Site Home   Archive Home   FAQ Home   How to search the Archive   How to Navigate the Archive   
Compare FPGA features and resources   

Threads starting:
1994JulAugSepOctNovDec1994
1995JanFebMarAprMayJunJulAugSepOctNovDec1995
1996JanFebMarAprMayJunJulAugSepOctNovDec1996
1997JanFebMarAprMayJunJulAugSepOctNovDec1997
1998JanFebMarAprMayJunJulAugSepOctNovDec1998
1999JanFebMarAprMayJunJulAugSepOctNovDec1999
2000JanFebMarAprMayJunJulAugSepOctNovDec2000
2001JanFebMarAprMayJunJulAugSepOctNovDec2001
2002JanFebMarAprMayJunJulAugSepOctNovDec2002
2003JanFebMarAprMayJunJulAugSepOctNovDec2003
2004JanFebMarAprMayJunJulAugSepOctNovDec2004
2005JanFebMarAprMayJunJulAugSepOctNovDec2005
2006JanFebMarAprMayJunJulAugSepOctNovDec2006
2007JanFebMarAprMayJunJulAugSepOctNovDec2007
2008JanFebMarAprMayJunJulAugSepOctNovDec2008
2009JanFebMarAprMayJunJulAugSepOctNovDec2009
2010JanFebMarAprMayJunJulAugSepOctNovDec2010
2011JanFebMarAprMayJunJulAugSepOctNovDec2011
2012JanFebMarAprMayJunJulAugSepOctNovDec2012
2013JanFebMarAprMayJunJulAugSepOctNovDec2013
2014JanFebMarAprMayJunJulAugSepOctNovDec2014
2015JanFebMarAprMayJunJulAugSepOctNovDec2015
2016JanFebMarAprMayJunJulAugSepOctNovDec2016
2017JanFebMarAprMayJunJulAugSepOctNovDec2017
2018JanFebMarAprMayJunJulAugSepOctNovDec2018
2019JanFebMarAprMayJunJulAugSepOctNovDec2019
2020JanFebMarAprMay2020

Authors:A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Custom Search

Messages from 119975

Article: 119975
Subject: Re: MPMC2 + flash bootloader problem
From: Patrick Dubois <prdubois@gmail.com>
Date: 30 May 2007 06:57:19 -0700
Links: << >>  << T >>  << A >>
On 29 mai, 17:31, Guru <ales.gor...@email.si> wrote:
> Patrick,
>
> I think I know what the problem is ! Did you specify in bootloader's
> linker script where to boot after reading from flash?
> It looks like his:
> .boot0   :
>    {
>        *(.boot0)
>        _end = .;
>    } > ddr
>
>   .boot : { *(.boot) } > ddr
>
> I hope it will boot properly :)
>
> Cheers,
>
> Guru

Thanks for the suggestion, I'll have to try that out. I think you
meant to say to modify the linker script of the _application_ loaded
from flash, not the bootloader?

I found an interresting note in the Embedded System Tools Reference
Manual:
"If your application is going to be loaded from a bootloader, you
might not want to
overwrite the bootloader's processor reset vector with that of your
application. This reexecutes
the bootloader on a processor reset instead of your application. To
achieve this,
your application must not bring in boot.o as a startup file. Unlike
other compiler startup
files, boot.o is not explicitly linked in by the compiler. Instead,
the default linker scripts
and the tools for generating the linker scripts specify boot.o as a
startup file. You must
remove the STARTUP directive in such linker scripts. You must also
modify the ENTRY
directive to be _start instead of _boot."

I tried that this morning and it works without modifications to the
bootloader code. The drawback is that one needs a different linker
script for loading directly in ram or for loading from flash.

Yesterday however I was also able to modify the bootloader code to
avoid the last memcpy to FFFFFFFC, along with a few other
modifications, and it works fine. The advantage is that now I can
avoid the modifications to the linker script mentionned above.

So I finally have a working solution, although it requires loading
another design to program the flash. Not a big deal since programming
the flash shouldn't happen often.

Anyway, all that to say problem solved. Thanks for your help.

Patrick



Article: 119976
Subject: Xilinx CIC core in Spartan 3?
From: cs_posting@hotmail.com
Date: 30 May 2007 07:53:34 -0700
Links: << >>  << T >>  << A >>
Does anyone know if the Xilinx CIC core will work in the Spartan 3?
It's not listed as a target (only Spartan IIE) but then the data sheet
is dated 2002, wheras S3 apparently was announced in 2003...


Article: 119977
Subject: Re: Spartan3 LVCMOS33 Slew rate
From: austin <austin@xilinx.com>
Date: Wed, 30 May 2007 08:25:06 -0700
Links: << >>  << T >>  << A >>
Yes,

You may have to hand edit the ucf file, and check that the VHDL or
verilog is not  instantiating any input.

