Diff [c1e66d966aadf6846330871033458d4a398bd576:deda7e6002cc711be5c0af09984cec24300d5990] for / – Cforall

doc/theses/colby_parsons_MMAth/local.bib

-                      rc1e66d9
+                      rdeda7e6
     year        = 2023,
+}
+@inproceedings{Harris02,
+  title={A practical multi-word compare-and-swap operation},
+  author={Harris, Timothy L and Fraser, Keir and Pratt, Ian A},
+  booktitle={Distributed Computing: 16th International Conference, DISC 2002 Toulouse, France, October 28--30, 2002 Proceedings 16},
+  pages={265--279},
+  year={2002},
+  organization={Springer}
+}
+@misc{kotlin:channel,
+  author = "Kotlin Documentation",
+  title = "Channel",
+  howpublished = {\url{https://kotlinlang.org/api/kotlinx.coroutines/kotlinx-coroutines-core/kotlinx.coroutines.channels/-channel/}},
+  note = "[Online; accessed 11-September-2023]"
+}

doc/theses/colby_parsons_MMAth/text/CFA_concurrency.tex

-                      rc1e66d9
+                      rdeda7e6
 \chapter{Concurrency in \CFA}\label{s:cfa_concurrency}
 The groundwork for concurrency in \CFA was laid by Thierry Delisle in his Master's Thesis~\cite{Delisle18}.
 In that work, he introduced generators, coroutines, monitors, and user-level threading.
 Not listed in that work were basic concurrency features needed as building blocks, such as locks, futures, and condition variables, which he also added to \CFA.
+The groundwork for concurrency in \CFA was laid by Thierry Delisle in his Master's Thesis~\cite{Delisle18}.
+In that work, he introduced generators, coroutines, monitors, and user-level threading.
+Not listed in that work were basic concurrency features needed as building blocks, such as locks, futures, and condition variables.
 \section{Threading Model}\label{s:threading}
-\CFA provides user-level threading and supports an $M$:$N$ threading model where $M$ user threads are scheduled on $N$ kernel threads, where both $M$ and $N$ can be explicitly set by the user.
-Kernel threads are created by declaring a @processor@ structure.
-User-thread types are defined by creating a @thread@ aggregate-type, \ie replace @struct@ with @thread@.
-For each thread type a corresponding @main@ routine must be defined, which is where the thread starts running once it is created.
-Examples of \CFA  user thread and processor creation are shown in \VRef[Listing]{l:cfa_thd_init}.
+\CFA provides user-level threading and supports an $M$:$N$ threading model where $M$ user threads are scheduled on $N$ kernel threads and both $M$ and $N$ can be explicitly set by the programmer.
+Kernel threads are created by declaring processor objects;
+user threads are created by declaring a thread objects.
+\VRef[Listing]{l:cfa_thd_init} shows a typical examples of creating a \CFA user-thread type, and then as declaring processor ($N$) and thread objects ($M$).
 \begin{cfa}[caption={Example of \CFA user thread and processor creation},label={l:cfa_thd_init}]
+@thread@ my_thread {...};                       $\C{// user thread type}$
+void @main@( my_thread & this ) {       $\C{// thread start routine}$
+@thread@ my_thread {                            $\C{// user thread type (like structure)}$
+        ... // arbitrary number of field declarations
+};
+void @main@( @my_thread@ & this ) {     $\C{// thread start routine}$
         sout | "Hello threading world";
+}
+int main() {
+int main() {                                            $\C{// program starts with a processor (kernel thread)}$
         @processor@ p[2];                               $\C{// add 2 processors = 3 total with starting processor}$
+        {
                 my_thread t[2], * t3 = new();   $\C{// create 3 user threads, running in main routine}$
+                @my_thread@ t[2], * t3 = new(); $\C{// create 2 stack allocated, 1 dynamic allocated user threads}$
                 ... // execute concurrently
                 delete( t3 );                           $\C{// wait for thread to end and deallocate}$
         } // wait for threads to end and deallocate
+}
+                delete( t3 );                           $\C{// wait for t3 to end and deallocate}$
+    } // wait for threads t[0] and t[1] to end and deallocate
+} // deallocate additional kernel threads
 \end{cfa}
+When processors are added, they are added alongside the existing processor given to each program.
+Thus, for $N$ processors, allocate $N-1$ processors.
+A thread is implicitly joined at deallocation, either implicitly at block exit for stack allocation or explicitly at @delete@ for heap allocation.
+The thread performing the deallocation must wait for the thread to terminate before the deallocation can occur.
+A thread type is are defined using the aggregate kind @thread@.
+For each thread type, a corresponding @main@ routine must be defined, which is where the thread starts running once when a thread object are is created.
+The @processor@ declaration adds addition kernel threads alongside the existing processor given to each program.
+Thus, for $N$ processors, allocate $N-1$ processors.
+A thread is implicitly joined at deallocation, either implicitly at block exit for stack allocation or explicitly at @delete@ for heap allocation.
+The thread performing the deallocation must wait for the thread to terminate before the deallocation can occur.
 A thread terminates by returning from the main routine where it starts.
+\section{Existing Concurrency Features}
+\section{Existing and New Concurrency Features}
 \CFA currently provides a suite of concurrency features including futures, locks, condition variables, generators, coroutines, monitors.
 Examples of these features are omitted as most of them are the same as their counterparts in other languages.
 It is worthwhile to note that all concurrency features added to \CFA are made to be compatible each other.
+The laundry list of features above and the ones introduced in this thesis can be used in the same program without issue.
+The laundry list of features above and the ones introduced in this thesis can be used in the same program without issue, and the features are designed to interact in meaningful ways.
+For example, a thread can inteact with a monitor, which can interact with a coroutine, which can interact with a generator.
 Solving concurrent problems requires a diverse toolkit.

doc/theses/colby_parsons_MMAth/text/CFA_intro.tex

-                      rc1e66d9
+                      rdeda7e6
 \CFA is a layer over C, is transpiled\footnote{Source to source translator.} to C, and is largely considered to be an extension of C.
 Beyond C, it adds productivity features, extended libraries, an advanced type-system, and many control-flow/concurrency constructions.
 However, \CFA stays true to the C programming style, with most code revolving around @struct@'s and routines, and respects the same rules as C.
+However, \CFA stays true to the C programming style, with most code revolving around @struct@s and routines, and respects the same rules as C.
 \CFA is not object oriented as it has no notion of @this@ (receiver) and no structures with methods, but supports some object oriented ideas including constructors, destructors, and limited nominal inheritance.
 While \CFA is rich with interesting features, only the subset pertinent to this work is discussed here.
 …
 References in \CFA are a layer of syntactic sugar over pointers to reduce the number of syntactic ref/deref operations needed with pointer usage.
 Pointers in \CFA differ from C and \CC in their use of @0p@ instead of C's @NULL@ or \CC's @nullptr@.
+References can contain 0p in \CFA, which is the equivalent of a null reference.
 Examples of references are shown in \VRef[Listing]{l:cfa_ref}.
 …
 This feature is also implemented in Pascal~\cite{Pascal}.
 It can exist as a stand-alone statement or wrap a routine body to expose aggregate fields.
+If exposed fields share a name, the type system will attempt to disambiguate them based on type.
+If the type system is unable to disambiguate the fields then the user must qualify those names to avoid a compilation error.
 Examples of the @with@ statement are shown in \VRef[Listing]{l:cfa_with}.

