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 143825

Article: 143825
Subject: Re: synplify question for FPGA
From: Mike Treseler <mtreseler@gmail.com>
Date: Tue, 27 Oct 2009 20:48:54 -0700
Links: << >>  << T >>  << A >>
skyworld wrote:

> thanks for your reply. To my understanding, syn_keep is used to keep
> wire signals and syn_preserve is used to keep sequential signals. The
> signals I want to keep is generated in process(clk, reset)

Synthesis works from a device input pin to an output pin.
If my process(clk, reset) does not look at an input pin
*and* drive an output pin in the top entity,
I get no gates and no flops, in any case.

 -- Mike Treseler




Article: 143826
Subject: Re: synplify question for FPGA
From: skyworld <chenyong20000@gmail.com>
Date: Tue, 27 Oct 2009 21:48:10 -0700 (PDT)
Links: << >>  << T >>  << A >>
On 10=D4=C228=C8=D5, =C9=CF=CE=E711=CA=B148=B7=D6, Mike Treseler <mtrese...=
@gmail.com> wrote:
> skyworld wrote:
> > thanks for your reply. To my understanding, syn_keep is used to keep
> > wire signals and syn_preserve is used to keep sequential signals. The
> > signals I want to keep is generated in process(clk, reset)
>
> Synthesis works from a device input pin to an output pin.
> If my process(clk, reset) does not look at an input pin
> *and* drive an output pin in the top entity,
> I get no gates and no flops, in any case.
>
>  -- Mike Treseler

Hi,
does this mean I don't have a way to monitor these signals? thanks.


regards
skyworld

Article: 143827
Subject: Re: synplify question for FPGA
From: Matthew Hicks <mdhicks2@uiuc.edu>
Date: Wed, 28 Oct 2009 04:50:58 +0000 (UTC)
Links: << >>  << T >>  << A >>
I had a similar problem with chipscope cores.  Try using the syn_noprune 
attribute which is an integer with a value of 1.  This worked for me where 
syn-keep failed.


---Matthew Hicks


> On 10月27日, 下午7时43分, skyworld <chenyong20...@gmail.com> wrote:
> 
>> Hi,
>> 
>> I inserted some circuit in source code for debug. These code aren't
>> ported as output and is monitored by chipscope. In order not to be
>> optimised by synplify, I use attribute syn_preserve to keep these
>> registers. But synplify still removed these signals, I can't insert
>> these signals into cdc file in ISE. Is there a way to keep these
>> signals so that it can be used in chipscope?
>> 
>> for example,
>> 
> signal a : std_logic;
> signal b : std_logic;
> attribute syn_preserve : boolean;
> attribute syn_preserve of a : signal is true;
> attribute syn_preserve of b: signal is true;
> process(clk, reset)
> begin
> a <= ...
> b <= ...
> end
> signal a/b are internal signals and not sent to output port. how can I
> keep these signals? thanks.
> 



Article: 143828
Subject: Re: HI.. Help Needed Its Urgent
From: Mark McDougall <markm@vl.com.au>
Date: Wed, 28 Oct 2009 17:40:39 +1100
Links: << >>  << T >>  << A >>
kclo4 wrote:

> do they do not teach politeness in school anymore those days?? no
> please, no care for syntax, only "do my homework quickly and i do not
> want to search by myself"...

A sign of the times, unfortunately.

And usually, it's "...and don't waste my time trying to teach me _how_ to
do it - I'm not at all interested. Just tell me where to find a
pre-packaged, working solution! Now!"

Regards,

-- 
Mark McDougall, Engineer
Virtual Logic Pty Ltd, <http://www.vl.com.au>
21-25 King St, Rockdale, 2216
Ph: +612-9599-3255 Fax: +612-9599-3266

Article: 143829
Subject: Re: Ideas for a pulse programmer needed
From: Paul Pham <paulpham@yahoo.com>
Date: Wed, 28 Oct 2009 02:01:58 -0700 (PDT)
Links: << >>  << T >>  << A >>
Hi jmariano,

I have implemented a similar FPGA-based pulse programmer system for my
master's thesis, but instead of a Microblaze, I ended up designing a
custom processor and instruction set architecture just for outputting
pulses, detecting input pulses, looping, executing subroutines, and
branching conditionally. The problem with using a general-purpose
processor is that precise timing of the output cannot be guaranteed,
so I assume you are just using the Microblaze to load your banks of
SRAM before letting the timer run and stream the pulses directly from
memory.

