Custom Search
Site Home Archive Home FAQ Home How to search the Archive How to Navigate the Archive
Compare FPGA features and resources
Threads starting:
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
Pasacco wrote: > Given a design placeed and routed in FPGA, I need to know "how much > wiring resources my implementation are using out of total wiring > resources". > 1. How can we know wiring resource utilization? > 2. Can "number of used switch box" out of "total number of switch box" > be a metric of wiring resource utilization? If yes, are there any > method to know that? If you are really worried about it ... take a look using the FPGA editor in the areas you have some concerns.Article: 107976
On 2006-09-03 05:00:08 -0700, Phil Hays <spampostmaster@comcast.net> said: > Alan Nishioka wrote: > >> As a data point, webpack just downloaded fine for me. > > As a data point, webpack downloaded just fine for me as well. I'm near > Seattle WA, using firefox 1.5.0.4 on Fedora Core 5 (Linux), Comcast > network connection. Time was under an hour. Not to do a 'me too', but ... well ... me too :-) In San Jose, took 45 minutes. SimonArticle: 107977
Bob Ferapples wrote: To save time and bandwidth, I have boiled down the vast majority of postings to this newsgroup into a quick and dirty little format that can speed things up. Please follow this format from now on in this group. -------------------------------------------- Please kind (insert platitudes here, or addreess titles i.e. Sirs or Madames), I am a new (insert job function or social status here, i.e. Student, Engineer, Electrician, Homeowner etc.) <optional> from (insert non-english speaking country) and I have this (insert task here, i.e. homework assignment, personal project, project assigned to me, problem with my appliance etc.). I know (insert qualifications here, i.e. next to nothing, very little, a little, enough to be dangerous etc.) about electricity, but was hoping that you all could help me to (restate task here, but without saying what you really mean, i.e. do my homework for me, get me out of a jam, help me to avoid paying a professional to do it correctly, get me through this day without having to think, avoid electrocuting myself or my loved ones etc). I hope to hear from you soon. Nooby Nooberson --------------------------------- Snort x-posted to s.e.d, s.e.b., comp.arch.fpga, comp.arch.embedded, comp.arch.dspArticle: 107978
Simon Gornall wrote: > On 2006-09-03 05:00:08 -0700, Phil Hays <spampostmaster@comcast.net> said: > > > Alan Nishioka wrote: > > > >> As a data point, webpack just downloaded fine for me. > > > > As a data point, webpack downloaded just fine for me as well. I'm near > > Seattle WA, using firefox 1.5.0.4 on Fedora Core 5 (Linux), Comcast > > network connection. Time was under an hour. > > Not to do a 'me too', but ... well ... me too :-) > > In San Jose, took 45 minutes. > > Simon I downloaded it too, here near Cambridge, UK. It took about 35 minutes. The only problem I had was *starting* the download because of the braindead code on the page. Cheers PeteSArticle: 107979
Hello Frank, >>> >>> So ... was "two days" bid as 16 hrs, or 48 hrs? >>> >>Our drill sergeant used to say that the day has 24 hours and when that >>ain't sufficient then there is still the night. > > Hahaha, that's a good one ;) > Yes, the guys in our company thought so as well. However, when someone couldn't help but chuckle he went ballistic and screamed into his face. Basic training is often no fun. The other things he told us probably dates back to the 18th century: "As civilians you came, as men you will leave" and "Anything that doesn't immediately lead to your death will make you tougher". -- Regards, Joerg http://www.analogconsultants.comArticle: 107980
On 3 Sep 2006 13:41:01 -0700, in sci.electronics.design "PeteS" <PeterSmith1954@googlemail.com> wrote: >Bob Ferapples wrote: >To save time and bandwidth, I have boiled down the vast majority of >postings to this newsgroup into a quick and dirty little format that >can speed things up. Please follow this format from now on in this >group. > > What is Guy Macon up to these days? Maybe worth redirecting to wherever he is hiding? martinArticle: 107981
martin griffith wrote: > On 3 Sep 2006 13:41:01 -0700, in sci.electronics.design "PeteS" > <PeterSmith1954@googlemail.com> wrote: > > >Bob Ferapples wrote: > >To save time and bandwidth, I have boiled down the vast majority of > >postings to this newsgroup into a quick and dirty little format that > >can speed things up. Please follow this format from now on in this > >group. > > > > > > What is Guy Macon up to these days? Maybe worth redirecting to > wherever he is hiding? > > > martin LMAO W|N>K You now owe me a link for cheap new keyboards Cheers PeteSArticle: 107982
