vdr

vdr@linuxtv.org

2 participants
6731 discussions

PCI GT610 vdpau card - 512 MB or 1GB memory ?
by Goga777 08 Dec '13

08 Dec '13

Hello is there any sense to buy PCI GT610 card with 1 GB memory for processing of 1080i/p HD video ? may be 512 MB will be enough ? I plan to use vdpau decoding of HD video + temporal spatial deinterlacing + upscaling of SD video. Goga

2 3

VDR user command ?
by Torsten Mohr 08 Dec '13

08 Dec '13

Hello, i use the xine lib output plugin to run VDR in background and connect to it. I configured VDRChannelPlus and VDRChannelMinus on Ctrl- Up and -Down. But i'd also like them to show the program menu, additionally. It seems not possible that way as the keymap is layed out command => key key => list of commands would be better, then i could maybe add some more commands. Is it maybe possible with VDR user commands to define a sequence of actions? Basically what i want is to have two keys that switch the channels up/down and also display the program menu (what is on on this channel). Best regards Torsten

1 0

Change vdr shutdown command (into pm-suspend)
by cedric.dewijs＠telfort.nl 06 Dec '13

06 Dec '13

Hi All, I'm trying to make my TV server faster by calling pm-suspend instead of shutdown. I have made a script that stops vdr, unloads the DVB-t driver, suspend, resume, load the DVB-t driver, and restart VDR. This works nicely., see here for details: http://forums.debian.net/viewtopic.php?f=5&t=109612 Now I'm trying to automate it. VDR is capable of shutting down the PC. This can be setup in /etc/default/vdr Is there any way to let VDR call pm-suspend instead of /sbin/shutdown? Where is the setting for it? Best regards, Cedric

3 2

start VDR without hardware for test and tutorial
by cedric.dewijs＠telfort.nl 03 Dec '13

03 Dec '13

Hi All, I'm writing a tutorial on VDR for debian. To make the individual steps as small as possible, I would like to begin without any special hardware, just a normal PC. It turns out VDR refuses to run when no DVB card is found. Is there a configuration option somewhere to let VDR run without hardware? Can I create a dummy DVB device somehow? The guide (unfinished) is here: https://wiki.debian.org/VDR Best regards, Cedric My versions: VDR 1.7.28 Debian 7.2.0 Stable (installed today) The error in /var/log/syslog ... No DVB device found ERROR: invalid primary device number: 1 ERROR: no primary device found = using first device! ERROR: invalid primary device number: 1 ... exiting, exit code 2 runvdr: stopping after fatal fail (vdr: no primary device found - using first device!)

4 6

Really stupid question about Full-Function DVB Cards
by Brian-Imap 02 Dec '13

02 Dec '13

Hi, dont kill me please.............. I'm using a 4 Tuner Cine 2 DVB S2 setup for recording TV programs. For playback I was using a DVB-C FF Card, so this was not being used to record as there were no Cable channel entries defined and no cable attached to it. Worked perfectly, but now the DVB C card has broken. So I need a replacement. But I have multiple FF DVB-S cards laid around here not in use. As the dvbttpci driver is only used for this one card, wouldn't it be possible to patch the driver so that it reports the DVB-S card as a DVB-C card? Maybe even by an optional parameter? That way I wouldn't need to buy a new DVB-C card. Stupid idea or not? Cheers Brian

