Jonathon, Wade, Regarding the Replay block controller and the recent change on 'master' to Drop property, action, & mtu forwarding, there is one other modification that I think makes sense. The Replay block is the only block controller that implements "issue_stream_cmd" yet doesn't actually register this function as an action handler for the stream_cmd action. What this means is that if I connect an rx_streamer to a Replay block output (Play port) and call issue_stream_cmd() on the rx_streamer, the Replay block controller does not respect this command. In fact, on UHD 4.1, it actually forwards it along the path causing real trouble. On 'master', it just drops the command but does nothing (which is better). Take a look at the null block controller or DDC block controller to see that these controllers register the stream_cmd action. Let me know what you think. Thanks. Rob
On Thu, Feb 10, 2022 at 5:01 PM Jonathon Pendlum <jonathon.pend...@ettus.com> wrote: > Hi Rob, > > Yup, that should fix the issue. > > Jonathon > > On Thu, Feb 10, 2022 at 4:56 PM Rob Kossler <rkoss...@nd.edu> wrote: > >> Thanks Jonathon! >> This sure seems like a smoking gun. It doesn't seem like this is an >> auto-generated file so it seems we could manually modify it to include >> ports 2 and 3 in the same fashion as the N310 and then rebuild. Is this >> true? >> Rob >> >> On Thu, Feb 10, 2022 at 4:22 PM Jonathon Pendlum < >> jonathon.pend...@ettus.com> wrote: >> >>> It looks like the problem is that while there is a 4 port interconnect >>> available, only ports 0 and 1 are hooked up: >>> https://github.com/EttusResearch/uhd/blob/2c7ce2dbf72414b64f8a477be614e23bc12f086d/fpga/usrp3/top/e320/e320_core.v#L1050 >>> >>> This actually brings up a design suggestion: the Replay Block could have >>> it's own internal AXI4 interconnect that scales based on NUM_PORTS. I've >>> done this myself in a custom RFNoC block using the Xilinx AXI4 Interconnect >>> IP and it worked out well. >>> >>> Jonathon >>> >>> On Thu, Feb 10, 2022 at 3:33 PM Wade Fife <wade.f...@ettus.com> wrote: >>> >>>> I would start by double checking the YAML. For example, make sure the >>>> MEM_ADDR_W parameter is correct (for E320 it should be 31, for 2 GiB) and >>>> make sure each in/out port is connected the way you want. >>>> >>>> If you share the YAML with me, I'm happy to take a look to see if >>>> anything jumps out at me. >>>> >>>> Thanks, >>>> >>>> Wade >>>> >>>> On Thu, Feb 10, 2022 at 10:34 AM Rob Kossler <rkoss...@nd.edu> wrote: >>>> >>>>> Thanks Wade, >>>>> I am helping Ofer Saferman with an issue with the E320 and a 4-port >>>>> replay block. Apart from your response (& Jonathon's response) indicating >>>>> that data rates should not be an issue, I have also come to the same >>>>> conclusion by trying some tests. After these tests, the issue now seems >>>>> to >>>>> be that Replay ports 0 and 1 work as expected, but Replay ports 2 and 3 do >>>>> not. I know that a 4-port Replay block works fine on an N310 because I >>>>> use >>>>> it often. So, I'm wondering why we have this issue on the E320. Perhaps >>>>> the E320 yml file is wrong - I am still waiting to take a look at this. I >>>>> have also requested that Ofer run the stock example >>>>> "rfnoc_replay_samples_from_file" and use the --replay_chan option to prove >>>>> that ports 0 and 1 work fine but ports 2 and 3 do not. >>>>> >>>>> Anyway, if you have any suggestions, I'd love to hear them. >>>>> Rob >>>>> >>>>> >>>>> On Thu, Feb 10, 2022 at 11:17 AM Wade Fife <wade.f...@ettus.com> >>>>> wrote: >>>>> >>>>>> The E320's DRAM is pretty fast. It should have no problem keeping up >>>>>> for your use case. >>>>>> >>>>>> Wade >>>>>> >>>>>> On Thu, Feb 10, 2022 at 1:56 AM Jonathon Pendlum < >>>>>> jonathon.pend...@ettus.com> wrote: >>>>>> >>>>>>> Hi Rob, >>>>>>> >>>>>>> As long as the DRAM can keep up throughput wise, you should be fine >>>>>>> in that configuration. I think the E320 has a BIST you can run that >>>>>>> reports >>>>>>> the throughput. >>>>>>> >>>>>>> Jonathon >>>>>>> >>>>>>> On Wed, Feb 9, 2022 at 3:29 PM Rob Kossler <rkoss...@nd.edu> wrote: >>>>>>> >>>>>>>> Hi, >>>>>>>> I am wondering if there are any data rate restrictions for using >>>>>>>> the Replay block on the E320. I have a 4-port Replay block for >>>>>>>> simultaneously playing two streams to the 2-port Radio and capturing >>>>>>>> two >>>>>>>> streams from the 2-port Radio. If the master_clock_rate is equal to >>>>>>>> the >>>>>>>> sample rate, does this imply that I will have a data throughput issue? >>>>>>>> Rob >>>>>>>> _______________________________________________ >>>>>>>> USRP-users mailing list -- usrp-users@lists.ettus.com >>>>>>>> To unsubscribe send an email to usrp-users-le...@lists.ettus.com >>>>>>>> >>>>>>> _______________________________________________ >>>>>>> USRP-users mailing list -- usrp-users@lists.ettus.com >>>>>>> To unsubscribe send an email to usrp-users-le...@lists.ettus.com >>>>>>> >>>>>>
_______________________________________________ USRP-users mailing list -- usrp-users@lists.ettus.com To unsubscribe send an email to usrp-users-le...@lists.ettus.com
