11-27-2008 08:19 AM - edited 11-27-2008 08:22 AM
12-12-2008 04:07 PM
I'm having the exact same problem with a Hitech Global board HTG-V5-PCIE-330T Rev 2.0. I wonder if the problem is the
programming core loaded into the fpga, in my case C:/Xilinx/10.1/ISE/virtex5/data/xc5vlx330t_bpi.cor.
Could anyone help out on this? How have you solved the problem? I have several boards and it seems to affect some boards
but not others. Any help much appreciated, or if there is an alternative programming core please let us know.
12-29-2008 06:08 AM
Were you able to resolve the issue with Indirect BPI programming? I have the same problem: Failed to query common flash interface!
01-11-2009 05:42 PM
01-16-2009 06:54 AM
02-04-2009 03:08 AM
We have used the same flash(XCF128X) on our board and we received the same error.
But we were able to solve the problem just by changung the way we generated the .mcs file for flash in impact.
I would like to give you the procedure for the same.
2.Generate PROM file.
3.Select Xilinx PROM
FF as hex code
4.My flash is Parallel and bit width same as FPGA
5.Select my flash(XCF128X)
Then generate the mcs file by right clicking on the flash.
Then initialize the JTAg chain.
1.Then Add the BIT file and enable the option enable BRI flash.
2.Add the mcs file and Program the flash.
We still had some problem in one of our brs and found out it was an assembling issue.
I use 10.1.03
Let me know if you need any other infomation.
02-08-2009 06:25 PM
Can you tell me the former way that you generated the .mcs file for flash in impact (received the same error)? So i can know the difference between the former way and the new way. We used the same way as you give as above but still can't query the common flash.
By the way, which FPGA configuration method you used in your board? Master SelectMAP, Slave SelectMAP or Master BPI-UP? In the user guide, there is no information about the bpi.cor. So i can't know which configuration method the bpi.cor supports.
02-09-2009 09:37 AM
I follow the documented method for generating the mcs file. I have 2 identical boards. On one board the BPI flash
programming works fine and on the other it fails with the "Can't query common flash interface" error.
The mcs file is the same, if the mcs file generation was the problem it should not work on either board.
02-09-2009 07:39 PM - edited 02-09-2009 07:39 PM
We manufactured five boards with
LX220T and we found that three of them had the Assembling issue for
FPGA.(The error occured at this stage)
Other two boards with the same flash worked fine.
We tried reprogramming the flash after the board underwent reflow process again and is working well.
FPGA balls had dry soldering in the middle of its core.
02-09-2009 10:34 PM
We have 3 boards with lx220t. All boards can't query the common flash interface.
But they all work well when they are configured by the JTAG chain. So i can hardly believe it is due to the assembling error of the FPGA balls.
02-27-2009 07:53 AM
I too am having this problem on an LX30T.
However mine is more sparatic from what I can tell.
I have not successfully at this point programmed the flash...that always fails.
However I have successfully erased the chip about 50% of the time (the other times, erase fails).
I also can verify the chip is empty.
I have suspected maybe its a power related problem during the erase or program cycles, but I really can't tell at this point. I would assume if it was an assembly issue it would be 100% consistent in where it fails and what message I get, but maybe I'm wrong.
and an Intel 28F640J3
04-02-2009 12:31 AM
yeah,I have same problem to config three vsx50t by daisychained FPGA applications. I have cut off done signal each, and pull up both. I can progam the mcs file into flash, but only zhe first 50t can run ,the others two 50t are not configration! what 's the problem? The first one is BPI mode ,the others is Slave SelectMAP mode!
06-09-2009 06:29 AM
Hi did anyone ever resolve this?
We're currently having a similar problem. We were using an LX50-FF676 with an Intel StrataFlash 128 P30. This was working no problem for over a year. All we've done is move to an LX85-FF676 and Impact can no longer program the Flash with the FPGA image. If we use Impact/JTAG to download a bitstream with a Flash controller in the design, we can access the Flash without a problem - so we don't think there's a solder problem with our boards.
In a nutshell Impact can program all of our LX50 boards' FLASH, but cannot program all of our LX85 boards' FLASH. Other than the FPGA, the boards are identical.
We've been using ISE 10.1.03
We've been wondering if anyone (Xilinx in particular) have successfully programmed a Flash via an LX85, however, now seeing this thread, I guess others might have been wondering about the LX50.
06-12-2009 06:06 AM
Well I've found a solution to my problem. Read on
At 10.1.03 and almost certainly 11.1 ISE, the bpi_cor for the LX85 is wrong. When you look at the file in your ISE installation it is the bpi_cor for the LX85T (if you go off filesize).
Each family member has a bpi_cor and an spi_cor which should be the same filesize.
I took the LX85 bpi_cor from my 9.2 ISE installation and dropped it into my 10.1.03 installation and now everything is fine.
We can now use Impact to program the FLASH.
Xilinx are now aware of this.
07-28-2009 12:36 PM
I had this problem as it says below. I was never able to resolve it before being pulled from that program and then laid-off. The instrumented the FPGA to flash interface and it had the correct data for a ways, then the data was bogus and then stopped. The one thing I did not have as the App Notes specify was the CCLK terminated external to the FPGA.
That is my presumption at this point, since it is highly unlikely I will have the oppurtunity to test it again.
I saw the suggestion below about how to generate the PROM file. Basically, to NOT left-click on the "Generate PROM" link on the left-hand side of the window, but instead right-click and select "generate PROM" from the drop-down menu. I have heard of problems with selecting the links on the left side, so it might be worth a try.
07-17-2012 06:44 PM
Hi. Have you resolved this problem.3 years later, i also meet the same error: Failed to query common flash interface.
07-18-2012 02:40 AM
Might I suggest that you start a new thread in the Design Tools - Others forum to get more traction on this (http://forums.xilinx.com/t5/Design-Tools-Others/bd-p/OTHER) ? That is where questions on impact are answered, and there should be iMPACT programming experts monitoring that forum.