Maybe if you provide a more detailed description of what you're trying to
accomplish, we can help direct you toward a path to get there. Do you just
want to get a timestamp into your UHD application? Do you require a
timestamp for the custom logic in your RFNoC block on the FPGA?

On Mon, Nov 16, 2020 at 10:26 PM George Smith via USRP-users <
usrp-users@lists.ettus.com> wrote:

> Dear community,
>
> I use an RFNoC block and I would like to use it in combination with the
> GPS time functionallity on a N310.
> My aim is to create and write out a GPS precise timestamp - in the first
> step only once but maybe later more regularly.
> Until Inow I have a RFNoC block, where I put the controlling stuff (of the
> main task) inside *test1_block_ctrl_impl.cpp *which is called by
> *test1_impl.cc*
> My idea is to put the GPS time function inside *test1_block_ctrl_impl.cpp*
> / *test1_impl.cc *as well. In my opinion it is not necessary to create a
> completelty new self-written RFNoC block since it is a small function.
>
> In the following I describe two different approaches how I try to achieve
> my goal, but I would like to use the easiest one.
> Mainly I try to use the integrated GPSDO function
> So if you know a better solution, don't hesitate to post it. :)
>
>
> --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
>
> *(1)*
>
> Initially; I thought I could use code of the *sync_to_gps.cpp *in the
> examples folder of uhd, but I realized that therefore an instance of an
> *usrp* object is created.
>
> *uhd::usrp::multi_usrp::sptr usrp = uhd::usrp::multi_usrp::make(args);*
>
> Since I use a RFNoC block, previously a "*rfnoc*" object is constructed.
> (I dont know the exact name and where it is created). So, I have two
> instances which want to access at the same resource (board) at the same
> time. This is obviously not functioning.
> So when I run it in gnuradio I got the following error:
>
> *[ERROR] [RPC] Someone tried to claim this device again*
>
> In order to avoid this problem I would like to apply the needed methods
> for using the GPS time functionallity at the *rfnoc *object (so I do not
> need to create an object *usrp *).
> In the *sync_to_gps.cpp *example the following methods are used:
>
> *set_clock_source*
> *set_time_source*
> *get_mboard_sensor_names*
> *get_mboard_sensor*
> *set_time_next_pps*
> *get_time_last_pps*
>
> So my question is, where in the RFNoC framework is an instance of the *rfnoc
> *object created, so I can apply these methods to it.
>
> *Another thing:*
> If I start the built program *sync_to_gps*, I get after the
> initialization, a *seg fault *error. I think, it is because I use an
> RFNoC block.
> Is it true, and where is the core dump stored?
>
> *(2)*
>
> During my research, I noticed that there is an *RFNoC: Radio block. *Looking
> at the source code I see interesting methods in it:
>
> set_clock_source
> set_time_next_pps
> get_time_last_pps
>
> but not (*set_time_source, get_mboard_sensor_names, get_mboard_sensor *)
>
> So it seems to me that a GPS time function is not integrated in the 
> *rfnoc_radio_impl
> *class*.*
> This astonishes me, because I would assume that the RFNoC framework could
> perform at least the same tasks as an usual USRP (radio) one.
> I think about using (copying) the missing methods in the *rfnoc_radio_impl
> *class. Are there any problems doing this or why they have not been
> implemented so far?
> If I implement it, how can I recompile the block in order to get an
> updated *RFNoC: Radio* block.
>
>
> --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
>
> Used software:
> I use GNURadio 3.7.14.0, UHD 3.15 with Python 2.7.
> Hardware:
> N310
>
> Thanks for the help :)
>
> Kind regards
>
> George
>
>
>
> _______________________________________________
> USRP-users mailing list
> USRP-users@lists.ettus.com
> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
>
_______________________________________________
USRP-users mailing list
USRP-users@lists.ettus.com
http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com

Reply via email to