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
Dear All, I have a Spartan 3A-DSP 3400A board http://www.xilinx.com/products/devkits/HW-SD3400A-DSP-DB-UNI-G.htm and a licence for the ISE System Edition. I am trying to implement the simple stop watch tutorial at page 19 of the ISE In-depth tutorial at here: http://www.xilinx.com/support/documentation/sw_manuals/xilinx11/ise11tut.pdf Unfortunately at page 35 I get stack because the Digital Clock Manager "Single DCM_SP" does not seem to be available, see screenshot here: http://topo.epfl.ch/research/indoor_nav/prj_rep/uploads/FpgaRadioPrj/DCM_problem.png Can somebody please help me out by telling me how to proceed on this tutorial? am I making a mistake by using a DCM_SP? is this a problem with my license? unfortunately I have googled this problem in all possible ways but I found no solution. best regards fabrizioArticle: 143651
On Oct 19, 1:49=C2=A0pm, "jerzy.g...@gmail.com" <jerzy.g...@gmail.com> wrote: > On 14 Pa=C5=BA, 16:14, "Nial Stewart" > > <nial*REMOVE_TH...@nialstewartdevelopments.co.uk> wrote: > > > =C2=A0Please advise me in any sense. > > > If you've any sense you'll pick a more realistic final year project! > > > Sorry to be blunt but from the questions you're asking there's no > > way you'll get this finished. > > > Nial > > Why the promotor - lecturer has approved this kind of project? > Where is the common sense? > Kolopipo, change the project if you can. > > Kind Regards, > > Jerzy Gbur Kolopipo, I am second to the other posters who responded to your mail. It is a big challenge for some one who lacks knowledge in advanced digital design to implement hand writing recognition engine. But I would suggest you to understand the algorithms and implement them in Matlab. Xilinx Accel DSP is a very helpful tool for problem like yours. Without going into the deeper details, you can get your prototype working hardware on the FPGA. Hope this helps, Regards, MoazzamArticle: 143652
rickman <gnuarm@gmail.com> writes: > On Oct 19, 9:58 am, Gabor <ga...@alacron.com> wrote: > Are you saying that on the Power PC they number bits in the data bus > with 0 as msb and 31 as lsb? How about address bits, is the lsb > numbered 31 and the msb numbered 0? Yes. And IIRC on the Motorola 88110 which preceded it, on which the PowerPC 601 was based (buswise anyway). > I would find that very confusing. > > I could care less if the byte addressing is big-endian or little- > endian. I don't see that having much import when defining the bits in > a bus. > If you look at address 0, it corresponds with bits 0 to 7 of the data bus, address 1 is bits 8 to 15. Just like on a little-endian processor. (Would you want address 0 on bits 31 downto 24? Maybe?) Anyway, once they've taken that convention, making the address bus numbered the same way makes more sense than making it opposite to the databus! (But nothing beats the SHARC DSP booting from an 8 bit EPROM on bits 16-23 of the 48 bit address bus :) Cheers, Martin -- martin.j.thompson@trw.com TRW Conekt - Consultancy in Engineering, Knowledge and Technology http://www.conekt.net/electronics.htmlArticle: 143653
"hvo" <hai.vo@synrad.com> writes: > Yes I see the limitation of GPIO interrupts. There are no parameters to > set the gpio to only generate an interrupt on a rising or falling edge. > > How else can I connect an external interrupt without using GPIO? > Create an external pin for it - give it an "INTERRUPT" type, and you can set it to be edge or level triggered, and what polarity it should be. You get an MHS line like this for example: PORT xps_epc_usb_HC_IRQ = USB_HC_IRQ, DIR = I, SIGIS = INTERRUPT, SENSITIVITY = LEVEL_LOW Cheers, Martin -- martin.j.thompson@trw.com TRW Conekt - Consultancy in Engineering, Knowledge and Technology http://www.conekt.net/electronics.htmlArticle: 143654
