On Tue, 3 Aug 2004 20:12:36 +0200
Johannes Stezenbach <[EMAIL PROTECTED]> wrote:
> Roberto Ragusa wrote:
> >
> > OK, so I propose this:
> >
> > - you use FE_SET_FRONTEND as usual
> > - then, you call FE_SET_AUTO_SEARCH passing some
can_stop
after that, your hardware is already tuned to the last found thing.
If the hardware needs a reset or mode switch it can be done
in the "stop" call (so "stop" would actually mean "stop scanning and
return to normal behaviour").
I think we're close to a good solution.
Comments? (even on the names of the structures and fields)
Regards.
--
Roberto Ragusamail at robertoragusa.it
on FE_GET_FRONTEND).
We want to fit two 32 bit values into a 32 bit space, that's
the core of the problem, IMHO.
The best I can think is
bit 31-30 00: normal sr, 01: interval, 10:rsvd, 11:rsvd
bit 29-15 15 bit, start sr
bit 14-0 15 bit, end sr
Regards.
--
Roberto Ragusamail at robertoragusa.it
s. :-)
Are you sure about this "//" rule? No mention in CodingStyle and
too much infringing code already accepted.
Regards.
--
Roberto Ragusamail at robertoragusa.it
r /aaa/bbb/ccc-1.3 /aaa/bbb/ccc-1.3_modified >mypatch.diff
(u=human readable, r=recursive)
It's almost easier to create a diff than to apply it (when you have
to choose between option "-p0" and "-p1").
Regards.
--
Roberto Ragusamail at robertoragusa.it
On Mon, 02 Aug 2004 00:09:35 +0200
Andreas Oberritter <[EMAIL PROTECTED]> wrote:
> Am So, den 01.08.2004 schrieb Roberto Ragusa um 17:59:
> +
> + if ((ret = mt312_read(i2c, LNB_FREQ_H, &buf, sizeof(buf))) < 0)
> + return ret;
>
> this s
On Mon, 2 Aug 2004 16:07:13 +0200
Roberto Ragusa <[EMAIL PROTECTED]> wrote:
> and use (SR_AUTO|SR_30M_25M|SR_6M_5M) in apps.
I mean SR_SEARCH_RANGES|SR_30M_25M|SR_6M_5M.
--
Roberto Ragusamail at robertoragusa.it
anges
(the mt312 can't search between 25M and 30M without locking on
a 22M)
Obviously we'd have to define FE_CAN_SR_AUTO.
No compatibility broken, good hardware independence.
I'm open to comments on this matter.
BTW, is there any other frontend on which auto sr can be achieved?
I suspect we could let the mt312 write the rules and stop worrying.
Regards.
--
Roberto Ragusamail at robertoragusa.it
e same program and detect anomalies (such as wrong
info from the broadcaster).
Comments, suggestions, successes, failures, ideas kindly welcome.
Have fun.
--
Roberto Ragusamail at robertoragusa.it
frontend.h.diff
Description: Binary data
mt312.c.diff
Description: Binary data
blindscan.c
Description: Binary data
lugins to use a software decoder, for example xine.
You need to compile vdr and the plugin from source, use
'vdr -Pxine' and run xine with a special syntax to decode the MPEG2
and display the OSD.
> Sorry that I use User-Questions on a Developers list, if that is
> depreciated.
vdr has its own mailing list.
--
Roberto Ragusamail at robertoragusa.it
ommands by default.
(BTW, current drivers disagree on the type of error: I've seen
-ENOIOCTLCMD, -EOPNOTSUPP, -EINVAL).
--
Roberto Ragusamail at robertoragusa.it
pports many codecs, including libmpeg2.
But it uses a 0.3.x libmpeg2 version, while 4:2:2 is only available in 0.4.x.
There are some patches for mplayer which have not been accepted yet.
