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
Tommy Thorn wrote: > alpha wrote: > > It's just for fun to run it in a FPGA rather than making an ASIC. I > > spent almost all Saturdays in the past 1 year for this project. > > I should say, it has lots challenge than a single-issue one. > > Very cool, congratulations. > > > True, > > clock speed is pretty low now now (30-60 Mhz). But I am planning to > > enhance it by using a smarter compiler to removing some corner case > > from the design. > > How much work would it be to make it fully R3000 compatible (say, > user-level only)? [I am not trying to make binary compatiable with R3000. No delay slot. Right now, total 24 instructions. No multiplication. Actully V4's DSP core can be used to do multiplication. Precise Interrupt works but not well tested, I had a timer interrupt. Seems still lots work to be done.] > > > Hopefully I cam make it to 100Mhz range. > > Also, looks I need to buy an evaluation kit from Altera. Xilinx V4 > > block ram gives me some trouble during instruction fetching. > > Care to elaborate? [The design was orignially tageted an Altera ApexII board, so the instruiction fetching unit fetch instruction from altera's async block memory and interface to reorder buffers. Then later on, I switched to bigger V4 chip. I have to play some trick to make it work with Xilinx's sync block memory. ] > > My design is from scratch. Its instruction set is almost same as MIPS > > 3000 (without Multiplication). Lcc C compilier was ported. > > I can publish the source verilog files, do we have public domain for > > this purpose? > > Picking a license (eg. GPL, LGPL, BSD, public domain, etc) is a separate > issue from publishing (opencores, sourceforge, someones own web site > [I'd be happy to host it for you fx.], etc). Just be careful about > third party components (such as LCC) which come with their own licence. > [I need think about it, thanks anyway. Of course, I do not want to get any trouble] > > (IMHO: Out-of-Order on an FPGA is very cool, but will probably not lead > to the best performance/LUT ratio - especially not if you can do > compiler work also. For single threaded performance, Nios II and > MicroBlaze are impressive, as is John Jakson's R16. A 2- or 4-way LIW > may also make sense). > [This design was started in Modelsim just for fun a year ago. The goal is to make a out-of-order superscalar processor. I even do not know if I can systhesis it for a FPGA. 9000 LE(s) is pretty big. 50Mhz * 1.5(IPC) = 75 MIPS. can not compete with Microblaze or Nios. But it is a good try. Agree, a simple 2 or 4 way static scheduled VLIW is really worth to try. ] > TommyArticle: 102901
the supply is capable of approx 3A fault current. however, the limit was set to a little over 1A when the fault was first detected.Article: 102902
Steve, the OP asked for SPI indirect jtag programming, this is possible with software from Altera Lattice Xilant Xilinx is not on the list - and none of the references from your posting has relevant info how to implemented the jtag indirect SPI programming we are using a very simply BSCAN to SPI gateway - available for download here http://xilant.com/component/option,com_remository/Itemid,53/func,select/id,4/ and rather simple JTAG programming using custom jtag host software, but also it would be possible to convert the programming file to SVF and just play it agains the SPI gateway and you get the SPI flash programmed AnttiArticle: 102903
John Adair schrieb: > Headline pricing often has a date sometime in the future against it. But they keep publishing the press release on the website, that's enough for legal action: http://www.xilinx.com/prs_rls/silicon_vir/0568fx60_spirent.htm Apparently: 1. An FX20 did cost 50$ in 25k quantities in december 2005 2. Xilinx has been shipping 10gbps MGTs as early as June 2005. Now, my FAE tells me that there never have been any Virtex-4 with 10gpbs MGTs. So Xilinx should really modify that web page. Kolja SulimmaArticle: 102904
sorry, sure JTAG Tech, etc also, I left the big JTAG companies out of the list as I assumed its obvious they support the SPI programming over boundary scan. AnttiArticle: 102905
"alpha" <zhg.liu@gmail.com> writes: > [I am not trying to make binary compatiable with R3000. No delay slot. [...] > [I need think about it, thanks anyway. Of course, I do not want to get > any trouble] It appears that you've avoided the unaligned load/store instructions, which are patented. Between that and not actually being binary compatible, I don't think you're likely to get in any trouble if you choose to publish your design. Of course, I am not a lawyer. EricArticle: 102906
