Re: [casper] How does one go about emphasizing timing over placement?

Hi Randy The options used in mapping, place and route etc are found in fast_runtime.opt in the etc folder in the XPS_IBOB_base folder containing the EDK project created by the CASPER toolflow. You will find a bunch of options that you can play with here. Save the changes and then re-run the

Re: [casper] Casper library basic setup

-- *From:* Andrew Martens [mailto:martens.and...@gmail.com] *Sent:* Tuesday, 4 November 2008 6:07 PM *To:* Cheng, Wan (ATNF, Marsfield) *Cc:* casper@lists.berkeley.edu *Subject:* Re: [casper] Casper library basic setup Hi Wan I had a look at your design and noticed the following

Re: [casper] about the shift input of FFT

Hi Wan The FFT applies gain to any repeating input signal. You can view this happening with the following MATLAB commands; x = 0:1:32; y = sin(4*x*2*pi/32); plot(abs(fft(y,32))); You will see that the output has spikes of height 16 instead of 1 which you would expect from a sinusoid of

Re: [casper] about the shift input of FFT

the FFT module has. I just give it 32 bits width '1'. Thanks Wan -- *From:* Andrew Martens [mailto:martens.and...@gmail.com] *Sent:* Monday, 22 June 2009 4:23 PM *To:* Cheng, Wan (ATNF, Marsfield) *Cc:* casper@lists.berkeley.edu *Subject:* Re: [casper] about

Re: [casper] about the shift input of FFT

high with the data that caused the overflow but a lot earlier and is to used as an indicator that an overflow occured 'somewhere' . I will implement the delays properly when I have more time again so that it lines up correctly. Regards Andrew 2009/6/22 Andrew Martens martens.and...@gmail.com Hi Wan

Re: [casper] about the shift input of FFT

-- *From:* Andrew Martens [mailto:martens.and...@gmail.com] *Sent:* Monday, 22 June 2009 8:48 PM *To:* Cheng, Wan (ATNF, Marsfield) *Cc:* casper@lists.berkeley.edu *Subject:* Re: [casper] about the shift input of FFT Hi Wan et al The 'convert with overflow indication

Re: [casper] about the shift input of FFT

with overflow indication'. Anyway, I could read the FFT output correctly now. Thanks again. Wan -- *From:* Andrew Martens [mailto:martens.and...@gmail.com] *Sent:* Monday, 22 June 2009 8:48 PM *To:* Cheng, Wan (ATNF, Marsfield) *Cc:* casper@lists.berkeley.edu

Re: [casper] questions about tut4 from ref_designs_tutorials (wide-band version of pocket correlator)

Hi Note that there has already been work done in making it possible to use Linux-64bit as the operating system of choice for the 11.1 toolflow (see David George's email from a few days ago). This should solve these memory issues as well as others (stability, block initialisation speed etc). The

Re: [casper] Questions on porting to ROACH

Hi John The FFT is a good place to optimise for various things. The ROACH has plenty of multipliers in comparison to the iBOB so be sure to optimise for logic in the FFT unless you are very short of multipliers and have logic to spare. Complex multiplication can be done with a lot of adders and 3

Re: [casper] Capser FFT output level with Matlab FFT output level

Hi Wan I am unable to understand the details of your question as there is not enough information available in your email (axis labels, are you dividing the power or voltage values by 4096 etc). However, I shall explain the difference between the CASPER FFT and Matlab FFT with respect to gain

Re: [casper] Timing error

Hi John Some pointers on improving timing generally in ROACH. Decrease the space your design occupies. Implement delays on data inputs to multiplier cores in behavioural logic as these are absorbed by the DSP48E core. Implement large adders and counters in DSP48Es. Use Truncation as your