The Vref pin is only used if you have a standard that uses an input
comparator in that bank.

Austin

Test01 wrote:
> Austin,
> 
> So those VREF pins can still be used as user I/O pins if I use GTL output? If so that will be great. I tried using one GTL output and the constraint editor assigned the output as GTLP and the I/O Pin corresponding to VREF was assigned as VREF PIN. This is why I am asking.
> 
> THanks.

Article: 119978
Subject: Re: spartan-iie
From: jonpry@gmail.com
Date: 30 May 2007 08:29:10 -0700
Links: << >>  << T >>  << A >>
On May 30, 5:39 am, Gabor <g...@alacron.com> wrote:
> On May 29, 8:10 pm, jon...@gmail.com wrote:
>
> > I recently built a prototype using a xilinx spartan-iie specifically
> > the xc2s150e-fgg456 which is programmed with an xcf02s prom. I have
> > the prom connected first in the jtag chain. I am unable to detect the
> > fpga. If i "tap" the tdo of the prom i can detect and program it fine,
> > but the tdo of the fpga is always floating. I found an error in my
> > design where the program pin, which is connected to the cf pin on the
> > prom, is pulled high with a 4k7 resistor. I have tried pulling it low
> > to no avail.
>
> The Program pin is active low, so pulling it high is the right
> thing to do.
>
> > I am able to access the tdo solderball from the side, so
> > i know it is not a bad connection. Also, while programming the prom,
> > the fpga seems to randomly pull outputs high/low about once every
> > second. I'm not sure how this is possible, but it would indicate to me
> > that the data stream is able to cause a reset of the chip. Does anyone
> > know how i can get further information on the problem, or have any
> > ideas what the problem could be? Thanks in advance for any help.
>
> Make sure you have all of the JTAG connections to the FPGA.  A
> bad TMS connection for instance could cause the FPGA to enter
> a boudary scan test mode rather than program mode.  This could
> also explain why you can't initialize the chain with the FPGA
> in place.

It is hard to tell if things are connected to the bga. I'm thinking
about setting up a fast rise time clock signal with a large series
resistor to each jtag pin. Then i can try to detect the gate
capacitance on a scope. Afaik unconnected balls are pretty unlikely,
but shorts are more likely. In my design tms and tck are only next to
gnd and nc pins. They are definately not shorted to ground. Somewhere
i read that any assertion of tms should make tdo come out of highz. Is
that true?
>
> Also check all of the power supplies to the FPGA.  You need
> core power and also Vcco for the appropriate banks.  Look
> at the data sheet to see which supply runs the TDO pin.
>
All vcco are powered with 3.3 and vccc is 1.8. The rails measure
correctly. The chip seems to be working as it is ever so slightly warm
while running and some things work. Like done is actively pulled low.
Also, when the chip does it;s restting while writing the prom, it is
able to source and sink current.

> And double check the PROG_B pin connection to be sure it is
> on the CF output of the prom.  Make sure the mode pins of the
> FPGA are set for master serial or JTAG mode (master serial
> if you want it to boot up from the PROM, JTAG mode really
> just disables other modes so you can debug the JTAG problem
> easier).
>
It would be really hard to change the mode pins. My next board might
have this as an option. Instead, i removed  the prom and am now
feading the fpga's tdi directly from my programmer.

> HTH,
> Gabor

Is it possible that my programmer was supplying a clean enough tck/tms
for the prom but not for the fpga? It seems like even dirty signals
would cause tdo to be asserted in some way.

Thanks,
Jon Pry


Article: 119979
Subject: Nexys by Digilen xbd file
From: mozilla <godzillalad@gmail.com>
Date: 30 May 2007 09:28:24 -0700
Links: << >>  << T >>  << A >>
Hi,

I recently got a nexys -1000 board for a project i'm working on and
would like to put a xilinx EDK design on the board, however Digilent
don't provide a .xbd (board description file) for this board. Rather
than re-invent the wheel i was wondering if anyone out there had found
one or had already compiled one?

Any help would be greatly appreciated.


Article: 119980
Subject: Spartan-3E DIG-3E1600 Development Board Kit
From: Sandro <sdroamt@netscape.net>
Date: 30 May 2007 09:38:36 -0700
Links: << >>  << T >>  << A >>
Hi,
I wonder if someone know where I can buy the "Spartan-3E DIG-3E1600
Development Board Kit"

   http://www.digilentinc.com/Products/Detail.cfm?Prod=S3E1600&Nav1=Products&Nav2=Programmable

   http://www.xilinx.com/xlnx/xebiz/designResources/ip_product_details.jsp?key=DIG-3E1600&sGlobalNavPick=PRODUCTS&sSecondaryNavPick=BOARDS

but not from xilinx (says to buy from digilent) and not from digilent
(says board not currently available).

