hello,

I get the following message while trying to connect to my de2-115
board (altera cyclone 4)
./adv_jtag_bridge -b ./ usbblaster
Found Altera USB-Blaster
Failed to reset USB Blaster
Found Altera USB-Blaster
USBBlaster driver failed to set configuration
Failed to initialize cable 'usbblaster', aborting.
I'm using ubuntu virtual machine on windows 7. Both libusb, libusb-0.1
compatible api, libftdi were compiled with the default commands:
./configure
./make
sudo ./make install
Can anyone tell me what seems to be the issue / has anyone seen such
errors before? Please help.

Regards,
AR

You sould try to use ./adv_jtag_bridge ft245. I've never get it works using usb_blaster option.

Hi,

You should try to use ./adv_jtag_bridge ft245

This config : ./adv_jtag_bridge usbblaster doesn't work for me.

Franck.

i've tried ft245 option ... this is what i get:

./adv_jtag_bridge ft245
'usb_blaster' interface using libftdi
unable to open ftdi device with VID 0x9FB, PID 0x6001: inappropriate permissions on device!
Failed to initialize cable 'ft245', aborting.

My usb cable is recognised as altera ...

lsusb
Bus 001 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 001 Device 002: ID 80ee:0021 VirtualBox USB Tablet
Bus 001 Device 006: ID 09fb:6001 Altera

:-(

Some success ... the problem is probably due to user priviledge when the adv_debug_sys tries to access the drivers in the /usr/local directory ... using root priviledges sorts this out ... sudo ./adv_jtag_bridge -b ~/bsd but now i get these new errors: sudo ./adv_jtag_bridge -b ~/bsd ft245 'usb_blaster' interface using libftdi current latency timer: 2 Enumerating JTAG chain... ftdi_read_data: usb bulk read failed ftdi_read_data: usb bulk read failed Failed to read stream from the EP1 FIFO (-1) ftdi_read_data: usb bulk read failed Failed to read stream from the EP1 FIFO (-1) Error 'USB' enumerating JTAG chain, aborting. sudo ./adv_jtag_bridge -b ~/bsd usbblaster Found Altera USB-Blaster Found Altera USB-Blaster firmware version = 0x0004 (4) Enumerating JTAG chain... Failed to read from the EP1 FIFO (-110): No errorError 'USB' enumerating JTAG chain, aborting. Any ideas what can I do now ??? Regards, AR

This is still a problem with your setup, something between USB and your VM. Libusb isn't able to do anything at all to the device. I don't use a VM, maybe someone else who has one working can give you some hints.

Hi Nathan,

Thanks for all your all your help. I believe it is a problem with the VM since sometimes it connects well and sometimes it doesn't do anything at all. And sometimes the connection breaks once I try to connect the GNU Debugger to port 9999.

However, the VM i'm using is the same ORPSoC Virtual Box Ubuntu 11.10 distibution that comes with the toolchain and debugger preinstalled. I haven't heard any other posts on the forums regarding this.

I have some specific questions and would be very grateful if you could tale some time answering them.

(a) Is there any other alternative way of connecting instead of using adv_dbg_sys software?

(b) I've enabled the JSP on my 'adv_dbg_defines.v' file as well as in the 'Makefile.classic'. I've compiled using the following commands:
autoconf
sudo ./configure
make install
make
'libusb', 'libusb 0.1 compatible api', 'libftdi' were compiled similarly. Advanced hardware whatchpoint wasn't compiled as this is not necessary in my case. None of the patches were applied.

Ideally, when the system connects to the 'Altera Virtual JTAG Tap' (which it does sometimes after 10 - 20 retries) I should be getting a msg saying 'JSP listening on host xxx port 9944' etc. But I don't get any such message despite having JSP enabled. Am I compiling with any missing parameters? Can you see any other issue with the flow?

Many thanks in advance!

Regards,
AR

Hi,

I can answer to the (a) question. It is still under test but I patched OpenOCD to support the Advanced Debug Interface. I works with the Altera Virtual JTAG.

You'll find the patch here http://openrisc.net/forum.html
It is to applied against the OpenOCD version here: http://git.openrisc.net/cgit.cgi/julius/openocd/

It looks more stable than adv_jtag_bridge...

Franck.

Franck, Thanks a lot ... Just to clarify a few points on the order of execution: [1] Patch the usb_blaster_driver with your patch file using linux "patch " command which should enable support for adv_debug_sys (as a `define statement) [2] run ./bootstrap [3] run ./configure with these arguments: '--enable-usb_blaster_libftdi --enable-adv_debug_sys --enable-maintainer-mode' [4] Compile the source code with 'make' and 'make install' [5] Put FPGA JTAG ID in vjtag.cfg (Should be obtainable through quartus. [6] Save the following as a startup.tcl file and use it as a startup script using 'source startup.tcl' [7] Run the compiled binary located in the '\src' folder I'm a linux novice hence require this clarification. Please do reply back. Regards, AR