doc/theses/colby_parsons_MMAth/text/actors.tex

-                      rc1e66d9
+                      rdeda7e6
 Actors are an indirect concurrent feature that abstracts threading away from a programmer, and instead provides \gls{actor}s and messages as building blocks for concurrency.
 Hence, actors are in the realm of \gls{impl_concurrency}, where programmers write concurrent code without dealing with explicit thread creation or interaction.
 Actor message-passing is similar to channels, but with more abstraction, so there is no shared data to protect, making actors amenable in a distributed environment.
+Actor message-passing is similar to channels, but with more abstraction, so there is no shared data to protect, making actors amenable to a distributed environment.
 Actors are often used for high-performance computing and other data-centric problems, where the ease of use and scalability of an actor system provides an advantage over channels.
 …
 \section{Actor Model}
 The \Newterm{actor model} is a concurrent paradigm where computation is broken into units of work called actors, and the data for computation is distributed to actors in the form of messages~\cite{Hewitt73}.
+The \Newterm{actor model} is a concurrent paradigm where an actor is used as the fundamental building-block for computation, and the data for computation is distributed to actors in the form of messages~\cite{Hewitt73}.
 An actor is composed of a \Newterm{mailbox} (message queue) and a set of \Newterm{behaviours} that receive from the mailbox to perform work.
 Actors execute asynchronously upon receiving a message and can modify their own state, make decisions, spawn more actors, and send messages to other actors.
 …
 For example, mutual exclusion and locking are rarely relevant concepts in an actor model, as actors typically only operate on local state.
+An actor does not have a thread.
+\subsection{Classic Actor System}
+An implementation of the actor model with a theatre (group) of actors is called an \Newterm{actor system}.
+Actor systems largely follow the actor model, but can differ in some ways.
+In an actor system, an actor does not have a thread.
 An actor is executed by an underlying \Newterm{executor} (kernel thread-pool) that fairly invokes each actor, where an actor invocation processes one or more messages from its mailbox.
 The default number of executor threads is often proportional to the number of computer cores to achieve good performance.
 An executor is often tunable with respect to the number of kernel threads and its scheduling algorithm, which optimize for specific actor applications and workloads \see{Section~\ref{s:ActorSystem}}.
-\subsection{Classic Actor System}
-An implementation of the actor model with a community of actors is called an \Newterm{actor system}.
-Actor systems largely follow the actor model, but can differ in some ways.
 While the semantics of message \emph{send} is asynchronous, the implementation may be synchronous or a combination.
 The default semantics for message \emph{receive} is \gls{fifo}, so an actor receives messages from its mailbox in temporal (arrival) order;
 however, messages sent among actors arrive in any order.
+The default semantics for message \emph{receive} is \gls{fifo}, so an actor receives messages from its mailbox in temporal (arrival) order.
+% however, messages sent among actors arrive in any order.
 Some actor systems provide priority-based mailboxes and/or priority-based message-selection within a mailbox, where custom message dispatchers search among or within a mailbox(es) with a predicate for specific kinds of actors and/or messages.
 Some actor systems provide a shared mailbox where multiple actors receive from a common mailbox~\cite{Akka}, which is contrary to the no-sharing design of the basic actor-model (and requires additional locking).
 For non-\gls{fifo} service, some notion of fairness (eventual progress) must exist, otherwise messages have a high latency or starve, \ie never received.
 Finally, some actor systems provide multiple typed-mailboxes, which then lose the actor-\lstinline{become} mechanism \see{Section~\ref{s:SafetyProductivity}}.
+Some actor systems provide a shared mailbox where multiple actors receive from a common mailbox~\cite{Akka}, which is contrary to the no-sharing design of the basic actor-model (and may require additional locking).
+For non-\gls{fifo} service, some notion of fairness (eventual progress) should exist, otherwise messages have a high latency or starve, \ie are never received.
+% Finally, some actor systems provide multiple typed-mailboxes, which then lose the actor-\lstinline{become} mechanism \see{Section~\ref{s:SafetyProductivity}}.
 %While the definition of the actor model provides no restrictions on message ordering, actor systems tend to guarantee that messages sent from a given actor $i$ to actor $j$ arrive at actor $j$ in the order they were sent.
 Another way an actor system varies from the model is allowing access to shared global-state.
 …
 Figure \ref{f:inverted_actor} shows an actor system designed as \Newterm{message-centric}, where a set of messages are scheduled and run on underlying executor threads~\cite{uC++,Nigro21}.
 This design is \Newterm{inverted} because actors belong to a message queue, whereas in the classic approach a message queue belongs to each actor.
 Now a message send must queries the actor to know which message queue to post the message.
+Now a message send must query the actor to know which message queue to post the message to.
 Again, the simplest design has a single global queue of messages accessed by the executor threads, but this approach has the same contention problem by the executor threads.
 Therefore, the messages (mailboxes) are sharded and executor threads schedule each message, which points to its corresponding actor.
 …
         @actor | str_msg | int_msg;@                    $\C{// cascade sends}$
         @actor | int_msg;@                                              $\C{// send}$
         @actor | finished_msg;@                                 $\C{// send => terminate actor (deallocation deferred)}$
+        @actor | finished_msg;@                                 $\C{// send => terminate actor (builtin Poison-Pill)}$
         stop_actor_system();                                    $\C{// waits until actors finish}\CRT$
 } // deallocate actor, int_msg, str_msg
 …
 Each executor thread iterates over its own message queues until it finds one with messages.
 At this point, the executor thread atomically \gls{gulp}s the queue, meaning it moves the contents of message queue to a local queue of the executor thread.
+Gulping moves the contents of the message queue as a batch rather than removing individual elements.
 An example of the queue gulping operation is shown in the right side of Figure \ref{f:gulp}, where an executor thread gulps queue 0 and begins to process it locally.
 This step allows the executor thread to process the local queue without any atomics until the next gulp.
 …
 Since the copy queue is an array, envelopes are allocated first on the stack and then copied into the copy queue to persist until they are no longer needed.
 For many workloads, the copy queues grow in size to facilitate the average number of messages in flight and there are no further dynamic allocations.
+For many workloads, the copy queues reallocate and grow in size to facilitate the average number of messages in flight and there are no further dynamic allocations.
 The downside of this approach is that more storage is allocated than needed, \ie each copy queue is only partially full.
 Comparatively, the individual envelope allocations of a list-based queue mean that the actor system always uses the minimum amount of heap space and cleans up eagerly.
 …
 To ensure sequential actor execution and \gls{fifo} message delivery in a message-centric system, stealing requires finding and removing \emph{all} of an actor's messages, and inserting them consecutively in another message queue.