martin griffith wrote: > On 3 Sep 2006 13:41:01 -0700, in sci.electronics.design "PeteS" > <PeterSmith1954@googlemail.com> wrote: > > >Bob Ferapples wrote: > >To save time and bandwidth, I have boiled down the vast majority of > >postings to this newsgroup into a quick and dirty little format that > >can speed things up. Please follow this format from now on in this > >group. > > > > > > What is Guy Macon up to these days? Maybe worth redirecting to > wherever he is hiding? > > > martin Can't post it - he has his own moderated (by him) newsgroup, I understand. Perhaps he's going to go into competition with Google Cheers PeteSArticle: 107983
PeteS wrote: > Bob Ferapples wrote: > To save time and bandwidth, I have boiled down the vast majority of > postings to this newsgroup into a quick and dirty little format that > can speed things up. Please follow this format from now on in this > group. > > -------------------------------------------- > > Please kind (insert platitudes here, or addreess titles i.e. Sirs or > Madames), > > I am a new (insert job function or social status here, i.e. Student, > Engineer, > Electrician, Homeowner etc.) > > <optional> from (insert non-english speaking country) > > and I have this (insert task here, i.e. homework assignment, personal > project, project > assigned to me, problem with my appliance etc.). > > I know (insert qualifications here, i.e. next to nothing, very little, > a > little, enough to be dangerous etc.) about electricity, > > but was hoping that you all could help me to (restate task here, but > without saying what you really mean, i.e. do my homework for me, get me > out of a jam, help me to avoid paying a professional to do it > correctly, get me through this day without having to think, avoid > electrocuting myself or my loved ones etc). > > I hope to hear from you soon. > > Nooby Nooberson > > --------------------------------- > > Snort > x-posted to s.e.d, s.e.b., comp.arch.fpga, comp.arch.embedded, > comp.arch.dsp I'm new to this group and been reading over the different articles, and a funny thing is I haven't seen any that was of the nature you have described. Maybe others have a similar attitude and that would explain why I have received no replies to a simple question that the Xilinx site does not answer. Somehow I feel really welcomed. -- Cecil KD5NWA www.qrpradio.com www.hpsdr.com "Sacred Cows make the best Hamburger!" Don Seglio BatunaArticle: 107984
On Sun, 03 Sep 2006 23:26:40 +0200, the renowned martin griffith <mart_in_medina@yahoo.esXXX> wrote: >On 3 Sep 2006 13:41:01 -0700, in sci.electronics.design "PeteS" ><PeterSmith1954@googlemail.com> wrote: > >>Bob Ferapples wrote: >>To save time and bandwidth, I have boiled down the vast majority of >>postings to this newsgroup into a quick and dirty little format that >>can speed things up. Please follow this format from now on in this >>group. >> >> > >What is Guy Macon up to these days? Maybe worth redirecting to >wherever he is hiding? > > >martin Maybe in little fiefdom, the product development newsgroup. Best regards, Spehro Pefhany -- "it's the network..." "The Journey is the reward" speff@interlog.com Info for manufacturers: http://www.trexon.com Embedded software/hardware/analog Info for designers: http://www.speff.comArticle: 107985
zwsdotcom@gmail.com wrote: > Is it actually possible to download WebPACK? > > I tried using my existing xilinx.com account to download it, and I get > into an endless loop where I'm shown a button that says "click here to > register" then another button that says "click here to download if > you're already registered". The second button just takes me to the same > page again, over and over. > > I tried creating a new account on xilinx.com and with this account I > can't even log into the abovementioned infinite-loop page; it doesn't > accept the new username/password. > > Has anyone ever worked out the magic combination of browser versions > and whatever other magic the PITAs at xilinx require? I seem to recall having this same problem once. I think it was because I had signed up, but before you are authorized to download, your application has to be reviewed by a human. I'm not sure, but I think you may get an email on Monday telling you where/how to download the file.Article: 107986
Brian Drummond wrote:
> >wget --user={username} --password={password}
> >http://www.xilinx.com/webpack/index.htm
>
> Creative enough! I usually resort to "view/page source" for the same
> purpose.
