Hi,
I have an USB DVB-C card (Reddo dvb-c, actually a relabelled Tongshi box), which works very well with the current Linux driver excluding channels with QAM-256 modulation. Would it be easy to check FE_CAN_QAM_256 in vdr before trying to use a device to tune to a particular channel? In such a case I could deactivate FE_CAN_QAM_256 in the drivers. I am using VDR 1.7.9, if it makes a difference.
Of course at the moment the proper solution would be to modify the channels.conf to se a preselected card to tune to a particular channel, but in this case there are two drawbacks with it. 1) I have more than one card that I can use for QAM-256 channels and would like to use them all, and 2) channels woth QAM-256 are also encrypted and I have yet to figure out how to set both the CA code and the number of the card in the "Conditional access" parameter in the channels.conf.
Actually now that I think of it... How about separating the explicit card selection and conditional access by introducing a new paremeter in the channels.conf that could be used to preselect the card(s) that can tune to the selected channel?
-Petri
I have 3 dvb-s cards in one of my boxes. 2 of those cards are able to tune all channels in my channels.conf. 1 of them can only tune some. So, the problem is there seems to be no way to specify something like 'use device X,Y,Z' for a certain channel in channels.conf... I had raised this issue before but it seems there's no plans to address the issue. Big problem for users in this situation. I'm constantly losing recordings because of this as well. :(
On 29/10/2009, Petri Helin phelin@googlemail.com wrote:
Hi,
I have an USB DVB-C card (Reddo dvb-c, actually a relabelled Tongshi box), which works very well with the current Linux driver excluding channels with QAM-256 modulation. Would it be easy to check FE_CAN_QAM_256 in vdr before trying to use a device to tune to a particular channel? In such a case I could deactivate FE_CAN_QAM_256 in the drivers. I am using VDR 1.7.9, if it makes a difference.
Of course at the moment the proper solution would be to modify the channels.conf to se a preselected card to tune to a particular channel, but in this case there are two drawbacks with it. 1) I have more than one card that I can use for QAM-256 channels and would like to use them all, and 2) channels woth QAM-256 are also encrypted and I have yet to figure out how to set both the CA code and the number of the card in the "Conditional access" parameter in the channels.conf.
Actually now that I think of it... How about separating the explicit card selection and conditional access by introducing a new paremeter in the channels.conf that could be used to preselect the card(s) that can tune to the selected channel?
-Petri
Is this not related to the sourcecap patch?
On 10/28/09 23:15, Petri Helin wrote:
Hi,
I have an USB DVB-C card (Reddo dvb-c, actually a relabelled Tongshi box), which works very well with the current Linux driver excluding channels with QAM-256 modulation. Would it be easy to check FE_CAN_QAM_256 in vdr before trying to use a device to tune to a particular channel? In such a case I could deactivate FE_CAN_QAM_256 in the drivers. I am using VDR 1.7.9, if it makes a difference.
Checking the frontend capabilities is certainly the right thing to do, and I will see that this gets implemented.
Klaus
On Thu, Oct 29, 2009 at 10:13 AM, Klaus Schmidinger Klaus.Schmidinger@tvdr.de wrote:
On 10/28/09 23:15, Petri Helin wrote:
Hi,
I have an USB DVB-C card (Reddo dvb-c, actually a relabelled Tongshi box), which works very well with the current Linux driver excluding channels with QAM-256 modulation. Would it be easy to check FE_CAN_QAM_256 in vdr before trying to use a device to tune to a particular channel? In such a case I could deactivate FE_CAN_QAM_256 in the drivers. I am using VDR 1.7.9, if it makes a difference.
Checking the frontend capabilities is certainly the right thing to do, and I will see that this gets implemented.
Klaus
Thanks!
-Petri
On Thu, Oct 29, 2009 at 1:13 AM, Klaus Schmidinger Klaus.Schmidinger@tvdr.de wrote:
On 10/28/09 23:15, Petri Helin wrote:
Hi,
I have an USB DVB-C card (Reddo dvb-c, actually a relabelled Tongshi box), which works very well with the current Linux driver excluding channels with QAM-256 modulation. Would it be easy to check FE_CAN_QAM_256 in vdr before trying to use a device to tune to a particular channel? In such a case I could deactivate FE_CAN_QAM_256 in the drivers. I am using VDR 1.7.9, if it makes a difference.
Checking the frontend capabilities is certainly the right thing to do, and I will see that this gets implemented.
This still doesn't resolve a major problem for North Americans. The situation is that the 2 biggest dvb-s providers use a modified modulation, 8psk turbo-fec. There is only one device that is capable of tuning this but unfortunately the device has to pretend to be a dvb-s2 device to do so. Apparently the v4l guys (from what I was told) didn't want to allow 8psk turbo-fec in dvb-s because it's a non-standard modulation. So the problem becomes that VDR sees a dvb-s stream and the frontend of say a nexus-s will say its capable of receiving it, even though that dvb-s stream may contain the 8psk turbo-fec which a nexus-s can not tune.