 This operation is $O(N)$ with a non-trivial constant.
 The only way for work stealing to become practical is to shard each worker's message queue, which also reduces contention, and to steal queues to eliminate queue searching.
+The only way for work stealing to become practical is to shard each worker's message queue \see{Section~\ref{s:executor}}, which also reduces contention, and to steal queues to eliminate queue searching.
 Given queue stealing, the goal of the presented stealing implementation is to have an essentially zero-contention-cost stealing mechanism.
 …
 The outline for lazy-stealing by a thief is: select a victim, scan its queues once, and return immediately if a queue is stolen.
 The thief then assumes it normal operation of scanning over its own queues looking for work, where stolen work is placed at the end of the scan.
+The thief then assumes its normal operation of scanning over its own queues looking for work, where stolen work is placed at the end of the scan.
 Hence, only one victim is affected and there is a reasonable delay between stealing events as the thief scans its ready queue looking for its own work before potentially stealing again.
 This lazy examination by the thief has a low perturbation cost for victims, while still finding work in a moderately loaded system.
 …
 % Note that a thief never exceeds its $M$/$N$ worker range because it is always exchanging queues with other workers.
 If no appropriate victim mailbox is found, no swap is attempted.
+Note that since the mailbox checks happen non-atomically, the thieves effectively guess which mailbox is ripe for stealing.
+The thief may read stale data and can end up stealing an ineligible or empty mailbox.
+This is not a correctness issue and is addressed in Section~\ref{s:steal_prob}, but the steal will likely not be productive.
+These unproductive steals are uncommon, but do occur with some frequency, and are a tradeoff that is made to achieve minimal victim contention.
 \item
 …
 \end{enumerate}
 \subsection{Stealing Problem}
+\subsection{Stealing Problem}\label{s:steal_prob}
 Each queue access (send or gulp) involving any worker (thief or victim) is protected using spinlock @mutex_lock@.
 However, to achieve the goal of almost zero contention for the victim, it is necessary that the thief does not acquire any queue spinlocks in the stealing protocol.
 …
 None of the work-stealing actor-systems examined in this work perform well on the repeat benchmark.
 Hence, for all non-pathological cases, the claim is made that this stealing mechanism has a (probabilistically) zero-victim-cost in practice.
+Future work on the work stealing system could include a backoff mechanism after failed steals to further address the pathological cases.
 \subsection{Queue Pointer Swap}\label{s:swap}
 …
 The \gls{cas} is a read-modify-write instruction available on most modern architectures.
 It atomically compares two memory locations, and if the values are equal, it writes a new value into the first memory location.
 A software implementation of \gls{cas} is:
+A sequential specification of \gls{cas} is:
 \begin{cfa}
 // assume this routine executes atomically
 …
 Either a true memory/memory swap instruction or a \gls{dcas} would provide the ability to atomically swap two memory locations, but unfortunately neither of these instructions are supported on the architectures used in this work.
+There are lock-free implemetions of DCAS, or more generally K-word CAS (also known as MCAS or CASN)~\cite{Harris02} and LLX/SCX~\cite{Brown14} that can be used to provide the desired atomic swap capability.
+However, these lock-free implementations were not used as it is advantageous in the work stealing case to let an attempted atomic swap fail instead of retrying.
 Hence, a novel atomic swap specific to the actor use case is simulated, called \gls{qpcas}.
+Note that this swap is \emph{not} lock-free.
 The \gls{qpcas} is effectively a \gls{dcas} special cased in a few ways:
 \begin{enumerate}
 …
 \end{cfa}
 \item
 The values swapped are never null pointers, so a null pointer can be used as an intermediate value during the swap.
+The values swapped are never null pointers, so a null pointer can be used as an intermediate value during the swap. As such, null is effectively used as a lock for the swap.
 \end{enumerate}
 Figure~\ref{f:qpcasImpl} shows the \CFA pseudocode for the \gls{qpcas}.
 …
 The concurrent proof of correctness is shown through the existence of an invariant.
 The invariant states when a queue pointer is set to @0p@ by a thief, then the next write to the pointer can only be performed by the same thief.
+This is effictively a mutual exclusion condition for the later write.
 To show that this invariant holds, it is shown that it is true at each step of the swap.
 \begin{itemize}
 …
 The intuition behind this heuristic is that the slowest worker receives help via work stealing until it becomes a thief, which indicates that it has caught up to the pace of the rest of the workers.
 This heuristic should ideally result in lowered latency for message sends to victim workers that are overloaded with work.
+It must be acknowledged that this linear search could cause a lot of cache coherence traffic.
+Future work on this heuristic could include introducing a search that has less impact on caching.
 A negative side-effect of this heuristic is that if multiple thieves steal at the same time, they likely steal from the same victim, which increases the chance of contention.
 However, given that workers have multiple queues, often in the tens or hundreds of queues, it is rare for two thieves to attempt stealing from the same queue.
 …
 \CFA's actor system comes with a suite of safety and productivity features.
 Most of these features are only present in \CFA's debug mode, and hence, have zero-cost in no-debug mode.
 The suit of features include the following.
+The suite of features include the following.
 \begin{itemize}
 \item Static-typed message sends:
 If an actor does not support receiving a given message type, the receive call is rejected at compile time, allowing unsupported messages to never be sent to an actor.
+If an actor does not support receiving a given message type, the receive call is rejected at compile time, preventing unsupported messages from being sent to an actor.
 \item Detection of message sends to Finished/Destroyed/Deleted actors:
 …
 \item When an executor is configured, $M >= N$.
 That is, each worker must receive at least one mailbox queue, otherwise the worker spins and never does any work.
+That is, each worker must receive at least one mailbox queue, since otherwise a worker cannot receive any work without a queue pull messages from.
 \item Detection of unsent messages:
 …
 \begin{list}{\arabic{enumi}.}{\usecounter{enumi}\topsep=5pt\parsep=5pt\itemsep=0pt}
 \item
 Supermicro SYS--6029U--TR4 Intel Xeon Gold 5220R 24--core socket, hyper-threading $\times$ 2 sockets (48 process\-ing units) 2.2GHz, running Linux v5.8.0--59--generic
 \item
 Supermicro AS--1123US--TR4 AMD EPYC 7662 64--core socket, hyper-threading $\times$ 2 sockets (256 processing units) 2.0 GHz, running Linux v5.8.0--55--generic
+Supermicro SYS--6029U--TR4 Intel Xeon Gold 5220R 24--core socket, hyper-threading $\times$ 2 sockets (96 process\-ing units), running Linux v5.8.0--59--generic
+\item
+Supermicro AS--1123US--TR4 AMD EPYC 7662 64--core socket, hyper-threading $\times$ 2 sockets (256 processing units), running Linux v5.8.0--55--generic
 \end{list}
 …
 All benchmarks are run 5 times and the median is taken.
 Error bars showing the 95\% confidence intervals appear on each point in the graphs.
+The confidence intervals are calculated using bootstrapping to avoid normality assumptions.
 If the confidence bars are small enough, they may be obscured by the data point.
 In this section, \uC is compared to \CFA frequently, as the actor system in \CFA is heavily based off of \uC's actor system.

