Search Results

I'm using a J-Link Base with Eclipse trying to load/debug a Kinetis K64F board. Mostly it works extremely well, however... It very often fails to program the device and I get the error dialog below. Sometimes I have to power-cycle the device and retry a dozen times, other times it just works. Any hints as to what might be causing this intermittent behavior? Its very painful! Thanks in advance, Best Regards, Dave PS: Windows 10, JLink_Windows_V722b driver/gdb/firmware package installed.

Hi Fabian - This problem seems to have been caused by a power supply problem (dubious 3.3v) on the prototype board I received (fabrication error, dope-slap to the assembly house). Another board works AOK. Amazing that the problem was reproducible many times! Apologies for wasting your time with this. Thanks as always for the help, Best Regards, Dave

Thanks Fabian - Log file is attached. This is a custom board, probably our 6th design with K64F. Haven't seen this problem with earlier efforts - not sure what's changed. Thanks, Best Regards, Dave

Under debugger, my program stops at breakpoint - but there is no breakpoint at or near the location it stops. Running program not under debugger works AOK. Always stops at the same spot, every second or two. Picture attached shows the stopped debugger. Jlink Base running 6.80d, NXP MCUXpresso, Kinetis K64F I'm using debug console for printf output which works fine. Thanks in advance for any ideas or advice, Best Regards, Dave forum.segger.com/index.php/Attachment/4325/

Apologies, I had not found segger.com/products/debug-prob…xpresso-ide-free-edition/ and created the launch configuration the normal Eclipse way, which fails horribly. Following your instructions, everything works fine!

Just upgraded to NXP MCUXpresso 11.1 No problems using JLINK with earlier MCUXpresso version. Target is Kinetis K64F (Freedom board). JFlash lite can see and erase the device, so connections etc.. would seem to be AOK. When I try to debug, a message flashes by saying The selected device -IF is unknown to this selection Image of message attached and here: nadler.com/backups/JLINK_unknown_message.PNG Any idea what's going wrong??? Thanks for any help! Best Regards, Dave

PPS: Mysteriously, it started accepting the CPU selection. I set the correct CPU clock speed. Now the choices for SWO baud rate are strange - can't enter 60 (for 60000 as SWO is setup). And cannot connect... Jlink base, STM32F437VI In Jlink SWO viewer, CPU window is blank. Selecting correct (connected, operating, debugable) CPU does not do anything, window remains blank. How do I get this to work? Thanks! Best Regards, Dave PS: Version 6.70, Windows 10 64-bit, USB-connected. Debugging via STMCub…

Update: If I create a NEW debug configuration from Segger (not STM), I see the "connect to target" option. However, even though I explicitly filled in the path C:\Program Files (x86)\SEGGER\JLink\JLinkGDBServerCL.exe, I get an error "Error while launching command: gdb --version" when I try to debug. The project and ELF settings all look correct, and match the working non-Segger STM debug configuration. Any ideas??? Original post: I'm using J-Link base to debug an STM32F413CG over SWD, in the STM…

Background: Atollic 9.1, F429 Nucleo with Segger debug stub loaded. We'll be using Segger J-Link when production boards show up. CubeMX-generated FreeRTOS application. 6.34h Segger software installed. Trying to debug an LwIP problem, but breakpoint does not operate properly. When the breakpoint is hit inside an ISR (and FreeRTOS idle, no active task), the breakpoint fails to operate properly. The application stops, and a debugging failure is reported in the console window (shown below). RTOS-awa…

Apologies Nino if I wasn't clear. Connection is AOK, but the error happens on starting debug. I've attached console output below in case that's helpful. Thanks again for the help, Best Regards, Dave Source Code (134 lines)

Hi - I'm developing on a STM Nucleo-F413ZH Nucleo-144 Discovery board using STM/Atollic Eclipse. Because Segger doesn't provide a cable to connect my JLink to the debug header STM puts on these boards (please make and sell these!), I'm using the Segger-provided software in the onboard debugger. On this board (but not other boards I use), when I start the debugger it fails in the startup script, and pauses the debugger someplace related to the previous run. How can I get debug launch to work prop…

Is emFile suitable for: - nor SQI flash 64Mb - use under FreeRTOS (customer requirement) Is a hook provided to flush any cache for use entering power-down? Or is it expected that external hardware resets the processor while giving the flash time to finish up any operations in progress? Thanks! Best Regards, Dave

OK, problem is solved by adjusting FreeRTOS configuration; no idea why this is required: #define configUSE_TRACE_FACILITY 1 // without this, FreeRTOS-aware debug task list can miss tasks and show bogus stack overflows

Hi Guys - I'm not sure if this is a Segger problem, but maybe you've seen this before... I'm using MCUXpresso from NXP on a Kinetis K64 (Cortex-M4F). FreeRTOS 9.0.0 and ARM GCC tools of 1Q2017. #define configUSE_PORT_OPTIMISED_TASK_SELECTION 0 // Note: 1 causes FreeRTOS Eclipse plug-in to crash... #define configTASK_RETURN_ADDRESS 0 // place 0 task return address on stack to help FreeRTOS-aware debugger (GDB unwind thread stack) #define configCHECK_FOR_STACK_OVERFLOW 2 // FreeRTOS vTaskSwitchCon…

forum.segger.com/index.php/Attachment/1472/Related to previous post: [SOLVED] JLink stack trace 0xA5A5A5A4 Freescale K64F, KDS 3.2, FreeRTOS-aware debugging on Windows 8.1, JLink_Windows_V612a Thanks to Arne I found and reset a well-hidden size-optimization setting, and now most FreeRTOS tasks show accurate call stacks. Unfortunately "BT input" task shows incomplete call stack. Same project as Arne received last week; just start it and pause. Any ideas? Thanks as always, Best Regards, Dave

Thanks Arne, sorry I missed that optimization setting... Works fine now! Thanks again, Best Regards, Dave

Thanks Arne - I can send you the project; how do I do send this so it is not public? Thanks, Best Regards, Dave

Hi Arne - I'm also using FreeRTOS 9.0.0 (with no changes). IIRC other tasks using xQueueGenericReceive show normal stack trace. I've attached the FreeRTOSConfig.h in case that is helpful. Thanks, Best Regards, Dave PS: Is 0xA5A5A4 a known special fill sequence?

It's a debug build Arne - no optimization. Any ideas?

forum.segger.com/index.php/Attachment/1436/Freescale K64F, KDS 3.2, FreeRTOS-aware debugging on Windows 8.1, JLink_Windows_V612a: No values are displayed for the peripheral registers (see attached image). Any idea what is going on? Thanks! Best Regards, Dave