Some month ago I contacted Digilent and their answer was
   "We will not be selling this board in the foreseeable future"

thanks
Sandro


Article: 119981
Subject: Re: Spartan-3E DIG-3E1600 Development Board Kit
From: "John_H" <newsgroup@johnhandwork.com>
Date: Wed, 30 May 2007 09:56:58 -0700
Links: << >>  << T >>  << A >>
"Sandro" <sdroamt@netscape.net> wrote in message 
news:1180543116.694676.24010@o5g2000hsb.googlegroups.com...
> Hi,
> I wonder if someone know where I can buy the "Spartan-3E DIG-3E1600
> Development Board Kit"
>
> 
> http://www.digilentinc.com/Products/Detail.cfm?Prod=S3E1600&Nav1=Products&Nav2=Programmable
>
> 
> http://www.xilinx.com/xlnx/xebiz/designResources/ip_product_details.jsp?key=DIG-3E1600&sGlobalNavPick=PRODUCTS&sSecondaryNavPick=BOARDS
>
> but not from xilinx (says to buy from digilent) and not from digilent
> (says board not currently available).
>
> Some month ago I contacted Digilent and their answer was
>   "We will not be selling this board in the foreseeable future"
>
> thanks
> Sandro

I thought the only reason the board was currently unavailable for sale 
directly by Digilent was because the more costly bundled version

http://www.xilinx.com/xlnx/xebiz/designResources/ip_product_details.jsp?key=DO-SP3E1600E-DK-UNI-G

used that same board.  If Xilinx has the raw board as "Buy from Digilent" it 
may be that Digilent can now sell the board even though the web page says 
"Not surrently available for sale."

Please try calling the folks at Digilent - they're pretty responsive but 
sometimes get a backlog in their email.  Phone calls can go more quickly!

If you can use the EDK bundle, you should be able to get that through 
distribution per the Xilinx page referenced above.

- John_H 



Article: 119982
Subject: Re: Xilinx Coregen 2.3 problem
From: Vijayant <vijayant.bhatnagar@gmail.com>
Date: 30 May 2007 10:14:29 -0700
Links: << >>  << T >>  << A >>
Hi Gurus,
I need help desperately. Please help.

Regards,
Vijayant

On May 29, 3:19 pm, Vijayant <vijayant.bhatna...@gmail.com> wrote:
> On May 29, 3:17 pm, Vijayant <vijayant.bhatna...@gmail.com> wrote:
>  Hi,
>  We are using FIFO in our design. The FIFO has been generated using
>  CoreGenerator 2.3. The programmable flags have been enabled. The
>  threshold values have been kept within the depth limit. However, when
>  we try to synthesize on Xilinx, we get warnings:
>  WARNING:NgdBuild:452 - logical net 'fifo1/fifo1/BU2/
> prog_empty_thresh_assert<9>' has no driver
>  WARNING:NgdBuild:452 - logical net 'fifo1/fifo1/BU2/
>  prog_empty_thresh_assert<8>' has no driver
>  WARNING:NgdBuild:452 - logical net 'fifo1/fifo1/BU2/
>  prog_empty_thresh_assert<7>' has no driver
>  WARNING:NgdBuild:452 - logical net 'fifo1/fifo1/BU2/
>  prog_empty_thresh_assert<6>' has no driver
>  WARNING:NgdBuild:452 - logical net 'fifo1/fifo1/BU2/
>  prog_empty_thresh_assert<5>' has no driver
>  WARNING:NgdBuild:452 - logical net 'fifo1/fifo1/BU2/
>  prog_empty_thresh_assert<4>' has no driver
>  WARNING:NgdBuild:452 - logical net 'fifo1/fifo1/BU2/
>  prog_empty_thresh_assert<3>' has no driver
>  WARNING:NgdBuild:452 - logical net 'fifo1/fifo1/BU2/
>  prog_empty_thresh_assert<2>' has no driver
>  WARNING:NgdBuild:452 - logical net 'fifo1/fifo1/BU2/
>  prog_empty_thresh_assert<1>' has no driver
>  WARNING:NgdBuild:452 - logical net 'fifo1/fifo1/BU2/
>  prog_empty_thresh_assert<0>'  has no driver
>
>  Similar warnings are got for 'prog_empty_thresh_assert',
>  'prog_full_thresh_negate', 'prog_full_thresh',
>  'prog_empty_thresh_negate',  'prog_empty_thresh',
>
>  Secondly, our top level entity of the FIFO generated by Coregen does
>  not have signals labeled rd_clk and wr_clk. Still we got the
> following
>  errors :
> NgdBuild:452 - logical net 'fifo1/fifo1/BU2/rd_clk' has no driver
> NgdBuild:452 - logical net 'fifo1/fifo1/BU2/wr_clk' has no driver
>
> Pls help.
>
> Regards,
> Vijayant



