to close the loop the issue is solved by using 'stm32l4x option_load' command this was confirmed in IRC ( https://libera.irclog.whitequark.org/openocd/2021-09-02#30779652) according to that, a hint is added to the doc in the following change : https://review.openocd.org/c/openocd/+/6533
Le mer. 1 sept. 2021 à 23:08, Tarek BOCHKATI <tarekbouchk...@gmail.com> a écrit : > Well, it seems like I have read your mail very quickly ! > > mainly locking and unlocking works well too, you just need to perform an > OBL Launch > > $ telnet localhost 4444 > ... > > flash probe 0 > device idcode = 0x10016497 (STM32WLEx - Rev 'unknown' : 0x1001 - Flash > single-bank) > RDP level 0 (0xAA) > flash size = 256kbytes > flash mode : single-bank > flash 'stm32l4x' found at 0x08000000 > > > > halt > target halted due to debug-request, current mode: Thread > xPSR: 0x41000000 pc: 0x1fff26d2 msp: 0x200014f0 > > > stm32l4x lock 0 > > > stm32l4x option_load 0 > Fail reading CTRL/STAT register. Force reconnect > stlink_dap_op_connect(reconnect) > SWD DPIDR 0x6ba02477 > Failed to write memory at 0xe00fffec > stm32l4x option load completed. Power-on reset might be required > stm32wlx.cpu: external reset detected > > > flash probe 0 > device idcode = 0x10016497 (STM32WLEx - Rev 'unknown' : 0x1001 - Flash > single-bank) > RDP level 1 (0x00) > flash size = 256kbytes > flash mode : single-bank > flash 'stm32l4x' found at 0x08000000 > > > > does this solve the issue ? > > BTW, the missing revision id should not harm any functionality in openocd. > > > > Le mer. 1 sept. 2021 à 22:40, Tarek BOCHKATI <tarekbouchk...@gmail.com> a > écrit : > >> Hi Jeremy, >> >> Thank you for reporting the issue. I'm happy you are enjoying STM32 >> support within OpenOCD :) >> >> Well, I have tried the following steps on the current master branch >> ref: a098816a6557e5882bf088ab12a06b94934f30ce >> >> $ telnet localhost 4444 >> ... >> >> > flash probe 0 >> device idcode = 0x10016497 (STM32WLEx/WL5x - Rev 'unknown' : 0x1001) >> RDP level 0 (0xAA) >> flash size = 256kbytes >> flash mode : single-bank >> flash 'stm32l4x' found at 0x08000000 >> >> >> > mdw 0x8000000 8 >> 0x08000000: ffffffff ffffffff ffffffff ffffffff ffffffff ffffffff >> ffffffff ffffffff >> >> >> > program /media/sf_vm-share/elfs/WL55JC-NUC/GPIO_IOToggle.axf >> Unable to match requested speed 500 kHz, using 200 kHz >> Unable to match requested speed 500 kHz, using 200 kHz >> target halted due to debug-request, current mode: Thread >> xPSR: 0x01000000 pc: 0xfffffffe msp: 0xfffffffc >> ** Programming Started ** >> Adding extra erase range, 0x080011f0 .. 0x080017ff >> ** Programming Finished ** >> > mdw 0x8000000 8 >> 0x08000000: 20000428 0800014d 08001009 08001001 08001005 0800015b >> 0800015d 00000000 >> >> >> The programming seems to work just fine. >> >> I have tried the same steps with the ref you specified >> (3f1c15d2a718c9d417c859172f2b1736a769d822), and I have got the same good >> results. >> >> In order to investigate the issue you are facing, could you please share >> with us the debug log (with -d3) >> also I would be interested in having your option bytes configuration. >> >> >> Le mer. 1 sept. 2021 à 21:25, Jeremy Willden < >> jer...@constellationlabs.com> a écrit : >> >>> OpenOCD Devs (and Tarek), >>> >>> I posed this question in the OpenOCD IRC channel and Oleksij Rempel >>> suggested I ask you about this issue. I reviewed the Gerrit changes that >>> are currently in process and perhaps this fix is already in the works but >>> not yet merged into the main code base. If so, please let me know and I'll >>> see if I can manually merge those changes in the meantime. >>> >>> I'm currently using commit 3f1c15d2a718c9d417c859172f2b1736a769d822 from >>> the master branch of OpenOCD with an STM32WLxx family chip (in a LoRa-E5-HF >>> module form SeeedStudio) but I cannot successfully unlock nor lock the >>> flash. Based on logic analyzer captures, writes to FLASH_OPTKEYR and >>> FLASH_KEYR are happening correctly, but there is not a corresponding write >>> to the FLASH_CR register (just 8 reads), which is necessary to lock or >>> unlock the flash. >>> >>> I can probably pull together an interim solution if you can point me to >>> the general commits that have already fixed this issue. If this problem is >>> not expected, I'm equipped to gather better information and assist in >>> fixing it. >>> >>> Please let me know if I can help and if there is any assistance you can >>> offer. I am assisting a client to put products into their manufacturing >>> supply chain in the next few weeks and using Windows in their production >>> environment is something we really want to avoid, so OpenOCD is a great >>> fit. We already use OpenOCD for programming other STMicro parts and are >>> very happy with it. >>> >>> Thank You >>> Jeremy Willden >>> Constellation Labs, LLC >>> direct: 801-877-1101 >>> main: 801-877-1132 >>> mobile: 801-318-8531 >>> http://constellationlabs.com >>> HAM callsign: KG7JW (formerly KG7ELA) >>> >>
