[vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-16 Thread serge pecher
Sorry, but is this the last version of the patch ?

I believe there was an additional one, but can't find it anymore.

 

http://www.linuxtv.org/pipermail/vdr//2008-January/014956.html
http://www.linuxtv.org/pipermail/vdr/2008-January/014956.html  

 

thanks,

 

sp

___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-16 Thread Morfsta
Well, I think there has been a number of fixes (e.g. to the speedup
patches and the nit.c file) that probably warrants a new patch for
1.5.13 that incorporates them all. I'm sure Reinhard will do that when
he has a moment, he probably is taking a well deserved break after
looking into so many issues and fixes for us over the last few weeks!

In the meantime, just look over Reinhard's postings since the first
ANNOUNCE for vdr dvb-s2 and h264 for vdr-1.5.12 and download and apply
all of the relevant patches.

With all of the required patches, fixes and tweaks for VDR / FFMPEG /
XINE-LIB / VDR-XINE I'm wondering if it would be nice to set-up a web
site that tracks all of the latest patches for VDR / S2 / h264. Would
anyone be interested in this? I know there is the vdrportal.de site
that discusses a lot of this stuff, but as its in German I find it
quite hard to follow.

On Jan 16, 2008 12:34 PM, serge pecher [EMAIL PROTECTED] wrote:




 Sorry, but is this the last version of the patch ?

 I believe there was an additional one, but can't find it anymore.



 http://www.linuxtv.org/pipermail/vdr//2008-January/014956.html



 thanks,



 sp
 ___
 vdr mailing list
 vdr@linuxtv.org
 http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr



___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-16 Thread Morfsta
sorry for the double post, but let's also not forget multiproto,
TT3200 and HVR4000 status! There is quite a lot to have to keep track
of.

PS Manu, are you considering implementing the HVR4000 patch directly
into the multiproto tree for direct support?

Thanks

On Jan 16, 2008 3:01 PM, Morfsta [EMAIL PROTECTED] wrote:
 Well, I think there has been a number of fixes (e.g. to the speedup
 patches and the nit.c file) that probably warrants a new patch for
 1.5.13 that incorporates them all. I'm sure Reinhard will do that when
 he has a moment, he probably is taking a well deserved break after
 looking into so many issues and fixes for us over the last few weeks!

 In the meantime, just look over Reinhard's postings since the first
 ANNOUNCE for vdr dvb-s2 and h264 for vdr-1.5.12 and download and apply
 all of the relevant patches.

 With all of the required patches, fixes and tweaks for VDR / FFMPEG /
 XINE-LIB / VDR-XINE I'm wondering if it would be nice to set-up a web
 site that tracks all of the latest patches for VDR / S2 / h264. Would
 anyone be interested in this? I know there is the vdrportal.de site
 that discusses a lot of this stuff, but as its in German I find it
 quite hard to follow.


 On Jan 16, 2008 12:34 PM, serge pecher [EMAIL PROTECTED] wrote:
 
 
 
 
  Sorry, but is this the last version of the patch ?
 
  I believe there was an additional one, but can't find it anymore.
 
 
 
  http://www.linuxtv.org/pipermail/vdr//2008-January/014956.html
 
 
 
  thanks,
 
 
 
  sp
  ___
  vdr mailing list
  vdr@linuxtv.org
  http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
 
 


___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-16 Thread VDR User
On Jan 16, 2008 7:01 AM, Morfsta [EMAIL PROTECTED] wrote:
 With all of the required patches, fixes and tweaks for VDR / FFMPEG /
 XINE-LIB / VDR-XINE I'm wondering if it would be nice to set-up a web
 site that tracks all of the latest patches for VDR / S2 / h264. Would
 anyone be interested in this? I know there is the vdrportal.de site
 that discusses a lot of this stuff, but as its in German I find it
 quite hard to follow.

I think that's a great idea considering all the patches and then
different versions of some of them.  It seems a lot of people are
getting confused by what to use.  I also agree that vdrportal.de is
hard to follow for non-German speakers.  Which is a shame for us NA
users since it seems to be a good vdr resource.

___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-13 Thread Hagen Schöbel
I have an problem with the original patch to vdr-1.5.12:

[EMAIL PROTECTED]:/usr/local/src/vdr-1.5.12-dvbs2# make
g++ -g -O2 -Wall -Woverloaded-virtual -c -DREMOTE_KBD 
-DLIRC_DEVICE=\/dev/lircd\ -DRCU_DEVICE=\/dev/ttyS1\ -D_GNU_SOURCE 
-DVIDEODIR=\/video\ -DCONFDIR=\/video\ -DPLUGINDIR=\./PLUGINS/lib\ 
-DLOCDIR=\./locale\ -I/usr/include/freetype2 audio.c
dvbdevice.h:38: Fehler: »dvbfe_delsys« bezeichnet keinen Typ
make: *** [audio.o] Fehler 1

Any hints ?

Thank you...

...Hagen

___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-13 Thread Stefan Lucke
On Sunday 13 January 2008, Hagen Schöbel wrote:
 I have an problem with the original patch to vdr-1.5.12:
 
 [EMAIL PROTECTED]:/usr/local/src/vdr-1.5.12-dvbs2# make
 g++ -g -O2 -Wall -Woverloaded-virtual -c -DREMOTE_KBD 
 -DLIRC_DEVICE=\/dev/lircd\ -DRCU_DEVICE=\/dev/ttyS1\ -D_GNU_SOURCE 
 -DVIDEODIR=\/video\ -DCONFDIR=\/video\ -DPLUGINDIR=\./PLUGINS/lib\ 
 -DLOCDIR=\./locale\ -I/usr/include/freetype2 audio.c
 dvbdevice.h:38: Fehler: »dvbfe_delsys« bezeichnet keinen Typ
 make: *** [audio.o] Fehler 1
 
 Any hints ?

The include path to your modified DVB-S2 drivers is missing.
I set this path in Make.config (see Make.config.template) to:
DVBDIR   = /usr/local/src/v4l-dvb/linux

..-DLOCDIR=\./locale\  -I/usr/include/freetype2 
-I/usr/local/src/v4l-dvb/linux/include dvbdevice.c ..

-- 
Stefan Lucke

___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-07 Thread Gregoire Favre
On Mon, Jan 07, 2008 at 05:19:17PM +0100, Gregoire Favre wrote:
 On Sun, Jan 06, 2008 at 09:34:32PM +0100, Reinhard Nissl wrote:
  Hi,
  
  Please add some debug code to cDvbDevice::ProvidesTransponder()
  to find out what's going wrong in your case, i. e. compare the
  provided and requested modulation systems.
 
 My cards :
 
 0: 101469280
 1: 101469280
 2: 101469280
 
 But the card 1 : Hauppauge HVR-4000 is the dvb-s2 one ???
 
 DVB: registering frontend 0 (ST STV0299 DVB-S)...
 DVB: registering frontend 1 (Conexant CX24116/CX24118)...
 DVB: registering frontend 2 (Conexant CX24123/CX24109)...
 Without the patch and with -D1 DVB-S2 works perfectly.