Article: 119983
Subject: Re: Xilinx Coregen 2.3 problem
From: "John_H" <newsgroup@johnhandwork.com>
Date: Wed, 30 May 2007 10:40:40 -0700
Links: << >>  << T >>  << A >>
"Vijayant" <vijayant.bhatnagar@gmail.com> wrote in message 
news:1180466343.252183.170500@p77g2000hsh.googlegroups.com...
> On May 29, 3:17 pm, Vijayant <vijayant.bhatna...@gmail.com> wrote:
> Hi,
> We are using FIFO in our design. The FIFO has been generated using
> CoreGenerator 2.3. The programmable flags have been enabled. The
> threshold values have been kept within the depth limit. However, when
> we try to synthesize on Xilinx, we get warnings:
> WARNING:NgdBuild:452 - logical net 'fifo1/fifo1/BU2/
> prog_empty_thresh_assert<9>' has no driver
> WARNING:NgdBuild:452 - logical net 'fifo1/fifo1/BU2/
> prog_empty_thresh_assert<8>' has no driver
> WARNING:NgdBuild:452 - logical net 'fifo1/fifo1/BU2/
> prog_empty_thresh_assert<7>' has no driver
> WARNING:NgdBuild:452 - logical net 'fifo1/fifo1/BU2/
> prog_empty_thresh_assert<6>' has no driver
> WARNING:NgdBuild:452 - logical net 'fifo1/fifo1/BU2/
> prog_empty_thresh_assert<5>' has no driver
> WARNING:NgdBuild:452 - logical net 'fifo1/fifo1/BU2/
> prog_empty_thresh_assert<4>' has no driver
> WARNING:NgdBuild:452 - logical net 'fifo1/fifo1/BU2/
> prog_empty_thresh_assert<3>' has no driver
> WARNING:NgdBuild:452 - logical net 'fifo1/fifo1/BU2/
> prog_empty_thresh_assert<2>' has no driver
> WARNING:NgdBuild:452 - logical net 'fifo1/fifo1/BU2/
> prog_empty_thresh_assert<1>' has no driver
> WARNING:NgdBuild:452 - logical net 'fifo1/fifo1/BU2/
> prog_empty_thresh_assert<0>'  has no driver
>
> Similar warnings are got for 'prog_empty_thresh_assert',
> 'prog_full_thresh_negate', 'prog_full_thresh',
> 'prog_empty_thresh_negate',  'prog_empty_thresh',
>
> Secondly, our top level entity of the FIFO generated by Coregen does
> not have signals labeled rd_clk and wr_clk. Still we got the
> following
> errors :
> NgdBuild:452 - logical net 'fifo1/fifo1/BU2/rd_clk' has no driver
> NgdBuild:452 - logical net 'fifo1/fifo1/BU2/wr_clk' has no driver
>
> Pls help.
>
> Regards,
> Vijayant

Can you find the instantiation of BU2 in your code?  You'll need to have 
ports on BU2 of rd_clk, wr_clk, and prog_empty_thresh_assert.  I can't 
comment on why this aren't connected in the coregen output according to your 
flow, only that there are ports (according to the software) and they don't 
appear to be connected (in BU2, according to the software).

Good luck on this very basic level of "looking at your code." 



Article: 119984
Subject: Re: Nexys by Digilen xbd file
From: cs_posting@hotmail.com
Date: 30 May 2007 10:53:08 -0700
Links: << >>  << T >>  << A >>
On May 30, 11:28 am, mozilla <godzilla...@gmail.com> wrote:
> Hi,
>
> I recently got a nexys -1000 board for a project i'm working on and
> would like to put a xilinx EDK design on the board, however Digilent
> don't provide a .xbd (board description file) for this board. Rather
> than re-invent the wheel i was wondering if anyone out there had found
> one or had already compiled one?

Haven't done anything with the EDK, but so far I've found very little
indication that many people have bought the board at all.  Suprising
as it's an interesting one...   perhaps because it's an independent
digilent product and not one made for Xilinx?





Article: 119985
Subject: Re: XS40 Download Cable
From: Andreas Ehliar <ehliar@lysator.liu.se>
Date: Wed, 30 May 2007 17:55:48 +0000 (UTC)
Links: << >>  << T >>  << A >>
On 2007-05-30, raxpeter@gmail.com <raxpeter@gmail.com> wrote:
> Dear Folks !
>
> I'm hvin a rather outdated XS40 board, which I'm planning to start
> experiment with. Unfortunately, I don't have the Parallel Download
> cable for it.
>
> Is there any circuitry inside the cable or is it just pin-to-pin
> connection ?? Can anybody clarify this for me ?

From what I remember [1] it is just a straight pin-to-pin connection.
www.xess.com do have old manuals available, you could always take
a look there if in doubt.

BTW, do you have software available which can handle the FPGA
on it? There hasn't been any support for the XC4K fpga for quite a
long time now in ISE... 

