[vdr] Re: cam-errors (@Klaus)

["Uwe Weissbach" <uwe.weissbach@web.de>]

On 6 Apr 2003 at 23:31, Klaus Schmidinger wrote:

> Uwe Weissbach wrote:
> > 
> > Hallo Klaus,
> > 
> > according to some topics about cam errors and an earlier topic
> > about differences in several versions I have checked last days
> > my computer and the difference between 1.1.23 and 1.1.26
> > (we talked earlier about it) and for me it's absolutly no doubt:
> > 1.1.23 has had this error with the slow menu but don't
> > makes significant am errors
> 
> Please help me out here: what "slow menu" do you mean?
> (Sorry if I've missed something here, I've read so many emails
> today...)

please don't be sorry, you do a hard job ...

there was a thread "strange problem with controlling VDR":

On 2 Mar 2003 at 21:14, Uwe Weissbach wrote:

> I have a very strange behavior of VDR,
> I saw a description of this problem first in a mail from 
> Adrian Carpenter from 11/03/03 but not anywhere else
> and also no solution...
> The behavior is the following:
> 
> After some minutes or hours after starting VDR,
> the control over LIRC is quite slowly (if I press a button
> it takes 10 or 20 seconds until the it's done) maybe setting
> markers or step through a movie, but:
> if I press MENU it needs also a quite long time to go inside it,
> but the menu itself runs in real-time, all sequences comes in time until
> I leave the menu. After that I have that slow motion control again.
> Next fact: This behavior affects also the SVDR-Port because
> if it occours, vdradmin and epg2timers throws timeout errors.
> If I restart VDR the effect disappears for short or long time...
> There are no unusual busy cpu or other performance eating processes 
> during the problem, no supicious entries in the logfile and nothing else, 
> the behavior of all other programms seems normally during the problem...
> I run 1.1.23 on 2.4.19-SUSE and I have this problem since 1.1.21 (the great CAM 
> changes) but I don't know if it has anything do do with it...
> 
> Anyone out there with the same problem or (better) with a reason or solution for 
> that?????????

On 2 Mar 2003 at 22:44, Klaus Schmidinger wrote:
> This has been fixed in VDR 1.1.25.

this was the reason why I'm overjumped the 1.1.24 but I'm able to test it also..

then in addition:

the thread "with 1.25 more crashes then with 1.23 ???" beginning at 11/03 ending with:

> On 14 Mar 2003 at 14:58, Klaus Schmidinger wrote:
> 
> > > with 1.25 I have significant more reboots during recording as with 1.23
> > > regardless (!) of the driver version...
> > > with 1.23 I had aprox. 1 reboot in 10 films,
> > > with 1.25 I have 1 or more reboots in every film:
> > 
> > VDR 1.1.25 with driver 1.0.0pre2 runs very stabe here, and I do quite
> > a lot of recording and replay...
> > 
> > Could it be a heat problem? Maybe the new driver causes more load on the ARM?
> > My DVB cards are cooled by a large, slow running fan right on top of them.

> > I have carefully excluded IRQ or temperature problems
> > (exclusive IRQ for both of my cards, open case, second fan, etc.):
> > with 1.1.23 nearly 100% of the recordings are error free
> > and with 1.1.26 I have errrors in nearly every recording, because
> > of the error-reading-cam-thing with following vdr-reloads...
> > 
> > I have switched severals times between the versions (vanilla of course and
> > with 1.0.0-pre2 driver) and it's replicable.
> > maybe it helps you to explore the cam read error problem...


whats about the things you wrote about it at 30/3:

> I'm afraid I'll have to postpone any further work on the LL CAM problems
> until after VDR version 1.2.0. Until then please use the Root-icam/Dpram-icam
> firmware. Note that in that case you should wait about 10 seconds after
> loading the driver before you start VDR (otherwise apparently VDR tries
> to connect to that CAM - which it can't).
> 
> I have asked Holger to keep the -icam firmware up-to-date with the LL
> version, so that you will have the OSD fix in the -icam FW, too.
> The -icam version that is in the CVS right now still has the OSD bug.

is this a chance to debug the read-cam-problem ???
and are there any further knowledge about it this time ???


> You wrote that you have compared versions 1.1.23 and 1.1.26 - what about
> 1.1.24 and 1.1.25? There were several changes in VDR/ca.c between these versions.
> Could you try each of those to maybe further narrow this down? And with
> all tests make sure you are always using the same driver. I understand
> you have done your tests with driver 1.0.0-pre2. Please repeat these tests
> with the same driver and VDR versions 1.1.24 and 1.1.25. Then please also
> do the test with the latest driver from the DVB CVS, since there were some
> changes since 1.0.0-pre2


ok I will test 1.1.24 meanwhile and the CVS driver 
(should I take original Root/Dpram ???)...

greetings,

Uwe Weissbach


--

    _    _
   (o)--(o)     
  /.______.\    
  \________/    
 ./        \.  
( .        , )  
 \ \_\\//_/ /   
  ~~  ~~  ~~    
(see it with monospaced font)


 Uwe Weissbach, Leipzig

mail:   uwe.weissbach@web.de
pgp:    send mail with "pgp-request" in subject

-----BEGIN GEEK CODE BLOCK-----
Version: 3.1
GCS/CC/IT d- s+:- a C++(+++) UL++> P 
L+> E- W+++ N+ o-- K- w O- M !V-- PS+ 
PE+ Y+ PGP++ t+ 5@ X? R tv- b++(+++) 
DI-? D---- G++ e* h---- r- y++ 
------END GEEK CODE BLOCK------




-- 
Info:
To unsubscribe send a mail to listar@linuxtv.org with "unsubscribe vdr" as subject.