omsocombb not receiving calls

Submitted by Harald Welte on March 21, 2016, 9:52 a.m.

Details

Message ID 20160321095208.GH10371@nataraja
State New
Headers show

Commit Message

Harald Welte March 21, 2016, 9:52 a.m.
Hi robert,

On Mon, Mar 21, 2016 at 11:43:20AM +0200, robert wrote:
> I can’t figure out a way to dynamically set the time slot, I’m always
> getting errors during compilation. When I solve these errors I will
> give you some feedback about any progress.

Feel free to post your patches  + error messages here.

The below patch doens't work for you?

Comments

robert March 21, 2016, 11:57 a.m.
Hi Harald,

The patch that you provided does’t give any error, but is it enough to choose the right time slot? Shouldn’t I use another variable instead of ’tn’ so that I can set the time slot related to the paging group?

Best regards,
Robert,

On Mar 21, 2016, at 11:52 AM, Harald Welte <laforge@gnumonks.org> wrote:

> Hi robert,
> 
> On Mon, Mar 21, 2016 at 11:43:20AM +0200, robert wrote:
>> I can’t figure out a way to dynamically set the time slot, I’m always
>> getting errors during compilation. When I solve these errors I will
>> give you some feedback about any progress.
> 
> Feel free to post your patches  + error messages here.
> 
> The below patch doens't work for you?
> 
> diff --git a/src/target/firmware/layer1/prim_rx_nb.c b/src/target/firmware/layer1/prim_rx_nb.c
> index 38c7b53..d186bff 100644
> --- a/src/target/firmware/layer1/prim_rx_nb.c
> +++ b/src/target/firmware/layer1/prim_rx_nb.c
> @@ -204,7 +204,7 @@ static int l1s_nb_cmd(__unused uint8_t p1, uint8_t burst_id,
> 		burst_id, tsc
> 	);
> 
> -	l1s_rx_win_ctrl(arfcn, L1_RXWIN_NB, 0);
> +	l1s_rx_win_ctrl(arfcn, L1_RXWIN_NB, tn);
> 
> 	return 0;
> }
> -- 
> - Harald Welte <laforge@gnumonks.org>           http://laforge.gnumonks.org/
> ============================================================================
> "Privacy in residential applications is a desirable marketing option."
>                                                  (ETSI EN 300 175-7 Ch. A6)
robert March 21, 2016, 12:43 p.m.
Hi,

when using the patch, the phone was not able to connect to any BTS.
I think the time slot number should be kept 0 but only changed when the phone is listening to the paging channel.

Best regards,
Robert,
 

On Mar 21, 2016, at 11:52 AM, Harald Welte <laforge@gnumonks.org> wrote:

> Hi robert,
> 
> On Mon, Mar 21, 2016 at 11:43:20AM +0200, robert wrote:
>> I can’t figure out a way to dynamically set the time slot, I’m always
>> getting errors during compilation. When I solve these errors I will
>> give you some feedback about any progress.
> 
> Feel free to post your patches  + error messages here.
> 
> The below patch doens't work for you?
> 
> diff --git a/src/target/firmware/layer1/prim_rx_nb.c b/src/target/firmware/layer1/prim_rx_nb.c
> index 38c7b53..d186bff 100644
> --- a/src/target/firmware/layer1/prim_rx_nb.c
> +++ b/src/target/firmware/layer1/prim_rx_nb.c
> @@ -204,7 +204,7 @@ static int l1s_nb_cmd(__unused uint8_t p1, uint8_t burst_id,
> 		burst_id, tsc
> 	);
> 
> -	l1s_rx_win_ctrl(arfcn, L1_RXWIN_NB, 0);
> +	l1s_rx_win_ctrl(arfcn, L1_RXWIN_NB, tn);
> 
> 	return 0;
> }
> -- 
> - Harald Welte <laforge@gnumonks.org>           http://laforge.gnumonks.org/
> ============================================================================
> "Privacy in residential applications is a desirable marketing option."
>                                                  (ETSI EN 300 175-7 Ch. A6)
robert Oct. 25, 2016, 7:51 a.m.
Hi,

Is it possible to have two or more timeslots receiving at the same time. It is necessary to have timeslot 0 at all time and another timeslot for paging.

Best regards, 

On Mar 21, 2016, at 11:52 AM, Harald Welte <laforge@gnumonks.org> wrote:

> Hi robert,
> 
> On Mon, Mar 21, 2016 at 11:43:20AM +0200, robert wrote:
>> I can’t figure out a way to dynamically set the time slot, I’m always
>> getting errors during compilation. When I solve these errors I will
>> give you some feedback about any progress.
> 
> Feel free to post your patches  + error messages here.
> 
> The below patch doens't work for you?
> 
> diff --git a/src/target/firmware/layer1/prim_rx_nb.c b/src/target/firmware/layer1/prim_rx_nb.c
> index 38c7b53..d186bff 100644
> --- a/src/target/firmware/layer1/prim_rx_nb.c
> +++ b/src/target/firmware/layer1/prim_rx_nb.c
> @@ -204,7 +204,7 @@ static int l1s_nb_cmd(__unused uint8_t p1, uint8_t burst_id,
> 		burst_id, tsc
> 	);
> 
> -	l1s_rx_win_ctrl(arfcn, L1_RXWIN_NB, 0);
> +	l1s_rx_win_ctrl(arfcn, L1_RXWIN_NB, tn);
> 
> 	return 0;
> }
> -- 
> - Harald Welte <laforge@gnumonks.org>           http://laforge.gnumonks.org/
> ============================================================================
> "Privacy in residential applications is a desirable marketing option."
>                                                  (ETSI EN 300 175-7 Ch. A6)
Sylvain Munaut Oct. 25, 2016, 8:35 a.m.
Hi,

> Is it possible to have two or more timeslots receiving at the same time. It is necessary to have timeslot 0 at all time and another timeslot for paging.