3 7

aufs instead of mhddfs with vdr 2.1.2
by Mike Constabel 02 Dec '13

02 Dec '13

Hi Klaus, i tried mhddfs since I've 5 disks with together 8 TB. The problem is that mhddfs didn't run reliable for me. I tried Debian Wheezy with kernel 3.2, 3.6, 3.10, but every time after starting an recording I got vdr: [19801] ERROR (tools.c,423): /video1: Der Socket ist nicht verbunden/var/log/syslog.6.gz:Nov 24 19:54:38 video1 vdr: [20399] ERROR (recording.c,856): /video1/Spielfilm/Romantik~Liebe/Ein_verlockendes_Spiel/2011-12-11.20.05.8-0.rec/info: Der Socket ist nicht verbunden vdr: [20399] ERROR (recording.c,914): /video1/Spielfilm/Romantik~Liebe/Ein_verlockendes_Spiel/2011-12-11.20.05.8-0.rec/summary.vdr: Der Socket ist nicht verbunden vdr: [20399] ERROR (tools.c,613): /video1/Spielfilm/Romantik~Liebe/Ein_verlockendes_Spiel/2011-12-11.20.05.8-0.rec: Der Socket ist nicht verbunden vdr: [20431] ERROR (recorder.c,159): /video1/Doku/Natur/Terra_X:_Faszination_Erde_-_mit_Dirk_Steffens/Teil_:_Wildes_Mittelmeer_-_Wiege_Europas/2013-11-24.19.28.51-0.rec/00001.ts: Der Socket ist nicht verbunden vdr: [20397] ERROR (tools.c,423): /video1: Der Socket ist nicht verbunden So I tried aufs. This works nearly great for me. mount -t aufs -o br=/video1.1=rw:/video1.2=rw:/video1.3=rw:/video1.4=rw:/video1.0=rw -o create=mfsrr:$((10*1024*1024*1024)):30 -o udba=reval,sum,verbose none /video1 But now I cannot delete recordings from within vdr. I found something about this: http://aufs.sourceforge.net/aufs.html#Incompatible%20with%20an%20Ordinary%2… To rename(2) directory may return EXDEV even if both of src and tgt are on the same aufs. When the rename-src dir exists on multiple branches and the lower dir has child(ren), aufs has to copyup all his children. It can be recursive copyup. Current aufs does not support such huge copyup operation at one time in kernel space, instead produces a warning and returns EXDEV. Generally, mv(1) detects this error and tries mkdir(2) and rename(2) or copy/unlink recursively. So the result is harmless. If your application which issues rename(2) for a directory does not support EXDEV, it will not work on aufs. Also this specification is applied to the case when the src directory exists on the lower readonly branch and it has child(ren). Is it possible that vdr supports aufs? aufs would be better than mhddfs: faster, not fuse, in kernel, under development... Regards, Mike

6 6

vdr-sxfe does not react to keyboard in debian, works in arch linux
by cedric.dewijs＠telfort.nl 01 Dec '13

01 Dec '13

Hi All, I have a fully functional VDR server running under arch linux. I'm trying to migrade to debian. I have installed VDR and vdr-plugin-xineliboutput in Debian , and I can watch nederland 1. So far so good. The problem is that vdr-sxfe does not respond to keypresses. When I press "m" I don't get the menu, when I press "2" I don't switch to nederland 2 and so on. I have looked true the configuration files on my arch linux box, but I could not find the option to let vdr-sxfe listen to the keyboard. On the arch linux box I run a newer version of vdr (1.7.31), then on debian (1.7.28). What configuration file holds the keyboard bindings for vdr-sxfe? Best regards, Cedric

2 1

DNS-323 NAS with 64MB ram, can this run VDR?
by cedric.dewijs＠telfort.nl 18 Nov '13

18 Nov '13

Hi All, I have been given a D-link DNS-323 NAS. This machine runs linux, but only has 64MB RAM. I have tested on my x86 machine. VDR uses 28MB RAM while idle, and this rises to 48MB while decrypting and recording one SD channel. Is it possible to run VDR on this machine? How can I keep the RAM usage low? Best regards, Cedric