In case someone want to try to log here a small diff for it :

--- device.c.orig   2008-01-07 17:17:05.0 +0100
+++ device.c2008-01-07 16:59:22.0 +0100
@@ -18,6 +18,7 @@
 #include receiver.h
 #include status.h
 #include transfer.h
+#include iostream
 
 // --- cLiveSubtitle -
 
@@ -433,6 +434,7 @@
  imp = 1; imp |= NumUsableSlots ? 0 : device[i]-HasCi();
  // avoid cards with Common Interface for 
FTA channels
  imp = 1; imp |= device[i]-HasDecoder();
  // avoid full featured cards
  imp = 1; imp |= NumUsableSlots ? 
!ChannelCamRelations.CamDecrypt(Channel-GetChannelID(), j + 1) : 0; // prefer 
CAMs that are known to decrypt this channel
+std::cerr  DVBS2-log:   i  :   imp  std::endl;
  if (imp  Impact) {
 // This device has less impact than any previous one, so we 
take it.
 Impact = imp;
--- dvbdevice.c.orig2008-01-07 17:17:58.0 +0100
+++ dvbdevice.c 2008-01-07 17:01:46.0 +0100
@@ -26,6 +26,7 @@
 #include receiver.h
 #include status.h
 #include transfer.h
+#include iostream
 
 #define DO_REC_AND_PLAY_ON_PRIMARY_DEVICE 1
 #define DO_MULTIPLE_RECORDINGS 1
@@ -805,12 +806,18 @@
 
 bool cDvbDevice::ProvidesTransponder(const cChannel *Channel) const
 {
-  if (!ProvidesSource(Channel-Source())) 
+  if (!ProvidesSource(Channel-Source())) {
+ std::cerr  DVBS2-log: - Doesn't provide source  std::endl;
  return false; // doesn't provide source
-  if (!cSource::IsSat(Channel-Source()))
+  }
+  if (!cSource::IsSat(Channel-Source())) {
+ std::cerr  DVBS2-log: + source is sufficient for non sat  std::endl;
  return true; // source is sufficient for non sat
-  if (!(frontendType  Channel-ModulationSystem()))
+  }
+  if (!(frontendType  Channel-ModulationSystem())) {
+ std::cerr  DVBS2-log: - requires modulation system which frontend 
doesn't provide  std::endl;
  return false; // requires modulation system which frontend doesn't provide
+  }
   return !Setup.DiSEqC || Diseqcs.Get(Channel-Source(), Channel-Frequency(), 
Channel-Polarization());
 }
 

-- 
Grégoire FAVRE  http://gregoire.favre.googlepages.com  http://www.gnupg.org
   http://picasaweb.google.com/Gregoire.Favre

___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-07 Thread Reinhard Nissl
Hi,

Gregoire Favre schrieb:

 Please add some debug code to cDvbDevice::ProvidesTransponder()
 to find out what's going wrong in your case, i. e. compare the
 provided and requested modulation systems.

 My cards :

 0: 101469280
 1: 101469280
 2: 101469280

Hmm, looks like you've reported the impact numbers from
GetDevice(). But I've wanted some log messages from
ProvidesTransponder().

Anyway, as the impact is the same for all devices, your driver
doesn't set DVBFE_DELSYS_DVBS2.

 But the card 1 : Hauppauge HVR-4000 is the dvb-s2 one ???

 DVB: registering frontend 0 (ST STV0299 DVB-S)...
 DVB: registering frontend 1 (Conexant CX24116/CX24118)...
 DVB: registering frontend 2 (Conexant CX24123/CX24109)...
 Without the patch and with -D1 DVB-S2 works perfectly.

I've patched multiproto with the HVR-4000 support patch you've
mentioned on the linux-dvb mailing list.

cx24116.c contains this function:
/* TODO: The hardware does DSS too, how does the kernel demux
handle this? */
static int cx24116_get_delsys(struct dvb_frontend *fe,
enum dvbfe_delsys *fe_delsys)
{
dprintk(%s()\n,__FUNCTION__);
*fe_delsys = DVBFE_DELSYS_DVBS | DVBFE_DELSYS_DVBS2;

return 0;
}

Therefore the above mentioned flag should be set. Please check
whether this function gets called when VDR reaches the end of
cDvbDevice::cDvbDevice().

In case the emulation layer of the multiproto tree kicks in, then
the function will not get called and therefore DVBFE_DELSYS_DVBS2
is not set.

Bye.
-- 
Dipl.-Inform. (FH) Reinhard Nissl
mailto:[EMAIL PROTECTED]

___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-07 Thread Gregoire Favre
On Mon, Jan 07, 2008 at 07:27:21PM +0100, Reinhard Nissl wrote:

Hello :-)

 I've patched multiproto with the HVR-4000 support patch you've
 mentioned on the linux-dvb mailing list.

I use this one :
http://www.mail-archive.com/[EMAIL PROTECTED]/msg28020.html 
but it don't seems to be the same as the one you used to check your
source. It miss this function... I have to look after it in other
patch...

Thank you very much, then the problem don't seems to be your patch but
my driver, which will take some time to investigate :-)

Have a nice evening,
-- 
Grégoire FAVRE  http://gregoire.favre.googlepages.com  http://www.gnupg.org
   http://picasaweb.google.com/Gregoire.Favre

___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-06 Thread Gregoire Favre
On Sat, Jan 05, 2008 at 06:53:48PM +0100, Reinhard Nissl wrote:

Hello,

with the last patch, VDR refuse to tune to any DVB-S2 channels.
I guess that VDR thinks all my card are only DVB-S ones.
I get Channel not available! on all DVB-S2 channels.

Any maybe not directly interesting, but with this patch, graphlcd
make vdr segault at start.

Thanks,

 The attached version implements this behavior now. You may want
 to experiment a bit with the decision logic as explained below.
 
 The decision is based on the number of modulation systems a card
 provides. For example, my NOVA-S provides one (DVB-S) and my
 SkyStar HD provides three (DVB-S, DVB-DSS, DVB-S2).
 
 The decision logic is implemented in cDevice::GetDevice(), so
 have a look into device.c. After applying the patch, you'll find
 two lines in that function marked with comments like /*1*/ and
 /*2*/, and the latter one was disabled by a line comment.
 
 In my scenario with just the above cards and vdr-xine as software
 device, watching a channel requires VDR to operate in transfer
 mode. Therefore it selects a device which provides for example a
 DVB-S channel. The patched version will choose the NOVA-S with
 either implementation /*1*/ or /*2*/.
 
 Let's then start a DVB-S recording on a different transponder.
 When implementation /*2*/ would be active, VDR would choose the
 SkyStar HD, as the NOVA-S is claimed by transfer mode. Then, try
 to switch to a DVB-S2 channel. It wouldn't work, as the SkyStar
 HD would be claimed by a DVB-S recording.
 
 That's why I've chosen implementation /*1*/ as default. For the
 above scenario, starting a DVB-S recording on a different
 transponder will choose the NOVA-S and kick off transfer mode.
 Then VDR will look for a card to set up transfer mode again and
 it will choose the only remaining card, the SkyStar HD. If you
 then try to switch to a DVB-S2 channel, it will work as the
 SkyStar HD is not claimed by a recording.
 
  The patch is incremental to the original dvbs2 patch from
  yesterday, i. e. you can simply apply it to your already patched VDR.
 
 Bye.
 -- 
 Dipl.-Inform. (FH) Reinhard Nissl
 mailto:[EMAIL PROTECTED]

 --- ../vdr-1.5.12-dvbs2-other/device.c2008-01-01 22:55:18.0 
 +0100
 +++ device.c  2008-01-05 18:06:15.0 +0100
 @@ -359,6 +359,21 @@ cDevice *cDevice::GetDevice(int Index)
return (0 = Index  Index  numDevices) ? device[Index] : NULL;
  }
  