I believe the best way to handle this situation is by allowing the user control over what devices may be used to tune what channels. VDR can't be smart enough to figure it out due to incorrect & bad design in v4l. Another strong reason to allow this to be configured at the user-level is what if the the user has 2 devices...both able to tune a channel but the signal is weak. Tuner 1 can tune it fine but tuner 2 has problems because of the weak signal. VDR is simply going to ask a free tuner 'can you tune this?', whereas the best scenario would be for the user to say 'use tuner 1 for this channel'.
I can think of two ways to better deal with this; adding a new field in channels.conf, adding a tuneroverride.conf
By adding a new field.. If the value is "*" then it means any device may tune it. If its an integer, then the values listed represent which devices should be used. For example: :*: use any device :2,3: use only device 2 or device 3
By using a tuneroverride.conf you can specify which devices to use with certain channels. If the channel isn't contained in tuneroverride.conf then VDR just uses it's own logic to figure it out (the capability check). The good thing about this method is you can create a simple shell script to auto-generate the tuneroverride.conf file using key indicators in the channels.conf... The file structure would be very simple:
<channel #>:<device,device,device,etc>
This is really the only way to satisfy _everyones needs_, by ultimately allowing the user to control device priorities. PLEASE take this into serious consideration as it's been a long standing problem that's only getting worse with more and more transponders move to 8psk turbo-fec.
Best regards, Derek
VDR User wrote:
I believe the best way to handle this situation is by allowing the user control over what devices may be used to tune what channels. VDR can't be smart enough to figure it out due to incorrect & bad design in v4l. Another strong reason to allow this to be configured at the user-level is what if the the user has 2 devices...both able to tune a channel but the signal is weak. Tuner 1 can tune it fine but tuner 2 has problems because of the weak signal. VDR is simply going to ask a free tuner 'can you tune this?', whereas the best scenario would be for the user to say 'use tuner 1 for this channel'.
I can think of two ways to better deal with this; adding a new field in channels.conf,
Doesn't the CA field in channels.conf already allow this?
Conditional access A hexadecimal integer defining how this channel can be accessed:
0000 Free To Air 0001...000F explicitly requires the device with the given number
I don't know if it works if you specify more than one device, but it certainly works to specify only one device.
Regards... Michael
The CA field can only be used for this or CA, not both.
----- Original Message ----- From: "Michael Mauch" michael.mauch@gmx.de To: vdr@linuxtv.org Sent: Sunday, November 01, 2009 4:26 PM Subject: Re: [vdr] Restricting a particular dvb card from tuning tochannels?with a selected modulation
VDR User wrote:
I believe the best way to handle this situation is by allowing the user control over what devices may be used to tune what channels. VDR can't be smart enough to figure it out due to incorrect & bad design in v4l. Another strong reason to allow this to be configured at the user-level is what if the the user has 2 devices...both able to tune a channel but the signal is weak. Tuner 1 can tune it fine but tuner 2 has problems because of the weak signal. VDR is simply going to ask a free tuner 'can you tune this?', whereas the best scenario would be for the user to say 'use tuner 1 for this channel'.
I can think of two ways to better deal with this; adding a new field in channels.conf,
Doesn't the CA field in channels.conf already allow this?
Conditional access A hexadecimal integer defining how this channel can be accessed:
0000 Free To Air 0001...000F explicitly requires the device with the given numberI don't know if it works if you specify more than one device, but it certainly works to specify only one device.
Regards... Michael
vdr mailing list vdr@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
There is the isue whereone tunner might be connected to an LNB on a rotor and another to a fixed. Knowing how the tunner can tune wouldn't help then.
----- Original Message ----- From: "Klaus Schmidinger" Klaus.Schmidinger@tvdr.de To: vdr@linuxtv.org Sent: Thursday, October 29, 2009 1:13 AM Subject: Re: [vdr] Restricting a particular dvb card from tuning to channels with a selected modulation
On 10/28/09 23:15, Petri Helin wrote:
Hi,
I have an USB DVB-C card (Reddo dvb-c, actually a relabelled Tongshi box), which works very well with the current Linux driver excluding channels with QAM-256 modulation. Would it be easy to check FE_CAN_QAM_256 in vdr before trying to use a device to tune to a particular channel? In such a case I could deactivate FE_CAN_QAM_256 in the drivers. I am using VDR 1.7.9, if it makes a difference.
Checking the frontend capabilities is certainly the right thing to do, and I will see that this gets implemented.
Klaus
vdr mailing list vdr@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
hello guys
yes i do think this is a good idea... as with different tuners they could be used for other lnbs like c -band and ku i could only use my s2/s card for services on a c and ku setup then tune and record from different types of services... then each tuner could live on a fixed satellite if you wanted
On 10/29/09, Timothy D. Lenz tlenz@vorgon.com wrote:
There is the isue whereone tunner might be connected to an LNB on a rotor and another to a fixed. Knowing how the tunner can tune wouldn't help then.
----- Original Message ----- From: "Klaus Schmidinger" Klaus.Schmidinger@tvdr.de To: vdr@linuxtv.org Sent: Thursday, October 29, 2009 1:13 AM Subject: Re: [vdr] Restricting a particular dvb card from tuning to channels with a selected modulation
On 10/28/09 23:15, Petri Helin wrote:
Hi,
I have an USB DVB-C card (Reddo dvb-c, actually a relabelled Tongshi box), which works very well with the current Linux driver excluding channels with QAM-256 modulation. Would it be easy to check FE_CAN_QAM_256 in vdr before trying to use a device to tune to a particular channel? In such a case I could deactivate FE_CAN_QAM_256 in the drivers. I am using VDR 1.7.9, if it makes a difference.