"Antti" <Antti.Lukats@xilant.com> writes: > Steve, the OP asked for SPI indirect jtag programming, > this is possible with software from > Altera > Lattice > Xilant Also add JTAG Technologies (www.jtag.com). > it would be possible to convert the programming file to SVF and just I've done this for microcontrollers, I2C devices etc (not SPI yet, but the principle is the same). and used Xilinx, Altera, and JTAG Technologies software to program the devices. Petter -- A: Because it messes up the order in which people normally read text. Q: Why is top-posting such a bad thing? A: Top-posting. Q: What is the most annoying thing on usenet and in e-mail?Article: 102907
hi after deleting unused xilinx webpack directory the main ISE doesnt start anymore is there some trick except full re-install of ISE and service packs? at the moment PN starts shows the main gui windows, an hourglass and consumes 99% of CPU time (winXP) I already tried to fix the registry to fix-remove the references to webpack install location but that did not help AnttiArticle: 102908
It is very strange. If I change from the USB cable to the Par IV, it works. And if I change back to the USB again it also works. But if I compile and download it might not work again and I have to switch to the Parallellcable again?.? In this case I can use the ParallellCable, but I would really like to know why it act so strange with the USB. (My ParallellCable does not fit on my second board (ML401) so I am most likely to run in to this problem again :( RaymondArticle: 102909
Has anyone used the OPB Timer on the MicroBlaze? I am trying to figure out how it works and by doing that I have inspected some adresses: in the xparameters.h #define XPAR_TESTTIMER_BASEADDR 0x41C00000 in the xtmrctr_l.h #define XTC_TCSR_OFFSET 0 /**< control/status register */ #define XTC_TLR_OFFSET 4 /**< load register */ #define XTC_TCR_OFFSET 8 /**< timer counter register */ After I have called the function XTmrCtr_mWriteReg(XPAR_TESTTIMER_BASEADDR, 0, XTC_TLR_OFFSET, 0x0F0F0F0F); I suspected the memory to look like this: 0x41C00000 00 00 00 00 00 00 00 00 0f 0f 0f0f 00 00 00 00 0x41C00010 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 But however it looks like this. Does anyone know whats wrong? 0x41C00000 00 00 00 00 0f 0f 0f 0f 00 00 00 00 00 00 00 00 0x41C00010 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 in the in the xtmrctr_l.h The function I am calling is defined like this: #define XTmrCtr_mWriteReg(BaseAddress, TmrCtrNumber, RegOffset, ValueToWrite) \ XIo_Out32(((BaseAddress) + XTmrCtr_Offsets[(TmrCtrNumber)] + \ (RegOffset)), (ValueToWrite)) The XTmrCtr_Offsets[(TmrCtrNumber)] is "defined" ekstern extern Xuint8 XTmrCtr_Offsets[]; Where is it ? Do I need to include some other headers? I suspect it to be the reason that the memory looks unexpected. RaymondArticle: 102910
Grrrr. The problem comes and goes on the ParalellCable too RaymondArticle: 102911
Eric Smith schrieb: > "alpha" <zhg.liu@gmail.com> writes: >>[I need think about it, thanks anyway. Of course, I do not want to get >>any trouble] > > > It appears that you've avoided the unaligned load/store instructions, > which are patented. Between that and not actually being binary > compatible, I don't think you're likely to get in any trouble if you > choose to publish your design. Of course, I am not a lawyer. To begin with, it is controversial whether publishing a VHDL text that describes a patented method infringes the patent. A patent grants the monopoly for "making, using, selling, offering for sale, or importing the patented invention for the term of the patent". Implementing the vhdl in an FPGA probably would be "making". But just describing the invention in a formal way on your web page should be OK, as the whole point of the patent system is to publish the invention. Trade secrets and patents are mutually exclusive. Kolja SulimmaArticle: 102912
Hi Raymond, "Raymond" <raybakk@yahoo.no> wrote in message news:1148374841.922586.212360@j33g2000cwa.googlegroups.com... > Has anyone used the OPB Timer on the MicroBlaze? > > I am trying to figure out how it works and by doing that I have > inspected some adresses: > #define XPAR_TESTTIMER_BASEADDR 0x41C00000 > #define XTC_TCSR_OFFSET 0 /**< control/status register */ > #define XTC_TLR_OFFSET 4 /**< load register */ > #define XTC_TCR_OFFSET 8 /**< timer counter register */ > After I have called the function > XTmrCtr_mWriteReg(XPAR_TESTTIMER_BASEADDR, 0, XTC_TLR_OFFSET, > 0x0F0F0F0F); I suspected the memory to look like this: > > 0x41C00000 00 00 00 00 00 00 00 00 0f 0f 0f0f 00 00 00 00 > 0x41C00010 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 > > But however it looks like this. Does anyone know whats wrong? > > 0x41C00000 00 00 00 00 0f 0f 0f 0f 00 00 00 00 00 00 00 00 > 0x41C00010 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 Nothing is wrong, the memory looks perfect. The "TLR" register is 4 bytes above the base address and that's where the data got written. Why do you think the memory should look different? If you wrote to the XTC_TCR_OFFSET then it would look like what you "suspected" before... did you make a typo? -Ben-Article: 102913
