Johannes Stezenbach wrote: > > Klaus Schmidinger wrote: > > I'd like to point to my posting from yesterday (see > > http://linuxtv.org/mailinglists/linux-dvb/2003/06-2003/msg00218.html). > > > > Maybe somebody with deeper knowledge about this can look into this > > and tell us whether there actually is something wrong in the driver > > in case of hw_sections=0? > > You wrote there and then: > > I looked a little deeper into this and found that in StartHWFilter() > > the 'mode' is initialized to 0xb96a, which is the value necessary to > > allow simultaneous recording and replay. In case of DMX_TYPE_SEC and > > hw_sections=1, this will be overwritten with mode=4. So I guess the > > root of the problem is that in case of hw_sections=0 'mode' remains at > > 0xb96a, which I assume causes the section data to be transferred via > > the same channels as TS data and thus results in disrupted recordings. > > That's the intended behaviour. Mode=4 -> use HW filters, > mode=0xb96a -> pass TS packet to SW filter (BTW, I hate magic numbers).
Well, I didn't invent that number ;-) > And yes, that means that high bandwidth section data has the same channel > and priority as AV data. But even pay TV EPG data is relatively low bandwidth > compared to video, so I see no problem. That's odd, because with all other things the same, I get disrupted recordings with hw_sections=0, while they are fine with hw_sections=1. That's why I asked about the "magic" behind the 'mode' value. Klaus -- _______________________________________________________________ Klaus Schmidinger Phone: +49-8635-6989-10 CadSoft Computer GmbH Fax: +49-8635-6989-40 Hofmark 2 Email: [EMAIL PROTECTED] D-84568 Pleiskirchen, Germany URL: www.cadsoft.de _______________________________________________________________ -- Info: To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as subject.
