Michael Krufky wrote: > struct dvb_frontend_ops contains a function pointer for reading signal > strength from the demodulator, however, it would also be useful to be > able to read the signal strength from the tuner, itself. > > As of now, in dvb_tuner_ops, we only have the following function for status: > > int (*get_status)(struct dvb_frontend *fe, u32 *status); > > The usability of this function is rather limited, and currently only > provides a mechanism for reading a limited amount of bits. The current > status bits are as follows: > > #define TUNER_STATUS_LOCKED 1 > #define TUNER_STATUS_STEREO 2 > > ...this doesn't really lend itself for a useful signal strength reading. > > I can cite two examples of how this would be useful in the current codebase: > > 1) Some demodulators do not directly provide signal strength readings. > For example, lgdt330x. In order to provide some sort of measurement, > other values are used to arrive at an estimated reading. > > If we had a read_signal_strength function available from the tuner > driver, such demodulator drivers that are otherwise unable to provide > this functionality directly would be able to read the signal strength > from the tuner driver, and either pass on that value, or use it in its > calculation of status readings. > > 2) The analog tuner system accesses the tuner modules directly, and > would benefit greatly from having such functionality available to be > able to read signal strength from the tuner. Manu and I spoke about this on irc today, and we decided that it would be better to name this function, "get_rf_strength" ... So, the change would be as follows: --- v4l-dvb.orig/linux/drivers/media/dvb/dvb-core/dvb_frontend.h +++ v4l-dvb/linux/drivers/media/dvb/dvb-core/dvb_frontend.h @@ -92,6 +92,7 @@ #define TUNER_STATUS_LOCKED 1 #define TUNER_STATUS_STEREO 2 int (*get_status)(struct dvb_frontend *fe, u32 *status); + int (*get_rf_strength)(struct dvb_frontend *fe, u16 *strength); /** These are provided seperately from set_params in order to facilitate silicon * tuners which require sophisticated tuning loops, controlling each parameter seperately. */ This is a rather trivial addition, so I doubt there will be any debate on this matter. I plan to move forward with this tonight, unless anybody is opposed. Any acks would still be appreciated, of course :-) Regards, Mike Krufky _______________________________________________ linux-dvb mailing list linux-dvb@xxxxxxxxxxx http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
