Antti Palosaari wrote:
> heips,
> hmm, now I am a little bit confused because there is this same id as MT352
> board. Maybe this is not just board id, but some other value that differs
> from design to design. Could it be firmware version...
>
> There is matrix I have now:
> board id model demod tuner module PCB version
> 0a 01 00 E30C Plus TDA10023 DTOS403IH102A 507DC (REV 0.2) Timo
> 02 02 01 E30 ZL10353 ? PCB509T REV 1.61 Heikki
> 06 01 00 E30 Plus ZL10353 DNOS404ZH103A ? Risto
> 02 02 01 E30 MT352 DNOS4D4ZH102A ? Mika (not sure
> if 02 02 02 because sniff is from other dev)
>
> CI board seems to be same design in both Plus models mentioned: 507_SM
> BOARD (V1.2).
>
> More reports needed.
Antti,
This wouldn't be the first time that a vendor released a device without changing the product id, whose earlier revisions used an mt352, and later revisions used a zl10353.
I recommend that you handle it in the following way:
static int anysee_zl35x_frontend_attach(struct dvb_usb_adapter *adap)
{
/* any device-specific stuff may go here */
if (((adap->fe = dvb_attach(mt352_attach, &anysee_mt352_config,
&adap->dev->i2c_adap)) != NULL) ||
((adap->fe = dvb_attach(zl10353_attach,
&anysee_zl353_config,
&adap->dev->i2c_adap)) != NULL))
return 0;
return -EIO;
}
The _attach functions of each module tests the id register to ensure that the proper demod is actually present -- a few drivers are already doing this, such as dvb-usb-cxusb and dvb-bt8xx. I think there are others as well...
HTH,
Mike
_______________________________________________
linux-dvb mailing list
linux-dvb@xxxxxxxxxxx
http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
