the IVTV driver.
> >
> > Let me know if that's the case :).
>
> Is this a bug with IVTV? Do I have too many PCI devices? Is IO a
> bottleneck and possibly causing one of the PVR150's to hang?
>
> What can I look at to resolve my problem?

Possibly. Did you report your kernel and IVTV versions; I don't recall
seeing them?

I do have this message in my own logs:

> *********************** WARNING ***********************
> ivtv drivers prior to 0.10.0 can cause lockups when
> reading VBI. Drivers between 0.10.5 and 1.0.3+ do not
> properly capture VBI data on PVR-250 and PVR-350 cards.

though I don't know if it's going to be related to your problem
or not, or whether you in fact see it.

That machine has kernel:

> Linux plaintain 2.6.34.7-0.3-desktop #1 SMP PREEMPT 2010-09-20 15:27:38 +0200 i686 athlon i386 GNU/Linux

and ivtvctl 0.10.5, though the driver version seems to be

> ivtv: Start initialization, version 1.4.1.

We don't see lockups (anymore, since I replaced the motherboard), but
we do get two flavors of stuttering, one which also stutters the audio,
and one which doesn't. I suspect bus mastering problems or XFS problems,
but haven't been able to pin it down yet.

Cheers,
-- jra

_______________________________________________
ivtv-users mailing list
ivtv-users@ivtvdriver.org
http://ivtvdriver.org/mailman/listinfo/ivtv-users

On Thu, 24 Mar 2011 18:41:35 -0400 (EDT)
Jay Ashworth <jra@baylink.com> wrote:

> ----- Original Message -----
> > From: "Dan Serban" <dserban@lodgingcompany.com>
>
> > On 03/21/11 19:56, Dan Serban wrote:
>
> > > Is there a patch against any current kernel release?
> > >
> > > I'll start there, but I do have a lot of time invested in gathering
> > > possible information on this, my backend has been unusable since
> > > November due to this bug. I've tried many different things to
> > > alleviate the issue, though I feel that maybe I'm overlooking
> > > something simple. Like a bug in the IVTV driver.
> > >
> > > Let me know if that's the case :).
> >
> > Is this a bug with IVTV? Do I have too many PCI devices? Is IO a
> > bottleneck and possibly causing one of the PVR150's to hang?
> >
> > What can I look at to resolve my problem?
>
> Possibly. Did you report your kernel and IVTV versions; I don't recall
> seeing them?
>

My apologies for the super late reply, I've been fighting other things
lately, not just this oddity.

My kernel is 2.6.32, debian squeeze standard kernel.

I'm not at the location where the server is, but I'll attempt to give you
exact specifics.

> I do have this message in my own logs:
>
> > *********************** WARNING ***********************
> > ivtv drivers prior to 0.10.0 can cause lockups when
> > reading VBI. Drivers between 0.10.5 and 1.0.3+ do not
> > properly capture VBI data on PVR-250 and PVR-350 cards.
>
> though I don't know if it's going to be related to your problem
> or not, or whether you in fact see it.
>

I do capture VBI, and do see that output, but my ivtv driver is much newer
than the ones its warning about. So I never gave it any importance.

> That machine has kernel:
>
> > Linux plaintain 2.6.34.7-0.3-desktop #1 SMP PREEMPT 2010-09-20 15:27:38
> > +0200 i686 athlon i386 GNU/Linux
>
> and ivtvctl 0.10.5, though the driver version seems to be
>
> > ivtv: Start initialization, version 1.4.1.
>
> We don't see lockups (anymore, since I replaced the motherboard), but
> we do get two flavors of stuttering, one which also stutters the audio,
> and one which doesn't. I suspect bus mastering problems or XFS problems,
> but haven't been able to pin it down yet.

Yeah, it's odd, though I'm really starting to contemplate a different
setup, I don't know at this point, there doesn't seem to be much out there
for information on this bug.

I've run debug via ivtv for ages, and still can't see anything out of the
ordinary.

>
> Cheers,
> -- jra
>
> _______________________________________________
> ivtv-users mailing list
> ivtv-users@ivtvdriver.org
> http://ivtvdriver.org/mailman/listinfo/ivtv-users



_______________________________________________
ivtv-users mailing list
ivtv-users@ivtvdriver.org
http://ivtvdriver.org/mailman/listinfo/ivtv-users