[vdr] Re: error channel switching

[Usenet-372111@zocki.toppoint.de (Rainer Zocholl)]

Klaus.Schmidinger@cadsoft.de(Klaus Schmidinger)  22.09.01 11:29

Once upon a time Klaus Schmidinger shaped the electrons to say...

>Andreas Hoelscher wrote:
>>
>> error messages :
>>
>> Sep 20 19:59:16 andreas2 vdr[1935]: ERROR (dvbapi.c,3280): Invalid
>> argument
>> Sep 20 19:59:21 andreas2 vdr[1935]: ERROR: timeout while tuning
>> Sep 20 19:59:21 andreas2 vdr[1935]: ERROR: channel 1 not sync'ed on
>> DVB card 1! Sep 20 19:59:21 andreas2 vdr[1935]: panic level: 0
>> Sep 20 19:59:21 andreas2 vdr[1935]: retrying
>> Sep 20 19:59:21 andreas2 vdr[1935]: ERROR (dvbapi.c,3280): Invalid
>> argument Sep 20 19:59:26 andreas2 vdr[1935]: ERROR: timeout while
>> tuning Sep 20 19:59:26 andreas2 vdr[1935]: ERROR: channel 1 not
>> sync'ed on DVB card 1! Sep 20 19:59:26 andreas2 vdr[1935]: panic
>> level: 0 Sep 20 19:59:26 andreas2 vdr[1935]: retrying

Similar error here with newer drivers:

Driver tunes clear to ntv
but VDR can not tune to ARD (back screen) and then restarts the card



Got those funny errors:
Sep 22 19:36:55 msi kernel: Linux video capture interface: v1.00
Sep 22 19:36:55 msi kernel: i2c-core.o: driver VES1893 DVB demodulator registered.
Sep 22 19:36:55 msi kernel: i2c-core.o: I2C adapter 40004: I2C level transfers not supported
Sep 22 19:36:55 msi kernel: i2c-core.o: I2C adapter 50000: I2C level transfers not supported
Sep 22 19:36:55 msi kernel: i2c-core.o: driver VES1820 DVB demodulator registered.
Sep 22 19:36:55 msi kernel: i2c-core.o: I2C adapter 40004: I2C level transfers not supported
Sep 22 19:36:55 msi kernel: i2c-core.o: I2C adapter 50000: I2C level transfers not supported
Sep 22 19:36:55 msi kernel: i2c-core.o: driver L64781 DVB demodulator registered.
Sep 22 19:36:55 msi kernel: i2c-core.o: I2C adapter 40004: I2C level transfers not supported
Sep 22 19:36:55 msi last message repeated 2 times

etc...


>I did some testing with the new "Inversion" frontend parameter Ralph
>has announced yesterday, but I don't think that this could be
>responsible for the errors you are observing.

>The current VDR version doesn't know about this new parameter, so the
>struct member in FrontendParameters may have some random value.
>However, the driver developers were smart enough to make the 'default'
>case in SetInversion() result to INVERSION_AUTO for the VES1893, and
>to INVERSION_OFF for the STV0299. In both cases all channels on Astra
>(as far as I have observed) should be receivable, because they all
>have INVERSION_OFF.

>For the moment I will do the following change to VDR's dvbapi.c, just
>to have the Inversion parameter initialized:

>-+- dvbapi.c    2001/09/21 14:14:28
>+++ dvbapi.c    2001/09/22 09:08:05
>@@ -3252,6 +3252,7 @@

>         FrontendParameters Frontend;
>         Frontend.Frequency = freq * 1000UL;
>+        Frontend.Inversion = INVERSION_AUTO;
>         Frontend.u.qpsk.SymbolRate = Srate * 1000UL;
>         Frontend.u.qpsk.FEC_inner = FEC_AUTO;

>@@ -3298,6 +3299,7 @@

>         FrontendParameters Frontend;
>         Frontend.Frequency = FrequencyMHz * 1000000UL;
>+        Frontend.Inversion = INVERSION_AUTO;
>         Frontend.u.qam.SymbolRate = Srate * 1000UL;
>         Frontend.u.qam.FEC_inner = FEC_AUTO;
>         Frontend.u.qam.QAM = QAM_64;


This patch is required!
After applying that the newer drivers (2001-09-22) runs here too.
Thanks! :-)