Further steps. I downloaded the repository (using ZIP, git-lfs fails on top_adm_pcie_9h7.xpr.zip) at https://github.com/casper-astro/mlib_devel/tree/m2019a It is supposed to work with Matlab2019a and Vivado 2020.2, but the test design for the zcu111 requires Matlab R2020b. I opened it in 2020b and exported in 2019a. The new design opens with 2019a, it compiles in sysgen but then jasper returns an error:
top: /home/comore/casper3/mlib_devel-m2019a/test_zcu111_r2019a/top.v axi4_lite/*.vhd ['/home/comore/casper3/mlib_devel-m2019a/jasper_library/hdl_sources/axi4_lite/axi4lite_pkg.vhd' /home/comore/casper3/casper_venv/bin/python: symbol lookup error: /home/comore/casper3/casper_venv/bin/python: undefined symbol: XML_SetHashSalt Il giorno martedì 4 gennaio 2022 alle 23:45:28 UTC+1 Mitch Burnett ha scritto: > Sorry, I’m referring to this branch from the capser-astro mlib_devel > repository: https://github.com/casper-astro/mlib_devel/tree/m2019a > > Try and give this one a shot. > > Mitch > > On Jan 4, 2022, at 2:16 PM, Gianni Comoretto <[email protected]> wrote: > > Sorry but I see the following branches: > > > rfsocs/m2019a-v2019.1 > rfsocs/merge-casper-astro/m2019a > > Which one are you referring? The first looks like using Matlab2019a + > VIvado 2019.1. Is it correct? > The LFS file requires an authorization, which failed. Do I have to specify > some credentials in retrieving git archive? > > Gianni > > Il 04/01/22 19:26, Mitch Burnett ha scritto: > > Hi Gianni, > > > Sorry you are having issues getting started. Thanks for taking the time to > go through the material and identify the problem areas where the > documentation can be updated/improved/clarified. New year, new goals :) > > > When I wrote the tutorial our BYU gitlab repository is where I was > pointing interested folks to get started. I am still pushing to and staging > current work and changes to that BYU gitlab repo (DAC, multi-tile/board > synch, etc.) but for now the support that is covered in the tutorials has > been merged back to casper-astro/mlib_devel/m2019a. The issue you are > having is a Git Large File Storage issue. I am not sure how you may be > getting that issue with `rfsocs/devel` I thought I had fixed that and have > tested building from my gitlab instance several times from different > machines with clean versions of the repo. But, looks like I need to poke > around some more and is probably specific to my Gitlab instance, you > shouldn’t have that problem if you clone from mlib_devel GitHub. I would > try that first. > > > As you noticed there are typo’s in the documentation the branch name > `rfsoc/devel` >> `rfsocs/devel`. This `rfsocs/devel` is where everything > will go before going back to casper-astro. > > > There are also several stale branches, the `rfsocs/zcu216` is stale, > R2020b will work with rfsocs but turns out what I didn’t realize is that > the main support for mlib_devel (like the FFT) is on m2019a and so I need > to roll back and export the R2020b blocks to R2019a to match > `casper-astro/mlib_devel/m2019a. Just need to cleanup and remove those > stale branches. > > > So to recap and answer your question more directly: > > > Start with trying `m2019a` from `casper-astro/mlib_devel`, and the > remaining requirements on that getting started with rfsoc page are correct: > > > Compatible Linux host operating system (tested on RHEL 7.9, 8.4 and Ubuntu > 18.04 LTS, 20.04 LTS) > > Vivado 2020.2 > > Matlab 2019a (with Simulink) > > Python 3 environment > > > If `m2019a` still gives you problems let me know and I can keep helping > debug this. > > > Best, > > > Mitch > > > On Jan 4, 2022, at 1:52 AM, Gianni Comoretto <[email protected]> wrote: > > > I have a ZCU111 board and I am trying to use it with Casper. > > I followed the instructions in the "getting started with RFSOC" page. > > > > https://casper-toolflow.readthedocs.io/projects/tutorials/en/latest/tutorials/rfsoc/tut_getting_started.html > > > I was able to compile and configure the board, so it can be accessed e.g. > with SSH at casper@board_ip_address. I can connect to the board using > casperfpga as described in the tutorial. > > > I cloned the git repository > https://gitlab.ras.byu.edu/alpaca/casper/mlib_devel but I am confused on > the requirements on OS, Python, and tools. > > > I cannot check out the branch rfsoc/devel (not existing). > > > Branch rfsocs/devel fails to check out as the LFS file > jasper_library/template_projects/pcie/top_adm_pcie_9h7.xpr.zip fails to > load (Repository or object not found). Same happens with rfsoc/devel/dac > > > Branch rfsocs/zcu216 uses Matlab R2020b, which is not supported by Vivado > 2020 > > > Branch master, or zcu111, is bound to python2. I use linux Ubuntu 20, > which by default does not support py2. yaml calls are outdated and scripts > appear very old anyway. > > > Could I have a list of which git branch and tool version (Python, Vivado, > Matlab) to use? > > > Gianni Comoretto > > > -- > > 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/a8574ca1-4657-32c4-a06f-8baf966c7baa%40inaf.it > . > > > -- > 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/10ca2256-0684-f001-396f-80300d9280ca%40inaf.it > . > > -- 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/3327db1a-d61e-4738-91dd-0c141c1eef03n%40lists.berkeley.edu.