[casper] Simulink/Xilinx integration

-- Forwarded message -- From: Andrew Martens martens.and...@gmail.com Date: 16 March 2010 14:51 Subject: Re: [casper] Simulink/Xilinx integration To: Steve Maher stephen.f.ma...@nasa.gov Hi Steve As Jason has mentioned, going with Linux is choosing the path less traveled within

Re: [casper] casper Digest, Vol 30, Issue 1

Hi Chao-Te To save on DSP48Es in the FFT try the following Un-tick the 'Use DSP48s for adders' (uses the adders in DSP48 blocks to implement the twiddle stage in the FFT). Should save you 4 DSP48Es per FFT stage. Set the 'Optimization Target' to 'Multipliers' (reduces the use of multipliers in

Re: [casper] mlib_devel_10_1 and simulink 7.4

Hi Dave It may be time to copy our libraries to an mlib_devel_11_1 revision and continue from there. ROACH2 uses Virtex6 and the 10.x and earlier tools do not support it. Disadvantages are that a lot of library maintainers will be working in mlib_devel_11_1 and bug fixes, changes etc may not make

Re: [casper] FFT drawing issues (wideband real)

Hi all Please search the mail archives, there has been a thread discussing this over the past few months. It is possible to manually turn off specific warnings in Matlab and I have done so in my startup.m file so that I am not flooded with unnecessary warnings. The thread discusses how to do

Re: [casper] Fixed FFT block status

FWIW, I've noticed another related problem in the pfb_fir and pfb_fir_real blocks. The mask callbacks for the TotalTaps parameter of these blocks assign and use a variable named size. Unfortunately, this seems to wind up in the global workspace and overrides the built-in size function. This

Re: [casper] The delay in pfb_real_fir

Hi Wan The libraries available in subversion are out of date and may contain some bugs as changes were being made before the move across to git. The CASPER DSP and XPS libraries have moved to git repositories so you need to install a git client to access the latest stuff. (apt-get install git in

Re: [casper] FFT on ROACH

Hi all The casper_library.mdl file seems to be misaligned relative to the scripts. The scripts are trying to set mask parameters that don't exist and are failing. I may be wrong but the casper git repository seems to be broken at the moment. Use the following to get a version that should work

Re: [casper] Optimum PFB / FFT settings to meet timing

Hi To add to Jason's suggestions: Conversion latency = 3 The documentation on the FFT blocks at http://casper.berkeley.edu/wiki/Block_Documentation is quite recent and contains some suggestions and explanations. Good luck Regards Andrew

Re: [casper] iADC clocking woes in 11.4 toolflow

Hi Danny It looks like you may have run into a similar problem that we (and GMRT) experienced a while ago. Basically, the clock and data coming from the iADC were misaligned (off by 90 degrees if I recall correctly) and weird data like this occurred at certain frequencies (and not at others).

Re: [casper] Simulation/Synthesis trouble

Hi Daniel Welcome to CASPER! ... but then I receive a lot of messages of the type: Warning: casper_library/Accumulators/qdr_vacc/acc_ctrl/rd_addr is a parameterized link. To view, discard, or propagate the changes for this link, use the Link Options menu item In load_system at 41 In

Re: [casper] Optimum PFB / FFT settings to meet timing

Hi all Just a follow up email relating to Danny's email from a while ago. * The pfb_fir_real caused timing issues due to its adders and its convert. Changing the adder latency to 4, then breaking the mask and init script, I manually went into the adder blocks, changed the implentation from

Re: [casper] iBOB freezing

Hi Laura I have seen strange iBOB behaviour due to an incorrectly plugged in CX4 cable. It must have been bumped at some point and, although still plugged in, was causing strange behaviour. The iBOB would report being successfully programmed but the LEDs would do strange things. Not sure if this

Re: [casper] wideband conversion and correlation

Hi Jonathan To start we are looking closely at the FPGA resource utilization of large PFBs. Something that probably is common knowledge amongst those experienced in FX correlator design is that the demux factor drives the utilization much faster than the size of the PFB. In that sense