 +inline int GetClippedModulationSystemCount(int AvailableBits, cDevice 
 *Device)
 +{
 +  int ModulationSystemCount = Device-ModulationSystemCount();
 +  int MaxModulationSystemCount = 1  AvailableBits;
 +  if (ModulationSystemCount  MaxModulationSystemCount) {
 + esyslog(ERROR: device %d supports %d modulation systems but 
 cDevice::GetDevice() currently only supports %d modulation systems which 
 should be fixed, Device-CardIndex() + 1, ModulationSystemCount, 
 MaxModulationSystemCount);
 + ModulationSystemCount = MaxModulationSystemCount;
 + }
 +  else if (ModulationSystemCount = 0) {
 + esyslog(ERROR: device %d reported an invalid number (%d) of supported 
 modulation systems. The device should be fixed to return at least 1, 
 Device-CardIndex() + 1, ModulationSystemCount);
 + ModulationSystemCount = 1;
 + }
 +  return ModulationSystemCount;
 +}
 +
  cDevice *cDevice::GetDevice(const cChannel *Channel, int Priority, bool 
 LiveView)
  {
// Collect the current priorities of all CAM slots that can decrypt the 
 channel:
 @@ -408,11 +423,13 @@ cDevice *cDevice::GetDevice(const cChann
   imp = 1; imp |= LiveView ? !device[i]-IsPrimaryDevice() || 
 ndr : 0;  // prefer the primary device for 
 live viewing if we don't need to detach existing receivers
   imp = 1; imp |= !device[i]-Receiving()  (device[i] != 
 cTransferControl::ReceiverDevice() || device[i]-IsPrimaryDevice()) || ndr; 
 // use receiving devices if we don't need to detach existing receivers, but 
 avoid primary device in local transfer mode
   imp = 1; imp |= device[i]-Receiving();   
 // avoid devices that are receiving
 +/*1*/imp = 2; imp |= GetClippedModulationSystemCount(2, device[i]) 
 - 1;// avoid cards which support multiple 
 modulation systems
   imp = 1; imp |= device[i] == 
 cTransferControl::ReceiverDevice();  // 
 avoid the Transfer Mode receiver device
   imp = 8; imp |= min(max(device[i]-Priority() + MAXPRIORITY, 
 0), 0xFF);   // use the device with the lowest 
 priority (+MAXPRIORITY to assure that values -99..99 can be used)
   imp = 8; imp |= min(max((NumUsableSlots ? SlotPriority[j] : 
 0) + MAXPRIORITY, 0), 0xFF);  // use the CAM slot with the lowest 
 priority (+MAXPRIORITY to assure that values -99..99 can be used)
   imp 

Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-06 Thread Reinhard Nissl
Hi,

Gregoire Favre schrieb:

 with the last patch, VDR refuse to tune to any DVB-S2 channels.
 I guess that VDR thinks all my card are only DVB-S ones.
 I get Channel not available! on all DVB-S2 channels.

Please add some debug code to cDvbDevice::ProvidesTransponder()
to find out what's going wrong in your case, i. e. compare the
provided and requested modulation systems.

 Any maybe not directly interesting, but with this patch, graphlcd
 make vdr segault at start.

Did you recompile all plugins as a virtual function had been added?

Please provide a backtrace otherwise.

Bye.
-- 
Dipl.-Inform. (FH) Reinhard Nissl
mailto:[EMAIL PROTECTED]

___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-06 Thread Reinhard Nissl
Hi,

Reinhard Nissl schrieb:

 The patches now include my recently released speedup patches as
 well as an unreleased speedup patch for cAudioRepacker and
 cVideoRepacker, because at least the latter one would have been
 hard to extract separately.
 
 cAudioRepacker speedup patch contains a typo. Please apply the
 attached fix.

The optimized DrawRectangle() will crash when called with
incorrect coordinates, i. e. x1  x2 or y1  y2.

More generally such issues should be handled in Intersects() and
Covers(). The attached patch adds sanity checks to them.

Thanks to Claus Meder for reporting this issue.

Bye.
-- 
Dipl.-Inform. (FH) Reinhard Nissl
mailto:[EMAIL PROTECTED]
--- ../vdr-1.5.12-dvbs2-other/osd.c	2008-01-01 22:55:18.0 +0100
+++ osd.c	2008-01-06 22:39:33.0 +0100
@@ -217,6 +217,8 @@ bool cBitmap::Contains(int x, int y) con
 
 bool cBitmap::Covers(int x1, int y1, int x2, int y2) const
 {
+  if (x1  x2 || y1  y2) // sanity check
+ return false;
   x1 -= x0;
   y1 -= y0;
   x2 -= x0;
@@ -226,6 +228,8 @@ bool cBitmap::Covers(int x1, int y1, int
 
 bool cBitmap::Intersects(int x1, int y1, int x2, int y2) const
 {
+  if (x1  x2 || y1  y2) // sanity check
+ return false;
   x1 -= x0;
   y1 -= y0;
   x2 -= x0;
___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-06 Thread Gregoire Favre
On Sun, Jan 06, 2008 at 09:34:32PM +0100, Reinhard Nissl wrote:
 Hi,

Hello :-)

 Please add some debug code to cDvbDevice::ProvidesTransponder()
 to find out what's going wrong in your case, i. e. compare the
 provided and requested modulation systems.

I'll do tomorrow ;-)

  Any maybe not directly interesting, but with this patch, graphlcd
  make vdr segault at start.
 
 Did you recompile all plugins as a virtual function had been added?
 
 Please provide a backtrace otherwise.

The two fix solve it :-) (not the non tunable DVB-S2).

Thanks,
-- 
Grégoire FAVRE  http://gregoire.favre.googlepages.com  http://www.gnupg.org
   http://picasaweb.google.com/Gregoire.Favre

___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-05 Thread Reinhard Nissl
Hi,

Reinhard Nissl schrieb:

 Well, I was in contact with Marco already and attached you'll
 find a minimalistic change which reports channel not available.
 Now VDR should already be able to kick a low priority DVB-S
 recording (or transfer thread) from a DVB-S2 device.
 
 The previous patch was wrong. Only DVB-S2 devices could provide
 channels. The revised patch works now as expected.
 
 Still missing is to prefer DVB-S devices for DVB-S recordings so
 that DVB-S2 devices remain available for DVB-S2 recordings of
 same priority.
 
 Still to do.

The attached version implements this behavior now. You may want
to experiment a bit with the decision logic as explained below.

The decision is based on the number of modulation systems a card
provides. For example, my NOVA-S provides one (DVB-S) and my
SkyStar HD provides three (DVB-S, DVB-DSS, DVB-S2).

The decision logic is implemented in cDevice::GetDevice(), so
have a look into device.c. After applying the patch, you'll find
two lines in that function marked with comments like /*1*/ and
/*2*/, and the latter one was disabled by a line comment.

In my scenario with just the above cards and vdr-xine as software
device, watching a channel requires VDR to operate in transfer
mode. Therefore it selects a device which provides for example a
DVB-S channel. The patched version will choose the NOVA-S with
either implementation /*1*/ or /*2*/.

Let's then start a DVB-S recording on a different transponder.
When implementation /*2*/ would be active, VDR would choose the
SkyStar HD, as the NOVA-S is claimed by transfer mode. Then, try
to switch to a DVB-S2 channel. It wouldn't work, as the SkyStar
HD would be claimed by a DVB-S recording.

That's why I've chosen implementation /*1*/ as default. For the
above scenario, starting a DVB-S recording on a different
transponder will choose the NOVA-S and kick off transfer mode.
Then VDR will look for a card to set up transfer mode again and
it will choose the only remaining card, the SkyStar HD. If you
then try to switch to a DVB-S2 channel, it will work as the
SkyStar HD is not claimed by a recording.

 The patch is incremental to the original dvbs2 patch from
 yesterday, i. e. you can simply apply it to your already patched VDR.

Bye.
-- 
Dipl.-Inform. (FH) Reinhard Nissl
mailto:[EMAIL PROTECTED]
--- ../vdr-1.5.12-dvbs2-other/device.c	2008-01-01 22:55:18.0 +0100
+++ device.c	2008-01-05 18:06:15.0 +0100
@@ -359,6 +359,21 @@ cDevice *cDevice::GetDevice(int Index)
   return (0 = Index  Index  numDevices) ? device[Index] : NULL;
 }
 