Good luck with your experimentation!

/Andreas

[1] I have rather fond memories of this board since it was the
    prototype board which introduced me to FPGAs a number of years
    ago.

Article: 119986
Subject: Re: Inverse of a matrix
From: Thomas Womack <twomack@chiark.greenend.org.uk>
Date: 30 May 2007 19:01:00 +0100 (BST)
Links: << >>  << T >>  << A >>
In article <1180507013.553103.267690@a26g2000pre.googlegroups.com>,
Venkat  <venkat.japan@gmail.com> wrote:
>Hi all,
>Can anyone suggest simple algorithms for implementation of finding the
>inverse of a matrix (4 X 4)? Even information of IP Cores for such
>functionality will be greatly appreciated.

How long have you got?

Gaussian elimination in software on an implemented microprocessor will
work, relatively slowly; if you've got enormous amounts of hardware,
each entry of the inverse is of the form [some sum of six products of
three entries from the matrix] divided by the determinant, which is a
sum of 24 products each of four entries from the matrix, and you can
compute everything in parallel and do the whole thing in a few cycles
limited by the number of multipliers and by the depth of your addition
trees.

Perhaps a good trade-off is Gaussian elimination in hardware; build a
circuit which takes inputs A0..A7, B0..B7 and K, and outputs A0-B0*K,
A1-B1*K ...

Then start off with

m00 m01 m02 m03 1 0 0 0
m10 m11 m12 m13 0 1 0 0
m20 m21 m22 m23 0 0 1 0
m30 m31 m32 m33 0 0 0 1

in four eight-wide registers, and do successively

row0 -> row0 * 1/m00
row1 -> row1 - m10 * row0
row2 -> row2 - m20 * row0
row3 -> row3 - m30 * row0
row1 -> row1 * 1/m11
row0 -> row0 - m01 * row1
row2 -> row2 - m21 * row1
row3 -> row3 - m31 * row1

and similarly scaling row2 and row3 to have a 1 in the diagonal
element then subtracting them from the other rows to kill off the
off-diagonal entries.

Or perhaps there are cleverer FPGA-appropriate methods; I'm a
mathematician not a hardware person.

Tom

Article: 119987
Subject: Re: FIR Filter ON FPGA
From: nico@puntnl.niks (Nico Coesel)
Date: Wed, 30 May 2007 18:05:41 GMT
Links: << >>  << T >>  << A >>
bngguy <bngguy@gmail.com> wrote:

>Hi,
>    I'm working on implementing an FIR Filter on a FPGA (Spartan 3E),
>here's what i want to accomplish -->
>
>The FIR Filter coefficients are generated on a host system using
>LabView, these coefficients are written to a RAM / PROM on a DSP
>card , the number of taps is constant but other parameters like
>sampling frequency and cut off frequencies can change according to
>requirements.
>
>The FPGA reads these coefficients from the RAM / PROM and implements
>the FIR Filter.There should be a single bit file that is downloaded to
>configure the FPGA.
>
>Any pointers in the right direction would be appreciated.

It is (probably) easy to have the DSP download the coefficients into
the fpga through some serial (SPI-like) interface. An SPI interface
can consist of a counter generating the bit addres (BRAM in 1 bit mode
on the configuration side). The SPI clock can be the clock for the
counter and the memory, the SPI select resets the counter and disables
writing to the memory when low.

-- 
Reply to nico@nctdevpuntnl (punt=.)
Bedrijven en winkels vindt U op www.adresboekje.nl

Article: 119988
Subject: Re: Xilinx Seminars in Wiesbaden, Berlin, Hannover
From: nico@puntnl.niks (Nico Coesel)
Date: Wed, 30 May 2007 18:09:52 GMT
Links: << >>  << T >>  << A >>
Peter Alfke <peter@xilinx.com> wrote:

>X-Fest is a very successful and highly technical seminar series, given 
>in 92 cities worldwide.
>For the June 5, 12, and 14 seminars in three German locations, I will 
>give the keynote address.
>Come and learn for a full day, “auf deutsch”
>For details, and to register, click on:
>
>http://eetimes.eu/design/199500991
>http://www.em.avnet.com/xfest/xlx
>
>Peter Alfke, Xilinx Applications, with apologies for this commercial 
>message.

Hmm, so many locations in Germany... Amsterdam is a much better choice
:-)

-- 
Reply to nico@nctdevpuntnl (punt=.)
Bedrijven en winkels vindt U op www.adresboekje.nl

Article: 119989
Subject: Re: Nexys by Digilen xbd file
From: austin <austin@xilinx.com>
Date: Wed, 30 May 2007 11:36:33 -0700
Links: << >>  << T >>  << A >>
cs,

Heck, we (APD FPGA Lab) have purchased Digilent boards for use here at
Xilinx in the past!  A lot cheaper than fiddling with making something
ourselves.