Re: [casper] fft overflow in tut4 and more questions

Hi Mandana fft_shift is set to 0x by default in tutorial 4 poco design. Any other suggestions? When I monitor my ADC input levels, they are about 4.6 bit used. Ok, well that should rule out overflows from the FFT. I assume that there is no clipping with terminated inputs? At what

Re: [casper] fft overflow in tut4 and more questions

get to the ADC. dan On 03/01/2011 10:44 PM, Andrew Martens wrote: Hi Mandana fft_shift is set to 0x by default in tutorial 4 poco design. Any other suggestions? When I monitor my ADC input levels, they are about 4.6 bit used. Ok, well that should rule out

Re: [casper] FFT twiddle factor: bug in twiddle_coeff_1?

Hi all On 30 March 2011 16:14, Jack Hickish jackhick...@gmail.com wrote: Hi Casperites, After a long day of debugging an FFT spectrum yesterday, I got down to basics and compared the output of the 8-pt direct fft block with matlab's fft function. When I used random numbers as the inputs it

[casper] commit eade890: rcs block added, revision_control block removed

Hi all Changes have eventually been made to the block allowing revision control information to be embedded in your design. The old revision_control block in Misc has been replaced with the rcs block. This allows revision information about the application and libraries to be embedded in the final

Re: [casper] Re-compiled spectrometer tutorial design gives me wrong spectrums

Hi Daniel Sorry for the nasty surprise after all the hard work. The good news is that I have very recently (Monday) compiled a design using those library blocks and do not notice the problems you have. There are a few things to do to figure out what the problem is; 1. Check that the block

Re: [casper] The casper library status

Hi Wan I did not see any capsper library update from git since last year august. Do you immigrate the latest library to other server? The main repository has not been updated for a while. If you are looking for more recent changes use the repositories at;

[casper] CASPER tools telecon

Hi All Most of the current major contributors/maintainers of the CASPER libraries attended a telecon to resolve a couple of issues and discuss short to medium term work to be done. It was rather long but there should be some interesting developments to come in the following months. The following

Re: [casper] Sync Signal ADC

Hi Jesus The sync input is not used by the ADC at all, it is passed through to the FPGA. It is therefore not necessary to attach anything to this port to acquire data using your ADC. The sync input is typically attached to a PPS signal and used as part of a synchronisation system where time is

Re: [casper] setting up the equalizer gains for poco

Hi Mandana It is important to ensure that sufficient signal is coming out after requanisation to ensure reasonable correlator efficiency while avoiding clipping. This requires a calibration step that can be done using autocorrelation products. Thanks for the useful comments. Is there any

Re: [casper] Inter-F-engine PPS sync hardware?

Hi Jon We have a system using 8 ROACHes for F-engines and 8 for X-engines. We distribute the PPS to all F-engines using a digital pulse distribution amplifier (the Timetech TT-10188 www.ptsyst.com/TT-10188-B.pdf). This pulse originates from a GPS unit. We also use length-matched cables to ensure

Re: [casper] HELP: There is no cable plugged into port 0 Software Registers

Hi guys See the thread below, it may help regarding the 'reading back zeros problem'. http://www.mail-archive.com/casper@lists.berkeley.edu/msg01938.html Regards Andrew

Re: [casper] Difficulty in reading BRAMs

Hi Vikas Does the code read back all zeros consistently after FPGA configuration? Your problem may be related to the thread below; http://www.mail-archive.com/casper@lists.berkeley.edu/msg01938.html Regards Andrew On 24 September 2011 01:48, Vikas Asthana vikasf...@gmail.com wrote: Hi Glenn,

[casper] toolflow/libraries/firmware update from CASPER workshop 2011

Hi all Various working groups were formed at the 2011 CASPER workshop. One of these were for the toolflow, libraries, software etc. Various decisions and actions were decided. This is a list. 1. Test Suite The current library test suite is to be updated. The current bit-for-bit tests will

Re: [casper] toolflow/libraries/firmware update from CASPER workshop 2011

Hi Glenn Regarding the test suite, one thing I've been thinking of that could help a lot with designing test benches is to invest in some code (or blocks) for the tedious boiler plate that is often needed for this work. One extremely useful function would automatically extract sync in and

Re: [casper] Beamformer block ???

