Hi Paul (and everyone who come across this because they have the same problem as me).
"stm32f1x mass_erase 0" works. I also have to do a "stm32f1x unlock 0" This is my procedure: "init; halt; wait_halt; stm32f1x mass_erase 0; stm32f1x unlock 0; shutdown" For the casual reader who come across this by searching: Background: I used to just do a "program $file_to_flash; shutdown", and relied on the "auto erase" feature in openocd. This worked fine with the older boards. It seems, however, that the manufacturers of the boards found a quicker way to flash-program the boards, and this now requires a mass_erase. -However, you only need to 'mass_erase : unlock' once. After that, you only need to use 'program $file_to_flash' each time. @Paul: If I do a "stm32f1x mass_erase 0" twice in a row (or more than twice), will it check if the chip is "already erased" or will it "slowly wear down the flash memory" ? -Thank you so much for the hint. I now suddenly have 4 more working boards and can make some automation for my brother. :) (one board out of 5 failed boards is failing mass-erase, this is a different error and is probably a problem with the board itself) Lova Jens On Thu, 22 Sep 2016 21:27:06 +0300, Paul Fertser wrote: > Hi Jens :) > > On Thu, Sep 22, 2016 at 08:20:27PM +0200, Jens Bauer wrote: >> Before I start sending log-files, I'd like to ask if anyone else has >> seen these problems ? > > Probably you just need "stm32f1x mass_erase 0" to fully unlock it? > > Yes, you should have sent the log along with this mail ;) > > -- > Be free, use free (http://www.gnu.org/philosophy/free-sw.html) software! > mailto:fercer...@gmail.com ------------------------------------------------------------------------------ _______________________________________________ OpenOCD-devel mailing list OpenOCD-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/openocd-devel
