Hello everyone,
I am going bananas on this one.
Maybe someone can help me before I jump out the window.
I am using the following setup:
. Windows 7 64 bits
. Bus Blaster v4.1a
. uRJTAG 0.10 (.exe replaced by the hacked one)
Here's a list of the things that are working fine:
. The BB board enumerates normally and I can see it on usbview
. I am able to program the internal xilinx buffer by using uRJTAG and setting jumper to "Update Buffer"
(I have programmed the board with the BBv4.svf)
. The board passes self test by putting all the necessary jumpers in (green-ish LED turns on)
Now some more background information:
I am trying to detect/program an AVR XMEGA using this board. I am using JTAG, not PDI.
This AVR XMEGA board is detected correctly using a USB Blaster under QuartusII programmer.
Every time I try to detect the board with BB I get the infamous "TDO seems to be stuck at 1".
I have tried to detect different boards which also detect normally with USB Blaster. Without success.
I have tried detecting the Xilinx FPGA on my Open Bench Logic Sniffer, also without success.
I have these cables going to the AVR board (which is powered by the BB):
BB AVR
VTG->3.3V
GND->GND
TDI->TDI
TMS->TMS
TCK->TCK
TDO->TDO
So at this point I have a few questions:
1 - By passing self test and programming the buffer did I rule out hardware issues on the BB?
2 - Would the BB work with AVR over JTAG? (3.3v - I powered the AVR from the BB with J4 on)
3 - Is there any latest/bleeding edge BB4.1a image I could try?
4 - Maybe I did something wrong with my cabling?
5 - Is the meaning of life really 42?
Thanks for any help.
Cheers,
Marcelo.
An update ... still no success.
I have tried completely cleaning up the FTDI drivers from my computer and reinstalling them after a reboot ... no success.
I have tried again to read the FPGA from my Open Logic Sniffer, no success either.
The connection from my BB4.1a to my Open Logic Sniffer are:
VTG -> 2V5
GND -> GND
TDI -> TDI
TMS -> TMS
TCK -> TCK
TDO -> TDO
Both boards plugged to USB ports on my laptop
I ran the self test again on the BB4 board just to make sure I was not seeing things, but the test still passes with zero errors and I am still able to "see" the internal CPLD buffer by selecting interface=1 when in "Update Buffer" mode.
I have simulated errors by taking jumpers out of the test and saw that the test fails accordingly.
So I still have no clue of what is going on.
Hoping someone can give some advice.
Cheers,
Marcelo.
Someone? Anyone?
I haven't still been able to figure out.
Haven't used my BBv2 with Windows but I encountered the same error on my Linux system. urJTAG is problematic to say the least, had to install/compile/apt-get it couple of times. Initially I used it on a 32 bit Linux system, then switched to a 64 bit Linux system so I'm guessing there is a problem there. Do you have access to a 32 bit machine?
Hi Tayken,
Good point. I will give it a try on a 32bit Windows XP I still have here.
It is still strange for me that the self test passes all the time. Meaning the signals are reaching the JTAG connector and coming back.
However a test on a 32 bit will definitely give some better clue or help rule out driver issues.
Good idea. I will test it and report back.
Cheers,
Marcelo.
Same thing, the only difference was that I was using Linux. Self test program is written by someone on the forum, so I know it works. But I don't have any insight about urJtag, there may be something wrong with it. It's been so long that I forgot what I did to make it work.
Hi Tayken,
Tested on a Windows XP 32 bits. Same results.
:-(
Thanks,
Marcelo.
Hey,
I believe you had to use a patch (or a patched version) for Windows. At least that's what I remember from Ian's posts waaay back during v2 times.
I have a 64 bit machine that doesn't have urJtag installed, so I'll try to install on that and see what the steps were. It will take a while though.
Yes, I've used the patched urjtag version provided in the forum.
So far I believe I am doing everything correctly. I must be missing some stupid detail.
Thanks immensely for the help.
Cheers,
Marcelo.
Marcelo, i posted my verilog files on the BBv3 stuck at 1 thread, but figured out my problem was actually my urjtag command. after flashing the jtagkey file:
jtag> cable ft2232 pid=0x6010 vid=0x0403 interface=0
gave me tdo stuck errors
jtag> cable jtagkey pid=0x6010 vid=0x0403 interface=0
worked just fine.
Perhaps this is or isn't your issue, if not post a log and i'll see if i can help at all.
I don't have a BBv4 to test anything but i built my tested bbv3 files for the bbv4 hardware and attached them. bbv4-jtaglink is logically identical to the standard jtaglink svf. bbv4-passthrough passes through all jtag signals.
good luck!
-joe
Hi Joe,
I have loaded your bbv4-jtagkey.svf stream into the internal buffer and made the following connections:
BBv4 -> OLS
VTG -> 2v5
GND -> GND
TDI -> TDI
TMS -> TMS
TDO -> TDO
TCK -> TCK
Left the JP4 open, meaning the target is self-powered.
The yellow(ish) LED blinks and then stays lit until I connect to the driver.
I have the following results:
UrJTAG 0.10 #1869
...
jtag> cable jtagkey pid=0x6010 vid=0x0403 interface=0
Connected to libftd2xx driver. (yellow LED turns off here)
jtag> bsdl path D:/bsdl
jtag> detect
Error: parse.c:208 urj_parse_file() no error: Cannot open file 'C:UsersMockba/
.jtag/rc' to parse
jtag>
I don't know what this .jtag/rc file should be. Not sure if this is an improvement or not.
If I use the ft2232 cable instead of jtagkey I get the "stuck at 0" message before the .jtag/rc one.