More info in my post on mplayer-dev ML with subject
"Re: [MPlayer-dev-eng] Win32 codecs NODAEMON"
an
tar2 Rev. 2.6C & hyperthreading) Reinhard Nissl
says hyperthreading causes problems, Sebastian Frei says he has
problems without SMP (what about preempt?).
If there is a bug around, we need more detailed info to isolate it.
--
Roberto Ragusamail at robertoragusa.it
--
Info:
T
n't know this card, but it looks like an entire DVB receiver
(MC68302FC16C is a 16MHz CPU) put on a PCI bus in the prehistory
of DVB broadcasts (press release in 1998).
I think this is not supported under Linux and nobody will ever bother
writing a driver for it.
No idea about apps for Windows, sorry.
Indeed, I unsubscribed my old address and subscribed my new one.
I get mails to the old one and my posts from the new one are
rejected (no posting for unsubscribed users).
Someone playing with a time machine? :-)
--
Roberto Ragusar.ragusa at libero dot it
--
Info:
To unsubsc
ur correction doesn't look right to me.
Maybe it "solves" your problem by pure luck.
If we don't find something, how can we free it?
I hope someone more familiar with this code will reply.
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as
subject.
quality encoding (side note: -vop is obsoleted by -vf).
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as
subject.
With the primary DVB-S
> Device from Siemens Rev. 1.3 the system works fine at all.
>
> Are there experiences with this card out there?
You should try the dvb-kernel driver.
It is included in kernel 2.6.1 but you can compile it for 2.4.x too.
--
Roberto Ragusar.ragusa at liber
wn when the switch occurs. Nothing to see here.
Aren't that "feed not in list" suspicious?
Is the application trying to free the pids twice or what? Never seen
that message here.
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as
subject.
On Sat, 24 Jan 2004 13:44:33 +0100
Helmut Gildein <[EMAIL PROTECTED]> wrote:
> Any hints ?
> Should I try to use kernel 2.6 ?
The skystar2 module has a debug option, turn it on.
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a mail to [EMAIL PRO
as we are not doing
it with analog TV, where disturbing ghosts and reflections could be
improved by using an antenna rotor.
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as
subject.
, but during make install
The errors could be related to other drivers. I use to copy the modules
manually into lib/modules/`uname -r`/misc (for my 2.3p skystar2 they are
skystar2.o mt312.o and dvb-core.o on kernel 2.4.22). You are not interested
in successfully compile and install everything.
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as
subject.
e part of it to smooth the stream flow. Isn't it reasonable
to have a huge buffer in such a case?
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as
subject.
e it is busy. (using insmod.sh unload).
The skystar2 driver can be compiled with debug turned on (look at the source)
and has some module parameters. Debug info could be useful to understand
what is happening; send it to the list if you need help with its interpretation.
--
Roberto Ragusa
it can't be even raised. The default is 10*1024*188 bytes
(which is only one second with HDTV at 15Mbit/s).
Am I wrong?
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as
subject.
ed for file
playback too, with a flag to enable "tail -f" behavior.
I can kind of workaround the problem with pipes, tail -f and other
tricks but it appears too complicated for such a simple thing.
Bye. Ciao.
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubsc
try avoiding this (small and not overlapping window, or
"mplayer >/dev/null 2>/dev/null").
The biggest problem with HDTV is that human affordable monitors don't
have enough resolution (1920 pixels at 0.25mm need a 48cm wide monitor,
that would be a 4:3 monitor with a diagona
On Wed, 10 Dec 2003 16:52:16 +0100
Holger Waechtler <[EMAIL PROTECTED]> wrote:
> > adapter->dmxdev.filternum = 32;
>
> You can safely set this number to 256 -- when you open the whole
Committed.
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe
On Wed, 10 Dec 2003 18:28:15 +0100
Wolfgang Thiel <[EMAIL PROTECTED]> wrote:
> I'ld also vote to use just native 'int' for real counters, like pid_rc:
Right.
We haven't done the full step away from "use higher pid bits" yet. :-)
Committed.
