Message-ID: <20181122180611.2e7f1123@coco.lan>
Date:   Thu, 22 Nov 2018 18:06:11 -0200
From:   Mauro Carvalho Chehab <mchehab+samsung@...nel.org>
To:     stakanov <stakanov@...ipso.eu>
Cc:     Takashi Iwai <tiwai@...e.de>, linux-media@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: DVB-S PCI card regression on 4.19 / 4.20

Em Thu, 22 Nov 2018 19:58:19 +0100
stakanov <stakanov@...ipso.eu> escreveu:

> Hi Mauro!
> 
> In data giovedì 22 novembre 2018 17:19:08 CET, Mauro Carvalho Chehab ha 
> scritto:
> > Stakanov,
> > 
> > Em Tue, 20 Nov 2018 14:08:55 -0200
> > 
> > Mauro Carvalho Chehab <mchehab+samsung@...nel.org> escreveu:  
> > > Em Tue, 20 Nov 2018 14:20:01 +0100
> > > 
> > > stakanov <stakanov@...ipso.eu> escreveu:  
> > > > In data martedì 20 novembre 2018 13:42:17 CET, Mauro Carvalho Chehab ha
> > > > 
> > > > scritto:  
> > > > > Em Tue, 20 Nov 2018 13:11:58 +0100
> > > > > 
> > > > > "Stakanov Schufter" <stakanov@...ipso.eu> escreveu:  
> > > > > > Sorry for the delay. Apparently my smtp exits are blocking me (for
> > > > > > whatever reason). I am trying now via web mailer.
> > > > > > 
> > > > > > 
> > > > > > In short, no it does not work. Only EPG, no pic no sound.
> > > > > > But the error message in dmesg is gone I think.
> > > > > > 
> > > > > > uname -a
> > > > > > Linux silversurfer 4.20.0-rc3-2.gfe5d771-default #1 SMP PREEMPT Tue
> > > > > > Nov 20
> > > > > > 09:35:04 UTC 2018 (fe5d771) x86_64 x86_64 x86_64 GNU
> > > > > > 
> > > > > > dmesg:
> > > > > > 
> > > > > > [    6.412792] b2c2-flexcop: B2C2 FlexcopII/II(b)/III digital TV
> > > > > > receiver
> > > > > > chip loaded successfully
> > > > > > [    6.416645] flexcop-pci: will use the HW PID filter.
> > > > > > [    6.416648] flexcop-pci: card revision 2
> > > > > > [    6.423749] scsi host10: usb-storage 9-3.1:1.0
> > > > > > [    6.423842] usbcore: registered new interface driver usb-storage
> > > > > > [    6.426029] usbcore: registered new interface driver uas
> > > > > > [    6.439251] dvbdev: DVB: registering new adapter (FlexCop Digital
> > > > > > TV
> > > > > > device)
> > > > > > [    6.440845] b2c2-flexcop: MAC address = 00:d0:d7:11:8b:58
> > > > > > 
> > > > > > [    6.694999] dvb_pll_attach: delsys: 0, frequency range:
> > > > > > 950000000..2150000000
> > > > > > [    6.695001] b2c2-flexcop: found 'ST STV0299 DVB-S' .
> > > > > > [    6.695004] b2c2_flexcop_pci 0000:06:06.0: DVB: registering
> > > > > > adapter 0
> > > > > > frontend 0 (ST STV0299 DVB-S)...
> > > > > > [    6.695050] b2c2-flexcop: initialization of 'Sky2PC/SkyStar 2
> > > > > > DVB-S rev
> > > > > > 2.6' at the 'PCI' bus controlled by a 'FlexCopIIb' complete  
> > > > > 
> > > > > Well, the Kernel bug is probably gone. I don't see any other recent
> > > > > changes that would be affecting the b2c2 flexcop driver.
> > > > > 
> > > > > If you're successfully getting EPG data from the transponders, then it
> > > > > should also be receiving audio and video channels too, as, for the
> > > > > Kernel,
> > > > > there's no difference if a given program ID (PID) contains EPG, audio
> > > > > or
> > > > > video.
> > > > > 
> > > > > At the BZ, you're saying that you're using Kaffeine, right?
> > > > > 
> > > > > There are a few reasons why you can't watch audio/video, but you're
> > > > > 
> > > > > able to get EPG tables:
> > > > > 	- the audio/video PID had changed;
> > > > > 	- the audio/video is now encrypted;
> > > > > 	- too weak signal (or bad cabling).
> > > > > 
> > > > > The EPG data comes several times per second on well known PIDs, via a
> > > > > low
> > > > > bandwidth PID and it is not encrypted. So, it is usually trivial to
> > > > > get
> > > > > it.
> > > > > 
> > > > > I suggest you to re-scan your channels on Kaffeine, in order to force
> > > > > it to get the new PIDs. Also, please check that the channels you're
> > > > > trying to use are Free to the Air (FTA).
> > > > > 
> > > > > You can also use libdvbv5 tools in order to check if you're not
> > > > > losing data due to weak signal/bad cabling. The newer versions
> > > > > of dvbv5-zap have a logic with detects and report data loses, when
> > > > > started on monitor mode (-m command line option). It also prints
> > > > > the transponder bandwidth, and check what PIDs are received.
> > > > > 
> > > > > It is very useful to debug problems.
