Messages

As an alternative to raspberryPi, maybe this one instead?
http://www.cnx-software.com/2012/04/04/mele-a2000-android-2-3-media-player-powered-by-allwinner-a10/

At least 1.2 GHz processor freq. ! Maybe makes software decoding of mpeg possible?

approx $75 at Aliexpress: http://www.aliexpress.com/product-fm/560499953-New-Arrive-Mele-A1000-TV-box-Allwinner-A10-hackable-device-Network-Intelligent-drive-HD-player-wholesalers.html

/Anders

Hi,

For maximum versatility and simplicity, I bought the OpenRD client (http://www.globalscaletechnologies.com/p-24-openrd-client-openrd-client-board-with-enclosure.aspx)
I have connected the TV receiver and the hard disk to the USB connectors. I will move the disk to eSATA later.
The OS is debian Linux. No otimization is performed with regards, to performance.
There's only 18 MB left of the 512 MB RA. 9% of the memory is used by vdr-kbd. What's it used for?
The load when recording one channel, and viewing another one, is approximately 0.9 (as shown by top)

BR
Anders

Hi,

I'm running on the same Marvell SoC, like the sheevaplug, except that I have some extra hardware like eSATA, VGA port, etc.
My previous i86 server broke so I had to find another one. I'm tired of noisy 150 Watt monsters so I bought this Marvell based one instead.
6 Watt + 12 for the Disk, no problem running  24x7.

Oh, and yes, you need to apply the patches that I posted on the thread referred to in a previous post in this thread.

Regards
Anders

Patch below:

anders@server1:/mnt/disk1/src/vdr-1.6.0/PLUGINS/src$ cat vompserver.patch
diff -Naur vompserver.org/responsepacket.c vompserver/responsepacket.c
--- vompserver.org/responsepacket.c   2009-11-20 05:34:28.075696779 +0000
+++ vompserver/responsepacket.c   2009-11-22 06:32:37.415696613 +0000
@@ -87,7 +87,8 @@
bool ResponsePacket::addULONG(ULONG ul)
{
   if (!checkExtend(sizeof(ULONG))) return false;
-  *(ULONG*)&buffer[bufUsed] = htonl(ul);
+  ULONG ulnw = htonl(ul);
+  memcpy(buffer + bufUsed, &ulnw, sizeof(ULONG));
   bufUsed += sizeof(ULONG);
   return true;
} 
@@ -103,7 +104,8 @@
bool ResponsePacket::addLONG(LONG l)
{
   if (!checkExtend(sizeof(LONG))) return false;
-  *(LONG*)&buffer[bufUsed] = htonl(l);
+  LONG lnw = htonl(l);
+  memcpy(buffer + bufUsed, &lnw, sizeof(LONG));
   bufUsed += sizeof(LONG);
   return true;
}
@@ -111,7 +113,8 @@
bool ResponsePacket::addULLONG(ULLONG ull)
{
   if (!checkExtend(sizeof(ULLONG))) return false;
-  *(ULLONG*)&buffer[bufUsed] = htonll(ull);
+  ULLONG ullnw = htonll(ull);
+  memcpy(buffer + bufUsed, &ullnw, sizeof(ULLONG));
   bufUsed += sizeof(ULLONG);
   return true;
}

I found a solution to my problem!
An analysis  of the response from the server, showed that the last character and NULL of a string were overwritten by the following ULONG.
Locking at addULONG() of responsepacket.c it does:

*(ULONG*)&buffer[bufUsed]=htonl(ul);

replacing the clause with :

ULONG ulnw = htonl(ul);
memcpy(buffer+bufUsed, &ulnw, sizeof(ULONG) );

solved the problem

I guess that the left hand side casting of the first clause, for alignment reasons, is evaluated differently depending on processor arcitecture.
The addLONG() and addULLONG() need to modified in a similar way.

Would it be possible to get CVS updated with a patch?

BR
Anders

Hi,
I recently deployed a new server running debian. It's an arm5TE based openrd client box.
Unfortunately the MVP and the Windows clients don't play very well with it.