There appear to be several Microblaze modules available on
http://www.opencores.org, maybe you can adapt some of them to your
needs.

All of my project's design documents are released as open source on
SourceForge. Perhaps the VHDL source code will give you some ideas,
although the board I used is a custom design (also released on the
website).

http://pulse-sequencer.svn.sourceforge.net/viewvc/pulse-sequencer/firmware/=
sequencer-vhdl/

I assume you are designing this pulse programmer for scientific
applications, and if so, I would like to hear about it by personal e-
mail.

Good luck,
Paul

On Oct 6, 8:42=A0am, jmariano <jmarian...@gmail.com> wrote:
> Dear Group,
>
> I have very little experience in FPGA (and in digital design!).
>
> As part of a research project I have to add to an existing microblaze
> system, implemented on spartan 3 starter kit board, apulseprogrammer
> (PP). A PP is a system that outputs a given pattern to a set of
> digital lines for a given time and then changes the pattern according
> to a program.



> There are several ways of implementing the PP but I have decided to
> use what seams to me to be the simpler on: two blocks of RAM, say 2K
> deep and 16 bits wide, pointed by the same address counter. One block
> holds the time duration and the other the bit pattern. The control
> block load's the contents of the first ram into a counter and latches
> the content of the second one to the output. When the counter reaches
> the end, the AC in incremented and the next time and pattern words are
> loaded.
>
> The IP access to RAM must be fast (this determines the time resolution
> of the PP), but the access of the processor can be slow, since this is
> done only once at the beginning of the experiment to write the
> programming words and then the IP works by is one.
>
> I was thinking on using BRAM to hold the data. Is this a good choice?
> My other question is, what is the easiest way to implement microblaze
> access the ram? I appreciate any commets on this.
>
> I also appreciate if you could point me to somme examples or
> application notes of a similar system (not a PP, but a system were
> memory is accessed by an IP and microblaze), were I can get somme
> ideas.
>
> Tank you very much,
>
> jmariano


Article: 143830
Subject: Re: synplify question for FPGA
From: skyworld <chenyong20000@gmail.com>
Date: Wed, 28 Oct 2009 05:35:25 -0700 (PDT)
Links: << >>  << T >>  << A >>
On 10=E6=9C=8828=E6=97=A5, =E4=B8=8B=E5=8D=8812=E6=97=B650=E5=88=86, Matthe=
w Hicks <mdhic...@uiuc.edu> wrote:
> I had a similar problem with chipscope cores. =C2=A0Try using the syn_nop=
rune
> attribute which is an integer with a value of 1. =C2=A0This worked for me=
 where
> syn-keep failed.
>
> ---Matthew Hicks
>
>
>
> > On 10=E6=9C=8827=E6=97=A5, =E4=B8=8B=E5=8D=887=E6=97=B643=E5=88=86, sky=
world <chenyong20...@gmail.com> wrote:
>
> >> Hi,
>
> >> I inserted some circuit in source code for debug. These code aren't
> >> ported as output and is monitored by chipscope. In order not to be
> >> optimised by synplify, I use attribute syn_preserve to keep these
> >> registers. But synplify still removed these signals, I can't insert
> >> these signals into cdc file in ISE. Is there a way to keep these
> >> signals so that it can be used in chipscope?
>
> >> for example,
>
> > signal a : std_logic;
> > signal b : std_logic;
> > attribute syn_preserve : boolean;
> > attribute syn_preserve of a : signal is true;
> > attribute syn_preserve of b: signal is true;
> > process(clk, reset)
> > begin
> > a <=3D ...
> > b <=3D ...
> > end
> > signal a/b are internal signals and not sent to output port. how can I
> > keep these signals? thanks.- =E9=9A=90=E8=97=8F=E8=A2=AB=E5=BC=95=E7=94=
=A8=E6=96=87=E5=AD=97 -
>
> - =E6=98=BE=E7=A4=BA=E5=BC=95=E7=94=A8=E7=9A=84=E6=96=87=E5=AD=97 -

In fact I have tried this. First I tried to use syn_noprune to keep
these signals, but synplify reports this is an error usage for
syn_noprune.