On Mon, 19 Oct 2009, Frank Buss wrote: |-----------------------------------------------------------------------| |"> Fist of all I hope to learn the performance/price. | |> Is available a price list for each FPGA device? | |> I find a selection guide on xilinx's websit, and | |> price information is not available. I search it | |> on the internet but no helpful information is | |> found. | | | |Try Digikey, it has prices for Xilinx parts. The Altera parts are nice,| |too, and the prices are in their online shop. | | | |-- | |Frank Buss, fb@frank-buss.de | |http://www.frank-buss.de, http://www.it4-systems.de " | |-----------------------------------------------------------------------| You could also try WWW.FPGA-FAQ.com/FPGA_Boards.shtml to get an idea but it is not for FPGAs on their own and it is probably old. Regards, Colin Paul GlosterArticle: 143655
On Oct 20, 8:32=A0am, Martin Thompson <martin.j.thomp...@trw.com> wrote: > rickman <gnu...@gmail.com> writes: > > On Oct 19, 9:58=A0am, Gabor <ga...@alacron.com> wrote: > > Are you saying that on the Power PC they number bits in the data bus > > with 0 as msb and 31 as lsb? =A0How about address bits, is the lsb > > numbered 31 and the msb numbered 0? =A0 > > Yes. =A0And IIRC on the Motorola 88110 which preceded it, on which the > PowerPC 601 was based (buswise anyway). > > > I would find that very confusing. > > > I could care less if the byte addressing is big-endian or little- > > endian. =A0I don't see that having much import when defining the bits i= n > > a bus. > > If you look at address 0, it corresponds with bits 0 to 7 of the data > bus, address 1 is bits 8 to 15. =A0Just like on a little-endian processor= . > (Would you want address 0 on bits 31 downto 24? =A0Maybe?) > > Anyway, once they've taken that convention, making the address bus > numbered the same way makes more sense than making it opposite to the > databus! > > (But nothing beats the SHARC DSP booting from an 8 bit EPROM on bits 16-2= 3 of > the 48 bit address bus :) > > Cheers, > Martin > > -- > martin.j.thomp...@trw.com > TRW Conekt - Consultancy in Engineering, Knowledge and Technologyhttp://w= ww.conekt.net/electronics.html By the way the "hard" big endian numbering was IBM's not Motorola's. The Power-PC originated at IBM and was later a joint venture with Motorola. Motorola's own processors, 68K series, use a mix of big endian byte numbering and little endian bit numbering, so in fact bits 31:24 of a 32-bit word on a 32-bit bus would be byte 0. How did we get on this topic? Regards, GaborArticle: 143656
On Sun, 18 Oct 2009, Rickman wrote: |---------------------------------------------------------------------| |"[..] | |[..] My copy of ispLever from Lattice will | |not run compile or simulate because neither of these tools will run | |due to the license expiring. [..] | | | |[..] | | | |So now I am trying to use the Xilinx Webpack to allow me to continue | |working until I can get a new license file, but it won't run either."| |---------------------------------------------------------------------| Xilinx Webpack used to be licensed such that it was forbidden to use it for Lattice FPGAs. I do not recall that prohibition being removed. |---------------------------------------------------------------------| |"I think that it somehow is using networking to communicate between | |processes and it won't go through the firewall! Talk about making | |simple things difficult! [..] | | | |[..]" | |---------------------------------------------------------------------| OpenSolaris had a very complicated way to simulate X-Windows of GNU/Linux (on top of X-Windows of OpenSolaris) via networking.Article: 143657
On Oct 20, 8:49=A0pm, Colin Paul Gloster <Colin_Paul_Glos...@ACM.org> wrote: > On Mon, 19 Oct 2009, Frank Buss wrote: > > |-----------------------------------------------------------------------| > |"> Fist of all I hope to learn the performance/price. =A0 =A0 =A0 =A0 = =A0 =A0 =A0 =A0 =A0| > |> Is available a price list for each FPGA device? =A0 =A0 =A0 =A0 =A0 = =A0 =A0 =A0 =A0 =A0 =A0| > |> I find a selection guide on xilinx's websit, and =A0 =A0 =A0 =A0 =A0 = =A0 =A0 =A0 =A0 =A0 | > |> price information is not available. I search it =A0 =A0 =A0 =A0 =A0 = =A0 =A0 =A0 =A0 =A0 =A0| > |> on the internet but no helpful information is =A0 =A0 =A0 =A0 =A0 =A0 = =A0 =A0 =A0 =A0 =A0 =A0| > |> found. =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0= =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 | > | =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0= =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 | > |Try Digikey, it has prices for Xilinx parts. The Altera parts are nice,| > |too, and the prices are in their online shop. =A0 =A0 =A0 =A0 =A0 =A0 = =A0 =A0 =A0 =A0 =A0 =A0 =A0| > | =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0= =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 | > |-- =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 = =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 | > |Frank Buss, f...