IBM Application Runtimes Come for answers. Stay for best practices. All we’re missing is you. Join / Log in Ask a question
Hi,We can't answer your question.The processor is not enough, we need to know what compiler version, debugger version, simulator/emulator, what is available to execute on the target (load and execute on the target) in your target environment.
However the first question consist to check the feasibility.You can open a ticket against IBM support to check the feasibility first and can also guide you to an existing TDP, with the latest release of Test RealTime you have a list of TDP you can use as base to my your own TDP.Regards,Yanis
For arm-none-eabi compiler, there are cGnuSTM32F411.xdp or cSTM32GnuOpenOcd.xdp in the latest delivery.
Do you have a way to get the buffer and transfer to the host
• Is there an OS on the target?
• Do you have socket?
below, by which method or option is used for Execution (execute on target/write buffer/get the buffer/transfer to host)
• Debugger can be started with macro/script on the command line as parameter
• File system is available on the target and can be used to write dynamic results.
• a TCP/IP socket is available to connect file server on the host machine
The most tricky part is the execution, there are basically 3 methods possible for the Data retrieval:
1. Standard means the target uses file system fopen/fwrite and the file will be moved at the end of the execution, if required.
2. User Mode means the target uses "custom code" to open the connection to the host, and write result data to an host machine.
one of the most often used is the TCP/IP socket connection for which TestRT provide and automatically launch its socket listener as explained in the TDP doc (SOCKET_UPLOAD). RS232 is also used.
3. BreakPoint means the target does nothing to open/write data.
This mode is used in conjunction to the debugger.
That means the application is stopped each time the ASCII buffer is full and a debugger macro will dump this buffer and restart the execution.
This mode has few impact on the execution behaviour because the result data are uploaded at the end of the application only (except for runtime trace).
And the answers to all of the questions below must be YES
• Does the debugger/simulator provide access to symbols or linker provide map file giving routines' addresses?
• Is there a command/script language?
• Is there a way to run commands from a script file?
• Can the command/script file be executed automatically when the debugger starts
• Is there a command to stop the debugger? (The execution process must be blocked until execution is terminated and the trace file is generated.)
• Is there a way to set software breakpoints?
• Is there a way to check the breakpoint/address reached?
• Is there a way to dump the contents of a variable in any format, or to dump a memory buffer and log the value?
For an example, there is the TDP clinuxgnu3breakpoint.xdp located in the xdp folder of Test RealTime delivery.