Hi Vikas We are working on beamforming for the KAT-7 telescope here at SKA-SA. The beamformer functionality will be in the same FPGAs as the X-engines in our packetised correlator. The operation will be similar to delay compensation/fringe tracking and the results from the different antennas

Re: [casper] 10 gbe packets received at a reduced rate

Hi Nimish If you are not losing any packets (I assume you have a counter or something in every packet to confirm this...?) then it seems that you must be accumulating for longer than expected. This could be because of a clock that is not running as fast as expected (as suggested by John), or

Re: [casper] 10GbE losing first packet

Hi Jia Is the whole first packet lost, or just part of it? I ask because the snap block should capture any data after a trigger, and does not 'know' about packets. I might expect a partial first packet, but not total loss. It may be that the first packet is not being transmitted or being lost

Re: [casper] 10 gbe packets received at a reduced rate

than that. We are using iADC and want to sample at 1 GHz. I would be glad to know if you have any suggestions to try. Thanks again, Nimish On Tue, Nov 15, 2011 at 2:01 AM, Andrew Martens martens.and...@gmail.com wrote: Hi Nimish If you are not losing any

Re: [casper] Requesting subscription

https://calmail.berkeley.edu/manage/list/ On Wed, 2011-11-16 at 08:54 -0800, Sharath Raju wrote: Please add me to the Casper mailing list. Thanks. Regards, Sharath Raju

Re: [casper] Difficulty in reading BRAMs

. Is there something else I could try ? Is this related to fault in design file or python script? Thanks. -Vikas. On Mon, Sep 26, 2011 at 2:18 AM, Andrew Martens martens.and...@gmail.com wrote: Hi Vikas Does the code read back all zeros consistently after FPGA

Re: [casper] Fresh set up with Ubuntu

Hi Laura First off. The official stance taken by CASPER is that we recommend installation on the Xilinx-supported operating systems (SUSE, RedHat etc). These are often clunky to use, but allow us to approach Xilinx if there is a problem. When running on another OS, it is unlikely that Xilinx will

Re: [casper] Shared BRAM yellow block

Hi Nimish We have been making a lot of changes to the ROACH2 infrastructure in the last few weeks. This includes fixes that improve timing scores considerably (software registers were being connected so that the FPGA clock was registering the value instead of the EPB clock and the EPB clock

Re: [casper] Shared BRAM yellow block

Hi Nimish commit 3282865 just pushed to ska-sa github repo. ROACH2 related changes should fix your timing issues. Let us know if you have any problems. Regards Andrew On Mon, 2012-03-05 at 15:38 -0500, Nimish Sane wrote: Hi all, We are using multiple shared bram yellow blocks in our FX

Re: [casper] Shared BRAM yellow block

as well — in particular, the shared bram block? Thanks, Nimish On Tue, Mar 6, 2012 at 10:17 AM, Andrew Martens and...@ska.ac.za wrote: Hi Nimish commit 3282865 just pushed to ska-sa github repo. ROACH2 related changes should fix your timing

Re: [casper] KatADC documentation and questions

Hi Nimish Assuming we have a software register connected to the KatADC atten0 input, how often/fast can we change the attenuation? The change in value triggers an IIC operation using a controller in the katADC yellow block. The rate at which you can change the attenuatation/enable is thus

Re: [casper] More Toolflows on the same machine

Hi Andrea There was a time (v7 tools) when both the iBOB and ROACH were supported by the toolflow. The Virtex5 architecture offers various optimisations not available in Virtex2. These optimisations required setting up various signal processing blocks in specific ways (e.g if a multiplier is