At last I found the solution: I use core generator to generate icon
and ila, I connect these signals to ila and use syn_noprune to keep
ila and icon. In this way I can observe those signals in chipscope.
But this is a complicated one because VHDL doesn't like verilog, you
can infer signals from different levels, such as
level1.level2.signal_a, I have to connect signals from different
module to the correct module which I put the ila and icon. So do
anybody has a good way to use those attributes to keep these signals?
thanks.

Article: 143831
Subject: Re: synplify question for FPGA
From: General Schvantzkoph <schvantzkoph@yahoo.com>
Date: 28 Oct 2009 12:42:17 GMT
Links: << >>  << T >>  << A >>
On Tue, 27 Oct 2009 18:28:11 -0700, skyworld wrote:

> On 10月27日, 下午9时31分, General Schvantzkoph <schvantzk...@yahoo.com> 
wrote:
>> On Tue, 27 Oct 2009 04:43:25 -0700, skyworld wrote:
>> > Hi,
>>
>> > I inserted some circuit in source code for debug. These code aren't
>> > ported as output and is monitored by chipscope. In order not to be
>> > optimised by synplify, I use attribute syn_preserve to keep these
>> > registers. But synplify still removed these signals, I can't insert
>> > these signals into cdc file in ISE. Is there a way to keep these
>> > signals so that it can be used in chipscope?
>>
>> > for example,
>>
>> Rather then rely on syn_preserve I generally use a spare pin and then
>> OR all the signals that I want to protect together and connect the OR
>> to the pin.
> 
> Hi,
> 
> thanks for your reply. This is a big design and I don't have an option
> to connect debug signals to a pin. That is the reason why I use
> chipscope and use those annoying attributes.

Another technique that I've used is to add a bit to a status register 
that accessed from a control/status register bus. Anything that would 
have no effect on the actual operation of the design but fools the 
synthesis tool into thinking that the registers are necessary will do the 
trick.

Article: 143832
Subject: Chipscope with Verilog
From: "maxascent" <maxascent@yahoo.co.uk>
Date: Wed, 28 Oct 2009 08:36:07 -0500
Links: << >>  << T >>  << A >>
I have a design with some level of hierarchy and I want to connect a
chipscope ILA core to the bottom level.

For example say I have 3 modules A, B and C with a signal temp in C. 
Module A is the top level with B inside A and C inside B.
I would of thought that I could just do the following

chipscope_ila U_ila(
   .CLK     (clk),
   .CONTROL (control),
   .TRIG0   (A.B.C.temp));

The ila is in the top level and I want to monitor temp.
When I try and synthesize it with Synplify the program just errors.
Can anyone tell me what I am ddoing wrong?

Thanks

Jon	   
					
---------------------------------------		
This message was sent using the comp.arch.fpga web interface on
http://www.FPGARelated.com

Article: 143833
Subject: Re: ISe 10.1 nightmare bug
From: Gabor <gabor@alacron.com>
Date: Wed, 28 Oct 2009 08:14:06 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Oct 27, 4:50=A0am, Antti <antti.luk...@googlemail.com> wrote:
> On Oct 27, 10:45=A0am, Matthieu Michon
>
>
>
> <matthieu.d.u.m.m.y.mic...@gmail.com> wrote:
> > On Mon, 26 Oct 2009 12:43:12 -0700 (PDT)
>
> > Mawa_fugo <cco...@netscape.net> wrote:
> > > On Oct 24, 11:51=A0am, kevin93 <ke...@whitedigs.com> wrote:
> > > > On Oct 23, 12:32=A0pm, Mawa_fugo <cco...@netscape.net> wrote:
>
> > > > > I have SP3 installed in the 10.1 - =A0but sometimes - once a whil=
e, the
> > > > > entire project just corrupted - when "rerun all" it TOOK the topm=
odule
> > > > > source from "nowhere" - nomatter how you change your topmodule it
> > > > > still lock the topmodule source fom that mystery source
>
> > > > > Oh my goodness
>
> > > > The "Cleanup Project Files" under the "Project" menu can solve many=
 of
> > > > these problems.
>
> > > > kevin
>
> > > Thanks for suggestion - it does something difference when I tried to
> > > clean the project but it still lock the source from "nowhere"
>
> > Hi
>
> > I also had to deal with the same kind of issue with ISE 10.1SP3: after =
checking out a project stored on a version control system (SVN), the top-le=
vel mark disappeared and the "Set as Top Module" action was disabled. =A0Ne=
edless to say that I tried all the usuals (amongst other things: "Cleanup P=
roject Files", "Check Syntax", removing/adding again the top-level source f=
ile) without much success.
>
> > After a few minutes fiddling around, I tried to change the "Top-Level S=
ource Type" project property from HDL to EDIF and back to HDL --define long=
 shoot ;) -- and YGTBK! my top-level came back!
>
> > > THis is a night mare
>
> > At the very least it must scare some EEs away from FPGA design =A0;) =
=A0One a more serious note, I'm still hoping that five years after the buyo=
ut of Hier Design, the people behind Planahead didn't get brainwashed and w=
ould suggest the rest of the Design Software Division at Xilinx to put some=
 common sense in ISE (such as using --again-- a text format for the project=
 file).
>
> > --
> > Matthieu Michon <prenom....@gmail.com>- Hide quoted text -
>
> > - Show quoted text -
>
> 11.x uses text format again
>
> so at least once Xilinx has listened, well I bet they just had no
> choice as the binary project file in their implementation is nothing
> else and pure nightmare
>
> Antti