> Ok, I'm fully admitting that I didn't read your post super-carefully > and I'm not looking at the datasheet, but there is a setting in the > mode bits for drive strength in those Micron DDR chips. > Yep, it's the 'Extended Mode register' which, in simulation at least, gets set to all 0 by the Altera DDR Controller and that should keep the drive strength at the 'Class II' level. Haven't checked yet to see if this is happening correctly on the real board but I suspect that it is. KJArticle: 102914
Hi all, I am exploring the possibility of using a FIFO between the OPB and my custom IP-core. I want to write data from my IP into the FIFO. I see that the FIFO has an output called RFIFO2IP_WrAck. This is an ackonowledge signal that the fifo asserts when it is ready to read data. According to some examples it seems like this signal can go low randomly, that means regardless of the fifo being full or not. To me this means that the data I want to write to the fifo must be buffered before it enters the fifo, resulting in 2 levels fifos. Then it is better to use a self written FIFO that can accept data all the time(unless when the fifo is full, of course). Is my assumptions wrong ? In real life, maybe this signal never goes low ?Article: 102915
>On the otherhand there are strong truth in advertising laws... >...you can issue a PO based on those terms and start designing... You have a touching notion of how widely these laws extend (if they exist at all). The general rule is "caveat emptor" and, in general, no contract is made until a clear offer has been both made and accepted. Instead of playing amateur lawyers, engineers should look to practical means to keep component costs under control. Suppliers are dismayed most (and hence most flexible) when they know you have other options. This is the best reason to keep proprietary aspects of your design to a minimum.Article: 102916
sorry, problem solved the hard way (it was 8.1SP3 not 4 as my posting heading was) attempts to revocer, all failing 1 windows registry clean and fix 2 ISE SP3 re-install 3 8.1 re-install 4 win reboot 5 8.1 de-install 6 8.1 install 7 reboot 8 path and env clean up and fix 9 8.1 webpack install to same path where it was initially 10 reboot at this point both webpack 8.1 and 8.1 all did FROZE after startup. I was joking that all that helps is Format [X] ilinx but then I tried to double click on .ise file, and ISE started normally so the all problem was damaged .ise project file that ISE tried to load as default project. nasty BUG in ISE - did cost me half a day. :( AnttiArticle: 102917
Hello everybody, Currently I am designing very accurate delay generator, which will be based on FPGA . This delay generator should have similar technical requirements with DG535 http://www.thinksrs.com/products/DG535.htm. The major Delay Generator requirements are =B7 2 ns (1ns is desired, but 2ns will be also ok) time resolution on delayed channel (it means that time differences between any delayed channels can be set in 2 ns steps) =B7 maximal 50 ps - 60ps (RMS) jitter on each output. =B7 14 delayed ECL channels =B7 Two high speed (PECL) inputs (500 Mz ECL clock signal and ECL trigger) =B7 Configurable via standard bus (Ethernet/USB/Serial bus) =B7 Internal trigger with variable rate (DDS) =B7 Internal clock oscillator =B7 Clock master or slave Does anybody know for commercial available FPGA boards (preferred ISA (PC104) or PCI (PC104 plus) standards) that can be suitable for my requirements? Sutiable FPGA for my design is V5 or Stratix 2 GX. Thank you and Regrads,=20 AmirArticle: 102918
No Ben no typos this time. This is my fault I counted from the left, not the right sorry. But thanks for answering :) RaymondArticle: 102919
Kolja Sulimma wrote: > John Adair schrieb: > > Headline pricing often has a date sometime in the future against it. > > But they keep publishing the press release on the website, that's enough > for legal action: > http://www.xilinx.com/prs_rls/silicon_vir/0568fx60_spirent.htm > > Apparently: > 1. An FX20 did cost 50$ in 25k quantities in december 2005 > 2. Xilinx has been shipping 10gbps MGTs as early as June 2005. Howdy Kolja, While the page is "on the web", I wouldn't call it an active page - you are looking at a historical document (clearly dated June 2005) about the current state (as of June 2005) and a future predicted price (as of June 2005). I think it would difficult to find any evidence that they are still making those claims, although I suppose you could argue that since there is no newer *published* information, by default one would easily assume that to still be accurate. Note that I'm not defending these nearly useless press releases, which are purely a marketing ploy to get an absurdly low price in the minds of engineers, 99% of whom