[linux-dvb] Re: [PATCH] Small fix for stv0299.c

[Andrew de Quincey <adq_dvb@lidskialf.net>]

On Wednesday 12 November 2003 16:45, Andrew de Quincey wrote:
> On Wednesday 12 November 2003 15:32, Johannes Stezenbach wrote:
> > Vadim Catana wrote:
> > > > Andrew de Quincey wrote:
> > > >
> > > > Would it not actually be more useful to make it count uncorrected
> > > > blocks,
> > >
> > > as
> > >
> > > > this is a measure of what people actually will see.. as opposed to
> > > > errors which may (or may not) have been corrected.
> > >
> > > The problem that the patch fixes is that ioctl command FE_READ_BER  is
> > > actually returning
> > > uncorrected blocks count for samsung tuner and Viterbi bit errors count
> > > for other tuners,
> > > while FE_READ_UNCORRECTED_BLOCKS returns -EOPNOTSUPP.
> > >
> > > If uncorrected blocks count is more useful the STV0299 should be
> > > initialized accordingly,
> > > and the ioctl function fixed. Should I can send a patch ?
> >
> > FE_READ_BER is more useful for displaying a "signal quality" status
> > bar in a GUI. FE_READ_UNCORRECTED_BLOCKS is not suitable for that
> > because it stays at 0 until the signal is bad enough and then jumps
> > between 0 and some small number.
> >
> > Too bad the stv0299 cannot do both at the same time.
>
> uncorrected blocks is also very useful for diagnosing problems in
> transmission equipment, (e.g. logging for a long period to look for
> periodic
> corruption)... maybe we need a way to choose which one should be used...
> perhaps a module parameter?

Would anyone have any objections if I implemented such a module parameter.. 
e.g. "stv0299_status" set to "ucblocks" or "ber" (default being ber)?



-- 
Info:
To unsubscribe send a mail to ecartis@linuxtv.org with "unsubscribe linux-dvb" as subject.