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
2017JanFebMarApr2017

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 59575

Article: 59575
Subject: V2Pro, ML300 Linux reference design
From: antti@case2000.com (Antti Lukats)
Date: 22 Aug 2003 01:08:38 -0700
Links: << >>  << T >>  << A >>
Hi

those who want to build a linux hardware ref. platfrom for Virtex 2Pro
ML300 the reference desing for Xilinx EDK does exist - just received it.
it should be sufficient to run Montavista distribution, but I guess
denx ELDK would also work with minor patching.

antti
this reference design really is available from your local Xilinx FAE!
(I didnt believe it until got it)

Article: 59576
Subject: Re: Xilinx XC3000 with Xilinx ISE student edition 4.2i
From: Terry <terry@nospam.coolmail.com>
Date: Fri, 22 Aug 2003 17:43:06 +0800
Links: << >>  << T >>  << A >>
Actually, the University that I am currently at uses the Xilinx to 
implement 'simple' circuits as our lab experiments.  We do things like 
decoder, ring counter and stuffs.  The XC3000 is more than enough for 
such jobs.  As students, we are at a learning stage, thus we wouldn't 
know how to use the more advanced Spartan or XC4000 or etc even if those 
are given.

Thanks for your comments.

Puting all those aside.  I am now having this ISE 4.2i student edition 
software which doesn't come with the XC3000 parts diagrams.  My school 
uses Xilinx Foundation Series 3.11.  Is there any way that I can import 
the XC3000 diagram parts, do the schematic at home, make sure that it is 
'implementable' and then transfer the .sch file back to school for the 
actual testing on the Xilinx chip?

Thank you.

Peter Alfke wrote:
> The XC3000 is to a modern FPGA what a '286 is to a Pentium4. Would your
> professors suggest you use a '286? It's a capable computer, but
> hopelessly obsolete and underpowered. So is the XC3000.
> Get with it and use Spartan or Virtex devices.
> "One year in the life of an FPGA is like 15 years in a human".
> By that rule, your XC3000s should be in the old folks' home or the cemetary.
> 
> Peter Alfke, Xilinx
> =====================
> Terry wrote:
> 
>>How to I design a circuit involving XC3000 with Xilinx ISE student
>>edition?  I look through the components and parts, but couldn't find the
>>XC3000 series.  I only found Virtex, Spartan, XC4000 but not XC3000 series.
>>
>>And also, my school uses Xilinx Foundation Series software for designing
>>  circuits with XC3000.  Is it possible for me to save the .SCH file
>>from school and open it in my Xilinx ISE student edition?  and vice versa?
>>
>>Thank you.


Article: 59577
Subject: Re: Question about slew rate for SpartanII using ISE5.1
From: jaxlau@yahoo.com (Jacques athow)
Date: 22 Aug 2003 04:45:04 -0700
Links: << >>  << T >>  << A >>
Isnt the proper way to do this...

---- 
# SLEW sets the speed of an IOB output rise/fall time. 
 
NET mysignal slew=FAST; 
 
# Legal values: FAST, SLOW.
# FPGA Families: All.
# Applies to output and I/O pads or the net connected to an output pad
---

The pad report indeed says that I have a fast slew rate on the pad ive
designated to have the fast slew rate attribute.


alternatively you could use the constrain editor to edit your ucf file
and add, with the advance option, the required constrain that you
want, in this case the fast slew rate. Both gives you the same end
result.

jacques.

stephen@postec.co.nz (Stephen du Toit) wrote in message news:<798f0979.0308211905.d2b451@posting.google.com>...
> Hi,
> 
> If someone can help me with this, I'll be very happy.  
> 
> I would like to change the slew rate of some of my output pins of a
> SpartanII to "FAST". Tried to put it in the UCF file, it obeys all the
> other constraints that I put in there, but not the "FAST" one. The
> syntax used was
> NET "c" FAST;
> 
> I also declared some stuff in my VHDL source as follows:
> attribute FAST : string; 
> attribute FAST of c : signal is "true"; 
> 
> I looked at the properties of the Place-and-route but cannot find
> anywhere where the option has to be enabled.
> 
> The software is happy with all of this, but the PAD report still says
> that pin c is "SLOW".
> 
> Thanks very much.
> 
> Stephen du Toit
> 
> 
> I tried

Article: 59578
Subject: Re: Partial Reconfiguration on Xilinx FPGA
From: "Eduardo Wenzel Brião" <briao@inf.pucrs.br>
Date: Fri, 22 Aug 2003 05:26:28 -0700
Links: << >>  << T >>  << A >>
Serial interface can be used to reconfigure partially the FPGA. 
I did some tests using reconfigurable modules generated by Modular 
Design flow and the FPGA has been reconfigured correctly. 
Regards 
Eduardo Wenzel Brião

Article: 59579
Subject: Problems with PAR tool in Modular Design flow
From: eduwenzel@yahoo.com.br (=?ISO-8859-1?Q?Eduardo_Wenzel_Bri=E3o?=)
Date: 22 Aug 2003 05:34:38 -0700
Links: << >>  << T >>  << A >>
Hi