--
zap causes a oops of the kernel.
The system remains useable (the DVB subsystem doesn't).
The trace symbols point to dvb_release doing something wrong.
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as
subject.
the pids were requested.
Let's test this version for a while, then I'll do it.
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as
subject.
de behaves
incorrectly.
Thank you.
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as
subject.
On Sat, 06 Dec 2003 00:27:00 +
Vincenzo Di Massa <[EMAIL PROTECTED]> wrote:
> I send this again becouse no one answered.
> Maybe the previous message went to the bit bucket...
Applied to CVS.
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a ma
code myself, but maybe someone else
> > with more knowledge of the dmx_core internals can do it more
> > easily than me.
>
> :) no - code should always get implemented by the one who needs it...
Hehe :-) I tried.
I'd continue with
...under the guide of the ones who know the environment better than him
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as
subject.
s
(note, only szap is an existing program).
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as
subject.
; demultiplexing code from user space to kernel space. But, as I said, I
> am not an expert (just a random user) hence I might be completely wrong.
Yes, some kernel support is needed (this is why this thread exists).
But kernel space is the right place for this work (you used the word
&q
On Fri, 5 Dec 2003 14:27:13 +0100
Ralph Metzler <[EMAIL PROTECTED]> wrote:
> Roberto Ragusa writes:
> [...]
> >
> > At that point dvrY is totally useless and it could be theorically
> > removed (but it's better to not break existing apps, of course).
>
suggestions on the matter.
I could try to hack the code myself, but maybe someone else
with more knowledge of the dmx_core internals can do it more
easily than me.
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as
subject.
in a previous mail.
I'll look at the source code if I find some time.
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as
subject.
econd
feature.
Some one cares to clarify these issues or point to docs?
Bye.
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as
subject.
s version should be functionally equivalent to Niklas+myself
and add extra debug stuff.
Wolfgang's patch looks like the best (even if I've not tested it here
yet, he said no functional change was done).
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a m
(so having an imperfect
driver for some time is admissable) or has it a sort of "release candidate"
meaning?
In a development environment the current patch looks good enough to be
committed in order to receive more feedback.
--
Roberto Ragusar.ragusa at libero.it
--
ids + pid 8192
- scan on a very populated transponder (Astra 12051V)
I vote for CVS inclusion and I'm looking forward to reshaping the ugly
parts of the driver (according to the old rule "if it works, mess it
up until it doesn't!").
--
Roberto Ragusar.ragusa at libero.it
ull ts feature should now
> work cleaner.
> Again, it would be good to have some more tests first.
Of course, I'll try your new patch.
I forgot to mention that I used 2 and 0 as arguments in the lines with
"fixme: right flags?" instead of 3 and 0.
--
Roberto Ragusar.
=0x704;printk("%s reg 0x%04x = %08x\n",__FUNCTION__,r,read_reg_dw(adapter,
r));
r=0x710;printk("%s reg 0x%04x = %08x\n",__FUNCTION__,r,read_reg_dw(adapter,
r));
r=0x714;printk("%s reg 0x%04x = %08x\n",__FUNCTION__,r,read_reg_dw(adapter,
r));
}
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as
subject.
set or '0' and bit 14 set or '1' in
> order to enable the PID.
So, a value of 0x4000, which is op=2 in the code we have.
It remains a mistery why the HW filters refuse to work (nobody reported a success yet).
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To uns
On Sat, 29 Nov 2003 15:59:21 +0100
Holger Waechtler <[EMAIL PROTECTED]> wrote:
> Roberto Ragusa wrote:
> > In relation to "one of the first checks", do you think that the swfilter
> > code is fast enough to let us just open_whole_bandwith when we're out of
>
hw filters?
Consider that the unwanted/wanted ratio can be 5 or 10 in typical
usage and much higher in particular cases (we have to open
everything when scan is asking 10 pids).
What do you think?