+inline int GetClippedModulationSystemCount(int AvailableBits, cDevice *Device)
+{
+  int ModulationSystemCount = Device-ModulationSystemCount();
+  int MaxModulationSystemCount = 1  AvailableBits;
+  if (ModulationSystemCount  MaxModulationSystemCount) {
+ esyslog(ERROR: device %d supports %d modulation systems but cDevice::GetDevice() currently only supports %d modulation systems which should be fixed, Device-CardIndex() + 1, ModulationSystemCount, MaxModulationSystemCount);
+ ModulationSystemCount = MaxModulationSystemCount;
+ }
+  else if (ModulationSystemCount = 0) {
+ esyslog(ERROR: device %d reported an invalid number (%d) of supported modulation systems. The device should be fixed to return at least 1, Device-CardIndex() + 1, ModulationSystemCount);
+ ModulationSystemCount = 1;
+ }
+  return ModulationSystemCount;
+}
+
 cDevice *cDevice::GetDevice(const cChannel *Channel, int Priority, bool LiveView)
 {
   // Collect the current priorities of all CAM slots that can decrypt the channel:
@@ -408,11 +423,13 @@ cDevice *cDevice::GetDevice(const cChann
  imp = 1; imp |= LiveView ? !device[i]-IsPrimaryDevice() || ndr : 0;  // prefer the primary device for live viewing if we don't need to detach existing receivers
  imp = 1; imp |= !device[i]-Receiving()  (device[i] != cTransferControl::ReceiverDevice() || device[i]-IsPrimaryDevice()) || ndr; // use receiving devices if we don't need to detach existing receivers, but avoid primary device in local transfer mode
  imp = 1; imp |= device[i]-Receiving();   // avoid devices that are receiving
+/*1*/imp = 2; imp |= GetClippedModulationSystemCount(2, device[i]) - 1;// avoid cards which support multiple modulation systems
  imp = 1; imp |= device[i] == cTransferControl::ReceiverDevice();  // avoid the Transfer Mode receiver device
  imp = 8; imp |= min(max(device[i]-Priority() + MAXPRIORITY, 0), 0xFF);   // use the device with the lowest priority (+MAXPRIORITY to assure that values -99..99 can be used)
  imp = 8; imp |= min(max((NumUsableSlots ? SlotPriority[j] : 0) + MAXPRIORITY, 0), 0xFF);  // use the CAM slot with the lowest 

Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-04 Thread Reinhard Nissl
Hi,

Reinhard Nissl schrieb:

 Well, I was in contact with Marco already and attached you'll
 find a minimalistic change which reports channel not available.
 Now VDR should already be able to kick a low priority DVB-S
 recording (or transfer thread) from a DVB-S2 device.

The previous patch was wrong. Only DVB-S2 devices could provide
channels. The revised patch works now as expected.

 Still missing is to prefer DVB-S devices for DVB-S recordings so
 that DVB-S2 devices remain available for DVB-S2 recordings of
 same priority.

Still to do.

 The patch is incremental to the original dvbs2 patch from
 yesterday, i. e. you can simply apply it to your already patched VDR.

Bye.
-- 
Dipl.-Inform. (FH) Reinhard Nissl
mailto:[EMAIL PROTECTED]
--- ../vdr-1.5.12-dvbs2-other/dvbdevice.c	2008-01-01 22:55:18.0 +0100
+++ dvbdevice.c	2008-01-05 00:44:24.0 +0100
@@ -798,7 +798,13 @@ bool cDvbDevice::ProvidesSource(int Sour
 
 bool cDvbDevice::ProvidesTransponder(const cChannel *Channel) const
 {
-  return ProvidesSource(Channel-Source())  (!cSource::IsSat(Channel-Source()) || !Setup.DiSEqC || Diseqcs.Get(Channel-Source(), Channel-Frequency(), Channel-Polarization()));
+  if (!ProvidesSource(Channel-Source()))
+ return false; // doesn't provide source
+  if (!cSource::IsSat(Channel-Source()))
+ return true; // source is sufficient for non sat
+  if (!(frontendType  Channel-ModulationSystem()))
+ return false; // requires modulation system which frontend doesn't provide
+  return !Setup.DiSEqC || Diseqcs.Get(Channel-Source(), Channel-Frequency(), Channel-Polarization());
 }
 
 bool cDvbDevice::ProvidesChannel(const cChannel *Channel, int Priority, bool *NeedsDetachReceivers) const
@@ -807,7 +813,7 @@ bool cDvbDevice::ProvidesChannel(const c
   bool hasPriority = Priority  0 || Priority  this-Priority();
   bool needsDetachReceivers = false;
 
-  if (ProvidesSource(Channel-Source())) {
+  if (ProvidesTransponder(Channel)) {
  result = hasPriority;
  if (Priority = 0  Receiving(true)) {
 if (dvbTuner-IsTunedTo(Channel)) {
___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-03 Thread Magnus Hörlin
Reinhard Nissl wrote:
 Hi,

 attached you'll find updated patches for VDR-1.5.12, which replace all
 formerly patches regarding this subject.

 The patch named *-dvbs2-* additionally adds DVB-S2 support to VDR
 (thanks to Marco Schlüßler) and requires to use the DVB drivers from the
 multi-proto tree (see URL below for further details).

 The other patch is without DVB-S2 support and therefore most suitable
 for DVB-C users.

 The patches have been extended to also include the recently released
 audioindexer patch. Furthermore, the field detection code for H.264 has
 been adopted to MPEG2, where the same issue (VDR's index.vdr addresses
 frame pictures, so an index entry must not be generated for the second
 field of a field picture pair) exists, though hardly used compared to H.264.

 Have a look at this page for more instructions on this concern:

 http://www.vdr-wiki.de/wiki/index.php/OpenSuSE_DVB-S2_-_Step_by_Step_Installationsanleitung_%28Achtung_Beta%29

 Bye.
   
 

 ___
 vdr mailing list
 vdr@linuxtv.org
 http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
   
Hi. Big thanks to Reinhard, Manu, Claus and others who has made this 
possible. Using the old multiproto tree from 2007-10-25 I have SVT HD 
(swedish) working quite well with an S2-3200 on a 2.3GHz AMD BE-2400. My 
problem is that with that multiproto tree, VDR won't tune to any dvb-t 
channels on my nova-t 500. And with the latest multiproto tree dvb-t 
works but only unless I load the S2-3200 modules. If I do that VDR just 
hangs at startup saying nothing and doesn't answer on svdrp. Is there a 
way to debug this and see what's happening? Does anybody have any ideas 
about what's going on? If I remove the channels.conf vdr complains about 
that, so I know it's coming that far at least. There's no difference if 
I load any plugins or not.

Again, I have followed all steps on the wiki and h.264+dvb-s2 works with 
the old multiproto, both using vdr-1.5.10 with your old patch and on 
vdr-1.5.12 with the new one. I'm running this on the 2.6.22-14-generic 
kernel that ships with Ubuntu 7.10.

Keep up the good work,
/Magnus H


___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-03 Thread Morfsta
On Jan 3, 2008 5:40 PM, Magnus Hörlin [EMAIL PROTECTED] wrote:
 Hi. Big thanks to Reinhard, Manu, Claus and others who has made this
 possible. Using the old multiproto tree from 2007-10-25 I have SVT HD
 (swedish) working quite well with an S2-3200 on a 2.3GHz AMD BE-2400. My
 problem is that with that multiproto tree, VDR won't tune to any dvb-t
 channels on my nova-t 500. And with the latest multiproto tree dvb-t
 works but only unless I load the S2-3200 modules. If I do that VDR just
 hangs at startup saying nothing and doesn't answer on svdrp. Is there a
 way to debug this and see what's happening? Does anybody have any ideas
 about what's going on? If I remove the channels.conf vdr complains about
 that, so I know it's coming that far at least. There's no difference if
 I load any plugins or not.

 Again, I have followed all steps on the wiki and h.264+dvb-s2 works with
 the old multiproto, both using vdr-1.5.10 with your old patch and on
 vdr-1.5.12 with the new one. I'm running this on the 2.6.22-14-generic
 kernel that ships with Ubuntu 7.10.

Sounds like you are having the same problem I am - when you say new
multiproto, what do you mean?

I am using the mercurial tree at http://jusst.de/hg/multiproto which
hasn't been updated now in 4 weeks.

I can't get the DVB-T devices working with it, even though Manu said
it should work. The devices I use haven't had their source code
modified from the v4l-dvb tree, so according to Manu the compatibility
layer should mean they work OK, but when I tune to a DVB-T channel in
VDR I just get a black screen.

As it seems there are quite a few people desperately hoping for some
kind of DVB-T support alongisde their beloved DVB-S2 are there any
other suggestions from Manu or Reinhard on this?