I developed a project using one fixed area and two reconfigurable
areas. But when the Assembly Phase of MD flow is executed, the PAR
tool issued the following error:

FATAL_ERROR:Guide:basgitaskphyspr.c:369:1.17.4.3:137 - Guide
encountered a
   Logic0 or Logic1 signal GLOBAL_LOGIC1_6 that does not have a driver
or load
   within the module boundary. This problem may be caused by having a
constant
   driving the input from outside the module boundary or because a
driver or
   load comp did not meet the par-guiding criteria.  The design will
not be
   completely placed and routed by Par-Guide   Process will terminate.
 To
   resolve this error, please consult the Answers Database and other
online
   resources at http://support.xilinx.com. If you need further
assistance,
   please open a Webcase by clicking on the "WebCase" link at
   http://support.xilinx.com

I am following all the steps of XAPP290 document. Modular Design flow
generated some partial bitstreams, but the PAR error is issued in the
final phase of M.D.

Can someone help me in this thread?

Eduardo Wenzel Brião

Article: 59580
Subject: Why can't Xilinx DCM's regain lock without a RESET??
From: haklis@hotmail.com (Hakon Lislebo)
Date: 22 Aug 2003 06:27:03 -0700
Links: << >>  << T >>  << A >>
I am in the telecom business where redundant clocks and clock
protection switchovers is big issues. That leads to several problems
when using Xilinx FPGA's an DCM's. One example is when clocks are lost
and comes back again, the DCM then need a reset to start the lock
procedure. It is possible to make a circuit which checks the DCM
status and give a reset when LOCKED is low, but it takes resources. We
have in fact seen a few times that the STATUS bits from the DCM
indicated OK, but the output clock was not OK until we gave the DCM
another reset.
So Xilinx: Why cannot the DCM itself start to "hunt" for a lock as
soon as lock is lost??

The DCM's are nice though, could not achieved my I/O timing without
them!

Article: 59581
Subject: LVPECL I/O and Fndtn4.2
From: Marlboro <>
Date: Fri, 22 Aug 2003 06:30:32 -0700
Links: << >>  << T >>  << A >>
Hi guys, 
How to get around this error, I try to instantiate a LVPECL output 
and always get the error during bitgen phase. The UCF file already 
constrained output net IOSTANDARD = LVPECL and LOC=P56 (XCV100E-PQ240) 
For my understand, we just need to instantiate the P site only. 
Is that true? 

Running DRC. 
ERROR:DesignRules:441 - Blockcheck: 
Illegal LVPECL configuration. Comp P_OUT is configured as a LVPECL IOB that uses the OUTMUX, but it's slave site P57 is empty. 


Article: 59582
Subject: Simulating single module of design in ModelSim (Xilinx)
From: "Tom Derham" <tderham@NOSPAM.ee.ucl.ac.uk>
Date: Fri, 22 Aug 2003 14:57:08 +0100
Links: << >>  << T >>  << A >>
Hi there

I am using Xilinx Webpack / Modelsim Starter Edition / Spartan IIE-7fg456
development board.

I have made a design in VHDL, consisting of various buffers, fifos, control,
timers and counters.
I want to start the timing simulation process by just looking at a single
module (the simplest - a downcounter - code below).

The functional simulation in Modelsim works fine (I wrote the simple
testbench myself), but despite it appearing to meet the timing requirements
I placed in the constaints file, the post translate/map/p&r models seem to
have longer delays than expected.

I.e. the counter starts counting after an enable goes high - this happens
on the next clock cycle as expected in the functional simulation, but is
delayed by several clock cycles in the timing simulations, despite meeting
constraints that are less than one half clock cycle.  I need to run the
clock at 100 MHz (10ns period).

Then I thought that the mapping process is mapping the ports of the module
to actual physical i/o with its associated delays, whereas in fact this
module is a component in the module the next level up.  Could this be
causing the apparent problem?  If so, is there any reliable method to
simulate a single module that forms part of a design, other than at the
functional level?
I apologise if this seems a basic question, I am trying to learn the best I
can.

Many thanks for your help

Tom

library IEEE;
use IEEE.STD_LOGIC_1164.ALL;
use IEEE.STD_LOGIC_ARITH.ALL;
use IEEE.STD_LOGIC_UNSIGNED.ALL;

entity downcounter is
    Port (
CLOCK_IN : in std_logic;
LOAD_IN : in std_logic_vector(13 downto 0);
LOAD_ENABLE_IN : in std_logic;
Q : out std_logic;
RST : in std_logic
   );
end downcounter;

architecture Behavioral of downcounter is

attribute uselowskewlines: string;
attribute uselowskewlines of Q : signal is "yes";

begin

process (CLOCK_IN, RST)
variable COUNT : std_logic_vector(14 downto 0);
begin
if RST='1' then
 COUNT(14 downto 0) := "100000000000000";
elsif (CLOCK_IN='1' and CLOCK_IN'event) then
  if COUNT(14) = '0' then
   COUNT := COUNT - 1;
  elsif LOAD_ENABLE_IN='1' then
       COUNT(13 downto 0) := LOAD_IN;
   COUNT(14) := '0';
   COUNT := COUNT - 1;
  end if;
end if;
Q <= not(COUNT(14));
end process;

end Behavioral;



