[linux-dvb] twinhan 1020A not working
linux at bohmer.net
Sat Jan 5 20:41:01 CET 2008
I also had a lot of problems with the Technisat Skystar 1 CI series in the past.
There are different cards out there, ALL named Skystar 1 CI, with
different Firmware contents and can be recognised on the different
colors of the PCB, but all are distributed under the same name. The <
rev3.0 cards are clones of VP1032/VP1030C and the firmware is DST-CI,
while the rev3.0 card is a Clone of VP1030A, with DSTMCI firmware
I had them all described on the
http://www.linuxtv.org/wiki/index.php/DVB-S_PCI_Cards, but I see that
the page is rearranged, and the stuff I added to get all these cards
to work properly is gone now. I also described what to do to get rid
of the kernel hang during boot, which only occured with the rev3.0 of
this card back then, where an invalid procedure to determine the
card-id in the driver caused the wrong code to be executed, resulted
in a kernel hang during boot. I also notice the pictures of the 1.4
cards are gone also, this version of this card worked properly in the
past. (receiver block is rotated compared to the rev3.0 cards, and PCB
is dark-blue) It is still wise to make clear in the wiki that
different versions exist, how they look like, and how to get them to
So, in my opinion this site is now not complete anymore, because
crucial information for these annoying cards seems to be lacking...
You mention that you fixed some problems... which problems? In which
kernel version? This is not listed in the wiki. ( what is minimal
required to get things to work? On what kernel version is the wiki now
based on? What if I have an older distro that does not have the latest
and greatest kernel, and I do not want to recompile the kernel?)
Besides, these cards were not really stable at all, so in the mean
time I replaced them with other cards, (like TT-1500's, which work
much, much, much better, especially in combination with MythTV and a
CAM in the CI-slot).
Coming back to the problem of Gregor:
Looking at your logging, I see exactly the same problem I had with the
rev3.0 card, where the card-id is set to 0x0, resulting in a kernel
hangup. The solution for this can be found somewhere in the history of
I was only able of working around this by telling the driver the
card-id explicitly, instead of implementing a decent solution for this
in the kernel, this because the firmware of this card does something
strange here, and I could not get my hands on the interface
documentation for this firmware...
I hope this points you in the right direction...
2008/1/5, CityK <CityK at rogers.com>:
> Gregor Fuis wrote:
> > I have 4 twinhan 1020A cards (Skystar1 CI).
> The Skystar 1 CI is actually a VP-1030A clone:
> > When I was inserting one by one card separately. I
> > had to fix them firmware with instructions from dvb wiki and then they
> > worked.
> Ahh -- good to know. I was unaware that VP-1030A was susceptible to the
> same problem as the 1020, 1020A, and 1030 models....but, then again, I
> guess the 1030A is simply more-or-less a 1020A with the CI interface on
> I've recently provided a good number of updates to the "kernel hang
> solution" instructions in the 1020A article --- there have been several
> people who have contacted me privately about this problem and were
> having difficultly implementing the solution (as, really, it wasn't too
> clearly outlined in the wiki), as well as seeing several people mention
> it on the #irc channel recently. .... I guess that, coincidentally,
> there has been an elevated occurrence in the number of users
> encountering the problem recently. .... anyway, after discussions with
> Manu, and by adopting some of the output one end user (who managed to
> slug it through solving the problem on his own) provided in the wiki
> article talk page, a much more informative set of instructions for
> solving the kernel hang problem are now found (presently only) in the
> 1020A article...I'll adapt/apply that information soon to the 1020, 1030
> and 1030A articles as well.
> A very small quantity of the older 1020 & 1030 models are also
> susceptible to another problem, that is similar in nature, but which
> requires a different fix .... (the fix is really a hack, so it can't be
> permenantly added to the drivers ... but, nonetheless, for those users
> experiencing this, this quick work around should be sufficient in most
> cases to do the trick). It isn't expected that many users will encounter
> this problem (apparently the problem could only surface in the case of
> <10,000 cards from circa 2001/2 ... so from the purchasers of that lot,
> Linux end users probably account for only a small fraction ... and you
> could probably figure that many of these cards may no longer be in
> use/deployed any longer either). Nonetheless, one user encountered this
> as recently as August last year....I've been meaning to update the 1020
> & 1030 wiki articles regarding this for a long time, but haven't yet got
> there -- perhaps soon, when I make the other changes mentioned above.
> Anyway, this other problem doesn't pertain to your case ... which makes
> me question why I even blabbed on about it in the first place, but
> whatever :P
> > But when I installed them all in the system, they all stopped to
> > work and even halt system at boot sometimes.
> > ....
> > [snip]
> > ....
> Just some random thoughts:
> - do you have any other dvb cards installed in this system (PCI/USB/or
> - are you using CAMs with all 4 of these cards ?
> (I'm already thinking that this is going to be somehow related to CAM
> use .... no real basis for that thought, just an intuition)
> linux-dvb mailing list
> linux-dvb at linuxtv.org
More information about the linux-dvb