doc/theses/colby_parsons_MMAth/text/channels.tex

-                      rc1e66d9
+                      rdeda7e6
 Neither Go nor \CFA channels have the restrictions of the early channel-based concurrent systems.
 Other popular languages and libraries that provide channels include C++ Boost~\cite{boost:channel}, Rust~\cite{rust:channel}, Haskell~\cite{haskell:channel}, and OCaml~\cite{ocaml:channel}.
+Other popular languages and libraries that provide channels include C++ Boost~\cite{boost:channel}, Rust~\cite{rust:channel}, Haskell~\cite{haskell:channel}, OCaml~\cite{ocaml:channel}, and Kotlin~\cite{kotlin:channel}.
 Boost channels only support asynchronous (non-blocking) operations, and Rust channels are limited to only having one consumer per channel.
 Haskell channels are unbounded in size, and OCaml channels are zero-size.
 These restrictions in Haskell and OCaml are likely due to their functional approach, which results in them both using a list as the underlying data structure for their channel.
 These languages and libraries are not discussed further, as their channel implementation is not comparable to the bounded-buffer style channels present in Go and \CFA.
+Kotlin channels are comparable to Go and \CFA, but unfortunately they were not identified as a comparator until after presentation of this thesis and are omitted due to time constraints.
 \section{Producer-Consumer Problem}
 …
 In the problem, threads interact with a buffer in two ways: producing threads insert values into the buffer and consuming threads remove values from the buffer.
 In general, a buffer needs protection to ensure a producer only inserts into a non-full buffer and a consumer only removes from a non-empty buffer (synchronization).
 As well, a buffer needs protection from concurrent access by multiple producers or consumers attempting to insert or remove simultaneously (MX).
+As well, a buffer needs protection from concurrent access by multiple producers or consumers attempting to insert or remove simultaneously, which is often provided by MX.
 \section{Channel Size}\label{s:ChannelSize}
 …
 Fixed sized (bounded) implies the communication is mostly asynchronous, \ie the producer can proceed up to the buffer size and vice versa for the consumer with respect to removal, at which point the producer/consumer would wait.
 \item
+Infinite sized (unbounded) implies the communication is asynchronous, \ie the producer never waits but the consumer waits when the buffer is empty.
+Since memory is finite, all unbounded buffers are ultimately bounded;
+this restriction must be part of its implementation.
+Infinite sized (unbounded) implies the communication is asymmetrically asynchronous, \ie the producer never waits but the consumer waits when the buffer is empty.
 \end{enumerate}
 …
 However, like MX, a buffer should ensure every value is eventually removed after some reasonable bounded time (no long-term starvation).
 The simplest way to prevent starvation is to implement the buffer as a queue, either with a cyclic array or linked nodes.
+While \gls{fifo} is not required for producer-consumer problem correctness, it is a desired property in channels as it provides predictable and often relied upon channel ordering behaviour to users.
 \section{First-Come First-Served}
 As pointed out, a bounded buffer requires MX among multiple producers or consumers.
+As pointed out, a bounded buffer implementation often provides MX among multiple producers or consumers.
 This MX should be fair among threads, independent of the \gls{fifo} buffer being fair among values.
 Fairness among threads is called \gls{fcfs} and was defined by Lamport~\cite[p.~454]{Lamport74}.
 …
 \section{Channel Implementation}\label{s:chan_impl}
 Currently, only the Go and Erlang programming languages provide user-level threading where the primary communication mechanism is channels.
 Both Go and Erlang have user-level threading and preemptive scheduling, and both use channels for communication.
 Go provides multiple homogeneous channels; each have a single associated type.
+The programming languages Go, Kotlin, and Erlang provide user-level threading where the primary communication mechanism is channels.
+These languages have user-level threading and preemptive scheduling, and both use channels for communication.
+Go and Kotlin provide multiple homogeneous channels; each have a single associated type.
 Erlang, which is closely related to actor systems, provides one heterogeneous channel per thread (mailbox) with a typed receive pattern.
 Go encourages users to communicate via channels, but provides them as an optional language feature.
+Go and Kotlin encourage users to communicate via channels, but provides them as an optional language feature.
 On the other hand, Erlang's single heterogeneous channel is a fundamental part of the threading system design; using it is unavoidable.
 Similar to Go, \CFA's channels are offered as an optional language feature.
+Similar to Go and Kotlin, \CFA's channels are offered as an optional language feature.
 While iterating on channel implementation, experiments were conducted that varied the producer-consumer algorithm and lock type used inside the channel.
 …
 The Go channel implementation utilizes cooperation among threads to achieve good performance~\cite{go:chan}.
 This cooperation only occurs when producers or consumers need to block due to the buffer being full or empty.
+After a producer blocks it must wait for a consumer to signal it and vice versa.
+The consumer or producer that signals a blocked thread is called the signalling thread.
 In these cases, a blocking thread stores their relevant data in a shared location and the signalling thread completes the blocking thread's operation before waking them;
 \ie the blocking thread has no work to perform after it unblocks because the signalling threads has done this work.
 …
 First, each thread interacting with the channel only acquires and releases the internal channel lock once.
 As a result, contention on the internal lock is decreased; only entering threads compete for the lock since unblocking threads do not reacquire the lock.
 The other advantage of Go's wait-morphing approach is that it eliminates the bottleneck of waiting for signalled threads to run.
+The other advantage of Go's wait-morphing approach is that it eliminates the need to wait for signalled threads to run.
 Note that the property of acquiring/releasing the lock only once can also be achieved with a different form of cooperation, called \Newterm{baton passing}.
 Baton passing occurs when one thread acquires a lock but does not release it, and instead signals a thread inside the critical section, conceptually ``passing'' the mutual exclusion from the signalling thread to the signalled thread.
 …
 the wait-morphing approach has threads cooperate by completing the signalled thread's operation, thus removing a signalled thread's need for mutual exclusion after unblocking.
 While baton passing is useful in some algorithms, it results in worse channel performance than the Go approach.
 In the baton-passing approach, all threads need to wait for the signalled thread to reach the front of the ready queue, context switch, and run before other operations on the channel can proceed, since the signalled thread holds mutual exclusion;
+In the baton-passing approach, all threads need to wait for the signalled thread to unblock and run before other operations on the channel can proceed, since the signalled thread holds mutual exclusion;
 in the wait-morphing approach, since the operation is completed before the signal, other threads can continue to operate on the channel without waiting for the signalled thread to run.
 …
 Thus, improperly handled \gls{toctou} issues with channels often result in deadlocks as threads performing the termination may end up unexpectedly blocking in their attempt to help other threads exit the system.
+\subsubsection{Go Channel Close}
 Go channels provide a set of tools to help with concurrent shutdown~\cite{go:chan} using a @close@ operation in conjunction with the \Go{select} statement.
 The \Go{select} statement is discussed in \ref{s:waituntil}, where \CFA's @waituntil@ statement is compared with the Go \Go{select} statement.
 …
 Hence, due to Go's asymmetric approach to channel shutdown, separate synchronization between producers and consumers of a channel has to occur during shutdown.
+\paragraph{\CFA channels} have access to an extensive exception handling mechanism~\cite{Beach21}.
+\subsubsection{\CFA Channel Close}
+\CFA channels have access to an extensive exception handling mechanism~\cite{Beach21}.
 As such \CFA uses an exception-based approach to channel shutdown that is symmetric for both producers and consumers, and supports graceful shutdown.