Once a board exists from a vendor, no reason to re-invent the wheel.

I bought a Digilent Spartan 3E board personally to teach a class, using
class materials from San Jose University, and my co-worker Eric Crabill.

I am so very pleased that there are enough students using Xilinx chips
that Digilent can make a successful business out of making and selling
the boards.

The concept is simple:  buy one board, and it lasts through a number of
EE and CS courses (introduction to logic, HDL programming, machine
architectures, embedded systems, IO/Memory/Interfacing, etc...).
The cost is less than one engineering textbook, the student owns it (so
they take care of it), and it is useful even after they graduate.

That and the ubiquitous student laptop pc, and they are "in business."

Austin

Article: 119990
Subject: Re: Xilinx Seminars in Wiesbaden, Berlin, Hannover
From: Peter Alfke <peter@xilinx.com>
Date: 30 May 2007 11:44:48 -0700
Links: << >>  << T >>  << A >>
On May 30, 11:09 am, n...@puntnl.niks (Nico Coesel) wrote:
> Peter Alfke <p...@xilinx.com> wrote:
> >X-Fest is a very successful and highly technical seminar series, given
> >in 92 cities worldwide.
> >For the June 5, 12, and 14 seminars in three German locations, I will
> >give the keynote address.
> >Come and learn for a full day, "auf deutsch"
> >For details, and to register, click on:
>
> >http://eetimes.eu/design/199500991
> >http://www.em.avnet.com/xfest/xlx
>
> >Peter Alfke, Xilinx Applications, with apologies for this commercial
> >message.
>
> Hmm, so many locations in Germany... Amsterdam is a much better choice
> :-)
>
> --
> Reply to nico@nctdevpuntnl (punt=.)
> Bedrijven en winkels vindt U opwww.adresboekje.nl



Article: 119991
Subject: Re: Xilinx Seminars in Wiesbaden, Berlin, Hannover
From: "HT-Lab" <hans64@ht-lab.com>
Date: Wed, 30 May 2007 18:44:56 GMT
Links: << >>  << T >>  << A >>

"Nico Coesel" <nico@puntnl.niks> wrote in message 
news:465dbd7f.165182269@news.planet.nl...
> Peter Alfke <peter@xilinx.com> wrote:
>
>>X-Fest is a very successful and highly technical seminar series, given
>>in 92 cities worldwide.
>>For the June 5, 12, and 14 seminars in three German locations, I will
>>give the keynote address.
>>Come and learn for a full day, "auf deutsch"
>>For details, and to register, click on:
>>
>>http://eetimes.eu/design/199500991
>>http://www.em.avnet.com/xfest/xlx
>>
>>Peter Alfke, Xilinx Applications, with apologies for this commercial
>>message.
>
> Hmm, so many locations in Germany... Amsterdam is a much better choice
> :-)

Sigh......I remember my students days in Amsterdam.....actually I can't, not 
sure what I was doing during those years.... must have been the many happy 
mushrooms I eat or was it the daily morning space cake?.... :-)

Hans
www.ht-lab.com



>
> -- 
> Reply to nico@nctdevpuntnl (punt=.)
> Bedrijven en winkels vindt U op www.adresboekje.nl 



Article: 119992
Subject: Re: Xilinx Seminars in Wiesbaden, Berlin, Hannover
From: Peter Alfke <peter@xilinx.com>
Date: 30 May 2007 11:50:57 -0700
Links: << >>  << T >>  << A >>
Avnet picked Eindhoven as the preferred site in the Netherlands.
But the X-Fest seminar there was a month ago...
But the traditional European FPGA conference will be in Amsterdam in
August.
Peter




Article: 119993
Subject: Re: Xilinx Seminars in Wiesbaden, Berlin, Hannover
From: Thomas Heller <theller@python.net>
Date: Wed, 30 May 2007 21:02:23 +0200
Links: << >>  << T >>  << A >>
Peter Alfke schrieb:
> X-Fest is a very successful and highly technical seminar series, given 
> in 92 cities worldwide.
> For the June 5, 12, and 14 seminars in three German locations, I will 
> give the keynote address.
> Come and learn for a full day, “auf deutsch”
> For details, and to register, click on:
> 
> http://eetimes.eu/design/199500991
> http://www.em.avnet.com/xfest/xlx
> 

Has anyone success to register on the avnet site?  It didn't work for me.
Tried IE and Firefox.

Thomas

Article: 119994
Subject: Re: Xilinx Seminars in Wiesbaden, Berlin, Hannover
From: austin <austin@xilinx.com>
Date: Wed, 30 May 2007 12:02:41 -0700
Links: << >>  << T >>  << A >>
Hans,

Like the joke here about attending UC Berkeley in the late 1960's
("summer of love"):  "if you can remember it, you weren't here."