That won't work in this case. If you load the page in a browser you
never get to see the page that contains the download URLs - the server
(NOT a META REFRESH!) redirects you instantly back to the "register or
download" page. Broken, massively broken.
> Now all you have to do is hope both ends stay up for a 970MB download.
That's not a problem. I got the Windows and Linux versions, and
downloaded ModelSim for good measure.
Article: 107987On Sun, 03 Sep 2006 17:00:20 -0500, Don Seglio <cbayona@cox.net> wrote: >I'm new to this group and been reading over the different articles, and >a funny thing is I haven't seen any that was of the nature you have >described. Try watching a little longer - there are enough... >Maybe others have a similar attitude and that would explain why I have >received no replies to a simple question that the Xilinx site does not >answer. Errrm, maybe not. Traffic on comp.arch.fpga is pretty light at weekends - many of the most active contributors are professionals in the field and regard monitoring the NG as a useful part of their work, whether for self-publicity, keeping up-to-date, or just generally being helpful. Such folk are more likely to respond to your Sunday-morning post some time on Monday. And I'm not sure that your question about Matlab version compatibility really rates as "simple" - it sounds hard to me, but I'll wager there are plenty of people here who can tell you. For the record, my perception is that comp.arch.fpga is pretty open to anyone who's prepared to make the effort to articulate a pertinent question. A nice contrast with some other groups I've looked at! -- Jonathan Bromley, Consultant DOULOS - Developing Design Know-how VHDL * Verilog * SystemC * e * Perl * Tcl/Tk * Project Services Doulos Ltd., 22 Market Place, Ringwood, BH24 1AW, UK jonathan.bromley@MYCOMPANY.com http://www.MYCOMPANY.com The contents of this message may contain personal views which are not the views of Doulos Ltd., unless specifically stated.Article: 107988
On 3 Sep 2006 13:41:01 -0700, "PeteS" <PeterSmith1954@googlemail.com> wrote: >Bob Ferapples wrote: >To save time and bandwidth, I have boiled down the vast majority of >postings to this newsgroup into a quick and dirty little format that >can speed things up. A few years ago, someone posted a very funny take on this to sci.electronics.design. It listed a large number of spurious acronyms for use as responses to irritating querants - I can't find it now, but it was stuff like WWTYBYMHY - We Won't Tell You Because You Might Hurt Yourself (for use in response to people asking how to make a Tesla coil out of paper clips and a car battery - you know the kind of thing.) Anyone got a link to the original? -- Jonathan Bromley, Consultant DOULOS - Developing Design Know-how VHDL * Verilog * SystemC * e * Perl * Tcl/Tk * Project Services Doulos Ltd., 22 Market Place, Ringwood, BH24 1AW, UK jonathan.bromley@MYCOMPANY.com http://www.MYCOMPANY.com The contents of this message may contain personal views which are not the views of Doulos Ltd., unless specifically stated.Article: 107989
Brian Drummond wrote:
> On 2 Sep 2006 08:32:07 -0700, zwsdotcom@gmail.com wrote:
>
> >
> >Brian Drummond wrote:
> >
>
> >> You can either get creative about finding or guessing the actual
> >> download URL
> ...
> >wget --user={username} --password={password}
> >http://www.xilinx.com/webpack/index.htm
> >
> >then poke in the HTML file to get the direct download links (you can't
> >really read the HTML file easily without the CSS). And here they are:
>
> Creative enough! I usually resort to "view/page source" for the same
> purpose.