10.x has a project restore file, which is a bit of a pain to use
but not as bad as copying all your sources and building a new
hierarchy and remembering all of the build settings.  Just run
the <project_name>.restore file in a Tcl shell (directions are
included in the .restore file - yes it's a text file).

10.x has a problem with certain background applications including
Carbonite backup for instance that insert themselves into the
file system.  I imagine your source control may fall in this
category.  You could try to disable it while running ISE from
the GUI and see if the problem goes away.

Regards,
Gabor

Article: 143834
Subject: ML605 Evaluation Kit and FPGA Mezzanine Connectors (FMC) ?
From: LucienZ <lucien.zhang@gmail.com>
Date: Wed, 28 Oct 2009 08:27:04 -0700 (PDT)
Links: << >>  << T >>  << A >>
Dear all,

Has anyone ordered and used the ML605 Evaluation Kit?
http://www.xilinx.com/products/devkits/EK-V6-ML605-G.htm
I noticed some attractive features like DDR3, PCIe Gen 2 and quite
high FPGA capacity (241,152 logic cells). Before my ordering I would
like to hear some comments.

Another feature of this board is the adoption of the FMC interfaces.
This is a new face to me, but it looks quite promising. However I have
a concern about it: if I just want to interface some simple IO devices
(e.g. a LCD panel and some push buttons), does that mean I still have
to do the PCB layout for them with a FMC connection (previously it
could be solved just by some IO pin headers)?

I also do not quite understand the pricing, I notice that the
XC6VLX240T-1FFG1156CES chip itself costs about $2,100, but the fledged
ML506 Kit only takes $1,995. Is this just a marketing scheme or there
are some other reasons?

Thanks very much for your time!
Lucien

Article: 143835
Subject: Re: Tcl in PlanAhead
From: krish <krishnans@hotmail.com>
Date: Wed, 28 Oct 2009 08:33:19 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Oct 26, 10:38=A0pm, Kastil Jan <ikas...@stud.fit.vutbr.cz> wrote:
> Hello all,
> I am hoping that someone had solved problem similar to mine. I want to ru=
n
> tcl
> script in planAhead to create project, run some implementation and import
> the results into the floorplan. So I run these commands:
>
> hdi::run add -name impl_1 -project project_bft_core_hdl -floorplan
> floorplan_1 -parentRun synth_2 -part {xc5vsx35tff665-1} -flow {ISE 11}
> -strategy {Strategy #1};
> hdi::run launch -project project_bft_core_hdl
> -runs impl_1 -jobs 2 -scriptsOnly no -allPlacement no -dir
> /.../Projects/project_bft_core_hdl/project_bft_core_hdl.runs;
> hdi::run import -name impl_1 -project project_bft_core_hdl -placement yes
>
> Unfortunatelly I receive error:
>
> ERROR: Run 'impl_1' appears to not be finished. Unable to import
>
> Is there any way how to force tcl script to wait for the result of the
> dhi::run launch?
>
> Thanks for every answer
>
> Jan

 hdi::run wait -name impl_1 -project project_bft_core_hdl

Article: 143836
Subject: Re: Teammates, interested?
From: Svenn Are Bjerkem <svenn.bjerkem@googlemail.com>
Date: Wed, 28 Oct 2009 12:12:49 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Oct 21, 5:49=A0pm, nobody <cydrollin...@gmail.com> wrote:

> me? Now if a collection of individuals got together and developed an
> account it looks more like a firm many hands producing on the capital
> someone spends to get the job done. That is actually the way it works
> together we get the job done quick and capable. But that is one idea
> among one individual, now put several individuals together and see
> what we can get done.