Cheers

___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


[vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-02 Thread Morfsta
Reinhard,

Wow! I am running vdr-1.5.12 with the new xine-lib CVS (with your loop
filter and speed over accuracy changes enabled) alongside FFMPEG built
for 64bit linux on a k8 (AMD dual core BE-2350 processor overclocked
at 2.7Ghz) and the results are nothing short of amazing! Firstly I
would like to say thank-you very much for all your hard work on H264 -
what you have achieved is fantastic.

I can view channels such as BBC HD, Channel 4 HD and others perfectly.

I think this might have been covered before, but the only problem I
now see is with interlaced and spatial direct mode (particularly on
the French and Polish HD channels): -

[h264 @ 0x2d0b0700]Interlaced pictures + spatial direct mode is
not implemented
[h264 @ 0x2d0b0700]Interlaced pictures + spatial direct mode is
not implemented
buffer usage: 492,  0,  6,  0, 0x97d390
buffer usage: 492,  0,  7,  0, 0x97d390
buffer usage: 444,  0,  7,  0, 0x97d390
buffer usage: 444,  0,  7,  0, 0x97d390
[h264 @ 0x2d0b0700]Interlaced pictures + spatial direct mode is
not implemented
[h264 @ 0x2d0b0700]Interlaced pictures + spatial direct mode is
not implemented
buffer usage: 411,  0,  7,  0, 0x97d390
buffer usage: 411,  0,  7,  0, 0x97d390

this results in a picture, but artifacting on motion and eventually in a crash.

I have a short clip recorded if you would like to work on it and need
an example?

Please let me know,

Kind Regards,

Morfsta


On Jan 1, 2008 10:15 PM, Reinhard Nissl [EMAIL PROTECTED] wrote:
 Hi,

 Reinhard Nissl schrieb:

  attached you'll find updated patches for VDR-1.5.12, which replace all
  formerly patches regarding this subject.
 
  The patch named *-dvbs2-* additionally adds DVB-S2 support to VDR
  (thanks to Marco Schlüßler) and requires to use the DVB drivers from the
  multi-proto tree (see URL below for further details).
 
  The other patch is without DVB-S2 support and therefore most suitable
  for DVB-C users.
 
  The patches have been extended to also include the recently released
  audioindexer patch. Furthermore, the field detection code for H.264 has
  been adopted to MPEG2, where the same issue (VDR's index.vdr addresses
  frame pictures, so an index entry must not be generated for the second
  field of a field picture pair) exists, though hardly used compared to H.264.

 The patches now include my recently released speedup patches as
 well as an unreleased speedup patch for cAudioRepacker and
 cVideoRepacker, because at least the latter one would have been
 hard to extract separately.


  Have a look at this page for more instructions on this concern:
 
  http://www.vdr-wiki.de/wiki/index.php/OpenSuSE_DVB-S2_-_Step_by_Step_Installationsanleitung_%28Achtung_Beta%29

 Bye.
 --
 Dipl.-Inform. (FH) Reinhard Nissl
 mailto:[EMAIL PROTECTED]

 ___
 vdr mailing list
 vdr@linuxtv.org
 http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr



___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-02 Thread Torgeir Veimo

On 2 Jan 2008, at 21:04, Morfsta wrote:

 I can view channels such as BBC HD, Channel 4 HD and others perfectly.


A howto outlining the steps you've taken to get this to work would be  
appreciated! :)


-- 
Torgeir Veimo
[EMAIL PROTECTED]




___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-02 Thread Morfsta
On Jan 2, 2008 11:08 AM, Torgeir Veimo [EMAIL PROTECTED] wrote:
 A howto outlining the steps you've taken to get this to work would be
 appreciated! :)


 --
 Torgeir Veimo
 [EMAIL PROTECTED]


Hi Torgeir,

I used a lot of information from: -

http://www.vdr-wiki.de/wiki/index.php/OpenSuSE_DVB-S2_-_Step_by_Step_Installationsanleitung_%28Achtung_Beta%29

Essentially: -

1) Download and make install the multiproto tree from
http://jusst.de/hg/multiproto. I had to apply the HVR4000 patch for my
Hauppauge Nova S2
(http://www.linuxtv.org/pipermail/linux-dvb/attachments/20071215/13bb7cb3/attachment-0001.bin).
If you are running a TT 3200 you probably don't need a patch.

2) Download and install latest ffmpeg cvs (see URL above for details
on config options)

3) Download and install latest xine-lib (see URL above for details on
config options)

4) Download and install latest xine-ui (see URL above for details on
config options)

5) Download and make vdr-1.5.12 and patch with Reinhard's patches in this thread

6) Download xine-0.8.1 from http://home.vrweb.de/~rnissl/ and put it
in the plugins directory and make it

7) Setup a channels.conf with some HD channels (here's some I have collected): -

BBCHD;EURO1080:10847:vS0Z0:S28.2E:22000:12327+2327:2329=NAR;2328=eng:2330:0:6940
:0:0:0
Channel 4 HD;EURO1080:11798:hC34M2S1Z35:S28.2E:29500:10513+8190:0;661=eng:0:960,
961:3875:0:0:0
PREMIERE HD,PREM HD;PREMIERE:11914:hC910M2S1Z35:S19.2E:27500:10767+767:0;771=deu
,772=eng:32:1801,1831,1830:129:133:6:0
DISCOVERY HD,DISC HD;PREMIERE:11914:hC910M2S1Z35:S19.2E:27500:11023+1023:0;1027=
deu:32:1801,1831,1830:130:133:6:0
D.Alb HD-1;DigitALB:11094:vS0Z0:S16.0E:27900:2001:2002:0:B00:1004:366:30402:0
D.Alb HD-2;DigitALB:11094:vS0Z0:S16.0E:27900:2021:2022:0:0:1005:366:30402:0
CANAL+ FILM HD;Telenor:11421:hC23M5S1Z35:S1.0W:25000:10513+513:644=eng;645=eng:0
:B00:3306:70:14:0
CANAL+ SPORT HD;Telenor:11421:hC23M5S1Z35:S1.0W:25000:10514+514:648=sve,649=nor:
0:B00:1404:70:14:0
SVT HD;Telenor:11421:hC23M5S1Z35:S1.0W:25000:10512+512:640=sve;641=sve:0:B00:380
1:70:14:0
CANAL+ HI-TECH HD;CSAT:12522:vC23M5S1Z35:S19.2E:22000:10160+160:0;82=fra,83=eng:
0:100:9201:1:1106:0
NATIONAL GEO HD;CSAT:12522:vC23M5S1Z35:S19.2E:22000:10161+161:0;86=fra:0:100:920
2:1:1106:0
TF1 HD;CSAT:12522:vC23M5S1Z35:S19.2E:22000:10163+163:0;94=fra:0:100:9204:1:1106:
0
CINE PREMIER HD;CSAT:12581:vC23M5S1Z35:S19.2E:22000:10160+160:0;82=fra,83=eng:0:
100:9301:1:1110:0
HD suisse;SRG SSR idee suisse:12399:hS0Z0:S13.0E:27500:10180+180:0;131=deu,132=f
ra,133=ita,134=eng:0:500:990:318:8500:0
SKY Sport HD 1;SkyItalia:11996:vC23M5S1Z35:S13.0E:27500:10164+164:0;416=ita,417=
eng:0:919,93B:11020:64511:6400:0
Next HD;SkyItalia:11996:vC23M5S1Z35:S13.0E:27500:10160+160:0;400=ita,401=eng:0:9
19,93B:11030:64511:6400:0
SKY Cinema HD;SkyItalia:11996:vC23M5S1Z35:S13.0E:27500:10161+161:0;404=ita,405=e
ng:0:919,93B:11031:64511:6400:0
NationalGeo HD;SkyItalia:11996:vC23M5S1Z35:S13.0E:27500:10163+163:0;412=ita,413=
eng:0:919,93B:11032:64511:6400:0
SKY Sport HD 2;SkyItalia:11996:vC23M5S1Z35:S13.0E:27500:10165+165:0;420=eng,421=
ita:0:919,93B:11033:64511:6400:0
13EME RUE HD;CSAT:12581:vC23M5S1Z35:S19.2E:22000:10161+161:0;86=fra:0:100:9302:1
:1110:0
DISNEY MAGIC HD;CSAT:12581:vC23M5S1Z35:S19.2E:22000:10162+162:0;90=fra,91=eng:0:
100:9303:1:1110:0
M6 HD;CSAT:12581:vC23M5S1Z35:S19.2E:22000:10170+170:0;122=fra:0:100:9310:1:1110:
0
Discovery HD Europe;Canal+:11434:vC23M5S1Z35:S1.0W:25000:10513+513:0;645:0:B00:3
804:70:38:0
Fox Sports Turkey;DigiTurk:10928:hC23M5S1Z35:S7.0E:3:10201+201:0;301:0:D00,6
64:201:126:21100:0
Lig TV;DigiTurk:10928:hC23M5S1Z35:S7.0E:3:10202+202:0;302:0:D00,664:202:126:
21100:0
ANIXE HD;BetaDigital:12722:hC23M5S1Z35:S19.2E:22000:11023+1023:0;1027=deu:0:0:10
203:1:1119:0
ASTRA HD;BetaDigital:12722:hC23M5S1Z35:S19.2E:22000:10767+767:768=deu:0:0:10202:
1:1119:0
ProSieben HD;ProSiebenSat.1:12722:hC23M5S1Z35:S19.2E:22000:10255+255:0;259=deu:0
:0:10200:1:1119:0
Sat.1 HD;ProSiebenSat.1:12722:hC23M5S1Z35:S19.2E:22000:10511+511:0;515=deu:0:0:1
0201:1:1119:0
Euro1080 HD1;Euro1080:10758:vS0Z0:S23.5E:22000:10308+308:256=eng:0:622,624,100:1
081::3104:0
Euro1080 HD2/5;Euro1080:10758:vS0Z0:S23.5E:22000:10307+307:255=eng;258=deu:1026:
622,624:1::3104:0
EXQI;Euro1080:10758:vS0Z0:S23.5E:22000:10034+34:35=eng:0:622,624,100:2::3104
:0
HD Retail Info;BSkyB:12324:VC34M2S1Z0:S28.2E:29500:10512+8190:640=NAR;660=eng:23
05:960,961:3801:2:2032:0
Discovery HD;BSkyB:12324:VC34M2S1Z0:S28.2E:29500:10514+8190:0;662=eng:0:960,961:
3803:2:2032:0
Sky One HD;BSkyB:12344:HC34M2S1Z0:S28.2E:29500:10512+8190:640=NAR;660=eng:2305:9
60,961:3861:2:2033:0
SkyMovies HD2;BSkyB:12344:HC34M2S1Z0:S28.2E:29500:10513+8190:641=NAR;661=eng:231
2:960,961:3862:2:2033:0
Anytime;BSkyB:12344:HC34M2S1Z0:S28.2E:29500:0:0:0:0:4148:2:2033:0
NVOD:12344:HC34M2S1Z0:S28.2E:29500:0:0:0:0:3894:2:2408:0
NVOD:12344:HC34M2S1Z0:S28.2E:29500:0:0:0:0:3895:2:2408:0
NVOD:12344:HC34M2S1Z0:S28.2E:29500:0:0:0:0:3896:2:2005:0

Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-02 Thread Torgeir Veimo


On 2 Jan 2008, at 21:59, Morfsta wrote:


1) Download and make install the multiproto tree from
http://jusst.de/hg/multiproto. I had to apply the HVR4000 patch for my
Hauppauge Nova S2


Is S2 currently required to get BBC HD or CH4 HD? I've got a dvb-s  
card set up against Sky FTA, but no S2 card atm.


--
Torgeir Veimo
[EMAIL PROTECTED]



___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-02 Thread Iwan Davies
On Wed, 2 Jan 2008 11:59:36 +, in gmane.linux.vdr Morfsta wrote:

 On Jan 2, 2008 11:08 AM, Torgeir Veimo [EMAIL PROTECTED] wrote:
 A howto outlining the steps you've taken to get this to work would be
 appreciated! :)

 
 I used a lot of information from: -
 
 http://www.vdr-wiki.de/wiki/index.php/OpenSuSE_DVB-S2_-_Step_by_Step_Installationsanleitung_%28Achtung_Beta%29
 
 Essentially: -
 
detailed instructions snipped /
 
 Enjoy HD channels!
 
Of course, there's one bit of *vital* information missing from your HOWTO,
at least in regard to enjoying Channel4HD and the various Sky HD channels.
Any explanations would be gratefully received. Please reply offlist, since
I gather discussion of such procedures is frowned on here. 

Thanks!

Iwan Davies

___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-02 Thread Gregoire Favre
On Tue, Jan 01, 2008 at 11:15:58PM +0100, Reinhard Nissl wrote:

Hello :-)

 The patches now include my recently released speedup patches as
 well as an unreleased speedup patch for cAudioRepacker and
 cVideoRepacker, because at least the latter one would have been
 hard to extract separately.

I found those patches tremendous, thank you very much, it works very
well here !!!

Is there a way to use more than a card with it when only one is DVB-S2
capable ?

Thank you very much,
-- 
Grégoire FAVRE  http://gregoire.favre.googlepages.com  http://www.gnupg.org
   http://picasaweb.google.com/Gregoire.Favre

___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-02 Thread Morfsta
 Is there a way to use more than a card with it when only one is DVB-S2
 capable ?

I just dropped a mail to Reinhard asking the same question. My
production system runs 2 DVB-T cards that currently don't work with
vdr-1.5.12 / multiproto. I would love to start using it full time but
until I can get support for DVB-T then it won't be possible.

There's two ways you can do this I believe, port the dvb-t driver
properly to the new multiproto tree or modify VDR to fall back to the
old interface to tune the driver if the new one fails.

I know that it is possible to use the old interface because when I use
vdr-1.4.7 with the multiproto tree it works fine for DVB-T, however I
am not sure where to start coding this into either VDR and I'm not
sure my skills are that hot at the driver level!

If someone could provide some pointers here, that would be great (Manu
or Reinhard?)

Thanks,

Phil

___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-02 Thread Manu Abraham
Morfsta wrote:
 Is there a way to use more than a card with it when only one is DVB-S2
 capable ?
 
 I just dropped a mail to Reinhard asking the same question. My
 production system runs 2 DVB-T cards that currently don't work with
 vdr-1.5.12 / multiproto. I would love to start using it full time but
 until I can get support for DVB-T then it won't be possible.
 
 There's two ways you can do this I believe, port the dvb-t driver
 properly to the new multiproto tree or modify VDR to fall back to the
 old interface to tune the driver if the new one fails.
 
 I know that it is possible to use the old interface because when I use
 vdr-1.4.7 with the multiproto tree it works fine for DVB-T, however I
 am not sure where to start coding this into either VDR and I'm not
 sure my skills are that hot at the driver level!
 
 If someone could provide some pointers here, that would be great (Manu
 or Reinhard?)

Over here, tests were done on STB0899, STV0299 and TDA10021 based
all work out of the same multiproto tree (http://jusst.de.hg/multiproto)

Regards,
Manu


___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-02 Thread Gregoire Favre
On Wed, Jan 02, 2008 at 07:12:52PM +0400, Manu Abraham wrote:

 Over here, tests were done on STB0899, STV0299 and TDA10021 based
 all work out of the same multiproto tree (http://jusst.de.hg/multiproto)

Do you mean VDR keep the DVB-S2 card for DVB-S2 channels and use the
others for DVB-S usage ?

Thanks,
-- 
Grégoire FAVRE  http://gregoire.favre.googlepages.com  http://www.gnupg.org
   http://picasaweb.google.com/Gregoire.Favre

___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-02 Thread Manu Abraham
Gregoire Favre wrote:
 On Wed, Jan 02, 2008 at 07:12:52PM +0400, Manu Abraham wrote:
 
 Over here, tests were done on STB0899, STV0299 and TDA10021 based
 all work out of the same multiproto tree (http://jusst.de.hg/multiproto)
 
 Do you mean VDR keep the DVB-S2 card for DVB-S2 channels and use the
 others for DVB-S usage ?


