https://bugs.kde.org/show_bug.cgi?id=374693
--- Comment #30 from Mauro Carvalho Chehab <mchehab+...@kernel.org> ---
(In reply to Mauro Carvalho Chehab from comment #23)
> (In reply to juliushar from comment #18)
> > ...tested with a different PC, running current Manjaro Linux (KDE) 64 Bit, a
> > new Technisat SkyStar2 DVB card, same LNB. The same problem showed up, on
> > the screenshots it can be seen that the number of channels in kaffeine
> > differs greatly after the workaround. Also, w_scan results will be attached.
>
> That sounds a Kernel driver bug. Looking at the flexcop-fe-tuner.c, it looks
> that there are several different models of SkyStar2 supported by the driver,
> each with different tuners and different LNB supply chips. The models
> supported by the Kernel are:
>
> SkyStar2 rev 2.3: mt312
> SkyStar2 rev 2.6: stv0299, tbmu24112
> SkyStar2 rev 2.7: s5h1420, isl6421, itd1000
> SkyStar2 rev 2.8: cx24123, isl6421
>
> What's the exact model that you have? What Kernel version? I also need the
> Kernel logs with debug info, without the workarond and with the workaround,
> in order to try to identify what's missing there.
Ok, your board is using the entry meant for SkyStar S2 PCI DVB-S/S2 rev 3.3:
static int skystarS2_rev33_attach(struct flexcop_device *fc,
struct i2c_adapter *i2c)
{
fc->fe = dvb_attach(cx24120_attach,
&skystar2_rev3_3_cx24120_config, i2c);
if (!fc->fe)
return 0;
fc->dev_type = FC_SKYS2_REV33;
fc->fc_i2c_adap[2].no_base_addr = 1;
if (!dvb_attach(isl6421_attach, fc->fe, &fc->fc_i2c_adap[2].i2c_adap,
0x08, 0, 0, false)) {
err("ISL6421 could NOT be attached!");
fc->fc_i2c_adap[2].no_base_addr = 0;
return 0;
}
info("ISL6421 successfully attached.");
if (fc->has_32_hw_pid_filter)
fc->skip_6_hw_pid_filter = 1;
return 1;
}
It uses a cx24120 frontend, with an ISIL 6421 to power up the LNBf. The volrage
control (with switches from V/H polarization) seems to be implemented by the
isl6421 driver. The datasheet for this device is at:
http://www.intersil.com/content/dam/Intersil/documents/isl6/isl6421a.pdf
One thing this chip provides is that it can add an extra 1V to be used when the
antenna cable it too long, in order to compensate for eventual loss.
Apparently, something is not right there, at least for your device. I'll take a
look at the Kernel driver.
--
You are receiving this mail because:
You are watching all bug changes.