Re: [casper] ROACH2 build environment

Hi Progress on toolflow (yellow blocks etc) support for ROACH2 can be found at https://github.com/casper-astro/mlib_devel/issues/2 Regards Andrew

Re: [casper] Fwd: ROACH - multicast

Hi Ramesh, Dan No, Jan's patch has not been applied to the the KAT branch. I shall add it to the list of nice things to have but we are in the middle of getting the toolflow ready for ROACH2 so it may have to wait a while. Regards Andrew On Mon, 2012-03-26 at 14:20 -0700, Dan Werthimer wrote:

Re: [casper] Tutorial 2 Compilation Problems

Hi Joe, all There were additional errors These two errors occurred multiple times: Error: Block const encountered an error during execution of xlLoadDynamicListContents. Some data may not have been re-loaded correctly. Error: Block manipulate encountered an error during execution of

Re: [casper] FFT block

Hi Jesus To go into a lot of detail on the FFT would take a _lot_ of time (and is not necessary for most users) so I will answer each of your questions briefly and hope that you can 'look under mask' to get any details you require. I would like to know what signal processing scheme is

Re: [casper] Bidirectional IO?

Hi John 1) add the inout choice to the mask pulldown 2) add it to the mask script, and check to make sure it's a good config for the bits specified, and redraw the block with the extra direction port. 3) modify xps_gpio.m to handle the inout case 4) Write/modify a vhdl module to implement

Re: [casper] ten_gbe tx overflow at 28MHz (156.25MHz)

Hi Laura Probably a silly question, but do you have Enable Large Tx Frames enabled as a parameter option on the 10Ge core? The ~8000 data word limit only applies when this option is chosen, otherwise it is a _lot_ smaller. When buffer overflows happen, bad things occur, the 10Ge core does not

Re: [casper] ten_gbe tx overflow at 28MHz (156.25MHz)

Hi all My bad, I was thinking about receive, not transmit. Just got back from holiday so my brain is probably not in gear yet, my opologies. Regards Andrew On Mon, 2012-06-11 at 08:29 +0200, Andrew Martens wrote: Hi Laura Probably a silly question, but do you have Enable Large Tx Frames

Re: [casper] FFT overflow

Hey Tom We recently found a bug in the FFT for a specific parameter set that causes data corruption that includes overflows. If you use 'embedded' as your multiplier implementation of choice you may have this bug. Basically, in the twiddle_general_dsp48e block, the slice blocks are configured

Re: [casper] Some questions

Hi Dale On another topic, Andrew, you state that we can calculate the delays since we know the position of the antennas and source. Does anyone know an accurate way to find the location of GPS satellites? ... Our system is not being used for satellites and I assume that stars have better

Re: [casper] Some questions

Hi Jesus There are two main types of delay in a system that combines signals from two antennas produced by a source; 1. Delays that are produced by the geometry of the system. The physical location of the source and antennas mean that there is a different path length the signal produced by the

Re: [casper] how to configure DEC_FIR's coefficients

Hi Li Jian I am including the casper list in the reply in case others find this useful. When I use dec_fir to design a bandpass filter I had meet many questions. I don't know how to configure dec_fir's coefficients.I could not get more information about dec_fir from CASPER's webpage.

