Skip to main content
MMax.3
MMax.3
Associate
January 20, 2022
Solved

Why does Stm32CubeIDE give me "Break at address 0x1ff0b4cc with no debug information available" (right after downloading code into chip)?

  • January 20, 2022
  • 2 replies
  • 4083 views

Chip: STM32H750VBT6 (custom pcb)

BOOT0: connected to GND with a 10k resistor.

Debugger: ST-LINK v2 USB Dongle (updated to latest firmware by CubeIDE)

CubeMX: 6.4.0 (updated today to latest)

CubeIDE: 1.8.0 (11526_20211125_0815, updated today to latest)

Verified the following (maybe overkill to mention):

  • Active build configuration is "Debug"
  • Interface: swd
  • Probe: ST-LINK (ST-LINK GDB server)
  • CubeMX: Serial Wire Debugging

I've also tried a Mass Erase with CubeProgrammer, but it had no effect.

The weird part is that CubeIDE gives me this error, but TrueStudio just works fine.

This error occurs immediately after downloading the code into the chip.

(The first instruction auto-breakpoint also isn't triggered).

What am I missing / doing wrong?

This topic has been closed for replies.
Best answer by Tesla DeLorean

Well it definitely off in ROM / SYSTEM LOADER space

BOOT0 being low should stop this, however on the H7 as reset is not the same thing as a power cycle, and the latched state of BOOT0 will persist across a reset, and the SYSTEM LOADER will run if the FLASH is blanked.

The L4's had some errata about a setting in Option Bytes, I don't think the H7's had the same thing, but review the bytes.

2 replies

Tesla DeLorean
Tesla DeLoreanBest answer
Guru
January 20, 2022

Well it definitely off in ROM / SYSTEM LOADER space

BOOT0 being low should stop this, however on the H7 as reset is not the same thing as a power cycle, and the latched state of BOOT0 will persist across a reset, and the SYSTEM LOADER will run if the FLASH is blanked.

The L4's had some errata about a setting in Option Bytes, I don't think the H7's had the same thing, but review the bytes.

Tips, Buy me a coffee, or three.. PayPal Venmo (See Profile) Up vote any posts that you find helpful, it shows what's working..
MMax.3
MMax.3Author
Associate
January 20, 2022

Hi Tesla DeLorean,

Thanks for the advice!

I've checked the option bytes:

Cube Programmer:

Revision ID = Rev V

OB-Boot addresses (boot areas):

BOOT_CM7_ADD0 [0x800] (0x08000000) when BOOT0=0

BOOT_CM7_ADD1 [0x1ff0] (0x1ff00000) when BOOT0=1

Errata sheet:

STM32H750xB V 0x2003

Not sure for what to look for in here.

I think I know why I had a '0x1ff0b4cc'-address.

When it didn't enter correctly in debug mode I started tinkering with BOOT0 (high/low), and prob. just resetting I think.

So now I powered off and on the board completely.

It's low now, but the problem still exists.

I see that TrueStudio runs at 0x080002de and waits on a breakpoint (first line of main).

CubeIDE also flashes 0x08**** address (seems correct now), then to 0xb*****? and then stops at 0xf3af4804.

The message is now: 'Break at address "0xf3af4804" with no debug information available'.

This is what I did next (start from scratch with minimal setup):

- Create Empty project STM32H750VBTx

- Setup Debug: Serial Wire

- Toolchain: STM32CubeIDE

- Save

Then copied the whole folder.

For the 2nd version I only changed the Toolchain to TrueStudio.

Tried to run CubeIDE version, it worked (and so did the TrueStudio version).

So it must be somewhere in my CubeMX settings, firmware version or code I guess.

I'll just keep going from there and I'll post an update if I find something that could have caused it (in case someone else stumbles on the same problem).

Terms & ConditionsAccessibility statement