doc/theses/colby_parsons_MMAth/text/conclusion.tex

-                      rc1e66d9
+                      rdeda7e6
 % ======================================================================
+The goal of this thesis was to expand the concurrent support that \CFA offers to fill in gaps and support language users' ability to write safe and efficient concurrent programs.
+The presented features achieves this goal, and provides users with the means to write scalable programs in \CFA through multiple avenues.
+Additionally, the tools presented include safety and productivity features from deadlock detection, to detection of common programming errors, easy concurrent shutdown, and toggleable performance statistics.
+Programmers often have preferences between computing paradigms and concurrency is no exception.
+If users prefer the message passing paradigm of concurrency, \CFA now provides message passing utilities in the form of an actor system and channels.
+For shared memory concurrency, the mutex statement provides a safe and easy-to-use interface for mutual exclusion.
+The @waituntil@ statement aids in writing concurrent programs in both the message passing and shared memory paradigms of concurrency.
+Furthermore, no other language provides a synchronous multiplexing tool polymorphic over resources like \CFA's @waituntil@.
+This work successfully provides users with familiar concurrent language support, but with additional value added over similar utilities in other popular languages.
+The goal of this thesis is to expand concurrent support in \CFA to fill in gaps and increase support for writing safe and efficient concurrent programs.
+The presented features achieve this goal and provide users with the means to write scalable concurrent programs in \CFA through multiple avenues.
+Additionally, the tools presented provide safety and productivity features including: detection of deadlock and other common concurrency errors, easy concurrent shutdown, and toggleable performance statistics.
+On overview of the contributions in this thesis include the following:
+For locking, the mutex statement provides a safe and easy-to-use interface for mutual exclusion.
+If programmers prefer the message-passing paradigm, \CFA now supports it in the form of channels and actors.
+The @waituntil@ statement simplifies writing concurrent programs in both the message-passing and shared-memory paradigms of concurrency.
+Finally, no other programming language provides a synchronous multiplexing tool that is polymorphic over resources like \CFA's @waituntil@.
+This work successfully provides users with familiar concurrent-language support, but with additional value added over similar utilities in other popular languages.
+On overview of the contributions made in this thesis include the following:
 \begin{enumerate}
+\item The mutex statement, which provides performant and deadlock-free multiple lock acquisition.
+\item Channels with comparable performance to Go, that have safety and productivity features including deadlock detection, and an easy-to-use exception-based channel @close@ routine.
+\item An in-memory actor system that achieved the lowest latency message send of systems tested due to the novel copy-queue data structure. The actor system presented has built-in detection of six common actor errors, and it has good performance compared to other systems on all benchmarks.
+\item A @waituntil@ statement which tackles the hard problem of allowing a thread to safely synch\-ronously wait for some set of concurrent resources.
+\item The mutex statement, which provides performant and deadlock-free multi-lock acquisition.
+\item Channels with comparable performance to Go, which have safety and productivity features including deadlock detection and an easy-to-use exception-based channel @close@ routine.
+\item An in-memory actor system, which achieves the lowest latency message send of systems tested due to the novel copy-queue data structure.
+\item As well, the actor system has built-in detection of six common actor errors, with excellent performance compared to other systems across all benchmarks presented in this thesis.
+\item A @waituntil@ statement, which tackles the hard problem of allowing a thread wait synchronously for an arbitrary set of concurrent resources.
 \end{enumerate}
+The features presented are commonly used in conjunction to solve concurrent problems.
+The @waituntil@ statement, the @mutex@ statement, and channels will all likely see use in a program where a thread operates as an administrator or server which accepts and distributes work among channels based on some shared state.
+The @mutex@ statement sees use across almost all concurrent code in \CFA, since it is used with the stream operator @sout@ to provide thread-safe output.
+While not yet implemented, the polymorphic support of the @waituntil@ statement could see use in conjunction with the actor system to enable user threads outside the actor system to wait for work to be done, or for actors to finish.
+A user of \CFA does not have to solely subscribe to the message passing or shared memory concurrent paradigm.
+As such, channels in \CFA are often used to pass pointers to shared memory that may still need mutual exclusion, requiring the @mutex@ statement to also be used.
+The added features are now commonly used to solve concurrent problems in \CFA.
+The @mutex@ statement sees use across almost all concurrent code in \CFA, as it is the simplest mechanism for providing thread-safe input and output.
+The channels and the @waituntil@ statement see use in programs where a thread operates as a server or administrator, which accepts and distributes work among channels based on some shared state.
+When implemented, the polymorphic support of the @waituntil@ statement will see use with the actor system to enable user threads outside the actor system to wait for work to be done or for actors to finish.
+Finally, the new features are often combined, \eg channels pass pointers to shared memory that may still need mutual exclusion, requiring the @mutex@ statement to be used.
 From the novel copy-queue data structure in the actor system and the plethora of user-supporting safety features, all these utilities build upon existing concurrent tooling with value added.
 Performance results verify that each new feature is comparable or better than similar features in other programming languages.
 All in all, this suite of concurrent tools expands users' ability to easily write safe and performant multi-threaded programs in \CFA.
+All in all, this suite of concurrent tools expands a \CFA programmer's ability to easily write safe and performant multi-threaded programs.
 \section{Future Work}
 …
 This thesis only scratches the surface of implicit concurrency by providing an actor system.
 There is room for more implicit concurrency tools in \CFA.
 User-defined implicit concurrency in the form of annotated loops or recursive concurrent functions exists in many other languages and libraries~\cite{uC++,OpenMP}.
+User-defined implicit concurrency in the form of annotated loops or recursive concurrent functions exists in other languages and libraries~\cite{uC++,OpenMP}.
 Similar implicit concurrency mechanisms could be implemented and expanded on in \CFA.
 Additionally, the problem of automatic parallelism of sequential programs via the compiler is an interesting research space that other languages have approached~\cite{wilson94,haskell:parallel} and could be explored in \CFA.
 …
 \subsection{Advanced Actor Stealing Heuristics}
 In this thesis, two basic victim-selection heuristics are chosen when implementing the work stealing actor system.
 Good victim selection is an active area of work stealing research, especially when taking into account NUMA effects and cache locality~\cite{barghi18,wolke17}.
+In this thesis, two basic victim-selection heuristics are chosen when implementing the work-stealing actor-system.
+Good victim selection is an active area of work-stealing research, especially when taking into account NUMA effects and cache locality~\cite{barghi18,wolke17}.
 The actor system in \CFA is modular and exploration of other victim-selection heuristics for queue stealing in \CFA could be provided by pluggable modules.
 Another question in work stealing is: when should a worker thread steal?
 Work stealing systems can often be too aggressive when stealing, causing the cost of the steal to be have a negative rather positive effect on performance.
+Work-stealing systems can often be too aggressive when stealing, causing the cost of the steal to be have a negative rather positive effect on performance.
 Given that thief threads often have cycles to spare, there is room for a more nuanced approaches when stealing.
 Finally, there is the very difficult problem of blocking and unblocking idle threads for workloads with extreme oscillations in CPU needs.
 …
 \subsection{Synchronously Multiplexing System Calls}
+There are many tools that try to synchronously wait for or asynchronously check I/O, since improvements in this area pay dividends in many areas of computer science~\cite{linux:select,linux:poll,linux:epoll,linux:iouring}.
+There are many tools that try to synchronously wait for or asynchronously check I/O.
+Improvements in this area pay dividends in many areas of I/O based programming~\cite{linux:select,linux:poll,linux:epoll,linux:iouring}.
 Research on improving user-space tools to synchronize over I/O and other system calls is an interesting area to explore in the world of concurrent tooling.
 Specifically, incorporating I/O into the @waituntil@ to allow a network server to work with multiple kinds of asynchronous I/O interconnects without using tradition event loops.
 …
 The semantics and safety of these builtins require careful navigation since they require the user to have a deep understanding of concurrent memory-ordering models.
 Furthermore, these atomics also often require a user to understand how to fence appropriately to ensure correctness.
 All these problems and more could benefit from language support in \CFA.
+All these problems and more would benefit from language support in \CFA.
 Adding good language support for atomics is a difficult problem, which if solved well, would allow for easier and safer writing of low-level concurrent code.

