[vdr] Re: lock handling in sections.c

[Jouni Karvo <jouni.karvo@tkk.fi>]

Klaus Schmidinger writes:
 > 
 > I must admit that I'm having a hard time understanding why it would
 > be so important "to make sure the info on lock is as fresh as possible"?

Well, as I wrote earlier, I did not fully understand (and probably do
not now, either) the code.

 > I believe it is important to check and remember the DeviceHasLock condition
 > _before_ actually reading the data. Otherwise we might read data that came
 > in while the device didn't have a lock yet (i.e. garbage), and after reading
 > the data, but before processing it, the device has obtained a lock. Then
 > we would process data that in fact is garbage.

You are right, so at least my last version (with two separate checks
instead of a variable) is actually something that should not be used.

Something there was, in any way, wrong with tuning stuff since after
the EPG scan VDR not was able to tune properly the second card.  After
doing this change, the problem seems to have either gone or at least
happens less frequently --- so if it seems that this change is not the
one responsible for that it might be something that changes some
timings which in turn make that problem in other parts of the code not
to happen.  My problem is not properly understanding the process here
so this is just blind poking.

yours,
		Jouni