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a mail to [EMAIL PROT
k,
we'll have all the pieces to reshape the driver for elegance, performance,
readability, correctness and so on.
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as
subject.
let them go?" (from the
speed point of view).
> > Remote MTA mgate.uni-hannover.de: SMTP diagnostic: 550 5.7.1 <[EMAIL
> > PROTECTED]>... Rejected: 193.70.192.127 listed at dnsbl.sorbs.net
> >
> That seems to indicate that your MTA is a spammer?!
Yes.
But it
ed 1, 2 and 3, but the new filters don't work here.
My chip has revision 195 (II-B).
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as
subject.
e
filters for 0x00-0x1f too, expecially if, as I'm reading, some chips
don't have the extra 0x310 filters.
I'm glad to see great interest in this driver.
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as
subject.
trying to do.
I will test.
The first thing I'll do is to allocate the new filters before the old ones,
so we stress the 0x20 even when only 2 pids are asked for.
Side note:
the MTA of your email address doesn't want me to email you :-)
Remote MTA mgate.uni-hannover.de: SMTP diagnostic: 550 5.7.1 <[EMAIL PROTECTED]>...
Rejected: 193.70.192.127 listed at dnsbl.sorbs.net
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as
subject.
but removing limits is always good).
I'm sure these discussions will bring us a better skystar2.c in a few
days.
BTW, does scan work reliably for you?
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as
subject.
On Wed, 26 Nov 2003 23:36:59 +0100
Holger Waechtler <[EMAIL PROTECTED]> wrote:
> Roberto Ragusa wrote:
> > Should I go on with cosmetics?
>
> Yes, definitely. But your patch did not applied, can you please generate
> your patchsets using 'cvs -q diff -pu' in
re is any specific reason to mask pid
numbers with 0x3fff or 0x (instead of 0x1fff). I converted everything
to 0x1fff, otherwise there are problems with spurious high bits
not cleared in the hardware registers (no packets match).
The patch is not very human-readable because diff didn't
if ( strcmp(info.name, "STV0299/TSA5059/SL1935 based") == 0) {
dprintk("%s: FE_SET_TONE for %s frontend\n", __FUNCTION__,
info.name);
/* allow the frontend driver to handle this command*/
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as
subject.
/?
What about the pid=8192 one? That is a bug (if serious or not, it's
questionable).
When is the next major release supposed to happen?
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as
subject.
ibility with my (not vanilla)
kernel?).
Patches are based on CVS at 2003-11-23 22:00 UTC.
Bye.
--
Roberto Ragusar.ragusa at libero.it
dvb-style.diff.bz2
Description: BZip2 compressed data
dvbk-style.diff.bz2
Description: BZip2 compressed data
ibility with my (not vanilla)
kernel?).
Patches are based on CVS at 2003-11-23 22:00 UTC.
Bye.
--
Roberto Ragusar.ragusa at libero.it
of course), so it's OK too.
I think the 0x27 comes from 0x20 (pids <=0x1f) and 7 (single pid filters).
>
> [ I have miserably failed to add 4:2:2 support to mplayer this weekend:
> at least I got a first idea how mplayer is working, now :-(
> ]
Are you using the developme
system to check correct
compilation at least.
> Despite this the packet_header checks around line 1858 are not
> endian-safe.
Right, that looks really bad.
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as
subject.
exactly 0x27 slots?
2) to advertize 32 pids max?
3) to have stream1/stream2/pcr/pmt/emm/ecm specific routines, given that
it looks like they have no particular specialization, they're just generic
filters?
4) to have different coding style than linux kernel? ("CheckPID" instead
of &quo
|| CheckPID(adapter, 0x2000) != 0)
{
dvb_dmx_swfilter_packets(dvbdmx, pbDMABufCurPos,
dwDefaultPacketSize/188);
} else {
--
Roberto Ragusar.ragusa at libero.it
--
Info:
To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as
subject.
64 matches
Mail list logo