Re: [casper] CLKIN1_PERIOD error when building for ROACH II

Checked around for git pull request but did not find one so did the change myself. Untested as do not have access to an install with the latest Xilinx version, could someone check it please at https://github.com/ska-sa/mlib_devel/commit/b40ad47a84507770f8a480ecdedda2bde5ecc133 Regards Andrew On

Re: [casper] Problem using shared bram/snap blocks for Roach 2

Hi Nimish However, the problem still remains with my main design. I believe it has to do with number of software registers and shared brams in the design. At the moment the EPB bus from the PowerPC is connected to an OPB bus (opb0 in your system.mhs). Each OPB bus can handle up to 32 devices

Re: [casper] compiling fft block

Hi Matt I'm trying to compile an old design with a newly installed setup with matlab 2011b/ise 13.4. The fft_init failed, so I replaced the old fft blocks with new ones from the current casper library browser, but the fft_init still doesn't compile: A good way to test if the init scripts

Re: [casper] compiling fft block

trace: entering reorder_init trace: reorder_init post same_state trace: exiting reorder_init trace: entering fft_init trace: exiting fft_init On Sat, Aug 18, 2012 at 12:38 PM, Andrew Martens and...@ska.ac.za wrote: Hi Matt I'm trying to compile an old design with a newly

Re: [casper] compiling fft block

: exiting fft_init On Sat, Aug 18, 2012 at 12:38 PM, Andrew Martens and...@ska.ac.za wrote: Hi Matt I'm trying to compile an old design with a newly installed setup with matlab 2011b/ise 13.4. The fft_init failed, so I replaced the old fft blocks with new ones from the current casper

Re: [casper] Toolflow Updates

Hi Dave, Wes We could probably write a little Wishbone-to-OPB module and plug that in front of the OPB yellow blocks for some things. Messing with the internals of yellow blocks that we don't have hardware for could be dangerous e.g ADCs. Ideally people would help out with yellow blocks they have

Re: [casper] Synchronization issue (zero channel jumping)

Hi Ricardo I am trying to build a 2-pol spectrometer in a ROACH 1. I started with the Casper mlib_devel PFB+FFT but I ran out of slices so I swap the PFB for the one used at VEGAS (taken from the monroe_library_latest) on the git repo. Just to let you know in case you didn't that there are

Re: [casper] 1Gbe Correlator

Hi Arturo It is possible to generate a correlator with data otuput through 1Gbit RJ45 Ethernet? We have 1 ROACH and 1 computer with no 10Gbe. It is possible. You can write some software on the computer (look at the python CORR package) that checks to see when the accumulator is done, copies

Re: [casper] Problem setting parameters in fft blocks using mlib_devel

Hi Kenneth I get the following simulink error if I place a new fft block from the library and try to set parameters. Error:Error invoking object method Error due to multiple causes. -- Error in 'new_adc_test/fft': Initialization commands cannot be evaluated. -- Invalid object name:

Re: [casper] Problem setting parameters in fft blocks using mlib_devel

Hey Ken Problem is a bit hard to get my head around, I don't seem to be getting the same results. It may be a version problem, I still am (rather ashamedly) using a very old Matlab version. It would be cool if someone with similar versions to you could try to replicate your results... I will

Re: [casper] Problem setting parameters in fft blocks using mlib_devel

post back if you're still having trouble. Regards Paul On 18 January 2013 17:02, Andrew Martens and...@ska.ac.za wrote: Hey Ken Problem is a bit hard to get my head around, I don't seem to be getting the same results. It may be a version problem, I still am

Re: [casper] Problem setting parameters in fft blocks using mlib_devel

Hi Ken There were some fixes that had not yet been pushed to the main casper git repository (mostly to do with stale blocks). I have pulled them in (commits 9940104 and 778d19f). Please check out the latest library and see if that helps you. Regards Andrew

Re: [casper] Can't update link of adc083000x2

Hi I did not commit this change because it's not backwards compatible. I have noticed more and more instances of backward compatibility issues lately as I am using older versions of the tools. At some point soon we are going to be forced to make everyone upgrade to the latest tools if they are

Re: [casper] number of coefficients needed in PFB and FFT

Hey I haven't seen the Goertzel algorithm before, but it looks like a great idea for this: we might be able to produce a coefficient DDS in just two DSPs! The Goertzel algorithm predates the FFT even and is used to calculate sparse FFTs. It basically just calculates each bin individually.