Checking the frontend capabilities is certainly the right thing to do, and I will see that this gets implemented.
Klaus
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
On 30/10/2009, abbe normal 1abenormal@gmail.com wrote:
hello guys
yes i do think this is a good idea... as with different tuners they could be used for other lnbs like c -band and ku i could only use my s2/s card for services on a c and ku setup then tune and record from different types of services... then each tuner could live on a fixed satellite if you wanted
On 10/29/09, Timothy D. Lenz tlenz@vorgon.com wrote:
There is the isue whereone tunner might be connected to an LNB on a rotor and another to a fixed. Knowing how the tunner can tune wouldn't help then.
----- Original Message ----- From: "Klaus Schmidinger" Klaus.Schmidinger@tvdr.de To: vdr@linuxtv.org Sent: Thursday, October 29, 2009 1:13 AM Subject: Re: [vdr] Restricting a particular dvb card from tuning to channels with a selected modulation
On 10/28/09 23:15, Petri Helin wrote:
Hi,
I have an USB DVB-C card (Reddo dvb-c, actually a relabelled Tongshi box), which works very well with the current Linux driver excluding channels with QAM-256 modulation. Would it be easy to check FE_CAN_QAM_256 in vdr before trying to use a device to tune to a particular channel? In such a case I could deactivate FE_CAN_QAM_256 in the drivers. I am using VDR 1.7.9, if it makes a difference.
Checking the frontend capabilities is certainly the right thing to do, and I will see that this gets implemented.
Klaus
Does the sourcecap patch not address this whole issue? Even if source cap or something similar is applied, how will VDR be able to tune to a 8psk turbo-fec type channel if v4l2 does not provide any support for it? It shouldn't be VDR's responsibility surely? unless 8psk turbo-fec becomes part of the new dvb-plugin archtitecture of future VDR.
On Fri, Oct 30, 2009 at 12:11 AM, Theunis Potgieter theunis.potgieter@gmail.com wrote:
Does the sourcecap patch not address this whole issue? Even if source cap or something similar is applied, how will VDR be able to tune to a 8psk turbo-fec type channel if v4l2 does not provide any support for it? It shouldn't be VDR's responsibility surely? unless 8psk turbo-fec becomes part of the new dvb-plugin archtitecture of future VDR.
Turbo fec isn't specified in the stream or in v4l, it's only reported as 8psk. I may be wrong but doesn't sourcecaps only assign devices to orbital positions? What happens about all the orbital positions that contain both qpsk and 8psk turbo? It was suggested that you make a fake orbital position for those with 8psk turbo transponders, then change your channels.conf & sources.conf accordingly.
Consider the following:
dvb devices being used: budget dvb-s, genpix dvb-s (supporting turbo fec), twinhan dvb-s/s2
sat1 tp1 (qpsk) dvb-s sat1 tp2 (8psk turbo) dvb-s
sat2 tp1 (qpsk) dvb-s sat2 tp2 (8psk turbo) dvb-s sat2 tp3 (qpsk) dvb-s
sat3 tp1 (8psk) dvb-s2
If you say sat1 should use the genpix because it has an 8psk turbo fec tp then you restrict the budget and twinhan ability to tune the qpsk on tp1. Same for sat2. If you want to tune sat3 tp1 by asking the devices 'are you s2 capable?' then you get a false hit by the genpix, which lies about being s2 capable (so it can provide 8psk). If you use sourcecaps and create a fake orbital position for sat1 tp2 and sat2 tp2 then it works but requires the user to create fake data and do more work.
You can see simply doing a capability query is not good enough because it can return unreliable results. This isn't VDR's fault, it's v4l & drivers but it can still be easily addressed within VDR which giving the user other advantages such as assigning certain devices to certain channels/transponders because they (maintain) lock better.
Now consider you just simply add support for an optional file called override.conf (or whatever). This file could look like this:
<channel>:<list of devices to use> 100:1,3 means for channel 100 use only devices 1 and 3. 101:2 means for channel 101 use only device 2.
You could also add support for orbital positions with: <orbpos>:<list of devices to use> S100.0:2,3 means for orbital position S100.0 use only devices 2 and 3. S109.0:3 means for orbital position S109.0 use only device 3.
So you see, simply doing a capability query is not enough because it can be unreliable. Using sourcecaps can solve the problem but requires non-existent orbital positions to be created and maintained in both channels.conf and sources.conf. By adding an override.conf, you can give full control to the user by using one simple _optional_ file. This seems like the easiest & best with the most flexibility for the user. It solves _everyones_ needs and doesn't require false data, patching, etc. As mentioned before you can also write a simple shell script to auto-generate the override.conf for you. I know Klaus doesn't like to add files but if the file is optional and adding support for it means satisfying everyones needs, I can't see a good argument against it. Especially when the alternative requires at least some users to add fake data in their channels.conf+sources.conf.