No, that seems silly. The DVB-S2 card works for DVB-S also. It should be able 
to use all the cards. Though you can't ask a DVB-S card to tune to DVB-S2,
that's all the limitation it poses.

Regards,
Manu


___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-02 Thread Gregoire Favre
On Wed, Jan 02, 2008 at 07:32:13PM +0400, Manu Abraham wrote:

 No, that seems silly. The DVB-S2 card works for DVB-S also. It should be able 
 to use all the cards. Though you can't ask a DVB-S card to tune to DVB-S2,
 that's all the limitation it poses.

:-)

I know DVB-S2 cards can do DVB-S that's the reason to force VDR to use
DVB-S only cards for DVB-S before the DVB-S2 ones, which if you want a
DVB-S2 channels, so is your DVB-S2 cards free.

But one should get a way to tell VDR no to try to tune DVB-S2 channels
on DVB-S card.

Thanks,
-- 
Grégoire FAVRE  http://gregoire.favre.googlepages.com  http://www.gnupg.org
   http://picasaweb.google.com/Gregoire.Favre

___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-02 Thread Morfsta
 Over here, tests were done on STB0899, STV0299 and TDA10021 based
 all work out of the same multiproto tree (http://jusst.de.hg/multiproto)

There is something wrong then as it does not tune in vdr-1.5.12. I
have the following DVB-T frontends: -

[ 9539.157598] DVB: registering frontend 1 (Conexant CX22702 DVB-T)...
[ 9539.197883] DVB: registering frontend 3 (Philips TDA10045H DVB-T)...

Should the TDA10045H work? If so, perhaps the problem is that VDR is
tuning frontend 1 first which is not supported yet.

How can I go about porting the Conexant to the new interface? Is it
straightforward (i.e. can I compare the code for the TDA10021 based
frontend in v4l-dvb and multiproto and make the changes for the
cx22702 frontend?)

Thanks

___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-02 Thread Manu Abraham
Gregoire Favre wrote:
 On Wed, Jan 02, 2008 at 07:32:13PM +0400, Manu Abraham wrote:
 
 No, that seems silly. The DVB-S2 card works for DVB-S also. It should be 
 able 
 to use all the cards. Though you can't ask a DVB-S card to tune to DVB-S2,
 that's all the limitation it poses.
 
 :-)
 
 I know DVB-S2 cards can do DVB-S that's the reason to force VDR to use
 DVB-S only cards for DVB-S before the DVB-S2 ones, which if you want a
 DVB-S2 channels, so is your DVB-S2 cards free.
 
 But one should get a way to tell VDR no to try to tune DVB-S2 channels
 on DVB-S card.

You can do that. In fact the application should do a DVBFE_GET_INFO. This will 
query
the capability of that specific driver. Based on the driver capabilities, issue 
a tuning
request. That way a tune will be attempted only on the capable device.

It is a simple check, though.

Reinhard,

The patched version doesn't look at the device capability flags, before issuing 
a tune ?
Or is something else wrong/missing ?

Regards,
Manu


___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-02 Thread Manu Abraham
Morfsta wrote:
 Over here, tests were done on STB0899, STV0299 and TDA10021 based
 all work out of the same multiproto tree (http://jusst.de.hg/multiproto)
 
 There is something wrong then as it does not tune in vdr-1.5.12. I
 have the following DVB-T frontends: -
 
 [ 9539.157598] DVB: registering frontend 1 (Conexant CX22702 DVB-T)...
 [ 9539.197883] DVB: registering frontend 3 (Philips TDA10045H DVB-T)...
 
 Should the TDA10045H work? If so, perhaps the problem is that VDR is
 tuning frontend 1 first which is not supported yet.

If the driver what you have in the multiproto tree, if you see it as old, 
ie: changes have gone into those drivers in between. The easiest thing to
get going is:

Get the latest tree or whichever for which that demodulator/card is working 
with. Just overwrite the relevant older files for your driver/card in the 
multiproto tree (your local copy) That way, the devices which have been
newly updated also will work.

 
 How can I go about porting the Conexant to the new interface? Is it
 straightforward (i.e. can I compare the code for the TDA10021 based
 frontend in v4l-dvb and multiproto and make the changes for the
 cx22702 frontend?)
 

No need to do any porting. The same drivers which worked with the old API
will work with the new API too, just that there is a thin translation layer in 
between. If you need the shortest call to those drivers, then you will need 
to port them to the new API. Till then, you can go ahead with copying the
old drivers to the new API tree.

Regards,
Manu


___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-02 Thread Reinhard Nissl
Hi,

Manu Abraham schrieb:

 I know DVB-S2 cards can do DVB-S that's the reason to force VDR to use
 DVB-S only cards for DVB-S before the DVB-S2 ones, which if you want a
 DVB-S2 channels, so is your DVB-S2 cards free.

 But one should get a way to tell VDR no to try to tune DVB-S2 channels
 on DVB-S card.
 
 You can do that. In fact the application should do a DVBFE_GET_INFO. This 
 will query
 the capability of that specific driver. Based on the driver capabilities, 
 issue a tuning
 request. That way a tune will be attempted only on the capable device.
 
 It is a simple check, though.
 
 Reinhard,
 
 The patched version doesn't look at the device capability flags, before 
 issuing a tune ?
 Or is something else wrong/missing ?

I must admit, I didn't have a look at the source code so far, but
from what I recall, I don't think that it is that easy. I think,
a simple check will only make VDR tell you that the channel in
question cannot be received.

I've experienced myself that VDR used the DVB-S2 card for a DVB-S
recording and then endlessly tried to switch to a DVB-S2 channel
using the DVB-S card -- obviously without success.

VDR's device selection logic is already quite complex. It has to
deal with FF cards which have so far been used to watch the DVB-S
channels, cards which provide CI interfaces and simple receiver
cards. Moreover it now would have to deal with cards capable of
doing DVB-S2, with or without CI interface respectively.

As I'm only distributing the DVB-S2 part which Marco Schlüßler
provided several months ago, I've contacted him and asked him for
an update.

Bye.
-- 
Dipl.-Inform. (FH) Reinhard Nissl
mailto:[EMAIL PROTECTED]

___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-02 Thread Manu Abraham
Reinhard Nissl wrote:
 Hi,
 
 Manu Abraham schrieb:
 
 I know DVB-S2 cards can do DVB-S that's the reason to force VDR to use
 DVB-S only cards for DVB-S before the DVB-S2 ones, which if you want a
 DVB-S2 channels, so is your DVB-S2 cards free.

 But one should get a way to tell VDR no to try to tune DVB-S2 channels
 on DVB-S card.
 You can do that. In fact the application should do a DVBFE_GET_INFO. This 
 will query
 the capability of that specific driver. Based on the driver capabilities, 
 issue a tuning
 request. That way a tune will be attempted only on the capable device.

 It is a simple check, though.

 Reinhard,

 The patched version doesn't look at the device capability flags, before 
 issuing a tune ?
 Or is something else wrong/missing ?
 
 I must admit, I didn't have a look at the source code so far, but
 from what I recall, I don't think that it is that easy. I think,
 a simple check will only make VDR tell you that the channel in
 question cannot be received.
 
 I've experienced myself that VDR used the DVB-S2 card for a DVB-S
 recording and then endlessly tried to switch to a DVB-S2 channel
 using the DVB-S card -- obviously without success.

How can you ask a DVB-S device to tune to DVB-S2 ? Poor demodulator,
it has to do what it is not even capable of. :)

 VDR's device selection logic is already quite complex. It has to
 deal with FF cards which have so far been used to watch the DVB-S
 channels, cards which provide CI interfaces and simple receiver
 cards. Moreover it now would have to deal with cards capable of
 doing DVB-S2, with or without CI interface respectively.

The logic wouldn't be much different. It is the same as requesting a DVB-C 
device to be tuned to DVB-S. The CI interface doesn't make the hardware
look any different. 
 