Article: 59583
Subject: Re: Old Xilinx FPGAs
From: rickman <spamgoeshere4@yahoo.com>
Date: Fri, 22 Aug 2003 10:39:41 -0400
Links: << >>  << T >>  << A >>


Joseph H Allen wrote:
> 
> In article <3F416E2C.12F31ABC@andraka.com>,
> Ray Andraka  <ray@andraka.com> wrote:
> >Add to the list:
> >    Cutting out half the LUT RAMs/SRL16s in SpartanIII
> 
> Yeah this was a bad idea.

Hey, it works for me.  My understanding is that this helps save a lot on
price.   It is a rare design indeed that needs the SRL16s or LUT RAMs
from more than half the chip.  

Or you could look at it a different way.  Think of the Spartan 3 chips
as having half the number of FF/LUTs (with full SRL/RAM function) that
the spec says and then they give you that many more (limited versions)
for free!  I expect the Spartan 3 will still be a much better price
point than the Virtex II.  

-- 

Rick "rickman" Collins

rick.collins@XYarius.com
Ignore the reply address. To email me use the above address with the XY
removed.

Arius - A Signal Processing Solutions Company
Specializing in DSP and FPGA design      URL http://www.arius.com
4 King Ave                               301-682-7772 Voice
Frederick, MD 21701-3110                 301-682-7666 FAX

Article: 59584
Subject: Signal within block
From: fpga_uk@yahoo.co.uk (Isaac)
Date: 22 Aug 2003 08:19:56 -0700
Links: << >>  << T >>  << A >>
HI mates, 
I am using 12 same entitites using Component decleration method. I am
giving input to 12 entities in such a way that the internal signal's
in each entity has different values from each other at any time. Now
the top vhdl final in which all the component decleration are defined
, I want to use the internal signal's of each block to perform some
calculation. The probelm is that in each of the 12 entities signal has
the same name (as I am using component decleration method to generate
same entity 12 time).

How to access the internal signal of say Component1 , Component2 or so
on.
in the top entity. Is there is any way to access these Signal in VHDL?

Any suggestions will be Appreciated .


Rdgs 

Isaac

Article: 59585
Subject: Re: Why can't Xilinx DCM's regain lock without a RESET??
From: Austin Lesea <Austin.Lesea@xilinx.com>
Date: Fri, 22 Aug 2003 08:20:01 -0700
Links: << >>  << T >>  << A >>
Hakon,

I would be intersted to know what the conditions were where the status
bits did not indicate that the DCM was not functioning properly.

The overflow bit, the DLL not in phase bit, and the CLKIN lost bits are
all required (in addition to the LOCKED signal) to determine that the DCM
is functioning properly (if it is likely or normal to have the incoming
clock mis-behave).  Also there is the CLKFX stopped bit if you are using
that output.

I spent 23 years in telecom, and clock signals can fail in the most
strange ways (as microwave links fade in and out, for example), so I
understand that we may not have covered 100% of the odd behaviour of the
world's transmission systems.  For that reason one adds user logic to
solve the unusual user problems that are out there.

In a system that I designed, I needed to monitor embedded switch clock
commands, AIS, frame errors, bit errors, clock loss, clock frequency, and
clock rate of change of phase.  If any of these parameters exceeded a user
defined threshold, we needed to switch to another (better) clock and
report the problem.  Once switched, we were not allowed to switch back
until the problem was not only resolved, but a user defined time had
elapsed.

There is no way to design all of this for you, but we will supply the
necessary status bits.  If any of these bits did not work as expected, I
would like to know.

Austin

Hakon Lislebo wrote:

> I am in the telecom business where redundant clocks and clock
> protection switchovers is big issues. That leads to several problems
> when using Xilinx FPGA's an DCM's. One example is when clocks are lost
> and comes back again, the DCM then need a reset to start the lock
> procedure. It is possible to make a circuit which checks the DCM
> status and give a reset when LOCKED is low, but it takes resources. We
> have in fact seen a few times that the STATUS bits from the DCM
> indicated OK, but the output clock was not OK until we gave the DCM
> another reset.
> So Xilinx: Why cannot the DCM itself start to "hunt" for a lock as
> soon as lock is lost??
>
> The DCM's are nice though, could not achieved my I/O timing without
> them!


Article: 59586
Subject: Re: Question about slew rate for SpartanII using ISE5.1
From: "John_H" <johnhandwork@mail.com>
Date: Fri, 22 Aug 2003 15:52:05 GMT
Links: << >>  << T >>  << A >>
Comments inline, additional at the end.

"Jacques athow" <jaxlau@yahoo.com> wrote in message
news:acc717b2.0308220345.6d058151@posting.google.com...
> Isnt the proper way to do this...