Best regards, Derek
A tunner that can do turbo can also do normal giving you extra tunner for recording one while watching another. Fake postions ends up adding lots more channels, dupes to take advantage and manual mixing and matching in recording timmers. Better to be able to say which tunners can do a channel and which lnb's to use be they fixed or on a rotor. People with a mix of rotors and fixed lnbs on ku dishes and buds need a matrix system to say where to get what channels and what tuners to use.
----- Original Message ----- From: "VDR User" user.vdr@gmail.com To: "VDR Mailing List" vdr@linuxtv.org Sent: Friday, October 30, 2009 10:18 AM Subject: Re: [vdr] Restricting a particular dvb card from tuning to channelswith a selected modulation
On Fri, Oct 30, 2009 at 12:11 AM, Theunis Potgieter theunis.potgieter@gmail.com wrote:
Does the sourcecap patch not address this whole issue? Even if source cap or something similar is applied, how will VDR be able to tune to a 8psk turbo-fec type channel if v4l2 does not provide any support for it? It shouldn't be VDR's responsibility surely? unless 8psk turbo-fec becomes part of the new dvb-plugin archtitecture of future VDR.
Turbo fec isn't specified in the stream or in v4l, it's only reported as 8psk. I may be wrong but doesn't sourcecaps only assign devices to orbital positions? What happens about all the orbital positions that contain both qpsk and 8psk turbo? It was suggested that you make a fake orbital position for those with 8psk turbo transponders, then change your channels.conf & sources.conf accordingly.
Consider the following:
dvb devices being used: budget dvb-s, genpix dvb-s (supporting turbo fec), twinhan dvb-s/s2
sat1 tp1 (qpsk) dvb-s sat1 tp2 (8psk turbo) dvb-s
sat2 tp1 (qpsk) dvb-s sat2 tp2 (8psk turbo) dvb-s sat2 tp3 (qpsk) dvb-s
sat3 tp1 (8psk) dvb-s2
If you say sat1 should use the genpix because it has an 8psk turbo fec tp then you restrict the budget and twinhan ability to tune the qpsk on tp1. Same for sat2. If you want to tune sat3 tp1 by asking the devices 'are you s2 capable?' then you get a false hit by the genpix, which lies about being s2 capable (so it can provide 8psk). If you use sourcecaps and create a fake orbital position for sat1 tp2 and sat2 tp2 then it works but requires the user to create fake data and do more work.
You can see simply doing a capability query is not good enough because it can return unreliable results. This isn't VDR's fault, it's v4l & drivers but it can still be easily addressed within VDR which giving the user other advantages such as assigning certain devices to certain channels/transponders because they (maintain) lock better.
Now consider you just simply add support for an optional file called override.conf (or whatever). This file could look like this:
<channel>:<list of devices to use> 100:1,3 means for channel 100 use only devices 1 and 3. 101:2 means for channel 101 use only device 2.
You could also add support for orbital positions with: <orbpos>:<list of devices to use> S100.0:2,3 means for orbital position S100.0 use only devices 2 and 3. S109.0:3 means for orbital position S109.0 use only device 3.
So you see, simply doing a capability query is not enough because it can be unreliable. Using sourcecaps can solve the problem but requires non-existent orbital positions to be created and maintained in both channels.conf and sources.conf. By adding an override.conf, you can give full control to the user by using one simple _optional_ file. This seems like the easiest & best with the most flexibility for the user. It solves _everyones_ needs and doesn't require false data, patching, etc. As mentioned before you can also write a simple shell script to auto-generate the override.conf for you. I know Klaus doesn't like to add files but if the file is optional and adding support for it means satisfying everyones needs, I can't see a good argument against it. Especially when the alternative requires at least some users to add fake data in their channels.conf+sources.conf.
Best regards, Derek
vdr mailing list vdr@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
Hi,
I bought a Reddo dvb-c usb card last week and wanted to make it a backup card without a risk of it accidentally tuning into QAM256 channel and spitting out garbage. It seems to be quite good value for money in Finland if you want dvb-c usb card and don't need QAM256.
I followed this conversation through but as it seems there is no support for not using QAM256 in the driver nor in the VDR I decided to make a short patch for VDR.
**WARNING** This patch was written in quite short time and it's not a good example of clean code, but it works (for me) and will be useful for anyone with Reddo card having QAM256 channels. It will be obsolete if/when QAM256 disable is done "the right way" (by asking the driver), but until then... Please try it out and let me know if there is a problem with it.
The patch is available here:
http://tvr.dy.fi/vdr/vdr-1.7.14-DisableReddoQAM256.diff
Teemu
2009/10/29 Petri Helin phelin@googlemail.com
Hi,
I have an USB DVB-C card (Reddo dvb-c, actually a relabelled Tongshi box), which works very well with the current Linux driver excluding channels with QAM-256 modulation. Would it be easy to check FE_CAN_QAM_256 in vdr before trying to use a device to tune to a particular channel? In such a case I could deactivate FE_CAN_QAM_256 in the drivers. I am using VDR 1.7.9, if it makes a difference.
Of course at the moment the proper solution would be to modify the channels.conf to se a preselected card to tune to a particular channel, but in this case there are two drawbacks with it. 1) I have more than one card that I can use for QAM-256 channels and would like to use them all, and 2) channels woth QAM-256 are also encrypted and I have yet to figure out how to set both the CA code and the number of the card in the "Conditional access" parameter in the channels.conf.