@frank-buss.de =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 = =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 | > |http://www.frank-buss.de,http://www.it4-systems.de" =A0 =A0 =A0 =A0 =A0 = =A0 =A0 =A0 =A0| > |-----------------------------------------------------------------------| > > You could also try > WWW.FPGA-FAQ.com/FPGA_Boards.shtml > to get an idea but it is not for FPGAs on their own and it is probably > old. > > Regards, > Colin Paul Gloster Thanks for your replies. Actually I am an DSP software engineer for video codec and image processing, but want to master skills to design application with FPGA. I don't have any application at hand planned to use FPGA. At first I hope to know its applications in practical applications besides in IC design. I think cost is an important factor when choice is made. Best Regards JoggingArticle: 143658
>"hvo" <hai.vo@synrad.com> writes: > >> Yes I see the limitation of GPIO interrupts. There are no parameters to >> set the gpio to only generate an interrupt on a rising or falling edge. >> >> How else can I connect an external interrupt without using GPIO? >> > >Create an external pin for it - give it an "INTERRUPT" type, and >you can set it to be edge or level triggered, and what polarity it >should be. > >You get an MHS line like this for example: > PORT xps_epc_usb_HC_IRQ = USB_HC_IRQ, DIR = I, SIGIS = INTERRUPT, SENSITIVITY = LEVEL_LOW > >Cheers, >Martin > >-- >martin.j.thompson@trw.com >TRW Conekt - Consultancy in Engineering, Knowledge and Technology >http://www.conekt.net/electronics.html > Thanks! That worked for me. This is definately more versatile than GPIO interrtupts, and it takes less slices to implement. Cheers HV --------------------------------------- This message was sent using the comp.arch.fpga web interface on http://www.FPGARelated.comArticle: 143659
I'd like to implement an SPI master to read an AD7924 ADC. The thing is that there are some setup and hold time I have to respect, ie CS'to SCLK setup time, SCLK to DOUT valid hold time, etc. They go from typically 10 ns to 50 ns. What's the best way to implement these delays?Article: 143660
>I'd like to implement an SPI master to read an AD7924 ADC. The thing is >that there are some setup and hold time I have to respect, ie CS'to SCLK >setup time, SCLK to DOUT valid hold time, etc. They go from typically 10 ns >to 50 ns. > >What's the best way to implement these delays? > > > I mean implementing them in VHDL. --------------------------------------- This message was sent using the comp.arch.fpga web interface on http://www.FPGARelated.comArticle: 143661
> What's the best way to implement these delays? You can easily implement them with a state machine that is clocked considerably faster than the SPI bit rate.Article: 143662
I have ISE 11.1 and the DCM does exist in the coregen for this release. Is it not in your version of coregen or is it grayed out? Your image doesnt seem to work. Jon --------------------------------------- This message was sent using the comp.arch.fpga web interface on http://www.FPGARelated.com From usenet+5@ladybug.xs4all.nl Tue Oct 20 10:36:44 2009 Path: unlimited.newshosting.com!s02-b77!filter01.iad!npeer02.iad.highwinds-media.com!news.highwinds-media.com!feed-me.highwinds-media.com!postnews.google.com!news2.google.com!news.glorb.com!feeder.erje.net!newsfeed.xs4all.nl!newsfeed6.news.xs4all.nl!xs4all!newsgate.cistron.nl!newsgate.news.xs4all.nl!post.news.xs4all.nl!not-for-mail From: Arlet <usenet+5@ladybug.xs4all.nl> Subject: Re: Dealing with SPI ADC timings Date: Tue, 20 Oct 2009 19:36:44 +0200 User-Agent: Pan/0.14.2.91 (As She Crawled Across the Table (Debian GNU/Linux)) Message-Id: <pan.2009.10.20.17.36.43.626195@ladybug.xs4all.nl> Newsgroups: comp.arch.fpga References: <to6dnegeVNwQQEDXnZ2dnUVZ_qmdnZ2d@giganews.com> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Lines: 