>
> Now all you have to do is hope both ends stay up for a 970MB download.
>
> (I pity anyone still in 56k-land who needs WebPack. I needed WebPack
> 6.1 before ADSL got here, and 240MB was quite bad enough...)
Hell, this download is so large it creates other problems. I tried to
download it at work once and found it was too large to fit on a CD!!!
It proved impossible to get IT to install a DVD burner on my PC so I
had to get the local FAE to burn me a copy. I could have bought a copy
of the DVD from Xilinx onlne, but I didn't have a project budget at
that point and I would have had to pay for it out of pocket. I was
going to do that, but the FAE got in the loop.
I suggested to Xilinx that they split the download in two parts so it
can be burned onto two CDs rather than a DVD. But obviously they don't
see the utility in that. They spend a few bazillion dollars to create
a web pack and keep it working with all the upgrades, but they can't be
bothered enough to make it fit on CDs rather than a DVD. I guess they
just don't know how difficult it can be to work for a defense
contractor. If I dipped into my own pocket every time the bureaucracy
got in my way of doing work, I would owe them money at the end of the
week!
BTW, this is your money too, and Xilinx's! They pay taxes too. :^)
Article: 107990rickman wrote: > ... > I suggested to Xilinx that they split the download in two parts so it > can be burned onto two CDs rather than a DVD. But obviously they don't > see the utility in that. They spend a few bazillion dollars to create > a web pack and keep it working with all the upgrades, but they can't be > bothered enough to make it fit on CDs rather than a DVD... As of ISE8.2 they no longer offer CDs to those of us actually paying for the full ISE (they gave notice of this change with 8.1). I personally prefer the DVDs, but can see where it might cause problems with others.Article: 107991
Nevo wrote: > "Nevo" <nevo_n@hotmail.com> wrote in message > news:zwrIg.2012$XD1.1382@trnddc01... >> Newbie alert... >> >> I would like to drive my Spartan 3 design with inputs from a 5V source. >> I've been unable to find anything in the Spartan 3 datasheet that lets me >> know if this is do-able. (The signal will be coming fom the parallel >> port on my laptop, which I suspect probably is 3.3V but I want to be on >> the safe side.) >> >> Any tips are appreciated. >> >> Thanks, >> >> -Nevo > > Hm... I found http://direct.xilinx.com/bvdocs/appnotes/xapp179.pdf, which > lists the 5V tolerant SelectIO standards for the Spartan 2 family (and > implicitly says none of the standards are 5V tolerant in the Spartan 2E > family), but I'm unable to find a simlar reference for the Spartan 3 > family. > > Table 27, Absoulte Maximum Ratings under DC Electrical Characteristics, > suggests that putting 5V into the chip is a no-no. > > Hm; I think I'm going to answer my own question here by pointing myself to > XAPP429 and a whole host of archived messages in comp.arch.fpga. Nevo, i applaud your ability to post (partial) answers to your own questions. More people need that ability. It shows the ability to recognize when you have bypassed the limits of your sure knowledge; and the perseverance to continue seeking yourself, instead of passively waiting for "an/the answer". -- JosephKK Gegen dummheit kampfen die Gotter Selbst, vergebens. --SchillerArticle: 107992
Jonathan Bromley wrote: > On Sun, 03 Sep 2006 17:00:20 -0500, Don Seglio > <cbayona@cox.net> wrote: > >> I'm new to this group and been reading over the different articles, and >> a funny thing is I haven't seen any that was of the nature you have >> described. > > Try watching a little longer - there are enough... > >> Maybe others have a similar attitude and that would explain why I have >> received no replies to a simple question that the Xilinx site does not >> answer. > > Errrm, maybe not. Traffic on comp.arch.fpga is pretty light > at weekends - many of the most active contributors are > professionals in the field and regard monitoring the NG > as a useful part of their work, > whether for self-publicity, keeping up-to-date, or just generally > being helpful. Such folk are more likely to respond to your > Sunday-morning post some time on Monday. And I'm not > sure that your question about Matlab version compatibility > really rates as "simple" - it sounds hard