Actually now that I think of it... How about separating the explicit card selection and conditional access by introducing a new paremeter in the channels.conf that could be used to preselect the card(s) that can tune to the selected channel?
-Petri
vdr mailing list vdr@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
Klaus added the ability to limit satellites to certain dvb card in VDR-1.7.13. However, a lot of users have the need to do device limiting at the transponder level, which VDR still can't do (and likely will never do I think, unfortunately). But, Klaus also added device hooks in VDR-1.7.13 as well:
- Implemented cDeviceHook to allow plugins more control over which device can provide which transponder (thanks to Reinhard Nissl).
This means that a plugin can be written that DOES provide device limiting at the transponder level! Rnissl, as a test for the device hook patch, threw together a quick skeleton plugin using hardcoded values and it worked great. To finish the plugin, it would just need support for a conf file where you define which transponder:polarity:satellite & devices you want to limit them to. I assume this file would be read into a table stored in memory, and then referenced during a channel change for live tv or recording.
If Rnissl see's this, maybe he'll elaborate further.
Cheers, Derek
I see... I actually noticed cDeviceHook but couldn't figure out if it could be used or not as there was no reference anywhere for it.
This patch could be very nicely implemented fully in a plugin. I'll try that...
Thanks,
Teemu
2010/4/3 VDR User user.vdr@gmail.com
Klaus added the ability to limit satellites to certain dvb card in VDR-1.7.13. However, a lot of users have the need to do device limiting at the transponder level, which VDR still can't do (and likely will never do I think, unfortunately). But, Klaus also added device hooks in VDR-1.7.13 as well:
- Implemented cDeviceHook to allow plugins more control over which device
can provide which transponder (thanks to Reinhard Nissl).
This means that a plugin can be written that DOES provide device limiting at the transponder level! Rnissl, as a test for the device hook patch, threw together a quick skeleton plugin using hardcoded values and it worked great. To finish the plugin, it would just need support for a conf file where you define which transponder:polarity:satellite & devices you want to limit them to. I assume this file would be read into a table stored in memory, and then referenced during a channel change for live tv or recording.
If Rnissl see's this, maybe he'll elaborate further.
Cheers, Derek
vdr mailing list vdr@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
Hi,
There's a new version of the patch implemented as a plugin. It seems to work, but there are few things to notice:
- Plugin does not cache which devices are Reddo devices, instead it probes sysfs every time QAM256 channel is being tuned into (on any device). It would be nice if cDeviceHook added a mechanism for a hook to store context for each device. Hooking into device probe (like full feature card plugin does) would be much nicer way, but it would interfere with other plugins which need the same mechanism.
- VDR crashes at very late in exit() when clearing some list (most likely cDeviceHook list), but I couldn't find a way to avoid that
- You may get syslog messages about trying to tune into QAM256 channel if/when EPG is being updated on the background.
- The sysfs probe code in full feature card plugin is buggy :-)
The plugin is available here:
http://tvr.dy.fi/vdr/vdr-disablereddoqam256-0.0.1.tgz
Teemu
2010/4/3 VDR User user.vdr@gmail.com
Klaus added the ability to limit satellites to certain dvb card in VDR-1.7.13. However, a lot of users have the need to do device limiting at the transponder level, which VDR still can't do (and likely will never do I think, unfortunately). But, Klaus also added device hooks in VDR-1.7.13 as well:
- Implemented cDeviceHook to allow plugins more control over which device
can provide which transponder (thanks to Reinhard Nissl).
This means that a plugin can be written that DOES provide device limiting at the transponder level! Rnissl, as a test for the device hook patch, threw together a quick skeleton plugin using hardcoded values and it worked great. To finish the plugin, it would just need support for a conf file where you define which transponder:polarity:satellite & devices you want to limit them to. I assume this file would be read into a table stored in memory, and then referenced during a channel change for live tv or recording.
If Rnissl see's this, maybe he'll elaborate further.
Cheers, Derek
vdr mailing list vdr@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
On 05.04.2010 00:55, Teemu Rantanen wrote:
Hi,
There's a new version of the patch implemented as a plugin. It seems to work, but there are few things to notice:
- Plugin does not cache which devices are Reddo devices, instead it
probes sysfs every time QAM256 channel is being tuned into (on any device). It would be nice if cDeviceHook added a mechanism for a hook to store context for each device. Hooking into device probe (like full feature card plugin does) would be much nicer way, but it would interfere with other plugins which need the same mechanism.
The proper way of doing this is to check the modulation types in cDvbDevice::ProvidesTransponder(), as in the attached patch (which will be part of VDR 1.7.15). If the "reddo" driver doesn't set the FE_CAN_QAM_256 flag correctly, it needs to be fixed there.
- VDR crashes at very late in exit() when clearing some list (most
likely cDeviceHook list), but I couldn't find a way to avoid that
From PLUGINS.html:
A plugin that creates a derived cDeviceHook ... shall not delete this object. It will be automatically deleted when the program ends
I have added a similar note to device.h now.
...
- The sysfs probe code in full feature card plugin is buggy :-)
Would you mind posting a patch that fixes this?
Klaus
The plugin is available here:
Hi,
I tried also without delete cDeviceHook but it still crashed...
