SEGGER Flasher ARM standalone fails on STM32H743 (ERR255), but J-Flash programs

[mrthanhapple]

Hi all,

I am using a SEGGER Flasher ARM with an STM32H743 target.

Programming from PC with J-Flash works normally.
Downloading the image/config into the Flasher also works.

But when I use the Flasher in standalone mode and press PROG, it fails with red LED.

Log shows:
STATUS: INITIALIZING
STATUS: CONNECTING
STATUS: ERASING
ERR255: Error while flashing

So J-Flash can program the same target, but standalone mode fails.

Has anyone seen this before?
What should I check first: VTref, SWD speed, reset strategy, connect under reset, erase method, or old Flasher firmware?

[hans]

From what I can tell, the support for the Flasher software is a distinct difference between the J-Link BASE en PRO models.

Also 2022 sounds pretty old firmware, if that screenshot is accurate (F427 vs H743?). Perhaps you can upgrade it for the newer version support, assuming its a PRO dongle?

Otherwise contact Segger... Non-EDU segger dongles have a steep price because they include a support team, trade-in programs, etc.etc

[mrthanhapple]

From what I can tell, the support for the Flasher software is a distinct difference between the J-Link BASE en PRO models.

Also 2022 sounds pretty old firmware, if that screenshot is accurate (F427 vs H743?). Perhaps you can upgrade it for the newer version support, assuming its a PRO dongle?

Otherwise contact Segger... Non-EDU segger dongles have a steep price because they include a support team, trade-in programs, etc.etc

Thanks. The target I am trying to program is STM32H743.

I also work with STM32F427 in another project, so there may be some confusion from my screenshots/logs. But for this issue, the actual standalone flashing target is H743.

J-Flash can program the H743 correctly from PC.
The failure only happens when the image is downloaded into the Flasher and I use standalone PROG mode.

I will check firmware version and Flasher software version again.

[Pseudobyte]

what options are you using for the standalone mode? Specifically those listed under "Actions performed by 'Production Programming'"

[mrthanhapple]

what options are you using for the standalone mode? Specifically those listed under "Actions performed by 'Production Programming'"

Current standalone settings under "Actions performed by Production Programming" are:

- Erase: enabled
- Program: enabled
- Verify: enabled
- Secure: disabled
- Start Application: enabled

I also tested with Start Application disabled, but the standalone flashing still fails with ERR255.
J-Flash programming from PC still works correctly.

[Pseudobyte]

I feel like you're bound to run into issues trying to talk to the micro at 50MHz. The default is 4MHz, you should probably start there.

Also can't say I'm too familiar with u-flash either.

[mrthanhapple]

I feel like you're bound to run into issues trying to talk to the micro at 50MHz. The default is 4MHz, you should probably start there.

Also can't say I'm too familiar with u-flash either.

Thanks. I already tested a wide range of target interface speeds in standalone mode, from a few kHz up to 1000 kHz and 4000 kHz.

Unfortunately, the behavior is unchanged:
- J-Flash from PC works correctly
- standalone flashing via the Flasher still fails with ERR255

So it seems the issue is not only related to interface speed.

Since this is a Flasher ARM V4.2 and J-Flash also reports:
"Flasher V4 or earlier detected. Storing *.pex files in the config file requires a Flasher V5 or newer."
I am wondering if this is a limitation of the Flasher V4 standalone mode for this device/project.

[Pseudobyte]

Since this is a Flasher ARM V4.2 and J-Flash also reports:
"Flasher V4 or earlier detected. Storing *.pex files in the config file requires a Flasher V5 or newer."
I am wondering if this is a limitation of the Flasher V4 standalone mode for this device/project.

That is certainly a possibility. I only have used Flasher Portable Plus's & Flasher ATE's.

I have had an issue with firmware not starting after a device was flashed in standalone mode, but that was due to a bug on segger's side where they were not restoring the registers correctly. Your issue however does not sound like this.