The MVP client, for example, crashes when reading the recordings list after the printouts below:
I can se that there are htonl/ntohl in the client and server code, but I would like to know wther anyone has run the server on anything else than i86 before?

0:55:31.300471 [debug]  32 Remote - Button 21
10:55:31.311836 [debug]  32 BoxStack - Update called
10:55:31.312646 [debug]  32 BoxStack - Locked for update
10:55:31.339986 [debug]  32 BoxStack - Unlocked for update
10:55:31.754882 [debug]  32 Remote - Button 21
10:55:31.767367 [debug]  32 BoxStack - Update called
10:55:31.768219 [debug]  32 BoxStack - Locked for update
10:55:31.794675 [debug]  32 BoxStack - Unlocked for update
10:55:36.756600 [debug]  32 Remote - Button 37
10:55:36.757496 [debug]  32 Boxx - Construct, now 5
10:55:36.758289 [debug]  32 Boxx - Construct, now 6
10:55:36.797757 [debug]  32 TBBoxx - Draw
10:55:36.869361 [debug]  32 BoxStack - add called
10:55:36.870160 [debug]  32 BoxStack - Locked for add
10:55:36.870925 [debug]  32 BoxStack - Unlocked for add
10:55:36.871686 [debug]  32 BoxStack - Update called
10:55:36.872446 [debug]  32 BoxStack - Locked for update
10:55:36.957161 [debug]  32 BoxStack - Unlocked for update
10:55:36.958224 [debug]  32 VDR - RR sleep - opcode 2
10:55:36.961141 [debug]  37 VDR - Rxd a response packet, requestID=15, len=248
10:55:36.962242 [debug]  32 VDR - RR unsleep

BR
Anders

Do you have any hints for me? Especially the possibility to replay dvd iso files would be nice to have...

Have you installed lsdvd?

To enable mediaplayer on ubuntu 9.04 I had to do:

sudo apt-get install lsdvd
sudo apt-get install mencoder
sudo apt-get install lame
sudo apt-get install ffmpeg

I can play ISO's, but not avi's though.

/Anders

Hi again,

I've tried out some values that seem to work reasonably good.

if (Video::getInstance()->getFormat() == Video::PAL)
    setPosition(40, 31);

and

       } else if (firstlineupdate==30) {

For the subtitle It works like this for SVT2 of Sweden:

The green button gives  two choices, for example 691, and 8ff. When selecting one of them there is no subtitle shown and the there will likely be a hangup, like previously for teletext with "firstlineupdate==10".
However, for sub-title I'm used to choose teletext and page 299 for SVT. This works for the Vomp as well, so one can say that sub-title works to some extent.


When the hangup occurs there seems to be a busy loop with the printouts below. This is similar to the printout during the hangup that occurs when "firstlineupdate==10"

