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 103250

Article: 103250
Subject: hard disk drivers problem
From: "bjzhangwn" <bjzhangwn@126.com>
Date: 29 May 2006 08:59:36 -0700
Links: << >>  << T >>  << A >>
I am writing a interface about the ata interface but I have a problem
that the dma request signal is sometimes always high when power up,and
the hardware reset didn't work.So I don't have idea that what is the
problem.Also I try the samsung hard disk drivers,and the hardware reset
can reset the dma request signal to '0'!


Article: 103251
Subject: Re: hard disk drivers problem
From: "dp" <dp@tgi-sci.com>
Date: 29 May 2006 09:21:39 -0700
Links: << >>  << T >>  << A >>
bjzhangwn wrote:
> I am writing a interface about the ata interface but I have a problem
> that the dma request signal is sometimes always high when power up,and
> the hardware reset didn't work.So I don't have idea that what is the
> problem.Also I try the samsung hard disk drivers,and the hardware reset
> can reset the dma request signal to '0'!

If by "hardware reset" you mean the reset signal on the ATA interace,
don't expect too much of it. The name is misleading, it does nothing
near what a power cycling reset will do - and it may vary between
vendors, especially when it comes to DMA and ATAPI combined
(fortunately, ATAPI provides fast transfer mechanisms in PIO mode...).

Dimiter

------------------------------------------------------
Dimiter Popoff               Transgalactic Instruments

http://www.tgi-sci.com
------------------------------------------------------


Article: 103252
Subject: Re: Remote Application delivery for EDA
From: Phil Hays <Spampostmaster@comcast.net>
Date: Mon, 29 May 2006 09:24:08 -0700
Links: << >>  << T >>  << A >>
On Mon, 29 May 2006 04:59:58 -0700, Ron <News5@spamex.com> wrote:

>Ben Jones wrote:
>> Plenty of people would pay good money to have their compile
>> times dropped from 12 hours to 12 minutes if the technology were available.
>
>Wow, I wish I had computers as powerful as yours. I've had a design 
>synthesizing using Xilinx Foundation 5.1 for 178 CPU hours thus far on a 
>Pentium 4 with 1.5 GBs RAM, and it's showing no signs of completing 
>anytime soon. Looks like I need to upgrade to a dual Opteron with many 
>gigs of RAM or something. Unfortunately those are *not* cheap.
>
>Can anyone recommend a computer system configuration (h/w and s/w) 
>specifically targeted at FPGA design and synthesis for under $5,000 or 
>so? A Sun Sparc workstation used to be considered top of the line, but 
>I'm not sure that's true anymore.

I'm not sure that faster hardware would solve your compile time issue.

What part?  What design flow?

Do you have excessive numbers of constraints?  Watch out for, among
other things, constraints with wildcards.  Automatically generated
constraints from third party synthesis tools can sometimes generate a
huge number of constraints.  Look for the <design_name>.ncf file.  To
debug this, try running the design with just a period constraint on
every clock.  If this completes in reasonable time, add constraints
until it no longer runs in reasonable time, and then look carefully at
the constraint(s) that breaks it.

Before upgrading the hardware, have you considered upgrading the
software?  While the older software does some things better, run time
and memory footprint have both gotten smaller.  And it is fairly easy
to have multiple versions on the same computer.

Is the design paging?  If the computer is paging, CPU utilization will
be small, under 10%, and the disks will be running all the time.  If
so, then adding 0.5 G of memory might do the trick.  That is about the
limit with standard Windows.  Standard WinXP has a 2 GB limit (3 GB
with some hacking), and WinXP 64 isn't supported.

As far as I know, all but the biggest parts should run with 2 GB or
less.

If you have to buy new higher end hardware, I'd suggest thinking about
an Athlon dual core system with 4 GB RAM and RedHat 64.  Yes, it is
only about twice what you currently have, but it should be 1/3 the
price of a 8GB Opteron system.


--
Phil Hays (Xilinx, but speaking for himself)


Article: 103253
Subject: Re: Peripheral connected to multiple OPB buses
From: "savs" <vidyutg@gmail.com>
Date: 29 May 2006 09:56:32 -0700
Links: << >>  << T >>  << A >>
Thanx Zara ..... this thing worked !! the peripheral is now connected
to two opb buses now....  though along with the mpd file corresponding
ports for the 2nd bus have to be added in the vhdl modules too ......


