Skip to main content ↑ ↓

Show Posts

This section allows you to view all Show Posts made by this member. Note that you can only see Show Posts made in areas you currently have access to.

Messages - r250r

1
Bus Blaster JTAG debugger / Re: BBv2: jtagkey + serial without reconfig?
[quote author="ian"]However, you would need to add that to every app that you use (OpenOCD, urJTAG, crossworks, etc, etc) or it might never switch to UART mode.[/quote]

I was hoping that there would be a transition on the ft2232's pins when it changed modes from rs232 to jtag - or that the real jtagkey hardware required initialization. If so, one cpld configuration would work with any program that supports the jtagkey.

I guess I'll just use an extra pin on jp2, or reprogram it as necessary.
2
Bus Blaster JTAG debugger / Re: BBv2: jtagkey + serial without reconfig?
[quote author="ian"]That's a good point. I find they are usually hi-Z unless the com port is open, but out of extra caution I will make a buffer that goes hi-z on the CPLD->FT2232 side when not enabled.[/quote]

So there's a known sequence the FT2232 output pins go through when switching to MPSSE/JTAG mode?  I was wondering if there was, but wasn't sure how to find out.

I want to configure the cpld so that it is in "serial passthrough mode"[1] on powerup, and switches to JTAG mode when the ft2232 is reconfigured.

Now that I think about it, the JTAG pins must not be a problem since the ft2232 is marketed for apps with jtag on one channel and serial on the other.

Oh, I did finally find sink/source current info in the datasheets - just not where I was expecting it. However, neither device's datasheet says how many pins can be shorted, for how long, etc.

[1] all serial lines pass straight through to either the jtag header or to JP2
3
Bus Blaster JTAG debugger / Re: BBv2: jtagkey + serial without reconfig?
[quote author="ian"]Hi r250r - No, I don't think that will work. The buffer uses various control signals that would have to be set correctly, and I think the RX/TX are actually a different direction (input/output) than the JTAG signals on the same pins.
[/quote]

You're right. It would be nice if they had used the same directions for both!

Quote
It would be possible to make a new buffer that gave access to the RX/TX pins.
Yes, I was thinking of doing that.

In the UCF, there are two pins that connect to the 2232 that aren't used. I was wondering about programming the cpld with logic that would allow either jtag mode or serial mode, configurable using one or both of those spare IOs.

If I do that, the IO directions will change. I want to make sure that both chips can handle short-circuited IO.

I cannot find anything about short-circuit tolerance, or even absolute maximum output current, for the CoolRunner-II. I looked at 4-5 different PDFs on xilinx.com.

The ft2232 datasheet doesn't seem to say anything either.

I hope the JTAG pins on the CoolRunner can handle abuse, as it looks like both channels of the ft2232 default to RS232 async serial.
5
Bus Blaster JTAG debugger / Re: urJTAG fails with libFTDI on Linux
I just tried the latest version of urJTAG (git 08e714ef68513) with libFTDI, and it works fine.  Thanks for the bus blaster, and thanks to whoever fixed the issue as well!

I edited the original post to reflect the fact that this is fixed.
6
Bus Blaster JTAG debugger / Re: troubleshooting bbv2
Quote
Looking at the output I guess you are using urJTAG on Linux with the libFTDI GPL drivers?
Yes, the GPL driver and 64-bit Debian (also tried 64-bit Ubuntu, same result).
Quote
Just to check - did you compile the latest SVN version without applying our patch? Our patch is no longer needed because it is included in the project.
I used the latest revision of urjtag, and no patches.

Quote
Would it be possible to test the FTDI drivers instead? You may have found a bug in our patch for GPL libFTDI support.

I downloaded the latest 64-bit driver from http://www.ftdichip.com/Drivers/D2XX.htm .  It works.