doc/theses/colby_parsons_MMAth/text/intro.tex

-                      rc1e66d9
+                      rdeda7e6
 % ======================================================================
 Concurrent programs are the wild west of programming because determinism and simple ordering of program operations go out the window.
 To seize the reins and write performant and safe concurrent code, high-level concurrent-language features are needed.
 Like any other craftsmen, programmers are only as good as their tools, and concurrent tooling and features are no exception.
+Concurrent programs are the wild west of programming because determinism and simple ordering of program operations go out the window.
+To seize the reins and write performant and safe concurrent code, high-level concurrent-language features are needed.
+Like any other craftsmen, programmers are only as good as their tools, and concurrent tooling and features are no exception.
+This thesis presents a suite of high-level concurrent-language features implemented in the new programming-language \CFA.
+These features aim to improve the performance of concurrent programs, aid in writing safe programs, and assist user productivity by improving the ease of concurrent programming.
+The groundwork for concurrent features in \CFA was implemented by Thierry Delisle~\cite{Delisle18}, who contributed the threading system, coroutines, monitors and other tools.
+This thesis builds on top of that foundation by providing a suite of high-level concurrent features.
+The features include a @mutex@ statement, channels, a @waituntil@ statement, and an actor system.
+All of these features exist in other programming languages in some shape or form, however this thesis extends the original ideas by improving performance, productivity, and safety.
+This thesis presents a suite of high-level concurrent-language features implemented in the new programming-language \CFA.
+These features aim to improve the performance of concurrent programs, aid in writing safe programs, and assist user productivity by improving the ease of concurrent programming.
+The groundwork for concurrent features in \CFA was designed and implemented by Thierry Delisle~\cite{Delisle18}, who contributed the threading system, coroutines, monitors and other basic concurrency tools.
+This thesis builds on top of that foundation by providing a suite of high-level concurrent features.
+The features include a @mutex@ statement, channels, a @waituntil@ statement, and an actor system.
+All of these features exist in other programming languages in some shape or form;
+however, this thesis extends the original ideas by improving performance, productivity, and safety.
 \section{The Need For Concurrent Features}
-Asking a programmer to write a complex concurrent program without any concurrent language features is asking them to undertake a very difficult task.
-They would only be able to rely on the atomicity that their hardware provides and would have to build up from there.
-This would be like asking a programmer to write a complex sequential program only in assembly.
-Both are doable, but would often be easier and less error prone with higher level tooling.
+Concurrent programming has many pitfalls that are unique and do not show up in sequential code:
+% Asking a programmer to write a complex concurrent program without any concurrent language features is asking them to undertake a very difficult task.
+% They would only be able to rely on the atomicity that their hardware provides and would have to build up from there.
+% This would be like asking a programmer to write a complex sequential program only in assembly.
+% Both are doable, but would often be easier and less error prone with higher level tooling.
+Concurrent programming has many unique pitfalls that do not appear in sequential programming:
 \begin{enumerate}
 \item Deadlock, where threads cyclically wait on resources, blocking them indefinitely.
+\item Race conditions, where thread orderings can result in arbitrary behaviours, resulting in correctness problems.
 \item Livelock, where threads constantly attempt a concurrent operation unsuccessfully, resulting in no progress being made.
 \item Race conditions, where thread orderings can result in differing behaviours and correctness of a program execution.
 \item Starvation, where threads may be deprived of access to some shared resource due to unfairness and never make progress.
+\item Starvation, where \emph{some} threads constantly attempt a concurrent operation unsuccessfully, resulting in partial progress being made.
+\item Deadlock, where some threads wait for an event that cannot occur, blocking them indefinitely, resulting in no progress being made.
 \end{enumerate}
 Even with the guiding hand of concurrent tools these pitfalls can still catch unwary programmers, but good language support can prevent, detect, and mitigate these problems.
+Even with the guiding hand of concurrent tools these pitfalls still catch unwary programmers, but good language support helps significantly to prevent, detect, and mitigate these problems.
 \section{A Brief Overview}
+\section{Thesis Overview}
+The first chapter of this thesis aims to familiarize the reader with the language \CFA.
+In this chapter, syntax and features of the \CFA language that appear in this work are discussed The next chapter briefly discusses prior concurrency work in \CFA and how this work builds on top of existing features.
+The remaining chapters each introduce a concurrent language feature, discuss prior related work, and present contributions which are then benchmarked against other languages and systems.
+The first of these chapters discusses the @mutex@ statement, a language feature that improves ease of use and safety of lock usage.
+The @mutex@ statement is compared both in terms of safety and performance with similar tools in \CC and Java.
+The following chapter discusses channels, a message passing concurrency primitive that provides an avenue for safe synchronous and asynchronous communication across threads.
+Channels in \CFA are compared to Go, which popularized the use of channels in modern concurrent programs.
+The following chapter discusses the \CFA actor system.
+The \CFA actor system is a close cousin of channels, as it also belongs to the message passing paradigm of concurrency.
+However, the actor system provides a great degree of abstraction and ease of scalability, making it useful for a different range of problems than channels.
+The actor system in \CFA is compared with a variety of other systems on a suite of benchmarks, where it achieves significant performance gains over other systems due to its design.
+The final chapter discusses the \CFA @waituntil@ statement which provides the ability to synchronize while waiting for a resource, such as acquiring a lock, accessing a future, or writing to a channel.
+The @waituntil@ statement presented provides greater flexibility and expressibility than similar features in other languages.
+All in all, the features presented aim to fill in gaps in the current \CFA concurrent language support, and enable users to write a wider range of complex concurrent programs with ease.
+Chapter~\ref{s:cfa} of this thesis aims to familiarize the reader with the language \CFA.
+In this chapter, syntax and features of the \CFA language that appear in this work are discussed.
+Chapter~\ref{s:cfa_concurrency} briefly discusses prior concurrency work in \CFA, and how the work in this thesis builds on top of the existing framework.
+Each remaining chapter introduces an additional \CFA concurrent-language feature, which includes discussing prior related work for the feature, extensions over prior features, and uses benchmarks to compare the performance the feature with corresponding or similar features in other languages and systems.
+Chapter~\ref{s:mutexstmt} discusses the @mutex@ statement, a language feature that provides safe and simple lock usage.
+The @mutex@ statement is compared both in terms of safety and performance with similar mechanisms in \CC and Java.
+Chapter~\ref{s:channels} discusses channels, a message passing concurrency primitive that provides for safe synchronous and asynchronous communication among threads.
+Channels in \CFA are compared to Go's channels, which popularized the use of channels in modern concurrent programs.
+Chapter~\ref{s:actors} discusses the \CFA actor system.
+An actor system is a close cousin of channels, as it also belongs to the message passing paradigm of concurrency.
+However, an actor system provides a greater degree of abstraction and ease of scalability, making it useful for a different range of problems than channels.
+The actor system in \CFA is compared with a variety of other systems on a suite of benchmarks.
+Chapter~\ref{s:waituntil} discusses the \CFA @waituntil@ statement, which provides the ability to synchronize while waiting for a resource, such as acquiring a lock, accessing a future, or writing to a channel.
+The \CFA @waituntil@ statement provides greater flexibility and expressibility than similar features in other languages.
+All in all, the features presented aim to fill in gaps in the current \CFA concurrent-language support, enabling users to write a wider range of complex concurrent programs with ease.
 \section{Contributions}
 This work presents the following contributions:
 \begin{enumerate}
 \item The @mutex@ statement which:
 \begin{itemize}[itemsep=0pt]
 \item
 provides deadlock-free multiple lock acquisition,
 \item
 clearly denotes lock acquisition and release,
 \item
 and has good performance irrespective of lock ordering.
 \end{itemize}
 \item Channels which:
+This work presents the following contributions within each of the additional language features:
+\begin{enumerate}[leftmargin=*]
+\item The @mutex@ statement that:
+    \begin{itemize}[itemsep=0pt]
+    \item
+    provides deadlock-free multiple lock acquisition,
+    \item
+    clearly denotes lock acquisition and release,
+    \item
+    and has good performance irrespective of lock ordering.
+    \end{itemize}
+\item The channel that:
 \begin{itemize}[itemsep=0pt]
     \item
 …
     and provides toggle-able statistics for performance tuning.
 \end{itemize}
 \item An in-memory actor system that:
+\item The in-memory actor system that:
 \begin{itemize}[itemsep=0pt]
     \item
 …
     gains performance through static-typed message sends, eliminating the need for dynamic dispatch,
     \item
     introduces the copy queue, an array based queue specialized for the actor use case to minimize calls to the memory allocator,
+    introduces the copy queue, an array-based queue specialized for the actor use-case to minimize calls to the memory allocator,
     \item
     has robust detection of six tricky, but common actor programming errors,
 …
     and provides toggle-able statistics for performance tuning.
 \end{itemize}
+\item A @waituntil@ statement which:
+\item The @waituntil@ statement that:
 \begin{itemize}[itemsep=0pt]
     \item
     is the only known polymorphic synchronous multiplexing language feature,
     \item
     provides greater expressibility of waiting conditions than other languages,
+    provides greater expressibility for waiting conditions than other languages,
     \item
     and achieves comparable performance to similar features in two other languages,
+    and achieves comparable performance to similar features in two other languages.
 \end{itemize}
 \end{enumerate}