11 NNTP-Posting-Host: 213.84.26.192 X-Trace: 1256060204 news.xs4all.nl 83251 [::ffff:213.84.26.192]:37343 X-Complaints-To: abuse@xs4all.nl Xref: unlimited.newshosting.com comp.arch.fpga:94612 X-Received-Date: Tue, 20 Oct 2009 17:36:44 UTC (s02-b77) On Tue, 20 Oct 2009 10:55:57 -0500, ElVale wrote: > I'd like to implement an SPI master to read an AD7924 ADC. The thing is > that there are some setup and hold time I have to respect, ie CS'to SCLK > setup time, SCLK to DOUT valid hold time, etc. They go from typically 10 ns > to 50 ns. > > What's the best way to implement these delays? If the FPGA is too fast, add some dummy states to your state machine where you don't change the outputs.Article: 143663
I have foundation so compiling for large FPGAs is not a problem. There seems to me to be a gap between the exisiting $200/$300 boards & the $2000 and up boards. This gap is probably not big enough for commercial vendors which is why I asked if other people are interested in joining up for a larger "naked" FPGA board with lot of IOs so that it can be customized for whatever usage. The large FPGAs really need assembly with X-ray inspection to verify the soldering so doing it with manual tools is not appealing to me. But professional assembly has a large setup charge so doing a number of boards is necessary to get the per board cost down. Low cost & fast IO is not a problem, spending the money on the FPGA rather than setup charges for PCB & assembly is the aim for a group buy. "nwreader" <noone@home.com> wrote in message news:BvcCm.1404$MZ1.582@newsfe11.iad... > Is there any interest in a group buy & design of a Xilinx fpga board ? > I would like a fairly large Spartan FPGA + lots of IO expansion + low cost > $200 - $250 usd. > The exisiting FPGA boards on the market seems either too expensive or > small FPGA or too little > expansion potential. > > The idea is to get a high capacity & low cost expandable board via a group > buy to save on the > PCB + assembly. Probably only for US participants. >Article: 143664
Is anyone in this group interested in forming a team of loosely collected engineers, not of exclusivity but of mutual benefit? I work alone and have been successful in a few endeavors, but at this pace things are becoming excruciating. I am thinking that a team could generate a robust solution to a problem at an accelerated rate by combining resources of its members. I find myself isolated and need to reach out to prevent myself from sinking. I am not looking to drag anyone else down, but am reaching out for stability and stabilization. Words sometimes really miss the point I am trying to make. Anyone know what I am talking about. Sincerely, Cy Drollinger Electronic Realization L.L.C. Bozeman, MT cy@montana.net 406-586-5502Article: 143665
newbie, I would be interested. Having just completed a similar project, having posted a similar topic, and found a lot of interest, what is your ideal outcome of this project? On the topic of daughter boards what if the they were connected to themselves board to board like, bga, no cost? Cy DrollingerArticle: 143666
On Oct 20, 8:32=A0am, Martin Thompson <martin.j.thomp...@trw.com> wrote: > rickman <gnu...@gmail.com> writes: > > On Oct 19, 9:58=A0am, Gabor <ga...@alacron.com> wrote: > > Are you saying that on the Power PC they number bits in the data bus > > with 0 as msb and 31 as lsb? =A0How about address bits, is the lsb > > numbered 31 and the msb numbered 0? =A0 > > Yes. =A0And IIRC on the Motorola 88110 which preceded it, on which the > PowerPC 601 was based (buswise anyway). > > > I would find that very confusing. > > > I could care less if the byte addressing is big-endian or little- > > endian. =A0I don't see that having much import when defining the bits i= n > > a bus. > > If you look at address 0, it corresponds with bits 0 to 7 of the data > bus, address 1 is bits 8 to 15. =A0Just like on a little-endian processor= . > (Would you want address 0 on bits 31 downto 24? =A0Maybe?) > > Anyway, once they've taken that convention, making the address bus > numbered the same way makes more sense than making it opposite to the > databus! > > (But nothing beats the SHARC DSP booting from an 8 bit EPROM on bits 16-2= 3 of > the 48 bit address bus :) Why would I care what the numeric name of the bits for a given byte are? I much prefer to be able to determine the weights of each bit without messy arithmetic. The part that would really confuse me is if the address bus were numbered with 0 as the msb. Do the big endian processors do that as well? Is the uBlaze really done this way? RickArticle: 143667