Re: [casper] number of coefficients needed in PFB and FFT

Hi Dan we used to use CORDIC for generating coefficients. not sure how cordic comares to goertzel. there are a few open source VHDL cordics. Goertzel uses the coefficient factor of the previous calculation stage to calculate the coefficient factor currently required i.e rotate the previous

Re: [casper] Can't update link of adc083000x2

Hi If we could figure-out a clean way of scripting mask generation, then we could regenerate the entire library on-the-fly, which'd go a long way towards compatibility across versions. This is possible. It requires that the block itself be generated, as well as the mask using scripts. The

Re: [casper] Matlab 2012b

Hi Katty See this page for more information https://casper.berkeley.edu/wiki/MSSGE_Toolflow_Setup Cheers Andrew On Sat, 2013-02-02 at 10:33 -0300, katherine viviana cortes urbina wrote: Dear Casperites I decided change to Matlab 2012b but I need to know what products basics need Matlab

Re: [casper] Virtex 6 BRAM timing

Hi Dave It would be good to add an optional Latency parameter to the Shared BRAM block and allow the user to select 1 (current value that does not use the BRAMs optional output register) or 2 (new value that does use the BRAM's optional output register). I think this would help ROACH2

Re: [casper] Disable KATADC 20dB Amplifier

Hi Nimish Andrew, Your response makes it very clear. Perhaps, the description enable gain block in the mask could be changed to something else more intuitive in the future versions. The name makes sense if you trace that parameter from the base verilog but is probably not intuitive for a

Re: [casper] Error reading registers with ROACH2

Hi Jason I'm having a hard time updating the .mdl file for tutorial1 after I tried to update to the newest CASPER tools. By 'CASPER tools' are you referring to recent git commits of the Simulink libraries from the ska-sa repo? If so, I know what your problem below is... All the links to

Re: [casper] bug in shared bram block?

Hi John Hi all. When trying to use the shared bram block in a recent (march 15ish) version of the casper xps blockset from mlib_devel, we have a problem. The munge_io block inside the yellow block doesn't work. It has a bad parameter list. It ends up with a box in the corner signifying

Re: [casper] Changing Clock Source on MSSGE ROACH Block

Hi Tim Are you using one of the CASPER DAC yellow blocks? You are also using dac1_clk and not dac0_clk - are you using 2 DAC boards? Henno To expand on what Henno has said, if you specify that you will be using a clock source besides those derived from the onboard oscillator (sys_clk,

Re: [casper] Latency in PFB and FFT

Hi Ross The guys at Berkeley and Ryan would probably have more detailed advice (especially regarding hand-placement) but the following are some general guidelines if trying to optimise timing from within System Generator; 1. Adding latency to an operation in System Generator results in the

Re: [casper] Problems with shared_memory blocks on ROACH2?

Hi Ryan Hey all, just an update on this: First, I reverted to an old copy of the libraries -- it didn't work, but had a different problem. It looks like after you have more than a certain number of yellow blocks in use, the tools start using the opb2opb_lite pcore, which can be had from the

Re: [casper] Simulation of tutorial3

Hi Ross Solved - too much gain in the quant causing all the problems seen - Need to work out why it had the effect it did but reducing it produces a perfect spectrum. We did see this when preparing for the last workshop. Tutorial 3 is set up for a wideband noise input where the input power is

Re: [casper] Fwd:connecting with .py this is register cnt_rst not found

Hi Katty This is running my designs: *first this is for modeinterleave * root@roach:/boffiles# ./s1_2013_May_02_1157.bof [1] 519 root@roach:/boffiles# ps PID TTY TIME CMD 506 ttyS000:00:00 login 507 ttyS000:00:00 bash 519 ttyS000:00:00 s1_2013_May_02_ 520 ttyS0

Re: [casper] Address Space Overlap

Hi Ioana, all It turns out I was trying to use more BRAM's than we had payed for (well, my design was nice and simple, except I had put two huge snaps. Once I decreased the size of one, no problem) I didn't expect the error message to take this form though I'm not sure that these

