Hi Lean, The missing links you see indicate that the block didn't finish drawing properly - or to say it failed to run the initialization script. If you look at the various *_init.m functions in the casper_library, you can see they are the functions that *draw* the block according to given parameters. If the script encounters some error before it can complete, we are left with half-drawn blocks. I'm actually a bit confused about the library version that you were referring to --- 7.4, 7.2 --- I wasn't aware that casper_library had numbered release. (!)
For this issue I recommend updating your mlib_devel to the current casper-astro/mlib_devel on github. (Link<https://github.com/casper-astro/mlib_devel>) This should be easy (just replace your old mlib_devel folder with this new one, but do remember to get the pcores needed for xilinx 13<https://casper.berkeley.edu/wiki/MSSGE_Toolflow_Setup#13.x_Development.2FProduction>). If the problem persists, maybe we can take a closer look at your parameter settings and figure out the problem - it might be a bug in the library! Or if you'd like to try something fun, like debugging the error in your current set-up, you can try this (you don't have to do this, nor is this any standard Casper-style trick, just something I would do to debug a block): > 1. add the two *.m files in the attachment to your matlab path > 2. left click on your fft_wideband_real block in your design > 3. switch to matlab command window, and type >>> run_init(gcb) I expect that you would see some error message that tells you what goes wrong at which line of which initialization script. Then we can look into it. (I still hope that updating the mlib_devel would solve the problem, though). Best, Hong On Sun, Nov 18, 2012 at 7:32 AM, shu.lean <shu.l...@163.com> wrote: > Hi Hong, > It still didn't work. It seems that all blocks of FFT have the same > problems. When I looked under mask of the block fft_biplex_real_4x. I saw > disabled link (seeing the attachment). I am confused. > The version of casper_library I am using is 7.4 and that of > casper_library_ffts is 7.2. > > Thanks > Lean > > > > > > > > At 2012-11-18 06:57:35,"Hong Chen" <chen.hong.1...@gmail.com> wrote: > > Hi Lean, > > The version of various components of the development environment can be > quite tricky to deal with - many of them are not back-ward compatible, > unfortunately (including Matlab, Xilinx, and Casper libraries). But your > Matlab & Xilinx versions look fine to me. Looking at the error message you > got, I suspect it's a Casper library related problem - which version of > CASPER library (mlib_devel) are you using? If you try a fresh copy of > fft_wideband_real from the library browser (drag the fft_wideband_real from > the library browser and drop it into your design, change its parameters to > those that you want, and delete the old one), and do a ctrl-D, do you still > see this error? > > Thanks, > Hong > > On Fri, Nov 16, 2012 at 7:55 AM, shu.lean <shu.l...@163.com> wrote: > >> >> Hi Hong， >> >> Thanks for your suggestion. >> >> Now I have a question about the version of the matlab simulink . Is any >> version newer than the casper libraries‘ Ok？ or I have to use the same >> version that designed the libraries？ I get xinlinx 13.4 /system generator >> 13.4 and matlab 7.12/simulink 7.7 ,but some blocks in the casper library >> are unavailable（e.g FFT, when I start the simulation , it shows >> block error :Error in ' >> FFT_undone/fft_wideband_real/fft_biplex_real_4x0/biplex_core/fft_stage_3/butterfly_direct/twiddle_general_4mult': >> Initialization commands cannot be evaluated >> model error :There is no block named >> 'casper_library_ffts_twiddle_coeff_gen/coeff_gen' >> but I see a block named coeff_gen in the library. >> >> >> At 2012-11-15 13:46:50,"Hong Chen" <chen.hong.1...@gmail.com> wrote: >> >> Hi Lean, >> >> I'm the current developer/maintainer of the correlator library. My >> apologize for the empty content, but we (I mean, I) have been having some >> issues with the library versions and compatibility in general, and most of >> the blocks in this library are newly designed and are all kind of in rough >> condition, so I don't think this library is ready to be released at this >> moment. I will keep you posted on the progress if you'd like that. >> >> If you are interested, there are various CASPER correlator design >> projects that are available, you might want to check out the wiki page for >> Casper correlator collaboration. Some projects are available on github >> (e.g. PAPER in Berkeley, and some Oxford projects, I believe), some are >> available on SVN repositories (e.g. some earlier South African correlator >> developments). You can probably find links on their project webpages, or >> email specific designers for information. There are a lot of resources for >> correlator in Casper, but unfortunately they are kind of scattered and >> lacking documentation. I'm working on making the library as general purpose >> as possible, and trying to do some relatively detailed documentation. But >> again, that's work in progress. >> >> If you have some specific questions with regard to correlator designs, >> please do not hesitate to email casper list. There are lots of correlator >> designers around that might be able to help you. Just when it comes to this >> correlator library, it's just me for now. Sorry for the delay (I intended >> to do first release in September), but I'm working on that. >> >> Regards, >> Hong >> >> On Fri, Nov 9, 2012 at 7:23 PM, shu.lean <shu.l...@163.com> wrote: >> >>> Dear Casperites, >>> I am a graduate student . And I am new in this field. I want to know >>> where can I get the casper library since I can't get any sources from the >>> link https://github.com/casper-berkeley/correlator_library . Would >>> you send me a link or an attachment? >>> >>> Thank you very much. >>> >>> lean >>> >>> >>> >> >> >> > > >
get_mask_setting.m
Description: Binary data
run_init.m
Description: Binary data