Thanks in advance for any ideas.
Cheers,
Marcelo.
PS.: Not sure how I could use the BBv4-passthrough.
[quote author="marcelo"]...
Error: parse.c:208 urj_parse_file() no error: Cannot open file 'C:UsersMockba/
.jtag/rc' to parse
jtag>
I don't know what this .jtag/rc file should be. Not sure if this is an improvement or not.
[/quote]
Hey Marcelo, I remember that Ian had the same error message,check here: viewtopic.php?f=37&t=1655&start=15#p15566 (http://dangerousprototypes.com/forum/viewtopic.php?f=37&t=1655&start=15#p15566) Adding driver=ftd2xx-mpsse fixed the problem for him that time.
The problem arises form Linux/Windows folder naming or the fact that one uses / and the other uses for directories. Not sure how to fix that. Maybe trying out the latest urjtag for windows can help?
Hi Tayken,
Thanks for the reply and for the ideas.
I don't think this is the same issue I am having. This is the message I got when I tried it:
jtag> cable jtagkey interface=0 pid=0x6010 vid=0x0403 driver=ftdi-mpsse
Couldn't connect to suitable USB device.
Error: parse.c:208 urj_parse_file() no error: Cannot open file 'C:UsersMockba/.jtag/rc' to parse
jtag>
At this point I am not sure anymore if I made the right decision when purchasing the latest BusBlaster. Maybe I should have purchased the v2.x. Or none at all. I basically see no support from Dangerous Prototypes other than occasional answers to my eMails from Ian that he doesn't tend to follow up afterwards.
Maybe he's too busy to deal with this, maybe the product doesn't work at all, I don't know. After all nobody replied to my post asking if anyone else was using the same version as I am. Maybe no one is.
Other than some valuable help like I am getting from you here, there's nothing else.
Maybe I will just throw this device on my "old boards" box and leave it there to scavenge for parts. My mistake buying it without researching first.
Anyways, thanks a lot for helping with all this. I am still believing that this issue may be caused by some simple mistake that I am making or some simple incompatibility that could be easily solved if Dangerous Prototypes spent some time troubleshooting it.
I think that when a company sells a prototype as a product they must understand that we stop being fellow experimenters and start being regular customers, which need some level of support or guidance from the people who develop it.
Sorry for venting out on a reply to you, but I needed to vend out to someone who is at least listening to me :)
Again, thanks a lot for the ideas.
Let me know if you have any other that may help get this to work.
Cheers,
Marcelo.
Hey Marcelo,
You have tried
driver=ftdi-mpsse
not
driver=ftd2xx-mpsse
Hey man, you're right, my bad.
It turns out that didn't help either, but by coincidence I ended up finding a solution:
jtag> bsdl path d:bsdl
jtag> cable jtagkey pid=0x6010 vid=0x0403
Connected to libftd2xx driver.
jtag> detect
IR length: 6
Chain length: 1
Device Id: 00010001110000011010000010010011 (0x11C1A093)
Filename: d:bsdl/XC3S250E.bsdl
jtag>
When trying to repeat different versions of the command and also combine different parameters I ended up forgetting to add the interface=0, and this seems to have solved the problem.
2 out of 3 times it will detect the device correctly, so I can consider this a big advance.
I didn't try to write a svf stream to it yet, gotta go outside to some sow shoveling, but I will certainly do it later today.
Thanks a lot for insisting on my, it was a big help.
All the best,
Marcelo.
Damn! My bad! cable ft2232 is used for flashing to the onboard CPLD only, for interfacing with other devices, you have to use jtagkey cable. interface=0 means interface with the onboard CPLD so it seems OK, seems like I overlooked some of the commands you wrote. :(
If you manage to stumble on it, can you post the output when the device is detected incorrectly? I wonder if that's a problem with urjtag or BB.
Happy showeling! :)
Hi Tayken,
Still lots of snow here. So now I am stuck at home with lots of time to play with the BBv4.
For some reason I was not able to replicate the previous success anymore. I have a feeling it is something related to the driver or maybe even UrJtag being messed up.
Today I was going to test the svf programming, but it detected the device one time and then not anymore.
I am not sure why.
This is the output I am having consistently today:
UrJTAG 0.10 #1869
Copyright (C) 2002, 2003 ETC s.r.o.
Copyright (C) 2007, 2008, 2009 Kolja Waschk and the respective authors
UrJTAG is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain conditions.
There is absolutely no warranty for UrJTAG.
jtag.c:518 main() Warning: UrJTAG may damage your hardware!
Type "quit" to exit, "help" for help.
jtag> bsdl path d:bsdl
jtag> cable jtagkey pid=0x6010 vid=0x0403
Connected to libftd2xx driver.
jtag> detect
Error: parse.c:208 urj_parse_file() no error: Cannot open file 'C:UsersMockba/
.jtag/rc' to parse
jtag>
I was able to detect the device once and the not anymore. Right now I am using the buffer code provided by Joe previously, which is the one that was able to detect before also.
I have also tried loading back the original buffer code and no success there either.
I am not sure how to use the bbv4-passthrough buffer code Joe provided, do I am back at square one again.
:(
Thanks,
Marcelo.
Her marcela,
There was a lot of snow here too, snowed on Friday and didn't melt. Had to shovel some ice to just get to the washing machine today. :)
The problem seems to be related to urJTAG. Can you try the original .exe from their website to see if it works? Latest source does not have that function and the hacked exe may be broken.
I'll dig up my BBv2 and see if it can work with the original urJtag under Windows. (I only used it with Linux)