Well, the fixes are basically available in the plugin, as I copied the probe method (and modified slightly to suit the plugin). Those are: - The sysfs filenames are idVendor (subsystem_vendor) and idProduct (subsystem_device) - You need to set strtoul() base as 16 as the files don't have 0x in front of the hex
What about making this an utility method in VDR? Give device file name and return id...
Teemu
2010/4/5 Klaus Schmidinger Klaus.Schmidinger@tvdr.de
From PLUGINS.html:
A plugin that creates a derived cDeviceHook ... shall not delete this object. It will be automatically deleted when the program ends
I have added a similar note to device.h now.
...
- The sysfs probe code in full feature card plugin is buggy :-)
Would you mind posting a patch that fixes this?
Klaus
On 05.04.2010 12:43, Teemu Rantanen wrote:
Hi,
I tried also without delete cDeviceHook but it still crashed...
Please try creating the cReddoDeviceHook in cPluginDisableReddoQAM256::Initialize(), as suggested in PLUGINS.html.
Well, the fixes are basically available in the plugin, as I copied the probe method (and modified slightly to suit the plugin). Those are:
- The sysfs filenames are idVendor (subsystem_vendor) and idProduct
(subsystem_device)
This is what I get here:
root@video:/home/kls/vdr/VDR > ls -l /sys/class/dvb/dvb0.frontend0/device/subsystem_* -r--r--r-- 1 root root 4096 2010-04-05 16:25 /sys/class/dvb/dvb0.frontend0/device/subsystem_device -r--r--r-- 1 root root 4096 2010-04-05 16:25 /sys/class/dvb/dvb0.frontend0/device/subsystem_vendor
root@video:/home/kls/vdr/VDR > ls -l /sys/class/dvb/dvb0.frontend0/device/id* ls: cannot access /sys/class/dvb/dvb0.frontend0/device/id*: No such file or directory
Maybe a bug in the driver?
- You need to set strtoul() base as 16 as the files don't have 0x in
front of the hex
root@video:/home/kls/vdr/VDR > cat /sys/class/dvb/dvb0.frontend0/device/subsystem_device 0x0000 root@video:/home/kls/vdr/VDR > cat /sys/class/dvb/dvb0.frontend0/device/subsystem_vendor 0x13c2
Maybe the "reddo" driver behaves differently than the FuSi driver?
What about making this an utility method in VDR? Give device file name and return id...
Will do.
Have you tried whether it works with my patch? In that case you wouldn't even need the whole plugin.
Klaus
2010/4/5 Klaus Schmidinger <Klaus.Schmidinger@tvdr.de mailto:Klaus.Schmidinger@tvdr.de>
From PLUGINS.html: A plugin that creates a derived cDeviceHook ... shall not delete this object. It will be automatically deleted when the program ends I have added a similar note to device.h now. > ... > - The sysfs probe code in full feature card plugin is buggy :-) Would you mind posting a patch that fixes this? Klaus
Ok, so the sysfs names of vendor/product are driver-dependent... All my usb devices have idVendor/idProduct even though only one of them is "Reddo".
Tried to create the hook on Initialize(), still crashes on exit().
Haven't tried your patch, because as far as I know the driver claims it can do QAM256, and even if that was disabled today it would take some time to get into all linux distributions... Btw it even claims to support QAM256 on the product package, but it still doesn't work. I guess that's the reason they are so inexpensive here...
Teemu
2010/4/5 Klaus Schmidinger Klaus.Schmidinger@tvdr.de
On 05.04.2010 12:43, Teemu Rantanen wrote:
Hi,
I tried also without delete cDeviceHook but it still crashed...
Please try creating the cReddoDeviceHook in cPluginDisableReddoQAM256::Initialize(), as suggested in PLUGINS.html.
Well, the fixes are basically available in the plugin, as I copied the probe method (and modified slightly to suit the plugin). Those are:
- The sysfs filenames are idVendor (subsystem_vendor) and idProduct
(subsystem_device)
This is what I get here:
root@video:/home/kls/vdr/VDR > ls -l /sys/class/dvb/dvb0.frontend0/device/subsystem_* -r--r--r-- 1 root root 4096 2010-04-05 16:25 /sys/class/dvb/dvb0.frontend0/device/subsystem_device -r--r--r-- 1 root root 4096 2010-04-05 16:25 /sys/class/dvb/dvb0.frontend0/device/subsystem_vendor
root@video:/home/kls/vdr/VDR > ls -l /sys/class/dvb/dvb0.frontend0/device/id* ls: cannot access /sys/class/dvb/dvb0.frontend0/device/id*: No such file or directory
Maybe a bug in the driver?
- You need to set strtoul() base as 16 as the files don't have 0x in
front of the hex
root@video:/home/kls/vdr/VDR > cat /sys/class/dvb/dvb0.frontend0/device/subsystem_device 0x0000 root@video:/home/kls/vdr/VDR > cat /sys/class/dvb/dvb0.frontend0/device/subsystem_vendor 0x13c2
Maybe the "reddo" driver behaves differently than the FuSi driver?
What about making this an utility method in VDR? Give device file name and return id...
Will do.
Have you tried whether it works with my patch? In that case you wouldn't even need the whole plugin.