nobody wrote: > Is anyone in this group interested in forming a team of loosely > collected engineers, not of exclusivity but of mutual benefit? I work > alone and have been successful in a few endeavors, but at this pace > things are becoming excruciating. I am thinking that a team could > generate a robust solution to a problem at an accelerated rate by > combining resources of its members. I find myself isolated and need to > reach out to prevent myself from sinking. I am not looking to drag > anyone else down, but am reaching out for stability and stabilization. > Words sometimes really miss the point I am trying to make. > Anyone know what I am talking about. > > Sincerely, > > Cy Drollinger > > Electronic Realization L.L.C. > Bozeman, MT > cy@montana.net > 406-586-5502 I was thinking something similar on a global scale. An organization of scientists who would group together and work on common problems. People who "request" problems to be solved. The idea would be similar to massive computing but with humans instead. By having a proper way to organize communications it could be very powerful. The organization would be more towards theoretical problems but potentially could be for practical issues. For example, a company could put out a work order for solving a problem. It's not necessarily about money of course but about giving people the ability to work together. There is power in numbers...Article: 143668
On Oct 20, 9:47=A0am, Colin Paul Gloster <Colin_Paul_Glos...@ACM.org> wrote: > On Sun, 18 Oct 2009, Rickman wrote: > > |---------------------------------------------------------------------| > |"[..] =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 = =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0| > |[..] =A0My copy of ispLever from Lattice will =A0 =A0 =A0 =A0 =A0 =A0 = =A0 =A0 =A0 =A0 =A0 =A0 =A0| > |not run compile or simulate because neither of these tools will run =A0| > |due to the license expiring. =A0[..] =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0= =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 | > | =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0= =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 | > |[..] =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0= =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 | > | =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0= =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 | > |So now I am trying to use the Xilinx Webpack to allow me to continue | > |working until I can get a new license file, but it won't run either."| > |---------------------------------------------------------------------| > > Xilinx Webpack used to be licensed such that it was forbidden to use > it for Lattice FPGAs. I do not recall that prohibition being removed. How is this about Lattice FPGAs? I'm trying to compile VHDL code and simulate it. RickArticle: 143669
rickman <gnuarm@gmail.com> wrote: (snip) > Why would I care what the numeric name of the bits for a given byte > are? I much prefer to be able to determine the weights of each bit > without messy arithmetic. > The part that would really confuse me is if the address bus were > numbered with 0 as the msb. Do the big endian processors do that as > well? Is the uBlaze really done this way? IBM consistently numbered the MSB 0 on S/360, S/370, XA/370, ESA/370, ESA/390, and z/Architecture. Note that z/ is the 64 bit extension, where all general registers and addresses are now 64 bits. The 32 bit instructions consistently operate on bits 32 to 63 of general registers, leaving 0 to 31 unchanged. Even more, z/ can operate on code using 64, 31, or 24 bit addressing as appropriate, and instructions change as the addressing mode changes. It is nice and consistent, and not so hard to get used to. -- glenArticle: 143670
