Hi Lawrence, The YAML file is the easy bit -- i.e., if you want some functionality (eg, a 100GbE core) that the toolflow supports for the Ultrascale+, adding in the pins to the YAML file is just a case of looking up the board documentation and copying the right pin numbers over. For example, I suspect if they're not already there you could add LED support in a couple of minutes.
However, if you want new functionality (eg, the RFSoC ADC) or functionality that might not be already supported by CASPER's libraries on the Ultrascale+ (eg, CASPER's 10GbE core probably won't work out the box) then the problem is much more involved, and requires a reasonable understanding of how the toolflow fits together, as well as writing some HDL. It's just possible that if you put in the pins into the ZCU111 yaml file required by the 1GbE core that you just might be able to talk to it over casperfpga using the tapcp protocol, which uses a tftp server running on microblaze. This works (I believe) with the VCU118. However, I suspect you still wouldn't be able to program the FPGA over casperfpga without some work. Register read/writes _should_ work without any modifications. However, even this requires that the instantiation template for 1GbE is the same on the ZCU111 as on the VCU118 and I wouldn't be surprised if there was some minor tweaking that was needed to the HDL to cope with small hardware implementation differences - for example, different reference clock frequencies / different choice of PHY chip / etc. I'd venture that a better way to support the ZCU in casperfpga is to use an OS image similar to the Red Pitaya. I.e., any old linux, with the tcpborphserver comms server included. In essence, I doubt that you'll be able to make the ZCU111 work with casperfpga without at least some digging into the guts of the toolflow to massage out some wrinkles. How much of a rabbit hole that turns out to be is hard to say without just charging in. I'd suggest you head over to the ZCU111 channel in slack where there are a bunch of folk using the board who might have already done some of the work you need. PM me for an invite to the slack workspace if you don't already have one. Cheers Jack On Mon, 6 Jul 2020 at 19:11, Lawrence Shing <[email protected]> wrote: > How much work is involved in creating the YAML file with more I/O > definitions? Or, does that open a can of worms to verify that the > toolflow is completely functional after the definitions are made? > > And, if the YAML file were complete, could casperfpga be used on the > testbed without requiring an OS? > > Lawrence > > On Thu, 2 Jul 2020 15:31:14 -0600 > Mitchell Burnett <[email protected]> wrote: > > Hi Lawrence, > > > > Yes, the platform file is sparse. Right now it is enough to get the > >test model working but would need to be extended to include other > >onboard clocks, GPIO, etc. > > > > No, PYNQ is not mandatory and is independent of what is supported > >in the tool flow right now for the ZCU111. However, to get > >communication with the MPSoC for things like casperfpga you would > >need an operating system and the PYNQ image is what I have used to > >install and run tcpborphsever. > > > > Best, > > > > Mitch > > > >> On Jul 2, 2020, at 3:24 PM, Lawrence Shing <[email protected]> wrote: > >> > >> I’ve got my hands on a ZCU111, and created a Centos box with the > >>CASPER toolflow. > >> > >> The zcu111.yaml file included in CASPER’s platforms directory seems > >>quite sparse. Is there an alternate YAML file that has more > >>definitions? > >> > >> In order to use the ZCU111, do I need to incorporate PYNQ to access > >>the capabilities of the testbed? > >> > >> Lawrence Shing > >> > >> -- > >> You received this message because you are subscribed to the Google > >>Groups "[email protected]" group. > >> To unsubscribe from this group and stop receiving emails from it, > >>send an email to [email protected]. > >> To view this discussion on the web visit > >> > https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/web-22441962%40lmsal.com > . > > > > -- > > You received this message because you are subscribed to the Google > >Groups "[email protected]" group. > > To unsubscribe from this group and stop receiving emails from it, > >send an email to [email protected]. > > To view this discussion on the web visit > > > https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/C37E33DD-7435-4EA6-804B-E927048F6181%40gmail.com > . > > -- > You received this message because you are subscribed to the Google Groups " > [email protected]" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/web-22444960%40lmsal.com > . > -- You received this message because you are subscribed to the Google Groups "[email protected]" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CAG1GKSnqK0X8oXZuYkz3kbE3LJffyEQFUF_1XYEbGG22i7LWHQ%40mail.gmail.com.