2 1

Valgrind warnings with LCARS OSD
by Marko Mäkelä 16 Nov '13

16 Nov '13

Hi all, I played a little with Valgrind: valgrind --vgdb=yes --vgdb-error=0 ./vdr ... and in gdb, (gdb) target remote |vgdb This seems to give me a false alarm for cRecording::cRecording(): Program received signal SIGTRAP, Trace/breakpoint trap. 0x0811951a in cRecording::cRecording (this=0xbee4d958, FileName=0x0) at recording.c:801 801 if (*(fileName + strlen(fileName) - 1) == '/') The machine code is doing some magic after the strdup() on the previous line. I suspect that it is gcc that is performing strlen() with some black magic that trips this warning: ==3212== Invalid read of size 4 ==3212== at 0x811951A: cRecording::cRecording(char const*) (recording.c:801) ==3212== by 0x80CEF77: cDvbPlayer::cDvbPlayer(char const*, bool) (dvbplayer.c:268) ==3212== Address 0x4454004 is 76 bytes inside a block of size 78 alloc'd ==3212== at 0x4028308: malloc (vg_replace_malloc.c:263) ==3212== by 0x4315E1F: strdup (strdup.c:43) ==3212== by 0x81194FE: cRecording::cRecording(char const*) (recording.c:800) Then I got and fixed many warnings in Softdevice, which forgot to initialize some class members in constructors. Finally, I got a real warning for VDR: ==3212== Conditional jump or move depends on uninitialised value(s) ==3212== at 0x4029654: __GI_strcmp (mc_replace_strmem.c:712) ==3212== by 0x813ACD8: cSkinLCARSDisplayReplay::DrawTrack() (skinlcars.c:1786) ==3212== by 0x813AE9F: cSkinLCARSDisplayReplay::Flush() (skinlcars.c:1861) ==3212== by 0x80FCA55: cReplayControl::ShowProgress(bool) (menu.c:4670) ==3212== by 0x80FCBF7: cReplayControl::ShowTimed(int) (menu.c:4583) ==3212== by 0x80FCDEF: cReplayControl::cReplayControl(bool) (menu.c:4510) ==3212== by 0x80AC745: main (vdr.c:1307) "(gdb) monitor get_vbits" tells me that all of lastTrackId is uninitialized: (gdb) up #1 0x0813acd9 in cSkinLCARSDisplayReplay::DrawTrack ( this=this@entry=0x457c3a0) at skinlcars.c:1786 1786 if (!Track && *lastTrackId.description || Track && strcmp(lastTrackId.description, Track->description)) { (gdb) p lastTrackId $27 = {id = 0, language = "\000\000\000\000\000\000\000", description = '\000' <repeats 31 times>} (gdb) p &lastTrackId $28 = (tTrackId *) 0x457c434 (gdb) p sizeof lastTrackId $29 = 42 (gdb) monitor get_vbits 0x457c434 42 ffffffff ffffffff ffffffff ffffffff ffffffff ffffffff ffffffff ffffffff ffffffff ffffffff ffff BTW, !Track && x || Track && y should IMO be simpler written Track ? y : x. It looks like a memset() is missing from the cSkinLCARSDisplayReplay constructor. cSkinLCARSDisplayChannel::cSkinLCARSDisplayChannel() is doing the right thing: memset(&lastTrackId, 0, sizeof(lastTrackId)); Adding the memset() made this message go away. (Patch attached.) The next problem is this one, which I get every time by pressing Play, Pause, Menu, Recordings after startup: ==3601== Conditional jump or move depends on uninitialised value(s) ==3601== at 0x810C0DB: cRect::Intersected(cRect const&) const (osd.h:411) ==3601== by 0x810E3D1: cPixmapMemory::DrawRectangle(cRect const&, unsigned int) (osd.c:1333) ==3601== by 0x810AA0B: cOsd::DrawRectangle(int, int, int, int, unsigned int) (osd.c:1922) ==3601== by 0x8130482: cSkinLCARSDisplayMenu::Clear() (skinlcars.c:1463) ==3601== by 0x810651F: cOsdMenu::Display() (osdbase.c:223) ==3601== by 0x80FA3D1: cMenuMain::Set() (menu.c:3432) ==3601== by 0x80FA9FD: cMenuMain::cMenuMain(eOSState, bool) (menu.c:3376) ==3601== by 0x80AC21B: main (vdr.c:1078) According to "monitor get_vbits", the cRect is totally uninitialized. Program received signal SIGTRAP, Trace/breakpoint trap. 0x0810c0db in IsEmpty (this=0xbeba08a0) at osd.h:411 411 bool IsEmpty(void) const { return Width() <= 0 || Height() <= 0; } (gdb) up #1 cRect::Intersected (this=this@entry=0xbeba08a0, Rect=...) at osd.c:912 912 if (!IsEmpty() && !Rect.IsEmpty()) { (gdb) up #2 0x0810e3d2 in cPixmapMemory::DrawRectangle (this=0x6d3fe78, Rect=..., Color=2566914048) at osd.c:1333 1333 cRect r = Rect.Intersected(DrawPort().Size()); As far as I can tell, the entirely uninitialized cRect is being passed as the Rect parameter to cPixmapMemory::DrawRectangle(). Unfortunately, gdb cannot show me the stack above that. It would seem to me that cSkinLCARSDisplayMenu::Clear() is passing uninitialized bounds to cOsd::DrawRectangle(), which will lead to funny values like this: (gdb) p *this $31 = {point = {x = 1418239204, y = 0}, size = {width = -1379480940, height = 201}, static Null = {point = {x = 0, y = 0}, size = {width = 0, height = 0}, static Null = <same as static member of an already seen type>}} As a workaround, I guess that I will be switching away from the LCARS skin for now. I got into this exercise because vdr sometimes crashed when I pressed the Recordings or Menu button when using the LCARS skin. I am also attaching my patch against softdevice CVS, in case someone finds it useful. I was unable to figure out how to clear the garbage at the bottom of the OSD screen. It goes away if the dfb:mgatv video display area is high enough (4:3 video instead of 16:9). Best regards, Marko

2 4

[PATCH] spelling in usage text
by Lars Hanisch 16 Nov '13

16 Nov '13

Hi, " --vfat for backwards compatibility (same as\n" " --dirnames=250,40,1\n" Either use a comma before "same" or a closing bracket in the second line. Whatever you prefer, I would take the comma. :) Patches attached for both variants. Lars.

1 0

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

vdr