to me, but I'll > wager there are plenty of people here who can tell you. > > For the record, my perception is that comp.arch.fpga is > pretty open to anyone who's prepared to make the effort > to articulate a pertinent question. A nice contrast with > some other groups I've looked at! It's simple in the sense that one has tried it and it works or it doesn't. Guessing won't help me. That is also why I asked about what software is shipping with the new kit's, if it a recent enough version it might work with the MatLab I have. Once I break the wrapper on the package, I'm stuck with it so it would be nice to save a ton of money to know if it has been tried and if it worked before I end up wasting money I can't afford to waste. Right now I tried to get it exchanged for R14 but my vendor says sorry all they can do is refund my money if it has not been opened, they did me a favor by sending me the new version since it cost a lot more than what I paid. I have about a week to figure out what I need to do. -- Cecil KD5NWA www.qrpradio.com www.hpsdr.com "Sacred Cows make the best Hamburger!" Don Seglio BatunaArticle: 107993
Hi, I need your help. My project uses BlockRAM by Xilinx. I used CoreGenerator 6.2 to generate BlockRAM bram64_8, and Xilinx free 8.1 ISE version. By using code generated for ModelSim simulation, it works well without any error. But while compiling with Xilinx 8.1 ISE, it generates the following errors: ERROR:NgdBuild:604 - logical block 'MG_x_A3/bram64_8_A/BU5' with type 'RAMB16' could not be resolved. A pin name misspelling can cause this, a missing edif or ngc file, or the misspelling of a type name. Symbol 'RAMB16' is not supported in target 'virtex2'. I couldn't find any 'RAMB16' in my vhdl files. What is wrong? How to correct it? Thank you. WengArticle: 107994
Joerg wrote: > > Hello Frank, > > >>> > >>> So ... was "two days" bid as 16 hrs, or 48 hrs? > >>> > >>Our drill sergeant used to say that the day has 24 hours and when that > >>ain't sufficient then there is still the night. > > > > Hahaha, that's a good one ;) > > > > Yes, the guys in our company thought so as well. However, when someone > couldn't help but chuckle he went ballistic and screamed into his face. > Basic training is often no fun. > > The other things he told us probably dates back to the 18th century: "As > civilians you came, as men you will leave" and "Anything that doesn't > immediately lead to your death will make you tougher". > > -- > Regards, Joerg And your DI wants you to believe that they know every trick in the book. ;-) -- Service to my country? Been there, Done that, and I've got my DD214 to prove it. Member of DAV #85. Michael A. Terrell Central FloridaArticle: 107995
Jonathan Bromley wrote: > On 3 Sep 2006 13:41:01 -0700, "PeteS" <PeterSmith1954@googlemail.com> > wrote: > > >Bob Ferapples wrote: > >To save time and bandwidth, I have boiled down the vast majority of > >postings to this newsgroup into a quick and dirty little format that > >can speed things up. > > A few years ago, someone posted a very funny take on this > to sci.electronics.design. It listed a large number of > spurious acronyms for use as responses to irritating > querants - I can't find it now, but it was stuff like > > WWTYBYMHY - We Won't Tell You Because You Might Hurt Yourself > > (for use in response to people asking how to make a Tesla coil > out of paper clips and a car battery - you know the kind of thing.) > > Anyone got a link to the original? > -- > Jonathan Bromley, Consultant > > DOULOS - Developing Design Know-how > VHDL * Verilog * SystemC * e * Perl * Tcl/Tk * Project Services > > Doulos Ltd., 22 Market Place, Ringwood, BH24 1AW, UK > jonathan.bromley@MYCOMPANY.com > http://www.MYCOMPANY.com > > The contents of this message may contain personal views which > are not the views of Doulos Ltd., unless specifically stated. I get irritated by those who post serious replies when the very content of the question shows that the person has not a clue. "Hi, I need to design a control system for a nuclear reactor. Can someone tell me what a diode does?" Aaaarrrrggg! LuhanArticle: 107996
rickman wrote:
> Brian Drummond wrote:
>
>>On 2 Sep 2006 08:32:07 -0700, zwsdotcom@gmail.com wrote:
>>
>>
>>>Brian Drummond wrote:
>>>
>>
>>>>You can either get creative about finding or guessing the actual
>>>>download URL
>>
>>...
