Menu
▾
▴
#41 Debugging with J-Link
Hello,
I have some problems in debugging with J-Link. Sometime it works and sometime not.
The first question: is there a manual for the J_link GDB plug-in available? I would be interested to try some settings to make the system more stable. What for example can I enter in the "Type" field which is two time available in the the startup page?
The next question: What is the best place to install a script file for the J_Link device. I would like to have a delay of 100ms between Jtag reset and start of transfers. This is perfect working at my other recent debugging (RDI) solution.
Best regards Erik
please refer to the pages in the project web; if you have questions not answered there, just let me know and I update the.
this is explained in the SEGGER J-Link manual
you can add any commands either before or after loading the elf, you have two large windows in the interface
generally if you need to insert a delay, there is something wrong.
if you are debugging complex cores, like Cortex-A, with multi-stage bootloaders, sometimes you can cheat with delays, but they are highly unreliable.
Hello, here the log for starting a debug session:
For me it seems ok. But the what I see the first in the debug perspective
seems a bit confuse to me. The PC is set correct but I get a empty window
"no sorce available for 0x0". Normally I would await no to see a little arrow
pointing to the "Reset_Handler" label.(see attachment)
SEGGER J-Link GDB Server V4.80g Command Line Version
JLinkARM.dll V4.80g (DLL compiled Feb 13 2014 20:50:02)
-----GDB Server start settings-----
GDBInit file: none
GDB Server Listening port: 2331
SWO raw output listening port: 2332
Terminal I/O port: 2333
Accept remote connection: localhost only
Generate logfile: off
Verify download: on
Init regs on start: on
Silent mode: off
Single run mode: on
------J-Link related settings------
J-Link script: none
Target interface: JTAG
Host interface: USB
Target endian: little
Target interface speed: 0kHz
Connecting to J-Link...
J-Link is connected.
Firmware: J-Link ARM V8 compiled Nov 25 2013 19:20:08
Hardware: V8.00
SAM-ICE found !
S/N: 28013812
OEM: SAM-ICE
Feature(s): RDI, GDB
Checking target voltage...
Target voltage: 3.35 V
Listening on TCP/IP port 2331
Connecting to target...
J-Link found 1 JTAG device, Total IRLen = 4
JTAG ID: 0x0792603F (ARM9)
Connected to target
Waiting for GDB connection...Connected to 127.0.0.1
Reading all registers
Read 4 bytes @ address 0x00000000 (Data = 0xE59FF060)
Target interface speed set to 30 kHz
Select auto target interface speed (8000 kHz)
Flash breakpoints disabled
Downloading 776 bytes @ address 0x00300000 - Verified OK
Downloading 1068 bytes @ address 0x00300308 - Verified OK
Writing register (PC = 0x00300284)
Writing register (CPSR = 0x000000f3)
Read 4 bytes @ address 0x00300284 (Data = 0x4A074906)
Read 2 bytes @ address 0x00300286 (Data = 0x4A07)
Read 2 bytes @ address 0x00300288 (Data = 0x4B07)
Read 2 bytes @ address 0x0030029A (Data = 0xF000)
Read 4 bytes @ address 0x00300280 (Data = 0xEAFFFFFE)
Read 4 bytes @ address 0x00300280 (Data = 0xEAFFFFFE)
Read 4 bytes @ address 0x00300280 (Data = 0xEAFFFFFE)
Read 4 bytes @ address 0x00300280 (Data = 0xEAFFFFFE)
Read 4 bytes @ address 0x00300280 (Data = 0xEAFFFFFE)
Read 4 bytes @ address 0x00300280 (Data = 0xEAFFFFFE)
I'm not used to your configuration, are you trying to run something from RAM? If so, you need to disable the pre-run reset.
did you fix the problem?
More or less. The little problem what I have found was the following. In the startup tab of the plugin is a continue checkbox. If this checkbox is unchecked the image will be load in the target and the 0x0 " no source available" windows is poping up. In this moment the state seems not corect. But if I press only one time F6(step) I see a correct picture with the little arrow at the correct position. With checked continue checkbox this very first incorrect state is invisible. I now can good live with this behaivor.
sorry, but I can not follow you and can not reproduce the problem.
debugging arm9 cores is tricky, and, since you probably run in ram, mandates to disable the secondary reset.
What I have done is the first time. Manually reset the board and worked with none of the both resets. This was working fine. With the first reset set it was not working. Then I wrote a script which overwrite the initTarget procedure of jlink. Simple manually reset the both lines NRST and NTRST for 10ms. So I work wiht this script and none of the resets and be happy.
ok, glad you found a solution.
is there any open issue? or we can close this ticket?
OK,close it.
I forgot to tell you, the new beta version sets the CWD to the project folder, so if you put a custom script in the project, you should be able to point gdb to it with a portable relative path.
what do you mean with this? What I do is set the path under Debugger Tab
-> Other Options -> -s -scriptfile C:/Scripts/myJLinkScript
This path is not a gdb path but a jlink path. Do you mean I could save
the absolute path in any way. That would be nice.
Am 18.03.2014 13:32, schrieb Liviu Ionescu (ilg):
Related
Support requests:
#41if the script is located in the project, you can specify it with a relative name.
if the script is located outside Eclipse, define a substitution variable to the path and specify the relative name to it.
these aspects are very important if you set the debug configuration to be stored in the project, and you want to share the project with someone else.
Erik, please do not reply to the email notifications, use the web page