Adam Greig

This came up again in #1157. The issue is that the jlink driver error about not supporting multidrop: https://github.com/probe-rs/probe-rs/blob/5a7f872aa4dde49db6a9df92f606e6db0ff49388/probe-rs/src/probe/jlink/arm.rs#L933-L935 is being masked by this `map_err` higher up: https://github.com/probe-rs/probe-rs/blob/5a7f872aa4dde49db6a9df92f606e6db0ff49388/probe-rs/src/architecture/arm/communication_interface.rs#L404-L409

(cc https://github.com/probe-rs/probe-rs/issues/995) The main problem with fixing this is that hidapi doesn't expose interface strings, only top-level product/vendor/serial strings, making it hard to check properly. rusb does expose these strings...

It also looks like pyOCD whitelists a bunch of device names and VID/PID pairs that are known to be cmsis-dap devices, which I guess we could also do... https://github.com/pyocd/pyOCD/pull/1399/files

Does it matter that the targets only have a single `Flash` entry at 08000000, even though they have two algorithms for the two ranges?

I _think_ the device has one flash region, but it is aliased at both 0x0800_0000 and 0x0C00_0000, with one alias for nonsecure access only and one access which may be...

What sort of probe is this with? The cmsis-dap probes should open in HID mode without permissions but require udev rules for v2 mode, but probably st-links (and perhaps jlinks)...

I've been looking into this a bit (it turns out if you apt install openocd on Ubuntu it adds its own rules file which allows access to all devices with...

(maybe @rust-embedded/embedded-linux ?)

I think this is a really nice way for specific PACs to implement chip-specific functionality like you've done, nice work! Is there any reason in your `update.sh` you have each...

Could you add an entry to the changelog for this, please? Also, I wonder if it's a bit unnatural to have command line arguments like `--flag a b c`: it...