Sure it is, there's just more than one way to do it.  Your way is valid for
the FPGAs but not the CPLDs.
The original poster's method of "NET yadayada FAST;" should work as should
"INST yada_obuf FAST;" both of which I use withouth problem.

> ---- 
> # SLEW sets the speed of an IOB output rise/fall time.
>
> NET mysignal slew=FAST;
>
> # Legal values: FAST, SLOW.
> # FPGA Families: All.

# CPLD Families:  None.

> # Applies to output and I/O pads or the net connected to an output pad
> ---
>
> The pad report indeed says that I have a fast slew rate on the pad ive
> designated to have the fast slew rate attribute.
>
>
> alternatively you could use the constrain editor to edit your ucf file
> and add, with the advance option, the required constrain that you
> want, in this case the fast slew rate. Both gives you the same end
> result.
>
> jacques.
>
> stephen@postec.co.nz (Stephen du Toit) wrote in message
news:<798f0979.0308211905.d2b451@posting.google.com>...
> > Hi,
> >
> > If someone can help me with this, I'll be very happy.
> >
> > I would like to change the slew rate of some of my output pins of a
> > SpartanII to "FAST". Tried to put it in the UCF file, it obeys all the
> > other constraints that I put in there, but not the "FAST" one. The
> > syntax used was
> > NET "c" FAST;
> >
> > I also declared some stuff in my VHDL source as follows:
> > attribute FAST : string;
> > attribute FAST of c : signal is "true";
> >
> > I looked at the properties of the Place-and-route but cannot find
> > anywhere where the option has to be enabled.
> >
> > The software is happy with all of this, but the PAD report still says
> > that pin c is "SLOW".
> >
> > Thanks very much.
> >
> > Stephen du Toit
> >
> >
> > I tried

Did you set the IOSTANDARD for those signals?  The propagation rules and
applicable elements are the same for those two attributes.  I'm wondering if
the net name ended up being something different by the time it got to the
pad, making the propagation rules inapplicable.

If you can see the net names in FPGA Editor but your IOBs are still SLOW,
try using the INST format on the OBUF, OBUFT, IOBUF, or IOBUFT primitives
that were generated by your VHDL code.

If all the other constraints are showing up properly, is there somewhere you
have a SLOW that might inadvertantly be overriding the constraint?  If the
timing constraints are succussful, the checkbox for "Ignore Timing
Constraints" in the GUI probably isn't checked so that shouldn't be a
problem either.

I hope you find what's going on!



Article: 59587
Subject: Re: DCM vs state machine
From: oen_br@yahoo.com.br (Luiz Carlos)
Date: 22 Aug 2003 10:01:45 -0700
Links: << >>  << T >>  << A >>
Peter Alfke <peter@xilinx.com> wrote in message news:<3F4550A6.DA275722@xilinx.com>...
> Grrrrr!   :-(
> 

Buaaaahhhhh!

C'mon Peter, everybody can make a mistake or two.
I think you don't want to pay that champagne! (An old post, if you don't remember.)

Luiz Carlos.

Article: 59588
Subject: Re: DA FIR filter vs. MAC FIR filter
From: Ray Andraka <ray@andraka.com>
Date: Fri, 22 Aug 2003 14:54:52 -0400
Links: << >>  << T >>  << A >>
You are missing some key pieces of information: the available clock frequency
(this is different than the sample rate), and the target FPGA family.

A MAC type filter will occupy a fixed amount of real-estate for any number of
taps, but will take additional clock cycles for each sample as taps are added.
A DA filter runs in the same number of clock cycles (equal to the number of
bits in the input for a serial DA) regardless of the number of filter taps, but
requires more area as the length of the filter is increased.  The DA filter
also computes the sum of the partial products to full precision where the MAC
generally does not, although it may not be significant in your filter.
Ideally, you should be clocking the FPGA much faster than the 5MHz so tht you
can take advantage of serial arithmetic to reduce the physical size.

CIC filters are merely a recursive solution to a boxcar or moving average
filter.  In those filters, you add the N previous samples and divide by the
number of samples.  In the case of the CIC, there is no division, so there is a
net gain of N in the basic non-decimating CIC.  Now getting a bit fancier, you
can suck the decimation into the filter so as to reduce the size of the
delays.  In that case, the filter is equivalent to a N*R tap boxcar filter
whose output is sampled every R input samples.  In this case, the gain is N*R.
Then, if you cascade CIC sections, you multiply that gain by the gain of each
section, so for an M section CIC, the gain is |N*R|^M.  The gains and the
filter responses can be easily derived using the equivalent boxcar filters and
decimation or interpolation stages.