Can the hw do it : yes

Can the software do it : no.
Is it really necessary: no it's not. For each frame you'll need either
ts0 or tsX but not both at once in the same frame.

Cheers,

    Sylvain
robert Oct. 25, 2016, 8:42 a.m.
Hi Sylvain,

Thanks for explaining. I couldn’t find a way to decide which ts to choose so I thought about having multiple ts at the same time.
So I understand that I should stick with only one ts.

Best regards,


On Oct 25, 2016, at 11:35 AM, Sylvain Munaut <246tnt@gmail.com> wrote:

> Hi,
> 
>> Is it possible to have two or more timeslots receiving at the same time. It is necessary to have timeslot 0 at all time and another timeslot for paging.
> 
> Can the hw do it : yes
> 
> Can the software do it : no.
> Is it really necessary: no it's not. For each frame you'll need either
> ts0 or tsX but not both at once in the same frame.
> 
> Cheers,
> 
>    Sylvain
Sylvain Munaut Oct. 26, 2016, 5:10 a.m.
> Thanks for explaining. I couldn’t find a way to decide which ts to choose so I thought about having multiple ts at the same time.
> So I understand that I should stick with only one ts.

Which TS to listen on for paging message will be dependent on the
paging group your sim card is in. It's a formula that's based on your
IMSI and the cell configuration. It's described somewhere in the spec,
GSM 04.08 should have at least reference to it if you search for
"paging group".

Curently we don't support configs that have a CCCH on any other ts than 0.

Cheers,

    Sylvain
robert Oct. 26, 2016, 6:39 a.m.
Hi,

On Oct 26, 2016, at 8:10 AM, Sylvain Munaut <246tnt@gmail.com> wrote:

>> Thanks for explaining. I couldn’t find a way to decide which ts to choose so I thought about having multiple ts at the same time.
>> So I understand that I should stick with only one ts.
> 
> Which TS to listen on for paging message will be dependent on the
> paging group your sim card is in. It's a formula that's based on your
> IMSI and the cell configuration. It's described somewhere in the spec,
> GSM 04.08 should have at least reference to it if you search for
> "paging group”.
> 
Yes I understand, my problem is in modifying the osmocombb code so that it listens to other TS. I have tried to change the code to make it work but with no success. I only need some instructions on where to start and what should be done.

Best regards,
Tomcsányi, Domonkos Oct. 26, 2016, 7:07 a.m.
As far as I remember the calypso-trx code was capable of receiving multiple timeslots, maybe check out that code (jolly/trx if I'm right) and watch Sylvain's talk about it.
https://youtu.be/B1od4x9L3t4

Cheers,
Domi

2016. okt. 26. dátummal, 8:40 időpontban robert <robert.steve07@gmail.com> írta:

> Hi,

> 

> On Oct 26, 2016, at 8:10 AM, Sylvain Munaut <246tnt@gmail.com> wrote:

> 

>>> Thanks for explaining. I couldn’t find a way to decide which ts to choose so I thought about having multiple ts at the same time.

>>> So I understand that I should stick with only one ts.

>> 

>> Which TS to listen on for paging message will be dependent on the

>> paging group your sim card is in. It's a formula that's based on your

>> IMSI and the cell configuration. It's described somewhere in the spec,

>> GSM 04.08 should have at least reference to it if you search for

>> "paging group”.

>> 

> Yes I understand, my problem is in modifying the osmocombb code so that it listens to other TS. I have tried to change the code to make it work but with no success. I only need some instructions on where to start and what should be done.

> 

> Best regards,

>
Sylvain Munaut Oct. 26, 2016, 7:32 a.m.
On Wed, Oct 26, 2016 at 9:07 AM, Tomcsányi, Domonkos <domi@tomcsanyi.net> wrote:
> As far as I remember the calypso-trx code was capable of receiving multiple
> timeslots, maybe check out that code (jolly/trx if I'm right) and watch
> Sylvain's talk about it.
> https://youtu.be/B1od4x9L3t4

That's only to act as a BTS. This screws with the DSP code and breaks
all normal phone code function. You loose all the channel
coding/decoding with this and have to do all that on the host.


Cheers,

   Sylvain

Patch hide | download patch | download mbox

diff --git a/src/target/firmware/layer1/prim_rx_nb.c b/src/target/firmware/layer1/prim_rx_nb.c
index 38c7b53..d186bff 100644
--- a/src/target/firmware/layer1/prim_rx_nb.c
+++ b/src/target/firmware/layer1/prim_rx_nb.c
@@ -204,7 +204,7 @@  static int l1s_nb_cmd(__unused uint8_t p1, uint8_t burst_id,
 		burst_id, tsc
 	);
 
-	l1s_rx_win_ctrl(arfcn, L1_RXWIN_NB, 0);
+	l1s_rx_win_ctrl(arfcn, L1_RXWIN_NB, tn);
 
 	return 0;
 }