15:55 <Willmish> Hey, i am currently working on a project on a QuickFeather board with EOS S3 SoC (which has an ARM Cortex M4-F). i am trying to reflash its bootloader with OpenOCD (as the one it came with is broken), however when I connect to the gdb server and issue `monitor program` command it fails, and it looks like only part of the flash commands are

15:55 <Willmish> available (flash bank, flash banks, flash init, flash list) and none of the other writing commands are available. I believe this is due to an incomplete target/eos_s3.cfg definition (https://sourceforge.net/p/openocd/code/ci/master/tree/tcl/target/eos_s3.cfg) . Is anyone able o point me in the right direction ? I am relatively new to Openocd so

15:55 <Willmish> feels like im falling deeper and deeper in the rabbit hole hah

15:56 <PaulFertser> Willmish: hi

15:57 <PaulFertser> Willmish: OpenOCD is lacking the flash driver for that chip.

15:58 <PaulFertser> Willmish: but there's a patch to try: https://review.openocd.org/c/openocd/+/6956

15:58 <Willmish> hm, is this something that is possible for someone like me to add? (I only have a STLINK v2 with me, so would need to probably buy JLINK somehow)

15:58 <Willmish> ooh

15:58 <PaulFertser> Willmish: no reason it shouldn't work now, and also you can take over the patch and finally make upstream accept it by rebasing it on current master, testing and nagging the maintainers :)

15:59 <PaulFertser> No need for jlink, no.

16:00 <Willmish> PaulFertser: oh that is super good news, definitely a much better direction to go towards than trying to compare configs with other boards and reading what every line with "flash" related word is

16:00 <PaulFertser> Willmish: you're on the right IRC channel now :)

16:02 <Willmish> Ok, will grab some dinner and try and rebase the patch with most recent version then, wish me luck :D

16:02 <PaulFertser> Willmish: sure I do :)

16:02 <PaulFertser> Willmish: ping me if you get stuck

16:03 <Willmish> will do! really glad I decied to find the IRC before spending 3 more hours going in the wrong direction!

20:16 <maribu[m]> Sorry for nagging again about https://review.openocd.org/c/openocd/+/5584 - is there anything I could do to push that patch forward? I really would love to cross that one in my todo list :)

20:54 <Willmish> PaulFertser: pushing this forward, got the project built and fixed main merge conflicts, now it seems that the function for communication with flash over SPI is buggy, as on probing the SoC its not returning a proper Flash device ID (getting an error: Error: Unknown flash device (ID 0x00000000) ). Have some reference code for this SPI probe from

20:54 <Willmish> quicklogic's FreeRTOS fork (https://github.com/QuickLogic-Corp/qorc-sdk/blob/6968338b8778cb129ae441f6944444c9a1298390/Libraries/SysFlash/src/spi_flash.c#L1048 vs https://review.openocd.org/c/openocd/+/6956/1/src/flash/nor/eoss3.c#294 ), so hopefully this will prove enough for me to debug it. In the meantime, can I push these changes somehow to a

20:54 <Willmish> draft PR? Haven't worked with Gerrit before so unsure.

21:00 <PaulFertser> Willmish: I suggest you get it working first at least to some degree

21:01 <PaulFertser> Willmish: then just --amend the commit keeping the Change-Id intact, and push it to Gerrit, it'll be created as a new changset.

21:01 <Willmish> Cool, sounds good

22:29 <borneoa__> maribu[m]: queued in my "next" branch. If no other reviews, I will merge it in two weeks