Sasa Bremec wrote:

> Hi!
>
> I have an question about the main differences between this two filter
> architecture.
>
> In my design I would like to minimize slice resources used by the filter
> also I have to take in consideration that my sampling frequency.
>
> The MAC filter is ideal for my requirements but I would like to hear some
> more opinion about that.
>
> Filter details:
>
> Filter type: FIR (raised-cosine)
>
> Filter order : 32
>
> Sampling freq.: 5 MHz
>
> Decimation rate: 4
>
> I also have a question about CIC filter and its related CIC Gain, if anyone
> can help me understand this issue, I would really like to know more about
> it.
>
> Best regards, Sasa

--
--Ray Andraka, P.E.
President, the Andraka Consulting Group, Inc.
401/884-7930     Fax 401/884-7950
email ray@andraka.com
http://www.andraka.com

 "They that give up essential liberty to obtain a little
  temporary safety deserve neither liberty nor safety."
                                          -Benjamin Franklin, 1759



Article: 59589
Subject: Re: ise 5.2 timing summary
From: Ray Andraka <ray@andraka.com>
Date: Fri, 22 Aug 2003 14:56:53 -0400
Links: << >>  << T >>  << A >>
Only if your clocking is perfect.  You need to add the maximum jitter on your
clock to the minimum period to determine the minimum period in your system.
Don't forget to add the additional jitter added by the DCM or DLL.

smu wrote:

> Hello,
>
> I'm new in the fpga design and have a question about the timing summary from
> ise 5.2 from xilinx.
>
> Are the minimum period of a final design announced by the software bundle
> realistic ?
>
> Thank you in advance
>
> smu

--
--Ray Andraka, P.E.
President, the Andraka Consulting Group, Inc.
401/884-7930     Fax 401/884-7950
email ray@andraka.com
http://www.andraka.com

 "They that give up essential liberty to obtain a little
  temporary safety deserve neither liberty nor safety."
                                          -Benjamin Franklin, 1759



Article: 59590
Subject: Re: Async logic in FPGAs
From: Ray Andraka <ray@andraka.com>
Date: Fri, 22 Aug 2003 15:07:08 -0400
Links: << >>  << T >>  << A >>
Use the command strobes like you are as the clock to an input register, but
also toggle a toggle flip-flop with the command.  Use the address and any
qualifiers to produce the clock enable to both the input register and the
toggle flip-flop.   The toggle flip-flop switches state each time a valid
write is performed.  Resync the toggle flip-flop output to your local clock,
and then run it into a simple state machine that generates a 1 cycle wide
pulse as a response to a change on the input.  That pulse , which is sync'ed
to your local clock can then be used to validate the contents of the input
registers.  I usually use it to copy the input register contents to a second
layer input register that is clocked by the local clock.  It is also used in
your control logic to let the rest of the circuit know that there is new
write data.  As long as the bus write frequency is less than about 3 cycles
of your local clock this works fine as described.  If the bus can write
faster, then you can write successive writes into a shift register or memory
and transfer to the local in parallel every Nth write.

rickman wrote:

> After reading a post by Jonathan Bromley on the VME interface, I thought
> I would see if anyone had any comments on a design I am doing.  I have
> to interface to the PC104 bus which is just the PC ISA interface on a
> small board.  After beating my head against the wall for a couple of
> trys, I decided to dump the BCLK since it is not really part of the spec
> and treat the command stobes as async lines.  My design is further
> complicated by the fact that my system clock ranges from 2 MHz to 50 MHz
> to allow power conservation.
>
> I finally decided that the only way to get the job done was to treat the
> command strobes as clock lines.  Read cycles are done normally by muxing
> the data from various registers based on the address lines and the
> tristate outputs are controlled with the strobes.  Writes to simple
> registers are done on the trailing edge of the write strobe.  For reads
> that need to update or clear bits, a strobe is generated that is sync'ed
> to the system clock using a circuit to remove the metastable problem.
> The same circuit works with writes that need to set flags or control
> FIFOs.
>
> I don't see any problem with any of the circuits I have used.  I don't
> even see the need to use a clock tree for the command strobes since
> there are no race conditions where a few ns will matter.  Unless there
> are *huge* differences in routing delays, this circuit should work just
> fine.
>
> Anyone had big problems with similar async circuits?  BTW, here is the
> simple sync circuit to generate a single pulse in the target clock
> domain regardless of the relative speed of the clocks.
>
>    |------- Metastable -------|
>             __________
>            |          |                        _____
>    |------O| inverter |-------|---------------|     | Pulse
>    |       |__________|       |               | XOR |---->
>    |    ______       ______   |   ______   |--|_____| Out
>    |   |      |     |      |  |  |      |  |
>    |---| D  Q |-----| D  Q |--|--| D  Q |--|
> Strobe |      |     |      |     |      |
> /Clock |      |     |      |     |      |
> -------|>     |  ---|>     | |---|>     |
>        |______|  |  |______| |   |______|
>                  |           |
>                  |___________|___________  Output Clock
>
> The pulse out should be clean by the next clock edge as long as the
> routing is kept short.  Or if the clock period is very short another FF
> can be added to feed the other leg of the XOR gate and assure a clean
> output.
>
> If the input clock is faster than the output clock, extra clock edges on
> the input will be ignored until the signal has clocked through the other
> clock domain.
>
> You can also tap the second FF for the feedback if you want to use it
> for a handshake.  A second FF on the input side and an XOR gate will
> give you a pulse on "acknowledge" of the pulse crossing the clock
> domain.
>
> --
>
> Rick "rickman" Collins
>
> rick.collins@XYarius.com
> Ignore the reply address. To email me use the above address with the XY
> removed.
>
> Arius - A Signal Processing Solutions Company
> Specializing in DSP and FPGA design      URL http://www.arius.com
> 4 King Ave                               301-682-7772 Voice
> Frederick, MD 21701-3110                 301-682-7666 FAX

