Re: [linux-dvb] NXP 18211HDC1 tuner

2008-03-09 Thread Michael Krufky
On Mon, Mar 10, 2008 at 12:46 PM, Michael Krufky <[EMAIL PROTECTED]> 
wrote:
>> Jarryd Beck wrote:
>>  > Would someone be interested in writing tuner drivers for the NXP
>>  > 18211HDC1 tuner?
>>  > I recently bought the Winfast DTV Dongle Gold which uses an AF9015
>>  > chip and the NXP tuner.
>>  > I've managed to get it working up to the point of needing the tuner,
>>  > after that nothing works.
>>  > I have no idea how to write tuner code, so if someone is interested, I
>>  > can supply all the
>>  > info I've got about the card and test whatever you write.
>>  >
>>  > Jarryd.
>>
>>  Try the tda18271 driver -- I am under the impression that the tda18211
>>  is a dvb-t only subset of the tda18271, but I dont have a tda18211 to
>>  test with and find out, nor do I have a tda18211 spec to look at.  :-(
>>
>>  Good Luck,
>>
>>  Mike
Jarryd Beck wrote:
> I tried that, but I wasn't sure about a few things, I was kind of making stuff
> up as I went along.
>
> Can you tell me if I've done this right?
>
> At the af9015_tuner_attach function I wrote a function
> tda18211_tuner_attach which
> calls dvb_attach. The one thing I'm not sure about is the function
> tda18271_attach
> has a parameter u8 addr. I don't know what that is supposed to do or where I 
> am
> supposed to get it from.
>
> You can look up a datasheet from the nxp site, it appears it goes under the 
> name
> tda18211HD, I don't know what the C1 at the end means, I'm hoping it's the 
> same
> thing. The datasheet isn't very useful though, it pretty much only has a
> circuit diagram and a couple of numbers on it.
>
> Jarryd.
>
>   

Jarryd,

Please don't drop cc to the mailing list (added back), and also remember 
not to top quote.

The addr parameter is the i2c address of the tuner.  It is most likely 
0x60 or 0x61.

For an example of how to attach the tda18271 driver, look in 
cx23885-dvb.c for CX23885_BOARD_HAUPPAUGE_HVR1800 where alt_tuner is 1.

The datasheet on the nxp site wont help me -- i need to see the register 
map.

I think that the tda18271 driver will work with your tuner, but we may 
need to make some small adjustments.  If you look in tda18271-fe.c , 
you'll find the code that autodetects between a TDA18271c1 and a 
TDA18271c2 ...   If the autodetection fails for your tuner, you might 
want to try hardcoding it to the tda18271c1.  If that works, then I'll 
ask you to enable the register dump debug option (debug = 4) in the 
tda18271 driver and send me a dmesg snippit.  That should help us to add 
the autodetection later.

hth,

Mike


___
linux-dvb mailing list
linux-dvb@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb


Re: [linux-dvb] NXP 18211HDC1 tuner

2008-03-09 Thread Michael Krufky
Jarryd Beck wrote:
> Would someone be interested in writing tuner drivers for the NXP
> 18211HDC1 tuner?
> I recently bought the Winfast DTV Dongle Gold which uses an AF9015
> chip and the NXP tuner.
> I've managed to get it working up to the point of needing the tuner,
> after that nothing works.
> I have no idea how to write tuner code, so if someone is interested, I
> can supply all the
> info I've got about the card and test whatever you write.
>
> Jarryd.

Try the tda18271 driver -- I am under the impression that the tda18211 
is a dvb-t only subset of the tda18271, but I dont have a tda18211 to 
test with and find out, nor do I have a tda18211 spec to look at.  :-(

Good Luck,

Mike


___
linux-dvb mailing list
linux-dvb@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb


[linux-dvb] NXP 18211HDC1 tuner

2008-03-09 Thread Jarryd Beck
Would someone be interested in writing tuner drivers for the NXP
18211HDC1 tuner?
I recently bought the Winfast DTV Dongle Gold which uses an AF9015
chip and the NXP tuner.
I've managed to get it working up to the point of needing the tuner,
after that nothing works.
I have no idea how to write tuner code, so if someone is interested, I
can supply all the
info I've got about the card and test whatever you write.

