Marko, Do you happen to have a link to the patches you applied to openocd? I am building my own, and would like to incorporate them.
Cheers, simon On Thu, May 25, 2017 at 3:39 PM, marko kiiskila <ma...@runtime.io> wrote: > Peter, > > What’s missing from stock 0.10.0 for nrf52 is the capability to write/erase > the flash. That’s why we have our own tarball for openocd here: > http://mynewt.apache.org/latest/os/get_started/cross_tools/ < > http://mynewt.apache.org/latest/os/get_started/cross_tools/> > > I have not tried using openocd with JLink, but we do use it with Arduino > Primo with CMSIS-DAP. For reference, check out the download/debug scripts > for that > https://github.com/apache/incubator-mynewt-core/blob/ > master/hw/bsp/arduino_primo_nrf52/arduino_primo.cfg < > https://github.com/apache/incubator-mynewt-core/blob/ > master/hw/bsp/arduino_primo_nrf52/arduino_primo.cfg> > > OPENOCD_DEBUG setting is not present for all BSP scripts. It does > exist for Arduino Primo; take a look at how it is checked here: > https://github.com/apache/incubator-mynewt-core/blob/ > master/hw/bsp/arduino_primo_nrf52/primo_download.sh < > https://github.com/apache/incubator-mynewt-core/blob/ > master/hw/bsp/arduino_primo_nrf52/primo_download.sh> > If you’re using nrf52dk, you can see from the matching file that it does > not have matching pieces there. > > I *think* I built the tarball with J-Link option present as well, but I > admit > that I have not tried it. If it’s not there, I can put up the sources for > it, > and you can build it in. And share how you built it, so I can do the same. > > If you get it to work, and want to have the openocd option for nrf52dk, > submit a patch! > > Good luck, > M > > > On May 25, 2017, at 3:24 PM, Chris Merck <me...@olibra.io> wrote: > > > > Hi Peter, > > > > I use an ST-LINK V2, which costs $3-$12 depending where you buy online. > > > > Loading and debugging works with the stock nrf52.cfg and stlink-v2.cfg > > files provided by OpenOCD 0.10.0. > > > > Though ST-LINK is not a FOSS hw design, it is cheap and doesn't require > > proprietary drivers. > > > > Regards, > > Chris > > > > On Thu, May 25, 2017 at 3:05 PM, Wayne Keenan <wayne.kee...@gmail.com> > > wrote: > > > >> Hi Peter, > >> > >> I've used OpenOCD on a Pi with just jumper wires to program an nRF51, > for > >> reference: > >> > >> https://chisight.wordpress.com/2016/03/22/openocd-for- > >> programming-nrf51822-via-nothing-but-wires-and-a-raspberry-pi/ > >> > >> https://learn.adafruit.com/programming-microcontrollers- > >> using-openocd-on-raspberry-pi/wiring-and-test > >> > >> > >> > >> Just as a FYI: I was doing the above and also using avrdude + jumper > wires > >> for AVR's at the same time and although I don't recall having many/any > >> issues using OpenOCD I was having to manually 'free' the Pi's GPIO pins > >> occasionally using something like: > >> > >> echo 4 22 23 18 | xargs -n 1 gpio unexport > >> > >> If you run into troubles then maybe try that (check the pins match your > >> OpenOCD config) > >> > >> > >> Regards > >> Wayne > >> > >> On 25 May 2017 at 19:39, Peter Jones <mlists+myn...@devalot.com> wrote: > >> > >>> Is it possible to use newt to flash a nRF52 without having to use > >>> J-Link? > >>> > >>> The getting started guide seems to imply that OpenOCD version 0.10.0 is > >>> an option but I can't seem to figure out how to make `newt load' use > >>> it. I tried setting `OPENOCD_DEBUG=1' without luck: > >>> > >>> $ newt target set nrf52_blinky syscfg=OPENOCD_DEBUG=1 > >>> $ newt load --verbose nrf52_boot > >>> > >>> ...still tries to use J-Link. > >>> > >>> I want to avoid using J-Link because: > >>> > >>> 1) Pragmatic reason: I'm having a hard time getting J-Link to work on > >>> my OS (NixOS) > >>> > >>> 2) Philosophical reason: I'm building a completely open product and > >>> would like to have the entire tool chain open as well. I'm a firm > >>> supporter of Open Source and relying on proprietary software like > >>> J-Link makes me uncomfortable. > >>> > >>> Thanks. > >>> > >>> -- > >>> Peter Jones, Founder, Devalot.com > >>> Defending the honor of good code > >>> > >> > >