--
--Ray Andraka, P.E.
President, the Andraka Consulting Group, Inc.
401/884-7930     Fax 401/884-7950
email ray@andraka.com
http://www.andraka.com

 "They that give up essential liberty to obtain a little
  temporary safety deserve neither liberty nor safety."
                                          -Benjamin Franklin, 1759



Article: 59591
Subject: Re: Win2k service packs for running Xilinx tools
From: Adric Frost <adric@web-rat.com>
Date: Fri, 22 Aug 2003 18:18:12 -0500
Links: << >>  << T >>  << A >>
Microwaving is definitley the way to go.  You probably want to
make sure it is an old microwave, just in case.

Ian Poole wrote:
> I know this is off topic, but MICROWAVE it.
> See http://www.hamjudo.com/notes/cdrom.html
> Obviously its extremely dangerous and bad for you, so don't do it because I
> said it sounded fun...
> 
> Ian
> 
> "Martin Euredjian" <0_0_0_0_@pacbell.net> wrote in message
> news:jJ%Za.709$0t7.279@newssvr25.news.prodigy.com...
> 
>>I would also like to find out what the status of W2K/SP4 support might be.
>>I'm getting ready to jettison Win XP (big mistake, don't ask) in favor of
>>going back to W2K.  I'm looking for a really creative way to destroy the
> 
> Win
> 
>>XP CD, something commensurate with what I think of the product.  One
> 
> thought
> 
>>I had was to attach it to a high speed motor and let centrifugal force
> 
> shred
> 
>>it to pieces.  That ought to be fun.
>>
>>
>>--
>>~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>>Martin Euredjian
>>
>>To send private email:
>>0_0_0_0_@pacbell.net
>>where
>>"0_0_0_0_"  =  "martineu"
>>
>>
>>
>>"rickman" <spamgoeshere4@yahoo.com> wrote in message
>>news:3F3808AB.4FEE8557@yahoo.com...
>>
>>>rickman wrote:
>>>
>>>>I seem to recall that at one time the Xilinx tools were not compatible
>>>>with Win2k SP3.  Anyone know if that is still an issue?  I assume it
> 
> is
> 
>>>>since I still see SP2 listed under the System Requirements.
>>>
>>>Oh, and I noticed on the MicroSoft website that Windows 2000 is up to
>>>SP4.  It seems like an update to the Xilinx support of W2k is needed.
>>>
>>>--
>>>
>>>Rick "rickman" Collins
>>>
>>>rick.collins@XYarius.com
>>>Ignore the reply address. To email me use the above address with the XY
>>>removed.
>>>
>>>Arius - A Signal Processing Solutions Company
>>>Specializing in DSP and FPGA design      URL http://www.arius.com
>>>4 King Ave                               301-682-7772 Voice
>>>Frederick, MD 21701-3110                 301-682-7666 FAX
>>
>>
> 
> 


Article: 59592
Subject: Re: LVPECL I/O and Fndtn4.2
From: Philip Freidin <philip@fliptronics.com>
Date: Fri, 22 Aug 2003 23:20:28 GMT
Links: << >>  << T >>  << A >>
On Fri, 22 Aug 2003 06:30:32 -0700, Marlboro <> wrote:
>Hi guys, <p>How to get around this error, I try to instantiate a LVPECL output <BR>
>and always get the error during bitgen phase. The UCF file already constrained output net IOSTANDARD = LVPECL and LOC=P56 (XCV100E-PQ240) <BR>
>For my understand, we just need to instantiate the P site only. Is that true? <p>Running DRC. <BR>
>ERROR:DesignRules:441 - Blockcheck: <BR>
>Illegal LVPECL configuration. Comp P_OUT is configured as a LVPECL IOB that uses the OUTMUX, but it's slave site P57 is empty.

Hey Marlboro,

How about learning to post via Google, so that you stop
dumping HTML into this news group. (caused by problems with the
Xilinx gateway to this news group)



===================
Philip Freidin
philip@fliptronics.com
Host for WWW.FPGA-FAQ.COM

Article: 59593
Subject: Re: Async logic in FPGAs
From: rickman <spamgoeshere4@yahoo.com>
Date: Fri, 22 Aug 2003 20:46:58 -0400
Links: << >>  << T >>  << A >>
Ray Andraka wrote:
> 
> Use the command strobes like you are as the clock to an input register, but
> also toggle a toggle flip-flop with the command.  Use the address and any
> qualifiers to produce the clock enable to both the input register and the
> toggle flip-flop.   The toggle flip-flop switches state each time a valid
> write is performed.  Resync the toggle flip-flop output to your local clock,
> and then run it into a simple state machine that generates a 1 cycle wide
> pulse as a response to a change on the input.  That pulse , which is sync'ed
> to your local clock can then be used to validate the contents of the input
> registers.  I usually use it to copy the input register contents to a second
> layer input register that is clocked by the local clock.  It is also used in
> your control logic to let the rest of the circuit know that there is new
> write data.  As long as the bus write frequency is less than about 3 cycles
> of your local clock this works fine as described.  If the bus can write
> faster, then you can write successive writes into a shift register or memory
> and transfer to the local in parallel every Nth write.