 As I'm only distributing the DVB-S2 part which Marco Schlüßler
 provided several months ago, I've contacted him and asked him for
 an update.

Ok, I will ask Marco on the relevant. Thanks for the feedback.

Regards,
Manu

___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-02 Thread Igor Nikanov
 I think this might have been covered before, but the only problem I
 now see is with interlaced and spatial direct mode (particularly on
 the French and Polish HD channels): -
 
 [h264 @ 0x2d0b0700]Interlaced pictures + spatial direct mode is
 not implemented
 [h264 @ 0x2d0b0700]Interlaced pictures + spatial direct mode is
 not implemented


ah, it's known problem with FFmpeg. Unfortunately in FFmpeg this feature is not 
implement.
Nobody knows when FFmpeg will support spatial direct mode  for Interlaced 
pictures 

Igor

___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2008-01-02 Thread Reinhard Nissl
Hi,

Manu Abraham schrieb:

 The patched version doesn't look at the device capability flags, before 
 issuing a tune ?
 Or is something else wrong/missing ?

 I must admit, I didn't have a look at the source code so far, but
 from what I recall, I don't think that it is that easy. I think,
 a simple check will only make VDR tell you that the channel in
 question cannot be received.

 I've experienced myself that VDR used the DVB-S2 card for a DVB-S
 recording and then endlessly tried to switch to a DVB-S2 channel
 using the DVB-S card -- obviously without success.
 
 How can you ask a DVB-S device to tune to DVB-S2 ? Poor demodulator,
 it has to do what it is not even capable of. :)
 
 VDR's device selection logic is already quite complex. It has to
 deal with FF cards which have so far been used to watch the DVB-S
 channels, cards which provide CI interfaces and simple receiver
 cards. Moreover it now would have to deal with cards capable of
 doing DVB-S2, with or without CI interface respectively.
 
 The logic wouldn't be much different. It is the same as requesting a DVB-C 
 device to be tuned to DVB-S. The CI interface doesn't make the hardware
 look any different. 
  
 As I'm only distributing the DVB-S2 part which Marco Schlüßler
 provided several months ago, I've contacted him and asked him for
 an update.
 
 Ok, I will ask Marco on the relevant. Thanks for the feedback.

Well, I was in contact with Marco already and attached you'll
find a minimalistic change which reports channel not available.
Now VDR should already be able to kick a low priority DVB-S
recording (or transfer thread) from a DVB-S2 device.

Still missing is to prefer DVB-S devices for DVB-S recordings so
that DVB-S2 devices remain available for DVB-S2 recordings of
same priority.

The patch is incremental to the original dvbs2 patch from
yesterday, i. e. you can simply apply it to your already patched VDR.

Bye.
-- 
Dipl.-Inform. (FH) Reinhard Nissl
mailto:[EMAIL PROTECTED]
--- ../vdr-1.5.12-dvbs2-other/dvbdevice.c	2008-01-01 22:55:18.0 +0100
+++ dvbdevice.c	2008-01-02 22:25:49.0 +0100
@@ -798,7 +798,13 @@ bool cDvbDevice::ProvidesSource(int Sour
 
 bool cDvbDevice::ProvidesTransponder(const cChannel *Channel) const
 {
-  return ProvidesSource(Channel-Source())  (!cSource::IsSat(Channel-Source()) || !Setup.DiSEqC || Diseqcs.Get(Channel-Source(), Channel-Frequency(), Channel-Polarization()));
+  if (!ProvidesSource(Channel-Source()))
+ return false; // doesn't provide source
+  if (!cSource::IsSat(Channel-Source()))
+ return true; // source is sufficient
+  if (Channel-ModulationSystem()  !(frontendType  DVBFE_DELSYS_DVBS2))
+ return false; // requires DVB-S2, but device doesn't provide it
+  return !Setup.DiSEqC || Diseqcs.Get(Channel-Source(), Channel-Frequency(), Channel-Polarization());
 }
 
 bool cDvbDevice::ProvidesChannel(const cChannel *Channel, int Priority, bool *NeedsDetachReceivers) const
@@ -807,7 +813,7 @@ bool cDvbDevice::ProvidesChannel(const c
   bool hasPriority = Priority  0 || Priority  this-Priority();
   bool needsDetachReceivers = false;
 
-  if (ProvidesSource(Channel-Source())) {
+  if (ProvidesTransponder(Channel)) {
  result = hasPriority;
  if (Priority = 0  Receiving(true)) {
 if (dvbTuner-IsTunedTo(Channel)) {
___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2007-11-25 Thread ShorTie
Trying to get this h.264 stuff going and
vdr-1.5.12-h264-syncearly-framespersec-audioindexer-fielddetection.diff and
vdr-1.5.10-genpix-usb.diff aren't playing nice during compile .. :(~

dvbdevice.c: In member function 'bool cDvbTuner::SetFrontend()':
dvbdevice.c:253: error: jump to case label
dvbdevice.c:184: error:   crosses initialization of 'unsigned int frequency'
dvbdevice.c:267: error: jump to case label
dvbdevice.c:184: error:   crosses initialization of 'unsigned int frequency'
dvbdevice.c:285: error: jump to case label
dvbdevice.c:184: error:   crosses initialization of 'unsigned int frequency'
make: *** [dvbdevice.o] Error 1

me lost, lol.
Thankz


No virus found in this outgoing message.
Checked by AVG Free Edition. 
Version: 7.5.503 / Virus Database: 269.16.6/1150 - Release Date: 11/24/2007
5:58 PM
 
  


vdr-1.5.10-genpix-usb.diff
Description: Binary data
___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [ANNOUNCE] DVB-S2 + H.264 support for VDR-1.5.12

2007-11-25 Thread Reinhard Nissl
Hi,

ShorTie schrieb:

 Trying to get this h.264 stuff going and
 vdr-1.5.12-h264-syncearly-framespersec-audioindexer-fielddetection.diff and
 vdr-1.5.10-genpix-usb.diff aren't playing nice during compile .. :(~
 
 dvbdevice.c: In member function 'bool cDvbTuner::SetFrontend()':
 dvbdevice.c:253: error: jump to case label
 dvbdevice.c:184: error:   crosses initialization of 'unsigned int frequency'
 dvbdevice.c:267: error: jump to case label
 dvbdevice.c:184: error:   crosses initialization of 'unsigned int frequency'
 dvbdevice.c:285: error: jump to case label
 dvbdevice.c:184: error:   crosses initialization of 'unsigned int frequency'
 make: *** [dvbdevice.o] Error 1

This is not a matter of applying both patches -- only
vdr-1.5.10-genpix-usb.diff is wrong. The error is the result of changes
like that:

@@ -173,13 +173,14 @@

 bool cDvbTuner::SetFrontend(void)
 {
-  dvb_frontend_parameters Frontend;
+  dvb_frontend_parameters_new Frontend;

   memset(Frontend, 0, sizeof(Frontend));

   switch (frontendType) {
-case FE_QPSK: { // DVB-S
-
+case FE_QPSK:  // DVB-S
+case FE_DVB_S:  // DVB-S
+case FE_DVB_S2: // DVB-S
  unsigned int frequency = channel.Frequency();

  if (Setup.DiSEqC) {

See that the variable frequency a few lines above is declared within a
switch statement which requires to use a separate block (= a pair of
curly braces) for this declaration or you will get the above error. The
original code correctly opened a curly brace after case FE_QPSK: while
the new one doesn't after case FE_DVB_S2:.

BTW: this is the matching curly brace that was removed from the orignal
code:

  tuneTimeout = DVBS_TUNE_TIMEOUT;
  lockTimeout = DVBS_LOCK_TIMEOUT;
- }
  break;
 case FE_QAM: { // DVB-C

@@ -280,10 +286,17 @@

Try to add those removed curly braces and give it a try. Don't know
whether it will work correctly afterwards, but it should compile (though
not tested).

Bye.
-- 
Dipl.-Inform. (FH) Reinhard Nissl
mailto:[EMAIL PROTECTED]

___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr