h503 .cfg Files to enable Debug

[LightningStalker]

Is your feature request/improvement related to a problem? Please describe.

STM32H503RB Nucleo
analogWrite(PA5, xxval) is non-return
HAL_DAC_ConfigChannel() (called by analogWrite) is also non-return
Unable to trace further without debugging
No compile error but MCU "lockup"

Describe the solution you'd like

.cfg files and others related to openocd Debug for STM32H5xx

Describe alternatives you've considered

CubeIDE having its own issues

Additional context

.cfg files not found in stm32duino github
DAC config state unknown but doing conversion
HAL functions must be used (except ConfigChannel()
Nice MCU need to try more peripherals

[fpistm]

Hi @LightningStalker

About H5 .cfg, it was not provided because openocd do not support it (yet?).
https://github.com/openocd-org/openocd/tree/master/tcl/target

If you want to debug, you can try using cmake and import the cmake project in STM32CubeIDE which include a customized openocd version.
Or you can change the platform.txt to use the openocd from the STM32CubeIDE and copy the .cfg in the variants/STM32H5xx/ folder.

Anyway this is advanced usage and do not support this, sorry.

Note:
Since Arduino IDE 2.3.x, the debug config has change and have to update it.

[LightningStalker]

After updates to stm32duino core still same behavior with HAL_DAC_ConfigChannel() on H503RB

Test code:

void setup() {
  pinMode(PA5, OUTPUT);

  // initialize serial communication at 9600 bits per second:
  Serial.begin(9600);
  analogReadResolution(12);
}

// the loop routine runs over and over again forever:
void loop() {
  static uint16_t dac_value = 0;

// CALL HAL_DAC_ConfigChannel():
  analogWrite(PA5, dac_value);
  dac_value = (dac_value < 4095) ? (16 + dac_value) : 0;
  // read the input on analog pin 0:
  int sensorValue = analogRead(A0);
  // print out the value you read:
  Serial.println(sensorValue);
  Serial.println(dac_value);
  delay(1);  // delay in between reads for stability
}

After some hours:

[2024-07-24T06:21:03.734Z] SERVER CONSOLE DEBUG: onBackendConnect: gdb-server session connected. You can switch to "DEBUG CONSOLE" to see GDB interactions.
/opt/st/stm32cubeide_1.15.1/plugins/com.st.stm32cube.ide.mcu.externaltools.openocd.linux64_2.3.100.202312181736/tools/bin/openocd -c "gdb_port 50000" -c "tcl_port 50001" -c "telnet_port 50002" -s /home/robert/Arduino/h503_DAC -f /tmp/.mount_arduintiYITR/resources/app/plugins/cortex-debug/extension/support/openocd-helpers.tcl -f /home/robert/.arduino15/packages/STMicroelectronics/tools/xpack-openocd/0.12.0-3/openocd/scripts/interface/stlink.cfg -f /home/robert/.arduino15/packages/STMicroelectronics/hardware/stm32/2.8.1/debugger/select_hla.cfg -f /home/robert/.arduino15/packages/STMicroelectronics/tools/xpack-openocd/0.12.0-3/openocd/scripts/target/stm32h5x.cfg
Open On-Chip Debugger 0.12.0-00029-gf77e7cb03 (2023-12-15-13:24) [https://github.com/STMicroelectronics/OpenOCD]
Licensed under GNU GPL v2
For bug reports, read
        http://openocd.org/doc/doxygen/bugs.html
CDRTOSConfigure
hla_swd
Error: The selected transport doesn't support this target
/home/robert/.arduino15/packages/STMicroelectronics/tools/xpack-openocd/0.12.0-3/openocd/scripts/target/stm32h5x.cfg:77: Error: 
in procedure 'script' 
at file "embedded:startup.tcl", line 28
at file "/home/robert/.arduino15/packages/STMicroelectronics/tools/xpack-openocd/0.12.0-3/openocd/scripts/target/stm32h5x.cfg", line 77
[2024-07-24T06:21:03.757Z] SERVER CONSOLE DEBUG: onBackendConnect: gdb-server session closed
GDB server session ended. This terminal will be reused, waiting for next session to start...

[LightningStalker]

stm32h5xx_hal_dac.c line 1355:

  if ((*(tmp_package) & 0x1FUL) == 0x0FUL)

Should evaluate FALSE but is instead TRUE bug maybe?

[fpistm]

This topic was for debug config ....

This line is part of the Cube H5 HAL.
If you think there is a bug on this line open an issue on the GitHub STMicroelectronics in the CubeH5 repository.

[LightningStalker]

Not sure if it is HAL bug or conflict between core setting.
Memory address test is new for me! Abstraction normally.

[fpistm]

Hi @LightningStalker

I've tested and met the issue too.
This is due to the ICACHE which is enabled preventing reading the PACKAGE version.
More info here:
https://community.st.com/t5/stm32-mcus/how-to-avoid-a-hardfault-when-icache-is-enabled-on-the-stm32h5/ta-p/630085

You can disable it adding a file named hal_conf_extra.h with:

#define HAL_ICACHE_MODULE_DISABLED