How would you make sure that every team member got his/her fair share
of the revenue?
How would licensing of necessary tools be made? What about lab
testing? Production and distribution of evaluation boards? How will
NDAs be handled?

I am a simple salaryman, but I see how big the bucks are in FPGA
programming. I would really like to have a piece of that cake, but it
is safer to stay in a paid job than taking the risk of self-employment
even though my FPGA coding make my company rich and prosperous. I
would join your team, but I would not give up my regular job and that
is going to limit me. I can't work on things that would be in direct
competition to my employer, all my code would be tainted, and sooner
or later I would face legal actions if I get too successful as a
contributor. I think your idea is good and even if others have failed
that is not a reason to not try to make it work.

Good luck and thumbs up,
--
Svenn

Article: 143837
Subject: Re: ML605 Evaluation Kit and FPGA Mezzanine Connectors (FMC) ?
From: Ed McGettigan <ed.mcgettigan@xilinx.com>
Date: Wed, 28 Oct 2009 18:05:10 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Oct 28, 8:27=A0am, LucienZ <lucien.zh...@gmail.com> wrote:
> Dear all,
>
> Has anyone ordered and used the ML605 Evaluation Kit?http://www.xilinx.co=
m/products/devkits/EK-V6-ML605-G.htm
> I noticed some attractive features like DDR3, PCIe Gen 2 and quite
> high FPGA capacity (241,152 logic cells). Before my ordering I would
> like to hear some comments.
>
> Another feature of this board is the adoption of the FMC interfaces.
> This is a new face to me, but it looks quite promising. However I have
> a concern about it: if I just want to interface some simple IO devices
> (e.g. a LCD panel and some push buttons), does that mean I still have
> to do the PCB layout for them with a FMC connection (previously it
> could be solved just by some IO pin headers)?
>
> I also do not quite understand the pricing, I notice that the
> XC6VLX240T-1FFG1156CES chip itself costs about $2,100, but the fledged
> ML506 Kit only takes $1,995. Is this just a marketing scheme or there
> are some other reasons?
>
> Thanks very much for your time!
> Lucien

Yes, the features do make it a pretty cool board. :-)

The main board already has push buttons, LEDS and a LCD so you are
covered there. If you do want add on additional peripherals then yes,
you will need to design a board as it isn't practical to attach wires
to the FMC connectors.

Xilinx will/should be releasing a FMC that is primarily intending for
simple debug and is comprised of a series of 100 mil headers that are
connected to the FMC connector.  This board could also be used for
prototyping purposes.

Ed McGettigan
--
Xilinx Inc.

Article: 143838
Subject: Re: ML605 Evaluation Kit and FPGA Mezzanine Connectors (FMC) ?
From: LucienZ <lucien.zhang@gmail.com>
Date: Wed, 28 Oct 2009 18:43:29 -0700 (PDT)
Links: << >>  << T >>  << A >>
> Yes, the features do make it a pretty cool board. :-)
>
> The main board already has push buttons, LEDS and a LCD so you are
> covered there. If you do want add on additional peripherals then yes,
> you will need to design a board as it isn't practical to attach wires
> to the FMC connectors.

Thanks Ed!

Maybe I would like to interface a color LCD and two CMOS image
sensors, which are not very demanding for the IO performance. Without
testing I do not want to make a custom PCB first...

> Xilinx will/should be releasing a FMC that is primarily intending for
> simple debug and is comprised of a series of 100 mil headers that are
> connected to the FMC connector.  This board could also be used for
> prototyping purposes.

This is what I would like to see for fast prototyping some simple IO
devices :). Do you already have such a plan? And when could such a
daughter board be available?