will never be able to get anywhere close to that quantity (and therefore, that price). > Now, my FAE tells me that there never have been any Virtex-4 with 10gpbs > MGTs. It depends on what he means by "never" :-) They were never shipped as generally available product to any customer that wanted them (lots of us wanted them, and still do!). But I assume that Xilinx had 10 Gbps prototypes working, and even shipped some of those prototypes to customers as early as June 2005. None of these statements, however, mention anything about general availablity or being in full production. They could have shipped just a few parts to a few alpha customers and that would have fulfilled the statements made. Unfortunately, these types of press releases are all too common in the technology world (actually this one isn't as bad as some, where press releases are put out for products that aren't even designed yet). > So Xilinx should really modify that web page. Again, in this particuar case, it's a press release, which is historical documentation of a public statement. You can't go back and change them - the best you could do would be to put a notice at the top saying that it had been superceeded by something else, or remove it completely. Have fun, MarcArticle: 102920
Xilinx block RAM is synchronous read. Is this the source of your problem?Article: 102921
MikeShepherd564@btinternet.com wrote: > >On the otherhand there are strong truth in advertising laws... > >...you can issue a PO based on those terms and start designing... > > You have a touching notion of how widely these laws extend (if they > exist at all). The general rule is "caveat emptor" and, in general, > no contract is made until a clear offer has been both made and > accepted. > > Instead of playing amateur lawyers, engineers should look to practical > means to keep component costs under control. Suppliers are dismayed > most (and hence most flexible) when they know you have other options. > This is the best reason to keep proprietary aspects of your design to > a minimum. Certainly ... on the other hand, I made the final "bait and switch" complaint that shut one vendors advertising rights down for a year after talking with the local DAs business and economic crime unit. So .... your milage may vary depending on your local officals ... the laws do exist in some areas which mandate clear "truth in advertising" requirements. If you have a business, I'd suggest checking up on them before making press releases and printing advertisements, as somebody may hold you to the printed word if you refuse a sale. Other references: http://www.mlmwatch.org/04C/CT/suit.html http://www.mlmwatch.org/04C/CT/ruling.html http://sacramento.bizjournals.com/sacramento/stories/2003/07/14/daily32.html http://64.233.167.104/search?q=cache:BPtVNZKZ7n8J:reclaimdemocracy.org/nike/pr_seek_truth.html+california+truth+in+advertising&hl=en&gl=us&ct=clnk&cd=49Article: 102922
I have solved my problem, the bmm hierarchy should be updated as the mb is changed to be a submodule. And at the end of every bram, a "Loc" attribute should be added to point out the location of the bram. After the two modification, I update the bit file without any warnings. And then I download the new bit file, and the changes in code take effect, which proves that this method works.Article: 102923
MikeShepherd564@btinternet.com wrote: > >On the otherhand there are strong truth in advertising laws... > >...you can issue a PO based on those terms and start designing... > > You have a touching notion of how widely these laws extend (if they > exist at all). The general rule is "caveat emptor" and, in general, > no contract is made until a clear offer has been both made and > accepted. By the way Mike ... being from the UK you might not understand that Xilinx is located in California, the state with the strict "Truth in advertising" law referenced ... just google calfifornia truth in advertising ... the law does exist, and is well tested. Even corporate giant Nike lost a few years back when their PR releases were challenged ... exactly the same issue in this thread. So, if the Xilinx price claims are not true, in Calif they can run but not hide from the liability.Article: 102924
Nigel wrote: > the supply is capable of approx 3A fault current. however, the limit > was set to a little over 1A when the fault was first detected. Was the chip hot after the failure? If not, latchup can be ruled out. The 1 A it may have had may or may not suffice to fry it if latchup has occured (for whatever reason). Some large peak current discharging the power filtering capacitors can have helped, do you have an idea about that (how much capacitance there, that is)? Dimiter ------------------------------------------------------ Dimiter Popoff Transgalactic Instruments http://www.tgi-sci.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