That is exactly what the diagram below does.  The FSM is mearly a couple
of sequential registers and the one clock wide strobe is gated at the
difference point.  

The circuit below does not require a fast clock if you have control over
the cycle length with a wait signal.  That is how I am able to work with
a clock range of some 20x the bus write rate to about 1.5x.  The wait
signal slows the bus rate to whatever rate the circuit will run at. 
With the async nature, sometimes it takes an extra clock before the wait
signal is seen by the bus.  

I am pretty sure I have the kinks worked out on paper.  But one problem
with using the control strobes as clocks is that the software wants to
use clock routing for them.  We will see what happens when all the clock
resources are used up.  I guess I will be able to control this
manually.  


> rickman wrote:
> 
> > After reading a post by Jonathan Bromley on the VME interface, I thought
> > I would see if anyone had any comments on a design I am doing.  I have
> > to interface to the PC104 bus which is just the PC ISA interface on a
> > small board.  After beating my head against the wall for a couple of
> > trys, I decided to dump the BCLK since it is not really part of the spec
> > and treat the command stobes as async lines.  My design is further
> > complicated by the fact that my system clock ranges from 2 MHz to 50 MHz
> > to allow power conservation.
> >
> > I finally decided that the only way to get the job done was to treat the
> > command strobes as clock lines.  Read cycles are done normally by muxing
> > the data from various registers based on the address lines and the
> > tristate outputs are controlled with the strobes.  Writes to simple
> > registers are done on the trailing edge of the write strobe.  For reads
> > that need to update or clear bits, a strobe is generated that is sync'ed
> > to the system clock using a circuit to remove the metastable problem.
> > The same circuit works with writes that need to set flags or control
> > FIFOs.
> >
> > I don't see any problem with any of the circuits I have used.  I don't
> > even see the need to use a clock tree for the command strobes since
> > there are no race conditions where a few ns will matter.  Unless there
> > are *huge* differences in routing delays, this circuit should work just
> > fine.
> >
> > Anyone had big problems with similar async circuits?  BTW, here is the
> > simple sync circuit to generate a single pulse in the target clock
> > domain regardless of the relative speed of the clocks.
> >
> >    |------- Metastable -------|
> >             __________
> >            |          |                        _____
> >    |------O| inverter |-------|---------------|     | Pulse
> >    |       |__________|       |               | XOR |---->
> >    |    ______       ______   |   ______   |--|_____| Out
> >    |   |      |     |      |  |  |      |  |
> >    |---| D  Q |-----| D  Q |--|--| D  Q |--|
> > Strobe |      |     |      |     |      |
> > /Clock |      |     |      |     |      |
> > -------|>     |  ---|>     | |---|>     |
> >        |______|  |  |______| |   |______|
> >                  |           |
> >                  |___________|___________  Output Clock
> >
> > The pulse out should be clean by the next clock edge as long as the
> > routing is kept short.  Or if the clock period is very short another FF
> > can be added to feed the other leg of the XOR gate and assure a clean
> > output.
> >
> > If the input clock is faster than the output clock, extra clock edges on
> > the input will be ignored until the signal has clocked through the other
> > clock domain.
> >
> > You can also tap the second FF for the feedback if you want to use it
> > for a handshake.  A second FF on the input side and an XOR gate will
> > give you a pulse on "acknowledge" of the pulse crossing the clock
> > domain.
> >
> > --
> >
> > Rick "rickman" Collins
> >
> > rick.collins@XYarius.com
> > Ignore the reply address. To email me use the above address with the XY
> > removed.
> >
> > Arius - A Signal Processing Solutions Company
> > Specializing in DSP and FPGA design      URL http://www.arius.com
> > 4 King Ave                               301-682-7772 Voice
> > Frederick, MD 21701-3110                 301-682-7666 FAX
> 
> --
> --Ray Andraka, P.E.
> President, the Andraka Consulting Group, Inc.
> 401/884-7930     Fax 401/884-7950
> email ray@andraka.com
> http://www.andraka.com
> 
>  "They that give up essential liberty to obtain a little
>   temporary safety deserve neither liberty nor safety."
>                                           -Benjamin Franklin, 1759

-- 

Rick "rickman" Collins

rick.collins@XYarius.com
Ignore the reply address. To email me use the above address with the XY
removed.

