Add native SWD (ARM Serial Wire Debugging) support

[agatti]

A script-driven approach has been tried here: http://www.willdonnelly.net/blog/bus-pirate-serial-wire/ . Something worthwhile would be to reimplement this natively without the need for bit-banging data lines.

[agatti]

A very simple initial implementation for synchronous SWD communication can be found here: https://gist.github.com/agatti/a92574bf726706f77fa26b9fc4934ee2 although I do not have the appropriate hardware to try it out at the moment...

[u1f35c]

Is there a corresponding patch to OpenOCD or similar? I have STM32F103 hardware I could test against (I'm currently using OpenOCD with bitbanging on the Bus Pirate for SWD support).

[agatti]

Unfortunately said implementation was more of a proof of concept to see how difficult would be adding SWD support to the BP. I believe that adding SWD transport to OpenOCD's BP driver shouldn't be too difficult once SWD support is in place, though.

Right now I'm a bit swamped with work and other things so I cannot really dedicate lots of time to BP firmware development this month, but I'd love to see somebody picking up my initial implementation from #29 (comment) and make it work :)

[u1f35c]

I haven't had any time to look at the SWD implementation on the BusPirate itself, but I have been using the patch at https://the.earth.li/~noodles/bp-swd.diff with OpenOCD 0.10.0 to enable SWD via bitbanging using the BusPirate. This is a slightly modified version of the diff at http://openocd.zylin.com/#/c/2444/ and has been working fine for me with an STM32F103 device.

[ChlorideCull]

I'll see if I can get on-board SWD support up and running once I get my Bus Pirate v4, I assume it will take a while to get here because of the Chinese New Year.

[ChlorideCull]

I was reminded by this issue by the board finally showing up in the mail. Needless to say, it's in rough shape after spending six months in the postal system, so I don't think I'll be able to do this any time soon.