Klaus
On 05.04.2010 18:14, Teemu Rantanen wrote:
Ok, so the sysfs names of vendor/product are driver-dependent... All my usb devices have idVendor/idProduct even though only one of them is "Reddo".
I'm not saying that what the FuSI driver does is correct, either ;-) Somebody with more knowledge should clarify this and fix the driver that is faulty.
Tried to create the hook on Initialize(), still crashes on exit().
Well, then I guess you'll need to do some debugging to find out what really goes wrong here.
Haven't tried your patch, because as far as I know the driver claims it can do QAM256, and even if that was disabled today it would take some time to get into all linux distributions... Btw it even claims to support QAM256 on the product package, but it still doesn't work. I guess that's the reason they are so inexpensive here...
At any rate, these things *must* be fixed in the driver! Even if it takes a while for a fix to get into the main kernel source.
Klaus
2010/4/5 Klaus Schmidinger <Klaus.Schmidinger@tvdr.de mailto:Klaus.Schmidinger@tvdr.de>
On 05.04.2010 12:43, Teemu Rantanen wrote: > Hi, > > I tried also without delete cDeviceHook but it still crashed... Please try creating the cReddoDeviceHook in cPluginDisableReddoQAM256::Initialize(), as suggested in PLUGINS.html. > Well, the fixes are basically available in the plugin, as I copied the > probe method (and modified slightly to suit the plugin). Those are: > - The sysfs filenames are idVendor (subsystem_vendor) and idProduct > (subsystem_device) This is what I get here: root@video:/home/kls/vdr/VDR > ls -l /sys/class/dvb/dvb0.frontend0/device/subsystem_* -r--r--r-- 1 root root 4096 2010-04-05 16:25 /sys/class/dvb/dvb0.frontend0/device/subsystem_device -r--r--r-- 1 root root 4096 2010-04-05 16:25 /sys/class/dvb/dvb0.frontend0/device/subsystem_vendor root@video:/home/kls/vdr/VDR > ls -l /sys/class/dvb/dvb0.frontend0/device/id* ls: cannot access /sys/class/dvb/dvb0.frontend0/device/id*: No such file or directory Maybe a bug in the driver? > - You need to set strtoul() base as 16 as the files don't have 0x in > front of the hex root@video:/home/kls/vdr/VDR > cat /sys/class/dvb/dvb0.frontend0/device/subsystem_device 0x0000 root@video:/home/kls/vdr/VDR > cat /sys/class/dvb/dvb0.frontend0/device/subsystem_vendor 0x13c2 Maybe the "reddo" driver behaves differently than the FuSi driver? > What about making this an utility method in VDR? Give device file name > and return id... Will do. Have you tried whether it works with my patch? In that case you wouldn't even need the whole plugin. Klaus
vdr mailing list vdr@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
On 04/05/2010 12:57 PM, Klaus Schmidinger wrote:
On 05.04.2010 00:55, Teemu Rantanen wrote:
Hi,
There's a new version of the patch implemented as a plugin. It seems to work, but there are few things to notice:
- Plugin does not cache which devices are Reddo devices, instead it
probes sysfs every time QAM256 channel is being tuned into (on any device). It would be nice if cDeviceHook added a mechanism for a hook to store context for each device. Hooking into device probe (like full feature card plugin does) would be much nicer way, but it would interfere with other plugins which need the same mechanism.
The proper way of doing this is to check the modulation types in cDvbDevice::ProvidesTransponder(), as in the attached patch (which will be part of VDR 1.7.15). If the "reddo" driver doesn't set the FE_CAN_QAM_256 flag correctly, it needs to be fixed there.
I used your patch as an example and created a simple test patch for dvb-c (I think yours is for dvb-s(2) only) in order to test the approach. I also disabled FE_CAN_QAM256 from the driver. After that VDR no longer used Reddo for QAM256 channels as expected. The approach is very limited: It disables QAM256 for the every TDA10023 frontend, not just for Reddo's, and it doesn't make VDR to prefer Reddo for non QAM256 channels, which would make sense in order to keep QAM256 channels available as much as possible. The patches are inline below:
--- dvbdevice.c.orig 2010-02-21 19:10:35.000000000 +0200 +++ dvbdevice.c 2010-04-05 17:20:06.080525344 +0300 @@ -872,8 +872,12 @@ { if (!ProvidesSource(Channel->Source())) return false; // doesn't provide source - if (!cSource::IsSat(Channel->Source())) + if (!cSource::IsSat(Channel->Source())) { + cDvbTransponderParameters dtp(Channel->Parameters()); + if (dtp.Modulation() == QAM_256 && !(frontendInfo.caps & FE_CAN_QAM_256)) + return false; return DeviceHooksProvidesTransponder(Channel); // source is sufficient for non sat + } cDvbTransponderParameters dtp(Channel->Parameters()); if (frontendType == SYS_DVBS && dtp.System() == SYS_DVBS2) return false; // requires modulation system which frontend doesn't provide
--- v4l-dvb/linux/drivers/media/dvb/frontends/tda10023.c.orig 2010-04-05 15:28:22.605844128 +0300 +++ v4l-dvb/linux/drivers/media/dvb/frontends/tda10023.c 2010-04-05 15:27:54.934343796 +0300 @@ -553,7 +553,7 @@ #endif .caps = 0x400 | //FE_CAN_QAM_4 FE_CAN_QAM_16 | FE_CAN_QAM_32 | FE_CAN_QAM_64 | - FE_CAN_QAM_128 | FE_CAN_QAM_256 | + FE_CAN_QAM_128 | //FE_CAN_QAM_256 | FE_CAN_FEC_AUTO },
-Petri
On 05.04.2010 20:55, Petri Helin wrote:
On 04/05/2010 12:57 PM, Klaus Schmidinger wrote:
On 05.04.2010 00:55, Teemu Rantanen wrote:
Hi,
There's a new version of the patch implemented as a plugin. It seems to work, but there are few things to notice:
- Plugin does not cache which devices are Reddo devices, instead it
probes sysfs every time QAM256 channel is being tuned into (on any device). It would be nice if cDeviceHook added a mechanism for a hook to store context for each device. Hooking into device probe (like full feature card plugin does) would be much nicer way, but it would interfere with other plugins which need the same mechanism.
