That is interesting. Not sure what it tells us, but interesting. There is a
potentially related issue posted on github:
https://github.com/EttusResearch/uhd/issues/514. The author suggested
bumping MIN_DUR to 100us. If you're able to modify UHD, then that might be
something to try.

--M

On Fri, Mar 21, 2025 at 12:01 PM <[email protected]> wrote:

> Hi Martin
>
> I am still trying to investigate this issue, recreating it is difficult as
> it can take long time.
>
> However, on a system that crashed I observed a thread “uhd_ctrl_ep0001”
> specifically its CPU utilisation (I have isolated this to use a single CPU
> and not share it). Normally it is below ~20%, (when first starting our
> application is is < 10%) once there is a failure (as per the discussion
> here), this thread has a higher utilisation (not 100%), around 30-40%. When
> kill our process, and restart, the CPU utilisation is still high from the
> start. The only way to recover is to power off the USRP-2974 and on again.
>
> Kind regards
>
> Marino
> _______________________________________________
> USRP-users mailing list -- [email protected]
> To unsubscribe send an email to [email protected]
>
_______________________________________________
USRP-users mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to