By the way, I was (for a short time) a counselor for the BCIPD (Berkeley
Committee for Information on Psychedelic Drugs):  one of the great
crimes, and wastes of good minds was the abuse of both legal, and
illegal drugs by the youth of that time.  These poor souls are now
cluttering up the mental hospitals, or wandering the streets, and have
become a burden to society.

Not that I have any solution to this (continuing) problem, but I at
least had the chance to talk to people and try to convince them not to
turn their brains into soup.

I am also fortunate I survived.

Austin

Article: 119995
Subject: Virtex4 Configuration Problem
From: msn444@gmail.com
Date: 30 May 2007 12:30:23 -0700
Links: << >>  << T >>  << A >>
Hello,

I'm trying to use a slightly unconventional way of configuring a
Xilinx Virtex4 FPGA that as far as I can tell should work, but
doesn't. The plan involves using a microcontroller to place an atmel
serial flash into continuous read mode, and then relinquish control of
the flash's clock and data output lines to the FPGA, and finally
release INIT. The FPGA, set to Master Serial mode, then begins
clocking the data out of the flash as if it were clocking out of a
xilinx PROM.

It all seems to work the way it should -- the right data appears on
the data line -- but DONE never goes high, it just keeps clocking as
if it's not getting any data. The one difference between the flash and
a xilinx PROM is that the flash outputs data on the falling edge of
the clock, while the PROM outputs on the rising edge. And indeed, if I
place an inverter in line with CCLK, the configuration works fine.

I don't understand why the clock polarity is causing a problem given
that the FPGA supposedly samples on the rise of CCLK. On the rise of
CCLK, the data out of the flash has been stable for a full cycle,
having changed on the fall of CCLK.

Anyone know what I'm missing?

Mike.


Article: 119996
Subject: Virtex4 Configuration Problem
From: msn444@gmail.com
Date: 30 May 2007 12:32:55 -0700
Links: << >>  << T >>  << A >>
Hello,

I'm trying to use a slightly unconventional way of configuring a
Xilinx Virtex4 FPGA that as far as I can tell should work, but
doesn't. The plan involves using a microcontroller to place an atmel
serial flash into continuous read mode, and then relinquish control of
the flash's clock and data output lines to the FPGA, and finally
release INIT. The FPGA, set to Master Serial mode, then begins
clocking the data out of the flash as if it were clocking out of a
xilinx PROM.

It all seems to work the way it should -- the right data appears on
the data line -- but DONE never goes high, it just keeps clocking as
if it's not getting any data. The one difference between the flash and
a xilinx PROM is that the flash outputs data on the falling edge of
the clock, while the PROM outputs on the rising edge. And indeed, if I
place an inverter in line with CCLK, the configuration works fine.

I don't understand why the clock polarity is causing a problem given
that the FPGA supposedly samples on the rise of CCLK. On the rise of
CCLK, the data out of the flash has been stable for a full cycle,
having changed on the fall of CCLK.

Anyone know what I'm missing?

Mike.


Article: 119997
Subject: XS40 Download Cable
From: raxpeter@gmail.com
Date: 30 May 2007 12:45:17 -0700
Links: << >>  << T >>  << A >>
Dear Folks !

I'm hvin a rather outdated XS40 board, which I'm planning to start
experiment with. Unfortunately, I don't have the Parallel Download
cable for it.

Is there any circuitry inside the cable or is it just pin-to-pin
connection ?? Can anybody clarify this for me ?

Regards,

Rakesh


Article: 119998
Subject: Re: Inverse of a matrix
From: Newman <newman5382@yahoo.com>
Date: 30 May 2007 13:02:21 -0700
Links: << >>  << T >>  << A >>
On May 30, 2:36 am, Venkat <venkat.ja...@gmail.com> wrote:
> Hi all,
> Can anyone suggest simple algorithms for implementation of finding the
> inverse of a matrix (4 X 4)? Even information of IP Cores for such
> functionality will be greatly appreciated.
>
> Thanks in advance,
> Venkat.

If you are trying to solve the equation

Ax = y

x = 1/A*y

Then there are simpler methods than finding the inverse and
multiplying.

Otherwise, a long time ago, I used a symbolical math package that
would symbolically give you the inverse of a matrix. That might give
you an idea of the complexity.  I'm sure that some of the newer ones
will do it too.