I was able to run urjtag without installing either libftd2xx or urjtag.  If someone else wants to do it, here's what I did:
Code: [Select]
$ cd libftd2xx1.0.4/
$ ln -s build/x86_64/libftd2xx.so.1.0.4 libftd2xx.so
$ cd ../urjtag/urjtag
$ mkdir build
$ cd build
$           # --with-ftd2xx needs an absolute path
$ ../configure --with-ftd2xx=/home/mark/projects/hw-dev/libftd2xx1.0.4/ --without-libftdi
$ make
$ sudo rmmod ftdi_sio
$ sudo rmmod usbserial
$           # symlink so that urjtag can find its manufacturer data
$ sudo ln -s /home/mark/projects/hw-dev/urjtag/urjtag/data /usr/local/share/urjtag

You must point LD_LIBRARY_PATH at libftd2xx/ when running urjtag:
Code: [Select]
$ LD_LIBRARY_PATH=/home/mark/projects/hw-dev/libftd2xx1.0.4 src/apps/jtag/jtag

The Bus Blaster is recognized now:

Code: [Select]
jtag> bsdl path ../../bsdl
jtag> cable ft2232 pid=0x6010 vid=0x403 interface=1
Connected to libftd2xx driver.
jtag> detect
IR length: 8
Chain length: 1
Device Id: 00000110111000011100000010010011 (0x06E1C093)
  Filename:     ../../bsdl/xc2c32a_vq44.bsd
jtag> svf ../../buffer_logic/BBv2-JTAGkey-v1.3.svf progress stop
Parsing    660/663 ( 99%)
Scanned device output matched expected TDO values.
jtag>
7
Bus Blaster JTAG debugger / [RESOLVED] urJTAG fails with libFTDI on Linux
EDIT: this is fixed as of urjtag git revision 08e714ef68513

I got my BBv2 yesterday.  I downloaded and compiled the latest version of urjtag.

If I try to autodetect, urjtag says my board is a flyswatter.

If I use the ft2232 driver, I get a warning that TDO seems to be stuck at 1. I get this no matter the interface, and also if I try to use the jtagkey driver (which I thought the board was supposed to come preconfigured for).

Quote
jtag> cable ft2232 vid=0x403 pid=0x6010 interface=2
Connected to libftdi driver.
jtag> detect
../../../src/tap/discovery.c:117 urj_tap_detect_register_size() Warning: TDO seems to be stuck at 1
Error: (null):0 (null)() no error:
jtag>
I've tried interface=0, =1, and =2 -- all with the same result.  Urjtag has been acting strange - for example, the error above, and parameters seem to persist, which doesn't make sense to me:
Quote
jtag> cable ft2232 vid=0x403 pid=0x6010 interface=b
Usage: cable DRIVER [DRIVER_OPTS]
<snipped rest of usage message>
Error: ../../../src/global/params.c:201 parse_param_lu() syntax: need unsigned int, not 'b'
jtag> cable ft2232 vid=0x403 pid=0x6010 interface=1
Connected to libftdi driver.
jtag> detect
../../../src/tap/discovery.c:117 urj_tap_detect_register_size() Warning: TDO seems to be stuck at 1
Error: ../../../src/global/params.c:201 parse_param_lu() no error: need unsigned int, not 'b'
jtag>

So I'm not sure if urjtag is the problem or if it's something else.

One thing I noticed about my board is that the FTDI chip looks different than the one pictured at http://dangerousprototypes.com/docs/Bus ... ew#FT2232H .  The last three letters of FTDI look like they cut deep into the package, instead of being on the surface. ?!

As an aside, there is confusion in your docs about whether to use interface=1 or interface=2 - compare the screenshot with the text on both of these pages:
http://dangerousprototypes.com/docs/Cla ... ith_urJTAG
and
http://dangerousprototypes.com/docs/Bus ... _resources

edit: add TDO paragraph
8
Bus Blaster JTAG debugger / Re: Bus blaster v2 eta?
[quote author="ian"]LOL, they should have 100 waiting to be tested. Not sure what that is about :)

They should be available any day, I guess it is our next project.[/quote]

I certainly hope so! I've been checking every day, since frank26080115 told me about BBv2 on reddit :)

( ! ) Fatal error: Uncaught exception 'Elk_Exception' with message 'Please try again. If you come back to this error screen, report the error to an administrator.' in /var/www/dangerousprototypes/forum/sources/database/Db-mysql.class.php on line 696
( ! ) Elk_Exception: Please try again. If you come back to this error screen, report the error to an administrator. in /var/www/dangerousprototypes/forum/sources/database/Db-mysql.class.php on line 696
Call Stack
#TimeMemoryFunctionLocation
10.01572362680session_write_close ( )...(null):0
20.01612494256ElkArte\sources\subs\SessionHandler\DatabaseHandler->write( )...(null):0
30.01612495032Database_MySQL->query( ).../DatabaseHandler.php:119
40.06002633752Database_MySQL->error( ).../Db-mysql.class.php:273