On Oct 20, 3:21=A0pm, glen herrmannsfeldt <g...@ugcs.caltech.edu> wrote: > rickman <gnu...@gmail.com> wrote: > > (snip) > > > Why would I care what the numeric name of the bits for a given byte > > are? =A0I much prefer to be able to determine the weights of each bit > > without messy arithmetic. > > The part that would really confuse me is if the address bus were > > numbered with 0 as the msb. =A0Do the big endian processors do that as > > well? =A0Is the uBlaze really done this way? > > IBM consistently numbered the MSB 0 on S/360, S/370, XA/370, > ESA/370, ESA/390, and z/Architecture. > > Note that z/ is the 64 bit extension, where all general registers > and addresses are now 64 bits. =A0The 32 bit instructions consistently > operate on bits 32 to 63 of general registers, leaving 0 to 31 > unchanged. > > Even more, z/ can operate on code using 64, 31, or 24 bit addressing > as appropriate, and instructions change as the addressing mode changes. > > It is nice and consistent, and not so hard to get used to. > > -- glen How are the bits in the address bus numbered? I would find it extremely confusing to deal with an address where the lsb is numbered N where N varies depending on the size of the address range. RickArticle: 143671
rickman <gnuarm@gmail.com> wrote: > On Oct 20, 3:21?pm, glen herrmannsfeldt <g...@ugcs.caltech.edu> wrote: (snip) >> IBM consistently numbered the MSB 0 on S/360, S/370, XA/370, >> ESA/370, ESA/390, and z/Architecture. (snip) >> Even more, z/ can operate on code using 64, 31, or 24 bit addressing >> as appropriate, and instructions change as the addressing mode changes. > How are the bits in the address bus numbered? I would find it > extremely confusing to deal with an address where the lsb is numbered > N where N varies depending on the size of the address range. If by 'address bus' you mean the physical hardware, that is a separate question from the architecture. Otherwise, the numbering changed in z/, but is internally consistent. For S/360, 24 bit addresses are in bits 8 to 31 of general registers, for example, and the address as stored in the PSW is bits 40 to 63. z/ extended to 64 bits, so 24 bit addresses are in bits 40 to 63, 31 bit in 33 to 63, and 64 bit addresses in bits 0 to 63. That only comes up in reading the descriptions of the instructions. In actual coding, (software) one can write a loop in whichever way one wants. The hardware will execute it independent of the thought process of the programmer. It never confused me. -- glenArticle: 143672
Hi, I am trying to connect an IO signal from my microblaze to the external top-level vhdl code without using xilinx GPIO IP. My question is, couldn't I make a port definition in the MHS file and connect to it? for example, PORT Test_IO = Test_IO, DIR = I // an external port not defined by gpio ip and then on my top-level VHDL I would connect to it in the port map by Test_IO => somesignal, My second question is how can I read the signal value in microblaze. With xilinx's GPIO, I could read base address to get the value. But now there's no base address associated with Test_IO. Best Regards HV --------------------------------------- This message was sent using the comp.arch.fpga web interface on http://www.FPGARelated.comArticle: 143673
>But I would suggest you to understand the algorithms and implement >them in Matlab. Xilinx Accel DSP is a very helpful tool for problem >like yours. Without going into the deeper details, you can get your >prototype working hardware on the FPGA. I think that's the smartest, fastest way to do it, I've worked with DSP Builder and Synplify DSP with great results. You may not have a clue of what's going low level but the job gets done. --------------------------------------- This message was sent using the comp.arch.fpga web interface on http://www.FPGARelated.comArticle: 143674
If such organization ever takes place I would recommend strict implementation of guidelines for coding (like STARC VHDL desing rules, Aldec ALINT) and documentation (doxigen), so we could all understand each other easily. --------------------------------------- This message was sent using the comp.arch.fpga web interface on http://www.FPGARelated.com
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