Jarryd.

___
linux-dvb mailing list
linux-dvb@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb


Re: [linux-dvb] ~stoth/cx23885-video compile failure

2008-03-09 Thread Michael Krufky
aldebaran wrote:
> I own an HP/Hauppauge WinTv885 mod 77001 with cx23885 and xc3028 chipsets.
>
> Following the threads on this mailing list I understood these chipsets were 
> supported by 
> http://linuxtv.org/hg/~stoth/cx23885-video code, however I cannot even get 
> past the 'make all'.
>
> [snip]
> Is it a bug?
> Were I supposed to do something?
> Thank you.
>   
your card is supported in the master repository:

http://linuxtv.org/hg/v4l-dvb

If you continue to see these types of errors, please post again.  While waiting 
for a response, you can get past those errors by disabling the offending driver 
from the build.  try 'make menuconfig'.

hth,

Mike




___
linux-dvb mailing list
linux-dvb@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb


[linux-dvb] ~stoth/cx23885-video compile failure

2008-03-09 Thread aldebaran
Hi to everybody!

I own an HP/Hauppauge WinTv885 mod 77001 with cx23885 and xc3028 chipsets.

Following the threads on this mailing list I understood these chipsets were 
supported by 
http://linuxtv.org/hg/~stoth/cx23885-video code, however I cannot even get past 
the 'make all'.

here is what I get:
> CC [M] /home/user/cx23885-video-85708d2698cd/v4l/em28xx-audio.o In 
> file included from 
> /home/user/cx23885-video-85708d2698cd/v4l/em28xx-audio.c:39: 
> include/sound/core.h:281: error: 'SNDRV_CARDS' undeclared here (not in 
> a function) 
> /home/user/cx23885-video-85708d2698cd/v4l/em28xx-audio.c:58: error: 
> array index in initializer not of integer type 
> /home/user/cx23885-video-85708d2698cd/v4l/em28xx-audio.c:58: error: 
> (near initialization for 'index') make[3]: *** 
> [/home/user/cx23885-video-85708d2698cd/v4l/em28xx-audio.o] Error 1 
> make[2]: *** [_module_/home/user/cx23885-video-85708d2698cd/v4l] Error 
> 2 make[2]: Leaving directory 
> `/usr/src/linux-headers-2.6.24-11-generic' make[1]: *** [default] Error 2 

Is it a bug?
Were I supposed to do something?
Thank you.


___
linux-dvb mailing list
linux-dvb@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb


[linux-dvb] [PATCH] DMX_OUT_TSDEMUX_TAP: record two streams from same mux, resend

2008-03-09 Thread Andrea
 > Hi there,

 > This is what you want for capturing multiple audio streams from the same
 > multiplex simultaneously: open demux0 several times and send
 > depacketised output there. And capturing a single video stream is fine
 > too: open dvr0. But for capturing multiple video streams, it's surely
 > not what you want: you want multi-open (so demux0, not dvr0), but you
 > want the TS nature preserved (because that's what you want on output, as
 > you're going to re-multiplex it with the audio).

Can I ask your opinion on the best solution for:

I would like to have 2 separate TS (audio + video + subtitles...) for 2 
channels (same frequency).
Is it possible to set a DMX_SET_PES_FILTER with many pids?

Ideally I would like to multi-open the demux, set a PES filter for a few pids 
(currently I can do 
only for 1), read from demux and save the TS.

So that I can have more the one applications running.

What I can do now is: multi-open the demux, set filter -1 (i.e. get 
everything), read, and only save 
the pids I want.

I was just wondering what is the best solution.
I don't like to get many times the whole TS. It's a bit of a waste.

 > At least one existing solution -- GStreamer -- sends all its streams
 > simultaneously via dvr0 and demuxes again in userland, but it seems a
 > bit of a shame to pick out all the PIDs in kernel, stick them back
 > together in kernel, and send them to userland only to get unpicked
 > again, when the alternative is such a small API addition.