> > > > > 
> > > > > Thanks,
> > > > > Mauro  
> > > > 
> > > > I checked again and:
> > > > [sudo] password di root:
> > > > [    6.412792] b2c2-flexcop: B2C2 FlexcopII/II(b)/III digital TV
> > > > receiver chip loaded successfully
> > > > [    6.440845] b2c2-flexcop: MAC address = 00:d0:d7:11:8b:58
> > > > [    6.695001] b2c2-flexcop: found 'ST STV0299 DVB-S' .
> > > > [    6.695004] b2c2_flexcop_pci 0000:06:06.0: DVB: registering adapter 0
> > > > frontend 0 (ST STV0299 DVB-S)...
> > > > [    6.695050] b2c2-flexcop: initialization of 'Sky2PC/SkyStar 2 DVB-S
> > > > rev
> > > > 2.6' at the 'PCI' bus controlled by a 'FlexCopIIb' complete
> > > > [ 6265.403360] b2c2_flexcop_pci 0000:06:06.0: DVB: adapter 0 frontend 0
> > > > frequency 10719000 out of range (950000..2150000)
> > > > [ 6265.405702] b2c2_flexcop_pci 0000:06:06.0: DVB: adapter 0 frontend 0
> > > > frequency 10723000 out of range (950000..2150000)
> > > > [ 6265.407120] b2c2_flexcop_pci 0000:06:06.0: DVB: adapter 0 frontend 0
> > > > frequency 10757000 out of range (950000..2150000)
> > > > [ 6265.408556] b2c2_flexcop_pci 0000:06:06.0: DVB: adapter 0 frontend 0
> > > > frequency 10775000 out of range (950000..2150000)
> > > > [ 6265.409754] b2c2_flexcop_pci 0000:06:06.0: DVB: adapter 0 frontend 0
> > > > frequency 10795000 out of range (950000..2150000)
> > > > [ 6399.837806] b2c2_flexcop_pci 0000:06:06.0: DVB: adapter 0 frontend 0
> > > > frequency 12713000 out of range (950000..2150000)
> > > > [ 6399.839144] b2c2_flexcop_pci 0000:06:06.0: DVB: adapter 0 frontend 0
> > > > frequency 12731000 out of range (950000..2150000)  
> > > 
> > > Ok. Now, min/max frequencies are at the same scale. For DVB-S,
> > > dvb_frontend_get_frequency_limits() returns both in kHz, so the frequency
> > > range is now OK.
> > > 
> > > The tuning frequency is wrong through. 10,719,000 kHz - e. g. 10,719 MHz
> > > seems to be the transponder frequency you're trying to tune, and not the
> > > intermediate frequency used at the DVB-S board.
> > > 
> > > That sounds to me either a wrong LNBf setting or a bug at libdvbv5 or
> > > at Kaffeine's side. What happens is that the typical European LNBFs are:
> > > 
> > > 1) the "old" universal one:
> > > 
> > > UNIVERSAL
> > > 
> > > 	Universal, Europe
> > > 	Freqs     : 10800 to 11800 MHz, LO: 9750 MHz
> > > 	Freqs     : 11600 to 12700 MHz, LO: 10600 MHz
> > > 
> > > 2) the "new" universal one, with seems to be used by most modern
> > > satellite dishes in Europe nowadays:
> > > 
> > > EXTENDED
> > > 
> > > 	Astra 1E, European Universal Ku (extended)
> > > 	Freqs     : 10700 to 11700 MHz, LO: 9750 MHz
> > > 	Freqs     : 11700 to 12750 MHz, LO: 10600 MHz
> > > 
> > > Assuming that your satellite dish is equipped with an Astra-1E compatible
> > > LNBf, as you're trying to tune 10,719 MHz, you need to setup Kaffeine
> > > to use the EXTENDED LNBf, with covers the extended frequency range.
> > > 
> > > Kaffeine/libdvbv5 will subtract this value from the LO frequency:
> > > 	10719 MHz - 9750 MHz = 969 MHz
> > > 
> > > And pass the 969 MHz frequency for the Kernel to tune. As this
> > > is between the 950MHz..2150MHz limit, it won't produce any Kernel
> > > messages.
> > > 
> > > Please notice that there were some bugs at v4l-utils and DVB-S/S2,
> > > so be sure to check the v4l-utils version you're using. The last
> > > one is 1.14.2 (released on Feb, 10 2018). I recommend you to have
> > > at least version 1.12.4 (released on May, 6 2017).  
> > 
> > Did it work after using the EXTENDED LNBf on Kaffeine?
> > 
> > 
> > Thanks,
> > Mauro  
> 
> 
> 
> I am not sure to tell the truth.
> I am using Italian language on my machines. So the only setting I have is 
> LNBf "alta tensione". Now this is like "high tension LNBf if this does tell 
> you something.
> There are three settings (that are not well documented to tell the truth. 
> First is:
> selection square empty
> then 
> half empty
> then 
> full
> 
> Now, I did not find an explanation on the web for the third position. 
> One 
> should use no reinforcement of the signal. One should offer support by 
> reinforcing the signal (via low tension on the antenna cable?). The middle 
> position I do not know. Maybe it is the degree of reinforcement?
> Now, I tried all three in a sequence. I did not encounter any change. 