>>
>>>wget --user={username} --password={password}
>>>http://www.xilinx.com/webpack/index.htm
>>>
>>>then poke in the HTML file to get the direct download links (you can't
>>>really read the HTML file easily without the CSS). And here they are:
>>
>>Creative enough! I usually resort to "view/page source" for the same
>>purpose.
>>
>>Now all you have to do is hope both ends stay up for a 970MB download.
>>
>>(I pity anyone still in 56k-land who needs WebPack. I needed WebPack
>>6.1 before ADSL got here, and 240MB was quite bad enough...)
>
>
> Hell, this download is so large it creates other problems. I tried to
> download it at work once and found it was too large to fit on a CD!!!
> It proved impossible to get IT to install a DVD burner on my PC so I
> had to get the local FAE to burn me a copy. I could have bought a copy
> of the DVD from Xilinx onlne, but I didn't have a project budget at
> that point and I would have had to pay for it out of pocket. I was
> going to do that, but the FAE got in the loop.
>
> I suggested to Xilinx that they split the download in two parts so it
> can be burned onto two CDs rather than a DVD. But obviously they don't
> see the utility in that.
They DO see the sense of split downloads in their datasheets -
just need to get their datasheet oublisher to talk to their SW publisher!!
> They spend a few bazillion dollars to create
> a web pack and keep it working with all the upgrades, but they can't be
> bothered enough to make it fit on CDs rather than a DVD. I guess they
> just don't know how difficult it can be to work for a defense
> contractor. If I dipped into my own pocket every time the bureaucracy
> got in my way of doing work, I would owe them money at the end of the
> week!
>
> BTW, this is your money too, and Xilinx's! They pay taxes too. :^)
>
Something like webpack _should_ partition very easily.
Here I can see ..\bin, ..\doc, and a host of device dependant resource
Wonder if they even looked at a better class of ZIP, or even a split
into an installer, and zip data files.
Should be easy to copy what their datasheet team do: one large file, and
also selective downloads.
-jg
Article: 107997Weng Tianxiang wrote: > Hi, > I need your help. > > My project uses BlockRAM by Xilinx. > > I used CoreGenerator 6.2 to generate BlockRAM bram64_8, > and Xilinx free 8.1 ISE version. > > By using code generated for ModelSim simulation, it works well > without any error. > > But while compiling with Xilinx 8.1 ISE, it generates the > following errors: > > ERROR:NgdBuild:604 - logical block > 'MG_x_A3/bram64_8_A/BU5' with type 'RAMB16' could not > be resolved. A pin name misspelling can cause this, > a missing edif or ngc file, or the misspelling of a type name. > Symbol 'RAMB16' is not supported in target 'virtex2'. > > I couldn't find any 'RAMB16' in my vhdl files. > > What is wrong? How to correct it? > > Thank you. > > Weng Do you have your library specified? I'm not a VHDL guy but I know the libraries are needed, in this case the unisim library in particular. The following is the first few lines from the control.vhd sample file for the Spartan3E Starter Kit: _____________________________________________________________________ -- -- Definition of a dual port ROM for KCPSM2 or KCPSM3 program defined by control.psm -- and assmbled using KCPSM2 or KCPSM3 assembler. -- -- Standard IEEE libraries -- library IEEE; use IEEE.STD_LOGIC_1164.ALL; use IEEE.STD_LOGIC_ARITH.ALL; use IEEE.STD_LOGIC_UNSIGNED.ALL; -- -- The Unisim Library is used to define Xilinx primitives. It is also used during -- simulation. The source can be viewed at %XILINX%\vhdl\src\unisims\unisim_VCOMP.vhd -- library unisim; use unisim.vcomponents.all; -- --Article: 107998
