2012-04-21 10:38:02 nibble.max@xxxxxxxxx >Em 20-04-2012 06:47, Antti Palosaari escreveu: >> On 20.04.2012 11:01, nibble.max wrote: >>> 2012-04-20 15:56:27 nibble.max@xxxxxxxxx >>> At first time, I check it exist so try to patch it. >>> But with new m88ds3103 features to add and ts2022 tuner include, find it is hard to do simply patch. >>> It is better to create a new driver for maintain. >>>> Hi Max, >>>> >>>> Em 15-04-2012 12:53, nibble.max escreveu: >>>>> Montage m88ds3103 demodulator and ts2022 tuner driver. >>>> >>>> It was pointed to me that this device were already discussed on: >>>> >>>> http://www.mail-archive.com/linux-media@xxxxxxxxxxxxxxx/msg43109.html >>>> >>>> If m88ds3103 demod is similar enough to ds3000, it should just add the needed >>>> bits at the existing driver, and not creating a new driver. >>>> >>>> Thanks, >>>> Mauro >> >> The main problem of these all existing and upcoming Montage DVB-S/S2 drivers are those are not split originally correct as a tuner and demod and now it causes problems. >> >> I really suspect it should be: >> * single demod driver that supports both DS3000 and DS3103 >> * single tuner driver that supports both TS2020 and TS2022 >> >> And now what we have is 2 drivers that contains both tuner and demod. And a lot of same code. :-( >> >> But it is almost impossible to split it correctly at that phase if you don't have both hardware combinations, DS3000/TS2020 and DS3103/TS2022. I think it is best to leave old DS3000 as it is and make new driver for DS3103 *and* TS2022. Maybe after that someone could add DS3000 support to new DS3103 driver and TS2020 support to new TS2022 driver. After that it is possible to remove old DS3000 driver. >> >> And we should really consider make simple rule not to accept any driver which is not split as logical parts: USB/PCI-interface + demodulator + tuner. > >Mixing tuner and demod is not good. Yet, dropping the current ds3000 doesn't >seem to be the best approach. > >IMO, Konstantin/Montage should split the ds3000 driver on two drivers, putting >the ts2020 bits on a separate driver. > >Then, Max should write a patch for ds3000 in order to add support for ds3103 on >it, and a patch for ts2020 driver, in order to add support for ts2022 on it. > >Of course, Konstantin should check if Max changes don't break support for the >DS3000/TS2020 configuration. > >Regards, >Mauro Actually, I have the following hardware combinations. 1)DS3000 and TS2020 2)DS3103 and TS2020 3)DS3103 and TS2022 Should I sumbit the driver for DS3103 and TS2022 in the split files? Or I must wait for Konstantin's work. How long should I wait for? BR, Max. -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
