Hi Andrea, On Wed, 21 Mar 2018 at 13:22 Andrea Mattana <[email protected]> wrote:
> Hi all, > > after many months of working on another board (SKA TPM for LFAA) today I'm > back to a project using the ROACH-1 but I figured out that my "longterm" > Matlab R2009b license expired, therefore I decided to try to install the > "new" toolflow for ROACH2. > > Following the instruction on the wiki web page and some hint hidden in the > mailing list I didn't get a working system. I have installed: > > Ubuntu 12.04 > Matlab 2012b > Xilinx 14.7 > dumped mlib_devel repository from git casper-astro > additional deprecated pbcores and patched the sh shell > finally modified the startsg.local (specially the jasper=ise!) with my > paths > > but Matlab crashed at the very beginning when launched from the system > generator (got also a message saying the the XILINX_DSP env doesn't match > the XILINX env, resetting...). > > After many attempts I tried also with more recently ubuntu release, I > found a mail from Jack saying that with the 14.04 it works. I got the same > behavior. > > At the end I decided to play with the repository version and with some > luck I got a working setup, haven't tested yet to call it "stable". > > My questions are: > > 1. which is the right repository hash version of a stable ROACH2 > mlib_devel toolflow? For istance the one used during the CASPER workshop > training. > You want the tip of the "roach2" branch at gitrhub.com/casper-astro/mlib_devel . Currently this is hash d6cd28b5d > > 2. Can be reported to the wiki page as well? I found that something > similar is done for the tutorials of the last workshop. > I suppose this should really be both on the wiki and in the mlib_devel readme. I'll put this on my todo list, but obviously if you feel like editing either the wiki (if you have an account) or the git readme (by raising a pull request) that'd be very much appreciated. > > 3. What are the differences (if exists) between the ska-sa and > casper-astro libs? > DON'T EVER USE THE SKA-SA LIBRARIES!!! Sorry, that's not very helpful :) More seriously, for the most part everything in the ska-sa library is periodically merged into the casper-astro fork. But the casper-astro fork has been deliberately split up to maintain a roach2 branch, and at least has the intention of supporting general casper users. The ska-sa branch is (obviously) centered around the work going on at SKA-SA / MeerKAT, and as such, may or may not be appropriate for your needs. For example, the roach2 branch on ska-sa is currently 194 commits behind casper-astro:roach2. If there is a feature / bugfix in ska-sa that is needed in casper-astro, you can flag this by either raising an issue on casper-astro's github, or (even better, if you have the git-fu) merging the feature into your local mlib_devel and submitting a pull request. > > 4. I would like to try to porting automatically the simulink design > developed by using the Xilinx 11.4 blockset to the 14.7, but I don't > remember what is the xilinx script to be used (xlmupdate? xlblablabla, > xlabracadabra?!?) :D > > I believe it is xlUpdateModel, or xlUpdateModelAbracadabraPlease There is also update_casper_blocks(...) which will replace any casper blocks in your design with fresh versions from the library, copying over the parameters. It's basically the automated version of deleting the block, and re-adding it back in. If you've changed mlib_devel versions, this can be super useful. update_casper_blocks(bdroot) %bdroot is a shortcut for the top-level of your model "block diagram root" will update *every casper library block" in your design. Be aware that it might take quite a while to run. > 5. I'm going to use the ADC-64 input (already used with the ROACH-1 but > using a oxford patched lib for the poliphase), any hint? > I would suggest you go to the latest mlib_devel, and then if we need to re-patch the polyphase FIR I can help you do that. Eventually I'm sure we'll want the feature in the current PFB-FIR anyway. > > Thanks in advance for the help! > > Hope to see you at the next CASPER workshop (if I'll find who will pay my > ticket :]) > > Hope to see you there Andrea. If not, I'm always happy to come visit you in Medicina :) Cheers Jack > > Cheers, > Andrea > > > > > > -- > Andrea Mattana > > I.N.A.F. - Istituto di Radioastronomia > Radiotelescopi di Medicina > Via Fiorentina, 3513 - 40059 Medicina (Bo) > Tel. 051/6965847 Fax. 051/6965810 > > -- > 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 post to this group, send email to [email protected]. > -- 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 post to this group, send email to [email protected].