Newman


Article: 119999
Subject: Re: PacoBlaze 2.2
From: "dscolson@rcn.com" <dscolson@rcn.com>
Date: 30 May 2007 13:14:19 -0700
Links: << >>  << T >>  << A >>
On May 29, 3:37 am, Pablo Bleyer Kocik <pabloble...@hotmail.com>
wrote:
>  Dear group:
>
>  PacoBlaze 2.2 has been released. This version solves some bugs that
> were still lurking in the stack and interrupt manipulation. The cores
> have also received more testing, and more debug information has been
> added to simulations enabled with the HAS_DEBUG macro. Instructions
> also have now a one-hot encoding format that is used when
> USE_ONEHOT_ENCODING is defined. Unfortunately, the compatibility with
> Icarus Verilog is broken but I am still working on it.
>
>  The KCAsm assembler has been tweaked a bit to accept constructs in
> the form of the original KCPSM assembler, so that most PSM files
> should be accepted with little or no modification.
>
>  The PacoBlaze web site is athttp://bleyer.org/pacoblaze, where you
> can find more information and links to the source distribution.
>
>  I can't end this post without referring to the tragic and unexpected
> loss of Rodney Smith. My deepest condolences to his family and
> friends. He was one of the pioneers and pillars of this industry that
> has vamped digital design in a new era. He will be really missed.
>
>  Warmest regards.
>
> --
> PabloBleyerKocik / "I wish I could sleep in a warm sleeping bag.
>  pablo          / No matter what happens I am going to climb
> McKinley."
>   @bleyer.org  / -- Naomi Uemura, 1984

Pablo,

I was also wondering if you fixed the issues with the assembler that I
wrote to you about last June?

>From email sent to you.

 Pablo,

I encountered the follow error:


java -Dkcpsm=3 -Dbram=18 -Dmodule=isp -jar c:/clcfpga2/pacoblaze-2.1b1/
KCAsm.jar isp.psm isp.rmh Exception in thread "main" TokenMgrError:
Lexical error at line 668, column 23.  Encountered:
 "(" (40), after : ""
        at KCAsmTokenManager.getNextToken(KCAsmTokenManager.java:539)
        at KCAsm.jj_scan_token(KCAsm.java:629)
        at KCAsm.jj_3_15(KCAsm.java:479)
        at KCAsm.jj_3R_9(KCAsm.java:466)
        at KCAsm.jj_3_17(KCAsm.java:381)
        at KCAsm.jj_2_17(KCAsm.java:340)
        at KCAsm.Arguments(KCAsm.java:213)
        at KCAsm.CommandExpression(KCAsm.java:152)
        at KCAsm.Expression(KCAsm.java:100)
        at KCAsm.Start(KCAsm.java:84)
        at KCAsm.main(KCAsm.java:41)


The line of code that cause this error is pointed to below.
	load	Areg, seedata0		; pointer to eeprom parameters
	fetch	accm, fpwmminus0	; pwmminus0
-->	store	accm, (Areg)		;
	add	Areg, 01		;

I also noticed that there is a compatibility issue with the Xilinx
kcpsm assembler. It defaults to hex format for numbers while yours
defaults to decimal. This caused some of my constant definitions that
were hex values to give an errors

thanks for you help.

Dave Colson




Site Home   Archive Home   FAQ Home   How to search the Archive   How to Navigate the Archive   
Compare FPGA features and resources   

Threads starting:
1994JulAugSepOctNovDec1994
1995JanFebMarAprMayJunJulAugSepOctNovDec1995
1996JanFebMarAprMayJunJulAugSepOctNovDec1996
1997JanFebMarAprMayJunJulAugSepOctNovDec1997
1998JanFebMarAprMayJunJulAugSepOctNovDec1998
1999JanFebMarAprMayJunJulAugSepOctNovDec1999
2000JanFebMarAprMayJunJulAugSepOctNovDec2000
2001JanFebMarAprMayJunJulAugSepOctNovDec2001
2002JanFebMarAprMayJunJulAugSepOctNovDec2002
2003JanFebMarAprMayJunJulAugSepOctNovDec2003
2004JanFebMarAprMayJunJulAugSepOctNovDec2004
2005JanFebMarAprMayJunJulAugSepOctNovDec2005
2006JanFebMarAprMayJunJulAugSepOctNovDec2006
2007JanFebMarAprMayJunJulAugSepOctNovDec2007
2008JanFebMarAprMayJunJulAugSepOctNovDec2008
2009JanFebMarAprMayJunJulAugSepOctNovDec2009
2010JanFebMarAprMayJunJulAugSepOctNovDec2010
2011JanFebMarAprMayJunJulAugSepOctNovDec2011
2012JanFebMarAprMayJunJulAugSepOctNovDec2012
2013JanFebMarAprMayJunJulAugSepOctNovDec2013
2014JanFebMarAprMayJunJulAugSepOctNovDec2014
2015JanFebMarAprMayJunJulAugSepOctNovDec2015
2016JanFebMarAprMayJunJulAugSepOctNovDec2016
2017JanFebMarAprMayJunJulAugSepOctNovDec2017
2018JanFebMarAprMayJunJulAugSepOctNovDec2018
2019JanFebMarAprMayJunJulAugSepOctNovDec2019
2020JanFebMarAprMay2020

Authors:A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Custom Search