Article: 143839
Subject: Re: ML605 Evaluation Kit and FPGA Mezzanine Connectors (FMC) ?
From: Ed McGettigan <ed.mcgettigan@xilinx.com>
Date: Wed, 28 Oct 2009 20:56:46 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Oct 28, 6:43=A0pm, LucienZ <lucien.zh...@gmail.com> wrote:
> > Yes, the features do make it a pretty cool board. :-)
>
> > The main board already has push buttons, LEDS and a LCD so you are
> > covered there. If you do want add on additional peripherals then yes,
> > you will need to design a board as it isn't practical to attach wires
> > to the FMC connectors.
>
> Thanks Ed!
>
> Maybe I would like to interface a color LCD and two CMOS image
> sensors, which are not very demanding for the IO performance. Without
> testing I do not want to make a custom PCB first...
>
> > Xilinx will/should be releasing a FMC that is primarily intending for
> > simple debug and is comprised of a series of 100 mil headers that are
> > connected to the FMC connector. =A0This board could also be used for
> > prototyping purposes.
>
> This is what I would like to see for fast prototyping some simple IO
> devices :). Do you already have such a plan? And when could such a
> daughter board be available?

Yes, we have this in the current plan and have the first article in
house.  It should be released by the end of the year, but it could
slip into January.  Pricing hasn't been set yet, but since it appears
that we are subsidizing the ML605 we'll probably have to make it up on
the modules.  Sort of the razor and razor blades model. :-)

I have been concerned about the hobbyist/academic/small-shop use of
the FMC interface.  I'm hoping that we can find the time to put
together some basic LPC and HPC 6 layer PCB examples using the free
tools from low cost, quick turn PCB fabricators like PCB123.  These
could then be used as the basis for a quick custom FMC board.

Ed McGettigan
--
Xilinx Inc.

Article: 143840
Subject: Best way to model a large external ROM in a simulation? (XST
From: TSMGrizzly <sbattazz@yahoo.co.jp>
Date: Thu, 29 Oct 2009 03:21:35 -0700 (PDT)
Links: << >>  << T >>  << A >>
Hi, I am working on designing a state machine which will read some
data from a relatively large external ROM and process it.

I expect the ROM size to be on the order of 32k x 32 bits, or 64k x 32
bits (quite likely the latter because my data table is just a little
larger than 32k entries at the moment).
Is there a good way to model this in a VHDL testbench in such a way
that XST simulator can handle it?

I don't really want to have to build a gigantic constant assignment
table, but I also suspect that declaring an array of std_logic_vectors
as a signal and then writing them one by one in from a file would
probably take a really long time to execute.
I searched around a while last week but didn't really find any
concrete answers on this...

Cheers,

TSM

Article: 143841
Subject: Trouble in booting V5 FPGA from SPI flash.
From: crescent <hanpei369@gmail.com>
Date: Thu, 29 Oct 2009 03:39:20 -0700 (PDT)
Links: << >>  << T >>  << A >>
Hi all,
I am now using an Atmel AT45DB321D SPI flash booting xc5vsx50t-ff1136
and there is something wrong. First, the FPGA can be programmed
successfully, and the flash also can be programmed successfully both
directly & indirectly via iMPACT. But the FPGA just can't boot from
the flash. I checked the timing diagram of the signals: PROG_B,
INIT_B, M[2:0], FS[2:0], FCS_B, CCLK, MOSI, DIN, the MOSI did not
function right compared to the timing diagram in XAPP951 page8 (M[2:0]
=001,FS[2:0]=111). Pin MOSI clocks out 001100000.... after FCS_B goes
low and it should be 000010110...
I have no idea what's wrong and what should I do to fix this. So any
help would be thankfull!

Article: 143842
Subject: Re: save data from adc in text file
From: TSMGrizzly <sbattazz@yahoo.co.jp>
Date: Thu, 29 Oct 2009 03:59:35 -0700 (PDT)
Links: << >>  << T >>  << A >>
I'm a little bit confused-- Why are you using the microblaze-xilinx-
elf targeted gcc to compile what should be the PC-side application
where you want to do fprintf?
What does your source code look like?

I've never used microblaze, but my assumption is that you'd want to
use the microblaze gcc only to compile that code that goes onto your
microblaze on the FPGA.
Then you could compile a regular C (or other language) program on the
computer to send commands to your FPGA board and listen for data
coming back, which can then be easily written to a text file for later
analysis.

Personally I like to use Python to do this, though, because once you
have pyserial installed, the serial port routines are a breeze to work
with.. and matplotlib is nice for graphing results, too.