The proper way of doing this is to check the modulation types in cDvbDevice::ProvidesTransponder(), as in the attached patch (which will be part of VDR 1.7.15). If the "reddo" driver doesn't set the FE_CAN_QAM_256 flag correctly, it needs to be fixed there.
I used your patch as an example and created a simple test patch for dvb-c (I think yours is for dvb-s(2) only) in order to test the approach.
You're right. I guess the attached version should cover all frontend types.
I also disabled FE_CAN_QAM256 from the driver. After that VDR no longer used Reddo for QAM256 channels as expected. The approach is very limited: It disables QAM256 for the every TDA10023 frontend, not just for Reddo's,
Well, then the driver needs to make a finer distinction and *properly* set FE_CAN_QAM256.
and it doesn't make VDR to prefer Reddo for non QAM256 channels, which would make sense in order to keep QAM256 channels available as much as possible.
First the driver needs to properly report whether a device can handle a given modulation type. Then VDR can decide whether to use that device for a channel requiring that modulation. *Then* we can talk sparing such devices for recordings ;-)
Klaus
Hi,
the right way, of course, is to fix the driver, and make vdr check what device can do. I'm glad Petri had time to test it, and I hope to see the patch in next version of vdr. But I couldn't use the driver patch even if I wanted to patch kernel myself, as I actually have 3 devices with TDA10023 frontend, and only one of them is "reddo" :-)
But i'm afraid my time for this project just run out. I'll continue later if/when the reddo driver patch arrives, I will test it, but until then I'll have to manage with the plugin, even if it crashes or not at exit(). I'll make an empty plugin with an empty hook, and if it still crashes at exit() it's not my fault :-)
One more thought. Would it be possible to add vendor/product id as a method for cDevice (as well as the utility method for filename->id). And would it be possible (for a plugin) to manage device feature list and remove QAM256 from the reddo device. This way the plugin wouldn't have to actually do anything anymore when vdr is running, only fix this one time at plugin startup, and everything else would work 100% the same way before/after the driver is changed. Even though this isn't the right way, but we aren't living in a perfect world :-)
And thanks,
Teemu
2010/4/5 Klaus Schmidinger Klaus.Schmidinger@tvdr.de
Well, then the driver needs to make a finer distinction and *properly* set FE_CAN_QAM256.
On 06.04.2010 14:07, Teemu Rantanen wrote:
Hi,
the right way, of course, is to fix the driver, and make vdr check what device can do. I'm glad Petri had time to test it, and I hope to see the patch in next version of vdr. But I couldn't use the driver patch even if I wanted to patch kernel myself, as I actually have 3 devices with TDA10023 frontend, and only one of them is "reddo" :-)
Well, as I said, then the driver needs to detect which one is the "reddo" and only clear the FE_CAN_QAM256 for that particular one.
But i'm afraid my time for this project just run out. I'll continue later if/when the reddo driver patch arrives, I will test it, but until then I'll have to manage with the plugin, even if it crashes or not at exit(). I'll make an empty plugin with an empty hook, and if it still crashes at exit() it's not my fault :-)
I'll be glad to apply any fix to the core VDR code you can come up with.
One more thought. Would it be possible to add vendor/product id as a method for cDevice (as well as the utility method for filename->id). And
First of all somebody should clarify the names. Is it idVendor or subsystem_vendor?
would it be possible (for a plugin) to manage device feature list and remove QAM256 from the reddo device. This way the plugin wouldn't have to actually do anything anymore when vdr is running, only fix this one time at plugin startup, and everything else would work 100% the same way before/after the driver is changed.
With a properly working driver you won't even need this at all ;-)
Even though this isn't the right way, but we aren't living in a perfect world :-)
I'm afraid that would only take the pressure off of fixing the driver.
2010/4/5 Klaus Schmidinger <Klaus.Schmidinger@tvdr.de mailto:Klaus.Schmidinger@tvdr.de>
Well, then the driver needs to make a finer distinction and *properly* set FE_CAN_QAM256.
vdr mailing list vdr@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
-
abbe normal -
Klaus Schmidinger -
michael.mauch＠gmx.de -
Petri Helin -
Teemu Rantanen -
Theunis Potgieter -
Timothy D. Lenz -
VDR User