Arius - A Signal Processing Solutions Company
Specializing in DSP and FPGA design      URL http://www.arius.com
4 King Ave                               301-682-7772 Voice
Frederick, MD 21701-3110                 301-682-7666 FAX

Article: 59594
Subject: Altera ACEX 1K IOE
From: rickman <spamgoeshere4@yahoo.com>
Date: Fri, 22 Aug 2003 22:40:37 -0400
Links: << >>  << T >>  << A >>
I am working with the ACEX 1K30 and I don't understand the structure of
the IOE.  The diagram clearly shows three FFs; data input, data output
and output enable.  The three FFs even have different types of
connections to the routing.  But in the text, they seem to be saying
that there is only a single FF in the IOE.  

Is there really only one FF in the IOE so that it can be either input or
output?  Is the OE FF always in an LE rather than the IOE?  Seems like
this would make for some trouble getting the timing in tight
situations.  

-- 

Rick "rickman" Collins

rick.collins@XYarius.com
Ignore the reply address. To email me use the above address with the XY
removed.

Arius - A Signal Processing Solutions Company
Specializing in DSP and FPGA design      URL http://www.arius.com
4 King Ave                               301-682-7772 Voice
Frederick, MD 21701-3110                 301-682-7666 FAX

Article: 59595
(removed)


Article: 59596
(removed)


Article: 59597
(removed)


Article: 59598
Subject: Thinking out loud about metastability
From: rickman <spamgoeshere4@yahoo.com>
Date: Sat, 23 Aug 2003 00:05:42 -0400
Links: << >>  << T >>  << A >>
I am designing a FPGA interface for an MCU which uses only async
strobes.  However they are not truely async, in reality the timing to
the clock is simply not spec'd.  So if I understand metastability
correctly, I can not use the standard circuit to reduce the failure rate
to insignificant rates.  Since there is a fixed relationship between the
clock and the command strobes (even though it is unknown) no matter how
I choose to clock the circuit, I may end up balancing the pencil well
enough on end that I will see a much higher failure rate than predicted
by the standard formula.  

My understanding is that the standard formula assumes that the two
signals are not related in frequency.  So the probability that they will
be at the right point to cause a failure depends on the rate of each and
the width of the "failure window".  In my case the probability of the
event being in the failure window can be very high if the timing is just
right.  

Other than the obvious of using a separate clock for the FPGA, how can I
design this circuit and get a low enough failure rate?  I guess adding
more time will reduce the failure rate, but how do I know how much time
is enough since the standard formula does not apply?  

In this particular circuit I expect the strobes to change on the rising
edge of the clock at about 50 MHz.  Since I know nothing about the
actual relationship between the clock and the strobes, I have picked the
falling edge to clock the strobes into the chip.  I assume that if the
outputs change slightly before the clock rising edge, it will not be
much and if they change later, it may be short enough to still give me
close to the 3 ns setup time I need.  Then a rising edge FF syncs the
signal to the rest of the circuit and provides some 7 ns (out of 10 ns
clock difference) for settling.  

I have no idea how good of an assumption this is.  The output delays on
the CPU may be enough to put the strobe edge right in the critical point
to cause a failure.  Even if I use an emperical approach and see which
edge to clock on to make the circuit work, I don't know that this will
operate over temperature, process, etc...

I believe that some of the posts here have indicated that the "failure
window" in terms of time is in the low ps (or was it fs?) range.  If my
circuit allows some 7 ns for settling, will this be enough to put the
MTBF above 100 years in a worse case situation?  Is it possible to
balance the pencil well enough to make a circuit like this fail
frequently?  

-- 

Rick "rickman" Collins

rick.collins@XYarius.com
Ignore the reply address. To email me use the above address with the XY
removed.

Arius - A Signal Processing Solutions Company
Specializing in DSP and FPGA design      URL http://www.arius.com
4 King Ave                               301-682-7772 Voice
Frederick, MD 21701-3110                 301-682-7666 FAX

Article: 59599
Subject: Re: Thinking out loud about metastability
From: Jon Elson <elson@pico-systems.com>
Date: Fri, 22 Aug 2003 23:59:24 -0500
Links: << >>  << T >>  << A >>


rickman wrote:

>I am designing a FPGA interface for an MCU which uses only async
>strobes.  However they are not truely async, in reality the timing to
>the clock is simply not spec'd.  So if I understand metastability
>correctly, I can not use the standard circuit to reduce the failure rate
>to insignificant rates.  Since there is a fixed relationship between the
>clock and the command strobes (even though it is unknown) no matter how
>I choose to clock the circuit, I may end up balancing the pencil well
>enough on end that I will see a much higher failure rate than predicted
>by the standard formula.  
>  
>
I think you need to make some measurements, if you can't get info from
the manufacturer.  Most likely, the timing will be pretty simple to figure
out, as signals, especially strobes, will almost certainly either be clocked
through a flip-flop right at the output, or have a minimum of gates between
the last FF and the output.

You really CAN'T work with so little information.  What is the setup time
of data and address signals from the CPU, relative to the strobes?  What is
the required setup and hold time of data and addresses presented to the
CPU from outside?

Or, do they provide all this in great detail, just never referencing any 
of this
to the CPU clock?

If you can't beat this information out of the manufacturer, then you 
should be
able to measure it fairly easily.

Jon




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
2017JanFebMarApr2017

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