Dave wrote: > I cannot simply write data to or read data from a GPIO port from the > MicroBlaze. The inputs and outputs go through conditioners. Since I have > to go through conditoners and the GPIO does not need to talk directly to the > outside world, my question is can I map the GPIO_IN port pins to the > conditoner output port pins without having to use actual FPGA pins? I believe the gpio_in, gpio_d_out, gpio_t_out ports are supposed to work like you want (with internal logic) but I seem to recall trying to use them this way and failing. I ended up adding logic to an existing opb peripheral that I had already written. Alan NishiokaArticle: 107999
John_H wrote: > Weng Tianxiang wrote: > > Hi, > > I need your help. > > > > My project uses BlockRAM by Xilinx. > > > > I used CoreGenerator 6.2 to generate BlockRAM bram64_8, > > and Xilinx free 8.1 ISE version. > > > > By using code generated for ModelSim simulation, it works well > > without any error. > > > > But while compiling with Xilinx 8.1 ISE, it generates the > > following errors: > > > > ERROR:NgdBuild:604 - logical block > > 'MG_x_A3/bram64_8_A/BU5' with type 'RAMB16' could not > > be resolved. A pin name misspelling can cause this, > > a missing edif or ngc file, or the misspelling of a type name. > > Symbol 'RAMB16' is not supported in target 'virtex2'. > > > > I couldn't find any 'RAMB16' in my vhdl files. > > > > What is wrong? How to correct it? > > > > Thank you. > > > > Weng > > Do you have your library specified? I'm not a VHDL guy but I know the > libraries are needed, in this case the unisim library in particular. > The following is the first few lines from the control.vhd sample file > for the Spartan3E Starter Kit: > _____________________________________________________________________ > > -- > -- Definition of a dual port ROM for KCPSM2 or KCPSM3 program defined by > control.psm > -- and assmbled using KCPSM2 or KCPSM3 assembler. > -- > -- Standard IEEE libraries > -- > library IEEE; > use IEEE.STD_LOGIC_1164.ALL; > use IEEE.STD_LOGIC_ARITH.ALL; > use IEEE.STD_LOGIC_UNSIGNED.ALL; > -- > -- The Unisim Library is used to define Xilinx primitives. It is also > used during > -- simulation. The source can be viewed at > %XILINX%\vhdl\src\unisims\unisim_VCOMP.vhd > -- > library unisim; > use unisim.vcomponents.all; > -- > -- Hi, Yes, I used it as following statements show: LIBRARY ieee; USE ieee.std_logic_1164.all; use ieee.numeric_std.all; use work.MG_x_Constant.all; LIBRARY unisim; USE UNISIM.VCOMPONENTS.ALL; The library is specified in the project and all modules are compiled with errors: WARNING:NgdBuild:486 - Attribute "INITP_07" is not allowed on symbol "BU8" of type "RAMB16". This attribute will be ignored. WARNING:NgdBuild:486 - Attribute "SRVAL_A" is not allowed on symbol "BU8" of type "RAMB16". This attribute will be ignored. WARNING:NgdBuild:486 - Attribute "SRVAL_B" is not allowed on symbol "BU8" of type "RAMB16". This attribute will be ignored. WARNING:NgdBuild:486 - Attribute "WRITE_MODE_A" is not allowed on symbol "BU8" of type "RAMB16". This attribute will be ignored. WARNING:NgdBuild:486 - Attribute "WRITE_MODE_B" is not allowed on symbol "BU8" of type "RAMB16". This attribute will be ignored. ERROR:NgdBuild:604 - logical block 'MG_x_A4/BlockRAM64_16_2_32k_00/bram64_16_A/BU8' with type 'RAMB16' could not be resolved. A pin name misspelling can cause this, a missing edif or ngc file, or the misspelling of a type name. Symbol 'RAMB16' is not supported in target 'virtex2'. I never used any signals appeared in the above warning and errors. What I am doing now is download latest 8.2 ISE and re-generate all Block RAM modules again and see if the above errors happens again. Thank you. Weng
Site Home Archive Home FAQ Home How to search the Archive How to Navigate the Archive
Compare FPGA features and resources
Threads starting:
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