Writing to a text file with VHDL is useful to log results from a
simulation, not from an actual running chip.

Article: 143843
Subject: Re: Best way to model a large external ROM in a simulation? (XST
From: KJ <kkjennings@sbcglobal.net>
Date: Thu, 29 Oct 2009 04:14:50 -0700 (PDT)
Links: << >>  << T >>  << A >>
On Oct 29, 6:21=A0am, TSMGrizzly <sbatt...@yahoo.co.jp> wrote:
>
> I don't really want to have to build a gigantic constant assignment
> table,

Why not?  That's the fastest way to accomplish the task...fast both
for development as well as fast sim.

> but I also suspect that declaring an array of std_logic_vectors
> as a signal and then writing them one by one in from a file would
> probably take a really long time to execute.

Well, the first order of business is that you won't be writing an
array of std_logic_vectors to a file in order model a ROM, you would
be reading them in from a file.

Second, your suspicions are not correct, the sim time to read in and
parse the ROM data would be negligible.  The development time to do
this however, would be considerably longer though then simply defining
the constant array.  It's up to you to decide on how to spend your
time.

> I searched around a while last week but didn't really find any
> concrete answers on this...

You may find an already written model for the part that you'll be
using on
http://www.freemodelfoundry.com/

Kevin Jennings

Article: 143844
Subject: error while opening hex file
From: "zss" <zarithss@yahoo.com>
Date: Thu, 29 Oct 2009 06:22:33 -0500
Links: << >>  << T >>  << A >>
Hi, I'm designing using ROM and hex file.I'm new at using this.
I've created this .hex file for the ROM and everytime I close the quartus
window and open the hex file again to edit it, it says

"cant open file -- file contains ni memory cell values"..??

what I did wrong?help



Article: 143845
Subject: Re: save data from adc in text file
From: glen herrmannsfeldt <gah@ugcs.caltech.edu>
Date: Thu, 29 Oct 2009 11:23:15 +0000 (UTC)
Links: << >>  << T >>  << A >>
TSMGrizzly <sbattazz@yahoo.co.jp> wrote:
> I'm a little bit confused-- Why are you using the microblaze-xilinx-
> elf targeted gcc to compile what should be the PC-side application
> where you want to do fprintf?
> What does your source code look like?

Well, one could connect up some device to an FPGA mircoblaze
that one could fprintf to.  A serial printer might not be so
hard, a little harder for a disk drive and file system.
Maybe NFS through ethernet wouldn't be so bad, either.
 
> I've never used microblaze, but my assumption is that you'd want to
> use the microblaze gcc only to compile that code that goes onto your
> microblaze on the FPGA.
> Then you could compile a regular C (or other language) program on the
> computer to send commands to your FPGA board and listen for data
> coming back, which can then be easily written to a text file for later
> analysis.

Yes, it doesn't seem that the OP has such attached to the microblaze.

-- glen

Article: 143846
Subject: Re: Best way to model a large external ROM in a simulation? (XST
From: TSMGrizzly <sbattazz@yahoo.co.jp>
Date: Thu, 29 Oct 2009 04:31:49 -0700 (PDT)
Links: << >>  << T >>  << A >>
> > I don't really want to have to build a gigantic constant assignment
> > table,
>
> Why not? =A0That's the fastest way to accomplish the task...fast both
> for development as well as fast sim.

Well, because I didn't really want such a long constant table as part
of my testbench, though I suppose I can keep it in package in a
separate file where it would be out of the way.

> Well, the first order of business is that you won't be writing an
> array of std_logic_vectors to a file in order model a ROM, you would
> be reading them in from a file.

Yes, this is what I meant.. writing the values into the
std_logic_vectors after having read them from the file.

>
> Second, your suspicions are not correct, the sim time to read in and
> parse the ROM data would be negligible. =A0The development time to do
> this however, would be considerably longer though then simply defining
> the constant array. =A0It's up to you to decide on how to spend your
> time.

I don't really see too much difference in the time spent either way. I
have some code from my last testbench where I read stimulus in from a
text file, so it should be pretty quick to adapt. Admittedly a script
to write the constant table would be a little quicker, so I will
probably go with your suggestion on that.


> You may find an already written model for the part that you'll be
> using onhttp://www.freemodelfoundry.com/

Awesome, I will take a look.

