Dave Vandervies wrote:
> If that doesn't work, the patch at
> https://marc.info/?l=openbsd-ports&m=159897831109275&w=2 enables the
> R/M profile multilib options to get library support for Cortex-M
> cores.

Thanks Dave, I've been using your suggested approach of the
--with-multilib-list=rmprofile configure flag. It works, but it
introduces overhead to keep modifying the port as it changes -- like
arm-none-eabi-gcc-linaro to arm-none-eabi-gcc or during OS upgrades.

> (The reasons why this is necessary for Cortex-M cores and not already
> in the ports tree are complicated but round to "ARM multilib config
> is complicated, and this port is critical system infrastructure in a
> way that does not care about embedded cross-compiling".)

Do you think using rmprofile multilibs for the arm flavor would affect
openbsd's internal use of the cross-compiler? Or could there perhaps be
another port flavor introduced for embedded developers?

I love openbsd and try to use it for everything, but this issue (plus
the lagging arm-none-eabi-gdb version) has finally made me set up a dual
boot with linux on my machine. Hope I can someday return to obsd for
Cortex-M firmware development.