usrp-users@lists.ettus.com
Discussion and technical support related to USRP, UHD, RFNoC
View all threadsusing rfnoc_graph and usrp
Hello,
I am currently getting the following error, but I am lost as to how to work around it…
[ERROR] [RPC] Someone tried to claim this device again (From: 192.168.10.1)
[WARNING] [MPM.RPCServer] Someone tried to claim this device again (From: 192.168.10.1)
It appears the problem comes from instantiating an object uhd::rfnoc_graph, then, when this object already exists, I try to create a uhd::usrp::multi_usrp object. For example…
auto graph = uhd::rfnoc::rfnoc_graph::make(args);
uhd::usrp::multi_usrp::sptr usrp = uhd::usrp::multi_usrp::make("addr0=192.168.10.2");
It appears that trying to create both causes the error above to happen. However, I want to be able to create an rfnoc_graph and use multi_usrp to get and set the TOD on the ettus N321. How exactly can I work around this? I saw the rfnoc_graph has a function called “synchronize_devices()”, however, it does not seem this function can get and set the TOD as I need.
Thanks,
Joe
On 24/02/2023 11:05, jmaloyan@umass.edu wrote:
Hello,
I am currently getting the following error, but I am lost as to how to
work around it…
|[ERROR] [RPC] Someone tried to claim this device again (From:
192.168.10.1)|
|[WARNING] [MPM.RPCServer] Someone tried to claim this device again
(From: 192.168.10.1)|
It appears the problem comes from instantiating an object
uhd::rfnoc_graph, then, when this object already exists, I try to
create a uhd::usrp::multi_usrp object. For example…
|auto graph = uhd::rfnoc::rfnoc_graph::make(args);|
|uhd::usrp::multi_usrp::sptr usrp =
uhd::usrp::multi_usrp::make("addr0=192.168.10.2");|
It appears that trying to create both causes the error above to
happen. However, I want to be able to create an rfnoc_graph and use
multi_usrp to get and set the TOD on the ettus N321. How exactly can I
work around this? I saw the rfnoc_graph has a function called
“synchronize_devices()”, however, it does not seem this function can
get and set the TOD as I need.
Thanks,
Joe
USRP-users mailing list --usrp-users@lists.ettus.com
To unsubscribe send an email tousrp-users-leave@lists.ettus.com
You can't have two objects that "own" sessions to the device.
Mixing-and-matching multi_usrp and rfnoc objects isn't,
near as I can tell, supported.
However, your use case of wanting to do "housekeeping" functions like
"set_time_next_pps()" is legitimate, and I'm not
sure how that is supported via RFNoC. Perhaps one of the RFNoC
gurus on here can shed some light?
Although, looking at the rfnoc_rx_to_file.cpp example, you can:
graph->get_mb_controller(0)->set_clock_source(ref);
So there's a "mb_controller" class that looks like it takes the same
"stuff" you'd expect from a multi_usrp.
On 24/02/2023 11:05, jmaloyan@umass.edu wrote:
Hello,
I am currently getting the following error, but I am lost as to how to
work around it…
|[ERROR] [RPC] Someone tried to claim this device again (From:
192.168.10.1)|
|[WARNING] [MPM.RPCServer] Someone tried to claim this device again
(From: 192.168.10.1)|
It appears the problem comes from instantiating an object
uhd::rfnoc_graph, then, when this object already exists, I try to
create a uhd::usrp::multi_usrp object. For example…
|auto graph = uhd::rfnoc::rfnoc_graph::make(args);|
|uhd::usrp::multi_usrp::sptr usrp =
uhd::usrp::multi_usrp::make("addr0=192.168.10.2");|
It appears that trying to create both causes the error above to
happen. However, I want to be able to create an rfnoc_graph and use
multi_usrp to get and set the TOD on the ettus N321. How exactly can I
work around this? I saw the rfnoc_graph has a function called
“synchronize_devices()”, however, it does not seem this function can
get and set the TOD as I need.
Thanks,
Joe
USRP-users mailing list --usrp-users@lists.ettus.com
To unsubscribe send an email tousrp-users-leave@lists.ettus.com
i found that I needed to create a timekeeper, within the mb_controller. i.e
uhd::rfnoc::mb_controller::timekeeper::sptr tk = graph->get_mb_controller(0)->get_timekeeper(0);
This seems to do the trick. Im now getting a late error, but it is probably another kind of issue.