Re: [casper] number of coefficients needed in PFB and FFT

Hi All I have just completed a rather large overhaul of the CASPER FFT family. Rigorous testing has yet to be performed but it works at the moment and has been pushed to the ska-sa repo on github for early adopters/testers. The main aim (and reason this email continues this thread) was to

Re: [casper] number of coefficients needed in PFB and FFT

Hi Dave Do you have any utilization comparisons of old vs new? Not yet, I hope to do a basic one soon though. Savings will depend on use case. Also, you can optimise for resources in a few ways (DSPs vs BRAMs when using fft_direct, logic vs BRAMS in the biplex stages and in reset point

Re: [casper] Error in Shared BRAM yellow block

Hey Jason When I try to set the Data Binary Point parameter in the Shared BRAM block I get the error: Illegal parametrization: Binary point Binary point must be less than or equal to the number of bits, X. X appears to be the index of the Data Width rather than the actual value selected by

Re: [casper] strange problem in the FFT

)) on the output. It is easy to see channel swapping as the output function is nice and smooth. We're using the casper library: [ptcs@ptcs /export/home/ptcs/scratch/guppi2/libraries/mlib_devel]$ git log -1 commit 6050c725f770f0bdb776fae107dc4fe26f7fce51 Merge: f50084e 0c15bf7 Author: Andrew Martens

Re: [casper] Error in Shared BRAM yellow block

Hi Rurik I tried the recent changes to Shared BRAM on ska-sa but I ran into an error during Update Diagram in munge_init.m resulting from commit 2c13dab where it's trying to index div_size which is just a integer. Thanks for the fix :) I attach a patch which fixes this issue; however I'm

Re: [casper] strange problem in the FFT

Hi Dave Thanks for sending your test model. Since it was saved using a newer Simulink version, I had to simulate it using a newer toolflow version. Sorry about that. Doing so did indeed show the FFT problem you described. I then re-simulated my test model in the newer Simulink version

Re: [casper] Idea for speeding up CASPER build times

Hi Ryan So I just finished a ROACH2 build, and everything from copy base package through bitgen took almost 2 hours. It looks like this is because I have 48 shared memories, plus about another 20 snapshots. The tools seem to be synthesizing each of these separately. I would expect

Re: [casper] strange problem in the FFT

Hi Dave The only way that the script could generate logic that would result in correct data would be correct if; 1. The Relational block could get a latency of 1 when no latency was specified i.e default operation has latency of 1 in that toolflow version. You're right! That's exactly the

Re: [casper] Resources not decreases significant

Hi Katty Hi all, When the number of parallel spectrometers increases when one decreases the number of channels, but surprisingly, this increase is not very significant, why resources not decreases significant from 32 to 2048 channels and where do the resources, FFt, PFB? and like to know

Re: [casper] Fwd: error in the slice

Hi Katty Hi All, I tested the scalability in the number of channels of existing digital spectrometers but when the size of pfb and FFT are 2 ^15 pntos and the other blocks as acc-cntrl and vacc. I have any error: In the console of matlab: Error using gen_xps_files (line 196) The

Re: [casper] Problem setting parameters in fft blocks using mlib_devel

Hi Tim Was the following issue (see below) ever solved? I have exactly the same problem with latest libraries at https://github.com/ska-sa Matlab R2012b System Generator is 14.2.4415 Linux- Redhat See below from Ken Treptow from Fermilab: Tim Madden I get the following simulink error if I

Re: [casper] Question on fft_wideband_real

Hi Tim I think the number of simultaneous streams refers to how many signals you want to process in parallel in the FFT. I think it makes multiple parallel FFTs that share twiddle factors. I could have maybe used 'simultaneous polarisations' to be more astronomy specific. Also, the way the

  1   2   >