Andrea

___
linux-dvb mailing list
linux-dvb@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb


Re: [linux-dvb] Trying to setup PCTV HD Card 800i

2008-03-09 Thread Chaogui Zhang
On Sun, Mar 9, 2008 at 8:40 AM, Timothy E. Krantz <[EMAIL PROTECTED]> wrote:
> Chaogui,
>  Any idea when your signifigant patch to this driver will be included in the
>  mainline repository.  The driver is quite unusable without your patch.
>

I assume you are referring to the xc5000 patch.

Steve mentioned a while back that he will include it once more tests
are done on a few other boards that also uses xc5000. So I believe it
will be merged at some point. Given that he might be very busy with
other tasks at hand, I do not know when.

-- 
Chaogui Zhang

___
linux-dvb mailing list
linux-dvb@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb


Re: [linux-dvb] Trying to setup PCTV HD Card 800i

2008-03-09 Thread Chaogui Zhang
On Sun, Mar 9, 2008 at 1:31 AM, Brandon Rader <[EMAIL PROTECTED]> wrote:
> On Sat, Mar 8, 2008 at 8:50 PM, Chaogui Zhang <[EMAIL PROTECTED]> wrote:
>
>
>
> > On Fri, Mar 7, 2008 at 1:18 AM, Brandon Rader <[EMAIL PROTECTED]>
>  > wrote:
>  > > Hello,
>  > >
>  > >  I bought the PCTV HD 800i tuner from woot.com, and waited until drivers
>  > had
>  > >  been developed for it. I followed the guide from
>  > >  LinuxTV<
>  > http://www.linuxtv.org/wiki/index.php/Pinnacle_PCTV_HD_Card_%28800i%29>
>  > >  .
>  > >
>  > >  My dmesg and lspci outputs are below. The dmesg output has some errors
>  > in
>  > >  it, and the lspci looks like it is an entry short compared to some of
>  > the
>  > >  other lspci outputs I've seen for this card. When I try to modprobe
>  > >  cx88_dvb:
>  > >
>  > >  $ sudo modprobe cx88_dvb
>  > >  FATAL: Error inserting cx88_dvb
>  > >  (/lib/modules/2.6.22-14-generic/kernel/drivers/media/video/cx88/cx88-
>  > dvb.ko):
>  > >  No such device
>  > >
>  > >  lspci output:
>  > >  lspci | grep -i cx
>  > >  06:07.0 Multimedia video controller: Conexant CX23880/1/2/3 PCI Video
>  > and
>  > >  Audio Decoder [MPEG Port] (rev 05)
>  > >  06:07.1 Multimedia controller: Conexant CX23880/1/2/3 PCI Video and
>  > Audio
>  > >  Decoder [Audio Port] (rev 05)
>  > >
>  > >  dmesg output:
>  > >  dmesg | grep -i cx
>  > > 
>  >
>  > Did you reboot after installing the new driver? If not, try that first.
>  > If you did reboot and got the error, can you paste your *complete*
>  > dmesg output to http://pastebin.com and post the generated link here?
>  >
>  > BTW, I added cc to the linux-dvb list, which is the more appropriate place
>  > for your question.
>  >
>  > --
>  > Chaogui Zhang
>  >
>
>
>
>  I did restart after installing it.  Here is the link to my dmesg output:
>  http://pastebin.com/m19aa6aa4
>

A couple of strange things showed up in the dmesg output:

1. The i2c bus seems somehow corrupted, as suggested by the repeated
i2c read error and the tveeprom read error
2. The xc5000 driver was never loaded, probably due to the i2c read problems.

At the moment, I don't know what might have gone wrong. Can you try to
use the following repo instead and see what happens?

http://linuxtv.org/hg/~stoth/v4l-dvb/

This is not the most up to date code, but contains most that are
needed for the 800i. I suspect some recent patch might have introduced
minor bugs for the i2c bus. If the above repo proves to be working for
you, then we can start looking at what caused the i2c problem.