Article: 103254
Subject: Re: Fast Serial I/O on Virtex-5
From: "Peter Alfke" <alfke@sbcglobal.net>
Date: 29 May 2006 12:10:54 -0700
Links: << >>  << T >>  << A >>
Note that somebody seems to have made up that far in the future
"sometime in 2007".
I know better...
Peter Alfke


Article: 103255
Subject: Re: Fast Serial I/O on Virtex-5
From: "Antti Lukats" <antti@openchip.org>
Date: Mon, 29 May 2006 21:25:00 +0200
Links: << >>  << T >>  << A >>
"Peter Alfke" <alfke@sbcglobal.net> schrieb im Newsbeitrag 
news:1148929854.747791.35190@i40g2000cwc.googlegroups.com...
> Note that somebody seems to have made up that far in the future
> "sometime in 2007".
> I know better...
> Peter Alfke
>
well all the public info on V5xxT parts lists them available 2007
or is there some new update on the actual dates?

sure the parts will come earlier internally but Xilinx had no
forecast to deliver V5xxT parts this year.

so we are all eyes and ears if there is an update on actual dates.

I would of course also welcome if an PPC V5 part would
be released 2006 even if no rocket io on it.

V4FX12 is a nice part, smallest V5FX would be even better

Antti 



Article: 103256
Subject: Re: Fast Serial I/O on Virtex-5
From: Jim Granville <no.spam@designtools.co.nz>
Date: Tue, 30 May 2006 07:39:28 +1200
Links: << >>  << T >>  << A >>
Antti Lukats wrote:
> "Peter Alfke" <alfke@sbcglobal.net> schrieb im Newsbeitrag 
> news:1148929854.747791.35190@i40g2000cwc.googlegroups.com...
> 
>>Note that somebody seems to have made up that far in the future
>>"sometime in 2007".
>>I know better...
>>Peter Alfke
>>
> 
> well all the public info on V5xxT parts lists them available 2007
> or is there some new update on the actual dates?
> 
> sure the parts will come earlier internally but Xilinx had no
> forecast to deliver V5xxT parts this year.
> 
> so we are all eyes and ears if there is an update on actual dates.
> 
> I would of course also welcome if an PPC V5 part would
> be released 2006 even if no rocket io on it.
> 
> V4FX12 is a nice part, smallest V5FX would be even better
> 
> Antti 

Antti,
Xilinx (naturally) will be very cautious this time around.
Things like fast serial IO do not take much silicon, but they
do take large amounts of testing, and may need a tuning pass.
Probably, they also want some field hours clocked up as well.

All that, can easily add 6 months to a release time line.

ES would be a different story, and it sounds like you and Peter
should simply have a quiet word off-line ? :)

-jg




Article: 103257
Subject: Re: Fast Serial I/O on Virtex-5
From: "Antti" <Antti.Lukats@xilant.com>
Date: 29 May 2006 13:05:40 -0700
Links: << >>  << T >>  << A >>
I am currently happy with the public information too 

Antti


Article: 103258
Subject: Re: Remote Application delivery for EDA
From: "Tim" <tim@rockylogiccom.noooospam.com>
Date: Mon, 29 May 2006 22:17:56 +0100
Links: << >>  << T >>  << A >>

Ron wrote

> Wow, I wish I had computers as powerful as yours. I've had a design 
> synthesizing using Xilinx Foundation 5.1 for 178 CPU hours thus far on a 
> Pentium 4 with 1.5 GBs RAM, and it's showing no signs of completing 
> anytime soon. Looks like I need to upgrade to a dual Opteron with many 
> gigs of RAM or something. Unfortunately those are *not* cheap.

178 hours sounds wrong. A few suggestions for exploration:

Relax the timing constraints and/or when target the design at a larger 
device.

Partition or parameterise the design and check that there isn't something 
nasty going on.

Beg or borrow a copy of Synplicity and see what that tool makes of the 
design. 



Article: 103259
Subject: System Generator cc1 error
From: "Kishore" <kishore2k4@gmail.com>
Date: 29 May 2006 21:25:23 -0700
Links: << >>  << T >>  << A >>
Hi,

    I am trying to use System Generator but it always shows the
following error no matter what type of blocks are used. Even demos are
showing the same error.