>
> Kevin Jennings

Thanks for the quick answer! Sorry to be bothering you with such a
trivial question.

Article: 143847
Subject: Re: ML605 Evaluation Kit and FPGA Mezzanine Connectors (FMC) ?
From: Brian Drummond <brian_drummond@btconnect.com>
Date: Thu, 29 Oct 2009 12:49:34 +0000
Links: << >>  << T >>  << A >>
On Wed, 28 Oct 2009 20:56:46 -0700 (PDT), Ed McGettigan
<ed.mcgettigan@xilinx.com> wrote:

>On Oct 28, 6:43pm, LucienZ <lucien.zh...@gmail.com> wrote:
>> > Yes, the features do make it a pretty cool board. :-)

>> > Xilinx will/should be releasing a FMC that is primarily intending for
>> > simple debug and is comprised of a series of 100 mil headers that are
>> > connected to the FMC connector. This board could also be used for
>> > prototyping purposes.
>>
>> This is what I would like to see for fast prototyping some simple IO
>> devices :). Do you already have such a plan? And when could such a
>> daughter board be available?
>
>I have been concerned about the hobbyist/academic/small-shop use of
>the FMC interface.  I'm hoping that we can find the time to put
>together some basic LPC and HPC 6 layer PCB examples using the free
>tools from low cost, quick turn PCB fabricators like PCB123.  These
>could then be used as the basis for a quick custom FMC board.

If you don't want to get involved with the fabrication etc, simply make the
PCB123 design files available in the download section.

One for a debug-type board (0.1" headers).

Possibly one for an 0.1" stripboard with power planes and FMC connector.

And DEFINITELY one for an empty FMC board with the correct outline and the
connector placed, possibly also tracked out. 

I gave up on the PCB123 schematic editor long ago; it's easier to edit the
netlist, so these board files should be accompanied by the netlist (and/or
schematic if appropriate).

- Brian

Article: 143848
Subject: Re: Best way to model a large external ROM in a simulation? (XST simulator)
From: Brian Drummond <brian_drummond@btconnect.com>
Date: Thu, 29 Oct 2009 13:45:46 +0000
Links: << >>  << T >>  << A >>
On Thu, 29 Oct 2009 04:31:49 -0700 (PDT), TSMGrizzly <sbattazz@yahoo.co.jp>
wrote:

>> > I don't really want to have to build a gigantic constant assignment
>> > table,
>>
>> Why not? That's the fastest way to accomplish the task...fast both
>> for development as well as fast sim.
>
>Well, because I didn't really want such a long constant table as part
>of my testbench, though I suppose I can keep it in package in a
>separate file where it would be out of the way.

Keep it as a separate file. The tool that creates it simply has to print a few
lines of boilerplate VHDL at each end.

>> Second, your suspicions are not correct, the sim time to read in and
>> parse the ROM data would be negligible. The development time to do
>> this however, would be considerably longer though then simply defining
>> the constant array. 

Seconded. If you need to read files created by other tools (e.g. a spreadsheet,
or a binary file like a .bmp) this is a reasonable way to do it, and more
flexible than the above approach. However ISE Simulator has some
not-well-documented obstacles(*) to reading binary files, which is more
straightforward in Modelsim.

(*) As of ISE10:
a) it is opposite-endian to Modelsim's treatment of the same file.
b) ISIM writes an undocumented header before the file contents, and expects to
see that header on any file it reads. You can convert a 3rd party file to/from
ISIM using head, tail and cat to extract/insert the header (on Linux, or the
Windows equivalent commands) but it's a bother.
c)  Xilinx have declined to document this header. However there is a Change
Request to fix these issues; I have no idea if/when it will happen.

- Brian

Article: 143849
Subject: Re: error while opening hex file
From: Brian Drummond <brian_drummond@btconnect.com>
Date: Thu, 29 Oct 2009 13:47:05 +0000
Links: << >>  << T >>  << A >>
On Thu, 29 Oct 2009 06:22:33 -0500, "zss" <zarithss@yahoo.com> wrote:

>Hi, I'm designing using ROM and hex file.I'm new at using this.
>I've created this .hex file for the ROM and everytime I close the quartus
>window and open the hex file again to edit it, it says
>
>"cant open file -- file contains ni memory cell values"..??
>
>what I did wrong?help

You may need to build a shrubbery.

- Brian



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