Dispatch interrupts asynchroneously

See merge request card10/firmware!377

It's 44 years since the release of vi and sadly lots of editors still
don't care about trimming off trailing spaces ...

Signed-off-by: Rahix <rahix@rahix.de>

Kai authored 4 years ago and rahix committed 4 years ago

Original work from @sistason in MR !121, rebased
on latest master.

Commit 1ea42e8eac0498a027f8c303e33a87f93df4dba5
Author: Kai <sistason@sistason.de>
Date:   Mon Aug 19 21:58:11 2019 +0200

    Adds Docker for firmware-building

    This commit adds a Dockerfile to build a docker
    container to build the firmware with.  It also
    updates the documentation on how to do that.

Changelog:

    - packaging: upgrade development status to beta
    - build: remove FIXME about VERSION
    - test: cover additional macro cases
    - parser: simplify macro arguments extraction
    - Re-use the looked up cursor of a token.
    - test: cover cases of functions with no parameters
    - parser: fix documentation of non prototyped functions
    - doc: add example with preprocessor

Signed-off-by: Rahix <rahix@rahix.de>

Signed-off-by: Rahix <rahix@rahix.de>

Instead of blocking the triggering task when core 1 is still busy
handling the previous interrupt, offload interrupt dispatching into
a separate task.  This is the first step towards making API-calls
interrupt safe.

Next to the async triggering, the synchroneous mechanism is retained for
special cases where async does not work (e.g. because of spin-locks).
Currently, there is only one such case when resetting core 1 (triggering
EPIC_INT_RESET).

Signed-off-by: Rahix <rahix@rahix.de>

Signed-off-by: Rahix <rahix@rahix.de>

Signed-off-by: Rahix <rahix@rahix.de>

Signed-off-by: Rahix <rahix@rahix.de>

api_interrupt_trigger() cannot fail except by a programmer error
(passing an invalid ID) in which case we should raise an assertion.
Thus, make it return `void` instead of `int`.

This is also in preparation for future work to schedule interrupts
asynchroneously.

Signed-off-by: Rahix <rahix@rahix.de>

See merge request card10/firmware!383

When opening an ACM device Linux sends the ACM_SET_CONTROL_LINE_STATE
with DTR and CTS high. We can use this re-enable the ACM device on our
side after a lockup.

BLE fixes 2020 1

See merge request card10/firmware!381

See merge request card10/firmware!379

See merge request card10/firmware!378

This speeds up file transfers from Linux machines. Looks like the impact
on power consumption is not too large. Android also does not seem to
take this into account.

Default main app selector

See merge request card10/firmware!375

I need unit tests :/