/Anders

21:43:42.901128 [debug]  1002 Core - SIGURG caught
21:43:42.924003 [debug]  1002 BoxStack - Update called
21:43:42.924819 [debug]  1002 BoxStack - Locked for update
21:43:42.928701 [debug]  1002 BoxStack - Unlocked for update
21:43:42.929506 [debug]  1002 Command - processing message 32
21:43:42.930285 [debug]  1002 Command - Sending message to boxstack
21:43:42.931061 [debug]  1002 BoxStack - sending message from box 0x102517c8 to box 0x10271a68 32
21:43:42.983941 [debug]  1016 Command - PMFOS called
21:43:42.984837 [debug]  1016 Command - PMFOS called
21:43:42.985934 [debug]  1016 Command - PMFOS called
21:43:42.986913 [debug]  1016 Command - PMFOS called
21:43:42.987757 [notice] 1002 Core - Signal 23 received
21:43:42.988693 [debug]  1002 Core - SIGURG caught
21:43:43.077008 [debug]  1016 Command - PMFOS called
21:43:43.078164 [notice] 1002 Core - Signal 23 received
21:43:43.078976 [debug]  1002 Core - SIGURG caught
21:43:43.163042 [debug]  1016 Command - PMFOS called
21:43:43.163946 [debug]  1016 Command - PMFOS called
21:43:43.168788 [notice] 1002 Core - Signal 23 received
21:43:43.169725 [debug]  1002 Core - SIGURG caught
21:43:43.194140 [debug]  1016 Command - PMFOS called
21:43:43.195174 [notice] 1002 Core - Signal 23 received
21:43:43.196110 [debug]  1002 Core - SIGURG caught
21:43:43.284060 [debug]  1016 Command - PMFOS called
21:43:43.285081 [debug]  1016 Command - PMFOS called
21:43:43.285639 [debug]  1016 Command - PMFOS called
21:43:43.286472 [debug]  1016 Command - PMFOS called
21:43:43.287320 [notice] 1002 Core - Signal 23 received
21:43:43.288256 [debug]  1002 Core - SIGURG caught
21:43:43.373659 [debug]  1016 Command - PMFOS called
21:43:43.374805 [debug]  1016 Command - PMFOS called
21:43:43.375829 [notice] 1002 Core - Signal 23 received
21:43:43.376635 [debug]  1002 Core - SIGURG caught
21:43:43.584509 [debug]  1016 Command - PMFOS called
21:43:43.584993 [debug]  1016 Command - PMFOS called
21:43:43.585820 [debug]  1016 Command - PMFOS called
21:43:43.586911 [debug]  1016 Command - PMFOS called
21:43:43.587873 [notice] 1002 Core - Signal 23 received

A higher value works better. I've tried 17, 20 and 22. I've got a hangup at 17. 22 seem to work at least as good as 20.
I've seen occasional flickering screens.
The text at the top and right of the screen is slightly cut.

Otherwise it works OK.

/Anders

As far as I can tell running makedevenv again recompiles a new dongle. (the modified .cc file is still there and a new .o file is created.)
I can see no change for neither the ==5 or ==1 comparison.

Anders

Hi,

I believe I've tried firstlineupdate==5, but I'm not sure how to re-compile an build a new dongle!
I used makedevenv-5 to do initial build, and after patching. It seems do download and recompile everything.

How can I patch and rebuild dongle without the need to download and build from scratch?

/Anders

Sorry, but the green button works similar to the yellow one. After some time the options are shown. When one of them is selected, it takes some time before the subtitle is shown.
Then, it is impossible to change anything with the remote.
This is for SVT4.

For TV4, as with the telextext, the responsiveness is higher, but it is slow. The teletext mode can be exited.

In addition, when in Teletext mode, how can I step pages. The "up", "left", etc. show the program info.

/Anders

I finally managed to compile from CVS.
As Sirwio I'm located in Sweden. In contrast to his result, SVT2 did work out rather well, at first. After having accessed 1 or to pages it stopped responding to button presses.
I logged in to the box and started vompclient -d. This time there were no response at all, after the first page was presented. Unfortunately the log is cluttered with thousands of:

Going to select
Back from select with success = 1

I filtered those. The log was still to large to be sent to this forum.
If you believe the log might be of value, tell me where to put it!
Is there anything else I can do provide information, please let me know!

It would be very much appreciated if teletext was working, since then we don't need the switch between the MVP and the DVB receiver anymore!

Regards
Anders

It turns out that the teletext is not frozen, but is updated with changes over time. The problems are: 1) I don't think I see the upper rows of the text because the page number is not shown.
I can't change page number. In fact I can't send any input to the MVP.

I tried makedevenv-5 and -6 but both fail with:

