Feature: qStartNoAckmode

[perigoso]

Detailed description

This is the start of no acknowledge mode (QStartNoAckMode) support.

The BMP uses a USB connection with the CDC class (Bulk transfers) for GDB data transfer, this means the connection should be reliable, with integrated integrity checks and data delivery guarantees (provided there are no bugs in the class handling side of things in the firmware), this means that acknowledgment and checksum verification is redundant and should be safe to ignore, hopefully providing a performance benefit, GDB provides support for this with QStartNoAckMode.

See GDB: Packet-Acknowledgment for context.

The intent is to test if this brings improvements in performance, particularly in the flashing routines.

As of right now, this consists only of refactoring of the gdb_getpacket to make use of a state machine, in preparation of the actual implementation, reviews are not necessary but feedback is always welcome, shallow testing showed this to work, so testing is also welcome

Your checklist for this pull request

• I've read the Code of Conduct
• I've read the guidelines for contributing to this repository
• It builds for hardware native (make PROBE_HOST=native)
• It builds as BMDA (make PROBE_HOST=hosted)
• I've tested it to the best of my ability
• My commit messages provide a useful short description of what the commits do

Closing issues

[perigoso]

QStartNoAckmode now implemented, testing didn't show any noticeable difference in speeds, but further testing is needed and welcome

[dragonmux]

A quick check over reveals just one style mistake and something we need to consider between this branch and another that we're contemplating for v1.10. Generally though this looks great and improves the GDB packet handler handily

[perigoso]

@dragonmux rebased over your organization commit, theres still one issue to solve and adding a buildsystem option to enable this feature, but its in a ok state to star reviewing

[dragonmux]

This is looking great, nice job integrating our refactor of the remote packet handling!

There are a few items we've spotted, mostly commented out dead code.

[dragonmux]

We're happy to take this feature when you're happy with the implementation. Update the PR title and mark it as ready for review when you're ready.

We've only seen the one remaining item having re-checked through the cleanups and fixes you've performed and it is only a suggestion not a requested change.

[dragonmux]

LGTM, We'll run a couple of tests locally just to be sure things work again GDB 13, and then we'll get this merged ASAP. Excellent work and an awesome contribution, thank you!