doc/theses/colby_parsons_MMAth/text/mutex_stmt.tex

-                      rc1e66d9
+                      rdeda7e6
 \end{figure}
 Like Java, \CFA monitors have \Newterm{multi-acquire} semantics so the thread in the monitor may acquire it multiple times without deadlock, allowing recursion and calling of other MX functions.
+Like Java, \CFA monitors have \Newterm{multi-acquire} (reentrant locking) semantics so the thread in the monitor may acquire it multiple times without deadlock, allowing recursion and calling of other MX functions.
 For robustness, \CFA monitors ensure the monitor lock is released regardless of how an acquiring function ends, normal or exceptional, and returning a shared variable is safe via copying before the lock is released.
 Monitor objects can be passed through multiple helper functions without acquiring mutual exclusion, until a designated function associated with the object is called.
 …
+}
 \end{cfa}
+The \CFA monitor implementation ensures multi-lock acquisition is done in a deadlock-free manner regardless of the number of MX parameters and monitor arguments. It it important to note that \CFA monitors do not attempt to solve the nested monitor problem~\cite{Lister77}.
+The \CFA monitor implementation ensures multi-lock acquisition is done in a deadlock-free manner regardless of the number of MX parameters and monitor arguments via resource ordering.
+It it important to note that \CFA monitors do not attempt to solve the nested monitor problem~\cite{Lister77}.
 \section{\lstinline{mutex} statement}
 …
 In detail, the mutex statement has a clause and statement block, similar to a conditional or loop statement.
 The clause accepts any number of lockable objects (like a \CFA MX function prototype), and locks them for the duration of the statement.
+The locks are acquired in a deadlock free manner and released regardless of how control-flow exits the statement.
+The locks are acquired in a deadlock-free manner and released regardless of how control-flow exits the statement.
+Note that this deadlock-freedom has some limitations \see{\VRef{s:DeadlockAvoidance}}.
 The mutex statement provides easy lock usage in the common case of lexically wrapping a CS.
 Examples of \CFA mutex statement are shown in \VRef[Listing]{l:cfa_mutex_ex}.
 …
 Like Java, \CFA introduces a new statement rather than building from existing language features, although \CFA has sufficient language features to mimic \CC RAII locking.
 This syntactic choice makes MX explicit rather than implicit via object declarations.
 Hence, it is easier for programmers and language tools to identify MX points in a program, \eg scan for all @mutex@ parameters and statements in a body of code.
+Hence, it is easy for programmers and language tools to identify MX points in a program, \eg scan for all @mutex@ parameters and statements in a body of code; similar scanning can be done with Java's @synchronized@.
 Furthermore, concurrent safety is provided across an entire program for the complex operation of acquiring multiple locks in a deadlock-free manner.
 Unlike Java, \CFA's mutex statement and \CC's @scoped_lock@ both use parametric polymorphism to allow user defined types to work with this feature.
 …
 thread$\(_2\)$ : sout | "uvw" | "xyz";
 \end{cfa}
 any of the outputs can appear, included a segment fault due to I/O buffer corruption:
+any of the outputs can appear:
 \begin{cquote}
 \small\tt
 …
 mutex( sout ) { // acquire stream lock for sout for block duration
         sout | "abc";
         mutex( sout ) sout | "uvw" | "xyz"; // OK because sout lock is recursive
+        sout | "uvw" | "xyz";
         sout | "def";
 } // implicitly release sout lock
 \end{cfa}
-The inner lock acquire is likely to occur through a function call that does a thread-safe print.
 \section{Deadlock Avoidance}\label{s:DeadlockAvoidance}
 …
 For fewer than 7 locks ($2^3-1$), the sort is unrolled performing the minimum number of compare and swaps for the given number of locks;
 for 7 or more locks, insertion sort is used.
+Since it is extremely rare to hold more than 6 locks at a time, the algorithm is fast and executes in $O(1)$ time.
+Furthermore, lock addresses are unique across program execution, even for dynamically allocated locks, so the algorithm is safe across the entire program execution.
+It is assumed to be rare to hold more than 6 locks at a time.
+For 6 or fewer locks the algorithm is fast and executes in $O(1)$ time.
+Furthermore, lock addresses are unique across program execution, even for dynamically allocated locks, so the algorithm is safe across the entire program execution, as long as lifetimes of objects are appropriately managed.
+For example, deleting a lock and allocating another one could give the new lock the same address as the deleted one, however deleting a lock in use by another thread is a programming error irrespective of the usage of the @mutex@ statement.
 The downside to the sorting approach is that it is not fully compatible with manual usages of the same locks outside the @mutex@ statement, \ie the lock are acquired without using the @mutex@ statement.
 …
 \end{cquote}
