eFrontend::setTerrestrialAntennaVoltage broken? frontend.cpp

Alles rund um das grafische Benutzerinterface Enigma
gusax
Neugieriger
Neugieriger
Beiträge: 5
Registriert: Samstag 7. April 2007, 19:06

eFrontend::setTerrestrialAntennaVoltage broken? frontend.cpp

Beitrag von gusax »

I have a Dreambox 500T (DVB-T). When searching for channels, there is an option to "Disable 5V" (if unchecked, it's supposed to enable +5V). No matter what this switch is set to, the 5V is always disabled.

When "Disable 5V" is clicked, it runs the function

Code: Alles auswählen

void eFrontend::setTerrestrialAntennaVoltage(bool state)
{
	if (type != eSystemInfo::feTerrestrial)
		return;
#if HAVE_DVB_API_VERSION < 3
	int secfd=::open(SEC_DEV, O_RDWR);
	if (secfd<0)
	{
		eDebug("open secfd failed(%m)");
		return;
	}
	if ( ::ioctl(secfd, SEC_SET_VOLTAGE, state ? SEC_VOLTAGE_OFF : SEC_VOLTAGE_13) < 0 )
		eDebug("SEC_SET_VOLTAGE (-T) failed (%m)");
	::close(secfd);
#endif
}
which is found in /apps/tuxbox/enigma/lib/dvb/frontend.cpp

If i telnet to the dreambox, pressing "Disable 5V" with the remote causes the following error message to be printed in the console: "open secfd failed(No such device)".

Considering the source, this fault has to be generated here:

Code: Alles auswählen

	int secfd=::open(SEC_DEV, O_RDWR);
	if (secfd<0)
	{
		eDebug("open secfd failed(%m)");
		return;
	}
SEC_DEV is defined as

Code: Alles auswählen

#define SEC_DEV "/dev/dvb/card0/sec0"
in the file /apps/tuxbox/enigma/include/lib/dvb/frontend.h

I've checked /dev/dvb/card0/sec0 and it does exist in the filesystem.

But I need help from here :gruebel:

The strange thing is that if I use an old image like gemini 3.1, +5V works and is sent to the antenna. But on the newer images and CVS, I get "no such device" and always 0V.

Any help in solving this matter would be greatly appreciated, because I can't figure out why this doesn't work. Also, feel free to answer in german if that's easier. Ich kann Deutsch ganz gut lesen, aber nicht so gut schreiben.
seife
Developer
Beiträge: 4189
Registriert: Sonntag 2. November 2003, 12:36

Beitrag von seife »

what does "ls -l /dev/dvb/card0/sec0" say exactly? If sec0 is a link, how does the file the link points to look like?
Does it also happen with an original dream-multimedia image for the dm500?
gusax
Neugieriger
Neugieriger
Beiträge: 5
Registriert: Samstag 7. April 2007, 19:06

Beitrag von gusax »

Code: Alles auswählen

~ > ls -l /dev/dvb/card0/sec0
crw-------    1 root     root     245,   8 Jan  1  1970 /dev/dvb/card0/sec0
Will try with the latest original image to see if it works.
seife
Developer
Beiträge: 4189
Registriert: Sonntag 2. November 2003, 12:36

Beitrag von seife »

This looks pretty much the same as on my dm500-s (where opening the device works fine), so i'd really try the latest DMM image and then complain at their forum - i suspect a driver bug, but i unfortunately cannot check the source, nor do i have a terrestrial box.
gusax
Neugieriger
Neugieriger
Beiträge: 5
Registriert: Samstag 7. April 2007, 19:06

Beitrag von gusax »

Thanks very much for the replies and suggestions.

I tried the latest DMM image (1.08) and both that and gemini 3.2 said "no such device". Tonight I'll try with the old gemini image that works and then I'll copy the drivers (head.ko) from that image into the new one I just compiled from CVS. If a new image works with older drivers, then it probably is a driver bug.
seife
Developer
Beiträge: 4189
Registriert: Sonntag 2. November 2003, 12:36

Beitrag von seife »

Ok, i see. Yes, that sounds like a plan :-) I see you know enough that you will be able to handle it :-)
gusax
Neugieriger
Neugieriger
Beiträge: 5
Registriert: Samstag 7. April 2007, 19:06

Beitrag von gusax »

An update on this matter for those interested:
I just compiled a new image from cvs, replaced its head.ko-driver with an old head.ko from 2006 and it worked perfectly. I now have the +5V that I need for my antenna.

I'm going to post a bug report over at the Dream multimedia forum, hopefully they'll fix the bug and update (or rather "downdate"...) their drivers.
gusax
Neugieriger
Neugieriger
Beiträge: 5
Registriert: Samstag 7. April 2007, 19:06

Bug fixed

Beitrag von gusax »

And another update: Dream has released an updated version of the dreamdrivers that should give the +5V-functionality back. Have not tried it yet (something is wrong with my cdk), but I'm hoping for the best :)

Edit: The bug is fixed, +5V works with new drivers.