| Message ID | 5275690A.3080108@narod.ru (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
Em Sat, 02 Nov 2013 23:05:14 +0200 CrazyCat <crazycat69@narod.ru> escreveu: > Fix dvb-c standard selection - qam8 for ANNEX_AC > > Signed-off-by: Evgeny Plehov <EvgenyPlehov@ukr.net> > diff --git a/drivers/media/tuners/tda18271-fe.c b/drivers/media/tuners/tda18271-fe.c > index 4995b89..6a385c8 100644 > --- a/drivers/media/tuners/tda18271-fe.c > +++ b/drivers/media/tuners/tda18271-fe.c > @@ -960,16 +960,12 @@ static int tda18271_set_params(struct dvb_frontend *fe) > break; > case SYS_DVBC_ANNEX_B: > bw = 6000000; > - /* falltrough */ > + map = &std_map->qam_6; > + break; > case SYS_DVBC_ANNEX_A: > case SYS_DVBC_ANNEX_C: > - if (bw <= 6000000) { > - map = &std_map->qam_6; > - } else if (bw <= 7000000) { > - map = &std_map->qam_7; > - } else { > - map = &std_map->qam_8; > - } > + bw = 8000000; > + map = &std_map->qam_8; This is wrong, as it breaks for 6MHz-spaced channels, like what's used in Brazil and Japan. What happens here is that, if the tuner uses a too wide lowpass filter, the interference will be higher at the demod, and it may not be able to decode. As the bandwidth is already estimated by the DVB frontend core, the tuners should be adjusted to get the closest filter for a given bandwidth. So, the driver is correct (and it is tested under 6MHz spaced channels). > break; > default: > tda_warn("modulation type not supported!\n"); > > -- > To unsubscribe from this list: send the line "unsubscribe linux-media" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html
Em Sat, 02 Nov 2013 23:31:58 +0200 CrazyCat <crazycat69@narod.ru> escreveu: > Mauro Carvalho Chehab ?????: > > This is wrong, as it breaks for 6MHz-spaced channels, like what's used > > in Brazil and Japan. > > > > What happens here is that, if the tuner uses a too wide lowpass filter, > > the interference will be higher at the demod, and it may not be able > > to decode. > > > > As the bandwidth is already estimated by the DVB frontend core, the > > tuners should be adjusted to get the closest filter for a given > > bandwidth. > > > > So, the driver is correct (and it is tested under 6MHz spaced channels). > > But usual applications only set cable standard (Annex A/C or B) and not set bandwidth. So for annex A/C default selected 6MHz ? Usual applications set the symbol rate, and symbol rate is easily converted into bandwidth. The DVB core does that. see dtv_set_frontend(): switch (c->delivery_system) { case SYS_ATSC: case SYS_DVBC_ANNEX_B: c->bandwidth_hz = 6000000; break; case SYS_DVBC_ANNEX_A: rolloff = 115; break; case SYS_DVBC_ANNEX_C: rolloff = 113; break; default: break; } if (rolloff) c->bandwidth_hz = (c->symbol_rate * rolloff) / 100;
diff --git a/drivers/media/tuners/tda18271-fe.c b/drivers/media/tuners/tda18271-fe.c index 4995b89..6a385c8 100644 --- a/drivers/media/tuners/tda18271-fe.c +++ b/drivers/media/tuners/tda18271-fe.c @@ -960,16 +960,12 @@ static int tda18271_set_params(struct dvb_frontend *fe) break; case SYS_DVBC_ANNEX_B: bw = 6000000; - /* falltrough */ + map = &std_map->qam_6; + break; case SYS_DVBC_ANNEX_A: case SYS_DVBC_ANNEX_C: - if (bw <= 6000000) { - map = &std_map->qam_6; - } else if (bw <= 7000000) { - map = &std_map->qam_7; - } else { - map = &std_map->qam_8; - } + bw = 8000000; + map = &std_map->qam_8; break; default: tda_warn("modulation type not supported!\n");
Fix dvb-c standard selection - qam8 for ANNEX_AC Signed-off-by: Evgeny Plehov <EvgenyPlehov@ukr.net> -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html