-- 
Chaogui Zhang

___
linux-dvb mailing list
linux-dvb@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb


Re: [linux-dvb] Patching MythTV in order to use it with multiproto

2008-03-09 Thread Gernot Pansy

the hack is against svn trunk (0.21 branch is also working).

here you will find the latest one:
http://notz.homelinux.com/99_mythtv_multiproto_hack.dpatch 

gernot

On Saturday 08 March 2008 12:47:00 David BERCOT wrote:
> Hi,
>
> On Debian, I've downloaded the source version of MythTV :
> v0.20.2.svn20080126. Then, I've downloaded the patch from
> http://pansy.at/gernot/mythtv-multiproto-hack.diff.gz
>
> But, when I try to apply it, I have these errors :
> # patch <./mythtv-multiproto-hack.diff -p0
> patching file libs/libmythtv/dvbchannel.cpp
> Hunk #1 FAILED at 138.
> Hunk #2 succeeded at 146 (offset -11 lines).
> Hunk #3 FAILED at 452.
> Hunk #4 FAILED at 509.
> Hunk #5 succeeded at 586 with fuzz 1 (offset 19 lines).
> Hunk #6 FAILED at 625.
> Hunk #7 succeeded at 670 (offset 19 lines).
> Hunk #8 succeeded at 800 (offset -33 lines).
> Hunk #9 FAILED at 850.
> 5 out of 9 hunks FAILED -- saving rejects to file
> libs/libmythtv/dvbchannel.cpp.rej
>
> Do have any idea to resolve these errors ?
>
> Thank you very much.
>
> David.



___
linux-dvb mailing list
linux-dvb@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb


[linux-dvb] How to start a DVB reception

2008-03-09 Thread Andrea
Hi,

What does exactly trigger the DVB architecture to start receiving data?

It seems that I can only receive DVB as long as there is a running application 
that has opened the 
fronted in read/write mode and tuned.

The following does not work (if it is the only application running)
(Even if it reports a LOCK)

1) open the fronted in read only
2) check for LOCK
3) open the demux
4) set some filter, output to DEMUX
5) start the demux
6) read from demux

The is no data to read.

I've tried test_pes in dvb-apps/test and it behaves the same.

Someone has to call FE_SET_FRONTEND and keep the fd open.

Is it true?

I would like to receive the whole TS of whatever is currently tuned. (possibly 
in read only).

Andrea


___
linux-dvb mailing list
linux-dvb@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb


Re: [linux-dvb] VP 1041: Is anybody able to tune to DVBS2 or DSS?

2008-03-09 Thread Tim Hewett
Just to clarify, the diseqc switches mentioned do not work reliably  
with any of the D-Sat cards in my PC (one Skystar HD2 DVB-S/DVB-S2,  
three Skystar 2 (DVB-S only)), the VP-1041 driver is not the issue -  
it was best to remove the switch as they can be troublesome, in my  
case they tend to block tuning to either horizontal or vertical  
transponders (apparently randomly over time). All three switches work  
fine with a Technomate TM1000 receiver, incidentally.

If of any help, these are the commands I am using to tune successfully  
and reliably to DVB-S2 transponders on Astra 2D (28East). I am using  
the current mantis tree with a Skystar HD2 card, using the replacement  
szap.c from the VP1041 wiki.

Entry in channels.conf:

DVBS2:11798:h:1:29500:513:641:2:2018


szap command line and output:

[EMAIL PROTECTED]:~/linuxtv-dvb-apps-1.1.1/util/szap# ./szap -r -p -a 5 -l  
UNIVERSAL -t 2 DVBS2
reading channels from file '/root/.szap/channels.conf'
zapping to 11 'DVBS2':
sat 1, frequency = 11798 MHz H, symbolrate 2950, vpid = 0x0201,  
apid = 0x0281 sid = 0x0002
Querying info .. Delivery system=DVB-S2
using '/dev/dvb/adapter5/frontend0' and '/dev/dvb/adapter5/demux0'
--> Using 'STB0899 DVB-S2' DVB-S2
do_tune: API version=3, delivery system = 2
do_tune: Frequency = 1198000, Srate = 2950
do_tune: Frequency = 1198000, Srate = 2950


couldn't find pmt-pid for sid 0002
status 1a | signal 05aa | snr 0030 | ber  | unc fffe |  
FE_HAS_LOCK
status 1a | signal 05aa | snr 002f | ber  | unc fffe |  
FE_HAS_LOCK
status 1a | signal 05aa | snr 002f | ber 00516155 | unc fffe |  
FE_HAS_LOCK
status 1a | signal 05aa | snr 002e | ber 00145855 | unc fffe |  
FE_HAS_LOCK
status 1a | signal 05aa | snr 002f | ber  | unc fffe |  
FE_HAS_LOCK
status 1a | signal 05aa | snr 002f | ber  | unc fffe |  
FE_HAS_LOCK
status 1a | signal 05aa | snr 002e | ber  | unc fffe |  
FE_HAS_LOCK
status 1a | signal 05aa | snr 002f | ber  | unc fffe |  
FE_HAS_LOCK