powerpc-405-linux-gnu-gcc  version.c -c -O -Wall -Winline -Wstrict-prototypes -Wwrite-strings -fno-unit-at-a-time -mcpu=405 -mnew-mnemonics      -I../include -I. -I/home/anders/src/vdr-vomp/crosstool/crosstool-0.43/build/powerpc-405-linux-gnu/gcc-3.4.5-glibc-2.2.5/build-glibc/csu -I.. -I../libio  -I/home/anders/src/vdr-vomp/crosstool/crosstool-0.43/build/powerpc-405-linux-gnu/gcc-3.4.5-glibc-2.2.5/build-glibc -I../sysdeps/powerpc/elf -I../linuxthreads/sysdeps/unix/sysv/linux -I../linuxthreads/sysdeps/pthread -I../sysdeps/pthread -I../linuxthreads/sysdeps/unix/sysv -I../linuxthreads/sysdeps/unix -I../linuxthreads/sysdeps/powerpc -I../sysdeps/unix/sysv/linux/powerpc -I../sysdeps/unix/sysv/linux -I../sysdeps/gnu -I../sysdeps/unix/common -I../sysdeps/unix/mman -I../sysdeps/unix/inet -I../sysdeps/unix/sysv -I../sysdeps/unix/powerpc -I../sysdeps/unix -I../sysdeps/posix -I../sysdeps/powerpc -I../sysdeps/wordsize-32 -I../sysdeps/ieee754/flt-32 -I../sysdeps/ieee754/dbl-64 -I../sysdeps/powerpc/soft-fp -I../sysdeps/ieee754 -I../sysdeps/generic/elf -I../sysdeps/generic  -nostdinc -isystem /home/anders/src/vdr-vomp/crosstool/crosstool-0.43/build/powerpc-405-linux-gnu/gcc-3.4.5-glibc-2.2.5/gcc-core-prefix/lib/gcc/powerpc-405-linux-gnu/3.4.5/include -isystem /home/anders/src/vdr-vomp/crosstool/gcc-3.4.5-glibc-2.2.5/powerpc-405-linux-gnu/powerpc-405-linux-gnu/include -D_LIBC_REENTRANT -include ../include/libc-symbols.h     -DHAVE_INITFINI -o /home/anders/src/vdr-vomp/crosstool/crosstool-0.43/build/powerpc-405-linux-gnu/gcc-3.4.5-glibc-2.2.5/build-glibc/csu/version.o
In file included from version.c:32:
/home/anders/src/vdr-vomp/crosstool/crosstool-0.43/build/powerpc-405-linux-gnu/gcc-3.4.5-glibc-2.2.5/build-glibc/csu/version-info.h:1: error: missing terminating " character
/home/anders/src/vdr-vomp/crosstool/crosstool-0.43/build/powerpc-405-linux-gnu/gcc-3.4.5-glibc-2.2.5/build-glibc/csu/version-info.h:2: error: missing terminating " character
/home/anders/src/vdr-vomp/crosstool/crosstool-0.43/build/powerpc-405-linux-gnu/gcc-3.4.5-glibc-2.2.5/build-glibc/csu/version-info.h:3: error: missing terminating " character
/home/anders/src/vdr-vomp/crosstool/crosstool-0.43/build/powerpc-405-linux-gnu/gcc-3.4.5-glibc-2.2.5/build-glibc/csu/version-info.h:4: error: missing terminating " character
version.c:36: error: syntax error before string constant
make[2]: *** [/home/anders/src/vdr-vomp/crosstool/crosstool-0.43/build/powerpc-405-linux-gnu/gcc-3.4.5-glibc-2.2.5/build-glibc/csu/version.o] Error 1
make[2]: Leaving directory `/home/anders/src/vdr-vomp/crosstool/crosstool-0.43/build/powerpc-405-linux-gnu/gcc-3.4.5-glibc-2.2.5/glibc-2.2.5/csu'
make[1]: *** [csu/subdir_lib] Error 2
make[1]: Leaving directory `/home/anders/src/vdr-vomp/crosstool/crosstool-0.43/build/powerpc-405-linux-gnu/gcc-3.4.5-glibc-2.2.5/glibc-2.2.5'
make: *** [all] Error 2

/Anders

Hi
I can't get Teletext working correctly.
Having pressed the yellow button nothing happends, at first. After quite some time (a number of seconds) the first Teletext page is shown.
After this, there is no response to any button press.
I have to hard reset the MVP. The teletext page looks quite OK, but can't find the page number.
Any Ideas? Any debug switches that I can turn on to get more info? I see nothing in the messages file.

Ubuntu 8.04
VDR 1.6.0-1ubuntu2
Vdr-vompserver 0.3.1
Vomp 0.3.1
I'm located in Sweden
Got 2 tuners

/Anders