Error: "gcc.exe: installation problem, cannot exec `cc1': No such file
or directory" I don't know what does that mean. EDK is also showing the
same problem.

Configuraiton:
Windows XP SP2
Xilinx ISE 8.1i SP3
EDK 8.1i SP1
System Generator 8.1
Matlab R14 SP3

    I checked and cc1 seems to be preprocessor or something also EDK
complains about cpp0. I checked the installation directory and found
cc1 and cpp0 at C:\Xilinx\gnu\MinGW\2.0.0-3\nt\lib\gcc-lib\mingw32\3.2
and tried setting the PATH variable to it but it only gave more errors.
What could be the problem here

thank you,
kishore.


Article: 103260
Subject: IOB IO Standards in Spartan 3
From: "=?iso-8859-1?B?R2FMYUt0SWtVc5k=?=" <taileb.mehdi@gmail.com>
Date: 29 May 2006 21:47:33 -0700
Links: << >>  << T >>  << A >>
What if I configure some IOB tu use a standard of 1.8v but vcco 3.8v?
Will the FPGA be burned?

Thanks in advace for answer
Mehdi


Article: 103261
Subject: Re: IOB IO Standards in Spartan 3
From: "=?iso-8859-1?B?R2FMYUt0SWtVc5k=?=" <taileb.mehdi@gmail.com>
Date: 29 May 2006 22:32:17 -0700
Links: << >>  << T >>  << A >>

GaLaKtIkUs=99 wrote:
> What if I configure some IOB tu use a standard of 1.8v but vcco 3.8v?
> Will the FPGA be burned?
>=20
> Thanks in advace for answer
> Mehdi
oups 3.3v and not 3.8v!


Article: 103262
Subject: Personalization of Xilinx ISE
From: "=?iso-8859-1?B?R2FMYUt0SWtVc5k=?=" <taileb.mehdi@gmail.com>
Date: 29 May 2006 22:47:38 -0700
Links: << >>  << T >>  << A >>
Hi all, Is there a method for personalization of ISE?
One of the personalizations:
      - For avoiding students to make errors in IOB proprties in UCF
files (the student's projects are for S3-Starter kit): is it possible
to add some step in the translate phase? I saw a perl interpreter
installed with ISE.

Cheers


Article: 103263
Subject: Re: IOB IO Standards in Spartan 3
From: "Antti Lukats" <antti@openchip.org>
Date: Tue, 30 May 2006 07:48:12 +0200
Links: << >>  << T >>  << A >>

"GaLaKtIkUsT" <taileb.mehdi@gmail.com> schrieb im Newsbeitrag 
news:1148964453.918613.319150@j33g2000cwa.googlegroups.com...
> What if I configure some IOB tu use a standard of 1.8v but vcco 3.8v?
> Will the FPGA be burned?
no

>
> Thanks in advace for answer
> Mehdi
>



Article: 103264
Subject: Re: IOB IO Standards in Spartan 3
From: "=?iso-8859-1?B?R2FMYUt0SWtVc5k=?=" <taileb.mehdi@gmail.com>
Date: 29 May 2006 23:02:39 -0700
Links: << >>  << T >>  << A >>

Antti Lukats wrote:
> "GaLaKtIkUsT" <taileb.mehdi@gmail.com> schrieb im Newsbeitrag
> news:1148964453.918613.319150@j33g2000cwa.googlegroups.com...
> > What if I configure some IOB tu use a standard of 1.8v but vcco 3.8v?
> > Will the FPGA be burned?
> no
>
> >
> > Thanks in advace for answer
> > Mehdi
> >
ok. but will the design work normally?
are the io vcco information stored in the bitstream and how does it
affect the correct functionning of the fpga?


Article: 103265
Subject: Re: Peripheral connected to multiple OPB buses
From: Zara <yozara@terra.es>
Date: Tue, 30 May 2006 09:13:34 +0200
Links: << >>  << T >>  << A >>
On 29 May 2006 09:56:32 -0700, "savs" <vidyutg@gmail.com> wrote:

>Thanx Zara ..... this thing worked !! the peripheral is now connected
>to two opb buses now....  though along with the mpd file corresponding
>ports for the 2nd bus have to be added in the vhdl modules too ......

Ah, yes! Of course!

I should rememeber to make answers less succint, more detailed.
Sometimes I feel I express like Tarzan ;-)

I'll try to improve my communication abilities.

Best regards,

zara

Article: 103266
Subject: Re: COREGEN: DCM
From: "srini" <g.shrinivasan@gmail.com>
Date: 30 May 2006 00:49:14 -0700
Links: << >>  << T >>  << A >>
Hi Joel,
  Goto Project->Project Options->Generation in Coregen.
  There is a "Design Entry" option in which you can choose between
Verilog and VHDL code generation. 

Regards,
Srini.


Article: 103267
Subject: PCI Header types !!!
From: "jpvarkey@gmail.com" <jpvarkey@gmail.com>
Date: 30 May 2006 01:19:19 -0700
Links: << >>  << T >>  << A >>
Could somebody give me the format of type 1 and type 2 configuration
headers.


Article: 103268
Subject: Power Up delay in FPGA !!!!!
From: "jpvarkey@gmail.com" <jpvarkey@gmail.com>
Date: 30 May 2006 01:23:14 -0700
Links: << >>  << T >>  << A >>
Iam currently working on carbus project.I am really concerned with
power delay factor in FPGA's especially Xilinx FPGA's .Will that create
a problem for cardbus.Could somebody explain on, what all factors will
affect the power up delay in SRAM based FPGA's.


Article: 103269
Subject: Re: PCI Header types !!!
From: "Antti" <Antti.Lukats@xilant.com>
Date: 30 May 2006 01:42:51 -0700
Links: << >>  << T >>  << A >>
this is described in the PCI specification


Article: 103270
Subject: Re: Personalization of Xilinx ISE
From: Aurelian Lazarut <aurash@xilinx.com>
Date: Tue, 30 May 2006 10:51:05 +0100
Links: << >>  << T >>  << A >>
you can create a script to check the pad report (from par)  to make sure 
pin locations and IO standards are correct for your particular demo design
Aurash

GaLaKtIkUs™ wrote:
> Hi all, Is there a method for personalization of ISE?
> One of the personalizations:
>       - For avoiding students to make errors in IOB proprties in UCF
> files (the student's projects are for S3-Starter kit): is it possible
> to add some step in the translate phase? I saw a perl interpreter
> installed with ISE.
> 
> Cheers
> 

Article: 103271
Subject: Re: PCI Header types !!!
From: "jpvarkey@gmail.com" <jpvarkey@gmail.com>
Date: 30 May 2006 03:06:22 -0700
Links: << >>  << T >>  << A >>
Do you have any pdfs or documents on PCI system architecture?


Article: 103272
Subject: Re: PCI Header types !!!
From: "Antti" <Antti.Lukats@xilant.com>
Date: 30 May 2006 03:27:10 -0700
Links: << >>  << T >>  << A >>
you have to purchase the specification from PCISIG


Article: 103273
Subject: Mains pick-up on I/O pins
From: m_oylulan@hotmail.com
Date: 30 May 2006 03:45:23 -0700
Links: << >>  << T >>  << A >>
Hello,

I am trying to program an RC-100 demo board, which contains a
Spartan-II chip. The board is supposed to send three logic outputs to
external devices through I/O pins provided on an expansion header. I am
pretty sure I've coded the program correctly (only to the extent that
it works as intended in simulation), but when I probe the I/O pins with
a scope all I'm getting is what I assume is mains pick-up (50 Hz
~5Vpk-pk).

My probes have been calibrated; the probe and scope have adequate
bandwidth (greater than 5x the signal bandwidth), and I get the same
result whether I use the 'Auto-scale' function or manually set the
'Volts' and 'Time' division scales to the capture the expected
waveforms. In the event that the source of the problem isn't my code,
does anyone have any suggestions as to what I might be doing wrong?

Thanks,
mees


Article: 103274
Subject: Re: PCI Header types !!!
From: Alan Myler <amyler@eircom.net>
Date: Tue, 30 May 2006 11:50:04 +0100
Links: << >>  << T >>  << A >>
jpvarkey@gmail.com wrote:

> Do you have any pdfs or documents on PCI system architecture?
> 

Google for pci local bus specification revision 2.2


You might find the PDF :-)

Alan





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