Some overzealous locking to track down RequestObject related crashes.
bc0fa4d689 wrongly locked the current event loop's
request_invalidation_lock instead of the invalidation's list lock.
Also Abstract UI is able to delete requests concurrently with with
EventLoop invalidation.
e.g. PortManager::PortRegisteredOrUnregistered and GlobalPortMatrixWindow
so the lock needs to be exposed.
If this solves various issues, mutexes should to be consolidated
(request_buffer_map_lock + request_invalidation_lock) and be chosen
such that there is as little contention as possible.
- add more debugging output detecting regions whose
beat and frame position do not align on a playlist.
this is required as a check as we have never used
frame rounding on constant tempi before 8884a5723dc
- Evoral::Beats operator!= would prevent an increment
of start_beats by intervals of less than a tick,
so its possible that other subtle problems
existed due to this kind of thing.
- one_source_per_region_fix no longer clobbers midi source
- allow user to override per-region fix and suggest they do so.
- fix multiple note divisor detection.
- provides an upgrade path for sessions affected by a nasty bug
present in some 5.0 - 5.3 MIDI sessions.
If testing from source, it can be run from the session_utils directory
using ./run ardour5-headless-chicken -h
uselib is no longer implicit (inherited by .use). This is still incomplete,
some uselibs for non-linux variants may be missing.
bld.is_defined("HAVE_XXX") also no longer works and will have to be
changed (I think to bld.env["HAVE_XXX"]) in countless places.