Attached you'll find the patches to apply. Julius is going to create a new branch (for testing purpose) with patches applied. So for now, this is how to proceed:

git://openrisc.net/julius/openocd
cd openocd
tar xvzf patch.tar.gz
git apply 000* (you'll have some whitespace errors which need to be fixed...)
./bootstrap
./configure --enable-usb_blaster_libftdi --enable-adv_debug_sys --enable-maintainer-mode
make

As you said: Put FPGA JTAG ID in vjtag.cfg

This is my startup script (altera-dev.tcl):

source [find ./tcl/interface/altera-usb-blaster.cfg]
usb_blaster pin6 1
usb_blaster pin8 1
source [find ./tcl/target/vjtag.cfg]

proc fpga_test_init {} {


}

gdb_port 50001

init
#echo "Halting processor"
halt

foreach name [target names] {
set y [$name cget -endian]
set z [$name cget -type]
puts [format "Chip is %s, Endian: %s, type: %s" \
$name $y $z]
}

Then :

sudo ./src/openocd -f ./tcl/interface/altera-usb-blaster.cfg -f altera-dev.tcl

Good luck :)

Franck.

First line of the procedure it is: git clone git://openrisc.net/julius/openocd

To your other two questions:

(b) Are you using the Makefile.classic to compile, or did you used the autotools ./configure process? If the latter, then you'll need to run ./configure again with the right option to enable JSP support in adv_jtag_bridge: --enable-jsp-server=yes. There are two other JSP-related options, make sure you set them to match your configuration (check ./configure --help for the full list).

You'll then need to 'make clean; make' to rebuild adv_jtag_bridge (this is true even if you're using the Makefile.classic). If the JSP support is compiled in, you'll see a message as part of the help (./adv_jtag_bridge -h) that says "Compiled with support for the JTAG Serial Port (JSP)."


Franck, I'm curious what you mean by "more stable"?

Franck,

Thanks a lot for your help and support - finally I'm able to connect and it is a lot more stable than adv_dbg_sys :-)

However, I'm getting the following msg when I try to load linux into SDRAM through GDB:

Output from GDB:
================

(gdb) load
Loading section .text, size 0x11847c lma 0x0
Loading section .rodata, size 0x132d0 lma 0x11a000
Loading section __ksymtab, size 0x2728 lma 0x12d2d0
Loading section __ksymtab_gpl, size 0xf80 lma 0x12f9f8
Loading section __ksymtab_strings, size 0x7323 lma 0x130978
Loading section __param, size 0x90 lma 0x137c9c
Loading section .data, size 0xf5c0 lma 0x138000
Loading section __ex_table, size 0x7f0 lma 0x1475c0
Loading section .head.text, size 0x4000 lma 0x148000
Loading section .init.text, size 0xdb3c lma 0x14c000
Loading section .init.data, size 0x14acc4 lma 0x159b40
Start address 0xc0000000, load size 2762583
Transfer rate: 7 KB/sec, 15696 bytes/write.

Output from openOCD:
====================

Open On-Chip Debugger 0.5.0-dev-gd27b377-dirty (2012-04-12-03:24)
Licensed under GNU GPL v2
For bug reports, read
http://openocd.berlios.de/doc/doxygen/bugs.html
Warn : Adapter driver 'usb_blaster' did not declare which transports it allows; assuming legacy JTAG-only
Info : only one transport option; autoselect 'jtag'
3000 kHz
Warn : Interface already configured, ignoring
3000 kHz
Info : No lowlevel driver configured, will try them all
Info : usb blaster interface using libftdi
Error: Translation from khz to jtag_speed not implemented
Error: Translation from khz to jtag_speed not implemented
Error: Translation from jtag_speed to khz not implemented
Info : adapter-specific clock speed value -1076490936
Info : JTAG tap: or1k.cpu tap/device found: 0x020f70dd (mfg: 0x06e, part: 0x20f7, ver: 0x0)
target state: halted
Chip is or1k.cpu, Endian: big, type: or1k
Info : accepting 'gdb' connection from 50001
Warn : acknowledgment received, but no packet pending
Warn : negative acknowledgment, but no packet pending
Warn : negative acknowledgment, but no packet pending
Warn : negative acknowledgment, but no packet pending
Warn : negative acknowledgment, but no packet pending
Warn : negative acknowledgment, but no packet pending

Does this indicate there's something wrong with the Synthesis / RTL code?

Also, is a there a way to telnet to openOCD to get the linux console ???

Many thanks and I promise not to bother you again ...

Regards,
AR

@Nathan,

Tried your suggestions and this is working wonderfully well now :-)

But I'm getting the following msg:

ERROR! WB bus error during burst write, address 0x0 (index 0x0), retrying!

There's something wrong with my synthesis ... I'm using positive reset for OR1200 (as defined in OR1200_defines.v) and nReset as an input to the orpSoC top module ... I've used the same files used in the DE2-115 port ...

have you come across anything like this before ??? Any hints ???

Many thanks though for your help :-)

Regards,
AR

@afr1g11



That's good to know.




I also have those warnings. It is not related to an RTL problem. It's comming from the OpenOCD port, this need to be fixed. However, it does work anyway....



Not yet, the JSP is not supported by OpenOCD for now.



You don't bother me at all :)

Franck.