[EMAIL PROTECTED]:~/linuxtv-dvb-apps-1.1.1/util/szap#


In a different shell, output of scan while the card is tuned using  
szap above:

[EMAIL PROTECTED]:# scan -c -a 5
using '/dev/dvb/adapter5/frontend0' and '/dev/dvb/adapter5/demux0'
0x 0x0ef9: pmt_pid 0x0102 BSkyB -- Sky Screen 1HD (running,  
scrambled)
0x 0x0f23: pmt_pid 0x0103 BSkyB -- Channel 4 HD (running, scrambled)
0x 0x1034: pmt_pid 0x BSkyB -- Anytime (???, scrambled)
dumping lists (4 services)
Sky Screen 1HD   (0x0ef9) 86:V 0514   A 0642
(NAR) TT 2305   AC3 0662
Channel 4 HD (0x0f23) 86:V 0513   AC3 0661
[0f38]   (0x0f38) 00:
Anytime  (0x1034) 04:
Done.
[EMAIL PROTECTED]:#


HTH,

Tim.


On 8 Mar 2008, at 22:00, Simeon Simeonov wrote:

> Just for the record: in my case eliminating the diseqc did not make  
> a difference.
>
> - Original Message 
> From: Manu Abraham <[EMAIL PROTECTED]>
> To: Tim Hewett <[EMAIL PROTECTED]>
> Cc: Simeon Simeonov <[EMAIL PROTECTED]>; linux-dvb@linuxtv.org
> Sent: Saturday, March 8, 2008 11:50:33 AM
> Subject: Re: [linux-dvb] VP 1041: Is anybody able to tune to DVBS2  
> or DSS?
>
> Tim Hewett wrote:
>> Simeon,
>>
>> It doesn't matter re: the diseqc commands, with no switch present  
>> will
>> be nothing listening, assuming there is nothing else connected such  
>> as
>> a rotor.
>
>
> The SEC equipment should work fine AFAICS. The bad aspect of the tests
> were that no one really came up with any real logs that could really
> reproduced  a driver bug or a hardware bug.
>
> Nevertheless, there had been an interesting post on the VDR mailing  
> list.
>
> http://www.linuxtv.org/pipermail/vdr/2008-March/016068.html
>
> Regards,
> Manu
>
>
>
>>
>> Tim.
>>
>>
>> On 8 Mar 2008, at 01:27, Simeon Simeonov wrote:
>>
>>> Thanks for the answer Tim!
>>>
>>> That's a good point - I do have a switch. I will try bypassing it  
>>> but
>>> do you still send the diseqc commands or I have to get rid of them?
>>>
>>> Simeon
>>>
>>> - Original Message 
>>> From: Tim Hewett <[EMAIL PROTECTED]>
>>> To: [EMAIL PROTECTED]
>>> Cc: Tim Hewett <[EMAIL PROTECTED]>; linux-dvb@linuxtv.org
>>> Sent: Friday, March 7, 2008 5:18:43 PM
>>> Subject: VP 1041: Is anybody able to tune to DVBS2 or DSS?
>>>
>>> Yes, I have the Skystar HD2 tuning to DVB-S2 transponders using the
>>> replacement szap.c mentioned on the wiki page, then listing the
>>> channels using 'scan -c -a ' to prove proper reception, all using
>>> the current mantis tree. No hacks were needed, it works out of the  
>>> box
>>> every time now that the mantis tree has been updated to support the
>>> HD2 card (same one as the VP-1041).
>>>
>>> If you are using a Diseqc switch then get rid of it for now, mine  
>>> was
>>> causing lots of unreliability. I tried three different types of  
>>> Diseqc
>>> switch, all were the same. G