las3_pub / predictable_parallel_patterns

The start_chain property does not make sense, as chains are purely 'virtual', i.e. they only fully exist when walking through the computation (by patching them on important events). We initially added the property as a helper for better runtime and simpler implementation, but we think without it we will not get as much inconsistency in the runtime state. Performance can be 're-added' later on.

Everything so far is untested. We only made sure tha fast path still seems to function correctly. Next up is writing tests for both the fast and slow path to then introduce the slow path. After that we can look at performance optimizations.

This showcases the expected performance when a task executes a sub-tree without inference from other threads. We target to stay about 6x slower than a normal function call.

This allows us to more easily handle them and makes their interface closer to std::thread.

The scheduler yields if it failed to steal any work due to the task
lists being empty. This should improve performance on multiprogrammed
systems, as it potentially makes room for other worker threads which
still have work to perform.

This allows the stack and deque class to use the same offset, making it work better with each other.

Threads must be joined on scheduler termination and tasks must be pushed onto the stack to allow better memory management.

This seems to be only possible with two different function names, as variadic arguments and default arguments wont work together.

Please note that this does currently not execute, but only removed the 'old aspecs'. We are still missing launching the root fork_join task.

We do this to properly separate the cache alginment logic in the next step, allowing us to port all cache aligned objects without worrying about portability.

This lead to some bug-fixes that covered move constructors not correctly assigning memory. We might clean this up further by only allowing in place creation of these types (placement new).

This will be used for storing thread tasks aligned in continous memory, allowing different theads to work on the data without interfearing.

This requires the user to explicitly manage the storage duration/place of the thread local state, making a fine grained memory allocation simpler.

The local storage per thread is only a pointer, this must be assigned AFTER the local storage object has moved to it's final location (e.g. on the stack of the caller). We might even change this a little further in the future to make the location of the state object more transparent.