(and now with the CC to ivtv-users :-)

On Wednesday 14 November 2007 08:13:53 Hans Verkuil wrote:
> On Wednesday 14 November 2007 03:41:46 you wrote:
> > On Sat, Nov 10, 2007 at 07:47:12AM -0500, Dan Ritter wrote:
> > > > > > 4MB compressed of ivtv log files, now available at
> > > > > >
> > > > > > http://tao.merseine.nu:81/~dsr/ivtv.log.gz
> > > > >
> > > > > I have several more DMA errors, killing the recordings and
> > > > > rendering the device useless, if you need them. Let me know!
> > > >
> > > > Yes, please. I think it might be related to a particular time
> > > > of DMA error, but I would have to check more logs to determine
> > > > if this is indeed the case.
> >
> > I went back to 0.10.3, where I get soft errors for a while as a
> > warning before the killing stroke. They go in this pattern:
> >
> > Nov 13 20:47:33 localhost kernel: ivtv1 warning: ENC DMA ERROR b
> > Nov 13 20:47:33 localhost kernel: ivtv1 warning: encoder VBI:
> > Couldn't find start of buffer within the first 256 bytes
> > Nov 13 20:47:33 localhost kernel: ivtv1 warning: encoder MPEG:
> > offset 0 -> 128
> > Nov 13 20:47:33 localhost kernel: ivtv1 warning: encoder VBI:
> > offset 0 -> 128
> > Nov 13 20:47:33 localhost kernel: ivtv1 warning: encoder MPEG:
> > offset 128 -> 0
> > Nov 13 20:47:33 localhost kernel: ivtv1 warning: encoder VBI:
> > offset 128 -> 0
> > Nov 13 20:47:48 localhost kernel: ivtv0 warning: ENC DMA ERROR b
> > Nov 13 20:47:48 localhost kernel: ivtv0 warning: encoder VBI:
> > Couldn't find start of buffer within the first 256 bytes
> > Nov 13 20:47:48 localhost kernel: ivtv0 warning: encoder MPEG:
> > offset 0 -> 128
> > Nov 13 20:47:48 localhost kernel: ivtv0 warning: encoder VBI:
> > offset 0 -> 128
> > Nov 13 20:47:48 localhost kernel: ivtv0 warning: encoder MPEG:
> > offset 128 -> 64
> > Nov 13 20:47:48 localhost kernel: ivtv0 warning: encoder VBI:
> > offset 128 -> 64
> > Nov 13 20:47:48 localhost kernel: ivtv0 warning: encoder MPEG:
> > offset 64 -> 0
> > Nov 13 20:47:48 localhost kernel: ivtv0 warning: encoder VBI:
> > offset 64 -> 0
>
> Actually, these DMA errors do not indicate impending doom :-)
>
> The real killer is not DMA ERROR b but DMA ERROR 3.
>
> > If you think there's hope in a later or experimental branch, let
> > me know and I'll try it out.
>
> The problem is that I need to reproduce this error. And that is
> something I can only do on the PC back home but not on the PC here in
> Oslo: that one is rock solid. Since I'm back home this weekend I'll
> try to reproduce it (and hopefully fix it as well).
>
> > If you think it's a hardware issue, let me know that, too,
> > please, so I can go get something else that will work more
> > reliably.
>
> Well, it's really the buggy DMA engine of the cx23416. But some
> chipsets work better with the cx23416 than others. In particular
> Intel chipsets are 'better' for this card than other chipsets. But
> it's the cx23416 that's to blame, really.
>
> The ivtv driver has to be quite creative in order to work around all
> these issues, and I know the same is true for the Hauppauge Windows
> driver. But it seems my handling of this particular DMA error is not
> quite right. Odd, since it has worked in the past, so I might have
> removed/modified some code at one time.
>
> Regards,
>
>       Hans



_______________________________________________
ivtv-users mailing list
[email protected]
http://ivtvdriver.org/mailman/listinfo/ivtv-users

Reply via email to