This setting controls an special ioctl that it is not available on all
devices. Except if you have a problem, and your device supports it, you
should let it at the "half empty" state, meaning that Kaffeine won't use
the ioctl. 

Anyway, this not where you set the LNBf. Look at the Kaffeine's handbook:
it should have some pictures showing where the LNBf setting is done, but
it is on a separate popup window[1]:

	https://git.linuxtv.org/mchehab/kaffeine.git/plain/doc/kaffeine_configure_tv_lnbf.png

There, you can select the EXTENDED LNBf item (it is the third one: Astra 1E).

[1] this picture is there at the Kaffeine's handbook, with can be activated
    on Kaffeine by using the F1 key). You can also see the full Kaffeine's
    manual at:
	https://docs.kde.org/trunk5/en/extragear-multimedia/kaffeine/index.html

    There's even an Italian translation for it (although it is outdated):

	https://docs.kde.org/trunk5/it/extragear-multimedia/kaffeine/index.html

> I have however a doubt that I am going to verify today. I am thinking that up 
> to 4.18.15-1 for some reasons you did not use the firmware that Linux TV org 
> does advice for the card. While it could be that it is needed now, I am going 
> to download and verify today. The only rational explanation that I could have 
> is that before it did inadvertently another firmware that comes with TW (there 
> are several) or that TW did take out the firmware but I would have been asked 
> before removal... 
> Anyway, that is the last experiment at least if you do not tell me that up to 
> now I did not do the right thing (in that case be very verbose to be sure I do 
> understand well were to change settings). 
> BTW the only other choice is between "disecq and rotor" and as we do not have 
> a rotor on the roof....
> 
> Comparison: I have in the machine in parallel a Hauppauge Card 5525 HD and so 
> I can confirm the signal is good. However this card too stayed black only EPG, 
> so I did download the firmware to /lib/firmware. I did reboot and it works now 
> perfectly. I will try the same with the Technisat. But the question remains 
> why it did work without firmware with the previous kernel. I am puzzled. 

Are you sure that the difference is just the Kernel version? Perhaps you
also updated some other package.

> I will update you tomorrow. Be patient if you should have found out that I did 
> not check for the right thing. I really appreciate your help given and 
> ignorance is, so to say, never intentional. ;-) 
> 
> Thank you, 
> Stak. 
> 
> 
> 
> _________________________________________________________________
> ________________________________________________________
> Ihre E-Mail-Postfächer sicher & zentral an einem Ort. Jetzt wechseln und alte E-Mail-Adresse mitnehmen! https://www.eclipso.de
> 
> 



Thanks,
Mauro

Powered by blists - more mailing lists