 Comparatively, if the @scoped_lock@ is used and the same locks are acquired elsewhere, there is no concern of the @scoped_lock@ deadlocking, due to its avoidance scheme, but it may livelock.
 The convenience and safety of the @mutex@ statement, \ie guaranteed lock release with exceptions, should encourage programmers to always use it for locking, mitigating any deadlock scenario versus combining manual locking with the mutex statement.
+The convenience and safety of the @mutex@ statement, \ie guaranteed lock release with exceptions, should encourage programmers to always use it for locking, mitigating most deadlock scenarios versus combining manual locking with the mutex statement.
 Both \CC and the \CFA do not provide any deadlock guarantees for nested @scoped_lock@s or @mutex@ statements.
 To do so would require solving the nested monitor problem~\cite{Lister77}, which currently does not have any practical solutions.
 …
 \section{Performance}
 Given the two multi-acquisition algorithms in \CC and \CFA, each with differing advantages and disadvantages, it interesting to compare their performance.
 Comparison with Java is not possible, since it only takes a single lock.
+Comparison with Java was not conducted, since the synchronized statement only takes a single object and does not provide deadlock avoidance or prevention.
 The comparison starts with a baseline that acquires the locks directly without a mutex statement or @scoped_lock@ in a fixed ordering and then releases them.
 …
 Each variation is run 11 times on 2, 4, 8, 16, 24, 32 cores and with 2, 4, and 8 locks being acquired.
 The median is calculated and is plotted alongside the 95\% confidence intervals for each point.
+The confidence intervals are calculated using bootstrapping to avoid normality assumptions.
 \begin{figure}
 …
+}
 \end{cfa}
 \caption{Deadlock avoidance benchmark pseudocode}
+\caption{Deadlock avoidance benchmark \CFA pseudocode}
 \label{l:deadlock_avoid_pseudo}
 \end{figure}
 …
 % sudo dmidecode -t system
 \item
 Supermicro AS--1123US--TR4 AMD EPYC 7662 64--core socket, hyper-threading $\times$ 2 sockets (256 processing units) 2.0 GHz, TSO memory model, running Linux v5.8.0--55--generic, gcc--10 compiler
+Supermicro AS--1123US--TR4 AMD EPYC 7662 64--core socket, hyper-threading $\times$ 2 sockets (256 processing units), TSO memory model, running Linux v5.8.0--55--generic, gcc--10 compiler
 \item
 Supermicro SYS--6029U--TR4 Intel Xeon Gold 5220R 24--core socket, hyper-threading $\times$ 2 sockets (48 processing units) 2.2GHz, TSO memory model, running Linux v5.8.0--59--generic, gcc--10 compiler
+Supermicro SYS--6029U--TR4 Intel Xeon Gold 5220R 24--core socket, hyper-threading $\times$ 2 sockets (96 processing units), TSO memory model, running Linux v5.8.0--59--generic, gcc--10 compiler
 \end{list}
 %The hardware architectures are different in threading (multithreading vs hyper), cache structure (MESI or MESIF), NUMA layout (QPI vs HyperTransport), memory model (TSO vs WO), and energy/thermal mechanisms (turbo-boost).
 …
 For example, on the AMD machine with 32 threads and 8 locks, the benchmarks would occasionally livelock indefinitely, with no threads making any progress for 3 hours before the experiment was terminated manually.
 It is likely that shorter bouts of livelock occurred in many of the experiments, which would explain large confidence intervals for some of the data points in the \CC data.
 In Figures~\ref{f:mutex_bench8_AMD} and \ref{f:mutex_bench8_Intel} there is the counter-intuitive result of the mutex statement performing better than the baseline.
+In Figures~\ref{f:mutex_bench8_AMD} and \ref{f:mutex_bench8_Intel} there is the counter-intuitive result of the @mutex@ statement performing better than the baseline.
 At 7 locks and above the mutex statement switches from a hard coded sort to insertion sort, which should decrease performance.
 The hard coded sort is branch-free and constant-time and was verified to be faster than insertion sort for 6 or fewer locks.
+It is likely the increase in throughput compared to baseline is due to the delay spent in the insertion sort, which decreases contention on the locks.
+Part of the difference in throughput compared to baseline is due to the delay spent in the insertion sort, which decreases contention on the locks.
+This was verified to be part of the difference in throughput by experimenting with varying NCS delays in the baseline; however it only comprises a small portion of difference.
+It is possible that the baseline is slowed down or the @mutex@ is sped up by other factors that are not easily identifiable.
 \begin{figure}

doc/theses/colby_parsons_MMAth/text/waituntil.tex

-                      rc1e66d9
+                      rdeda7e6
 Go's @select@ has the same exclusive-or semantics as the ALT primitive from Occam and associated code blocks for each clause like ALT and Ada.
 However, unlike Ada and ALT, Go does not provide guards for the \lstinline[language=go]{case} clauses of the \lstinline[language=go]{select}.
 As such, the exponential blowup can be seen comparing Go and \uC in Figure~\label{f:AdaMultiplexing}.
+As such, the exponential blowup can be seen comparing Go and \uC in Figure~\ref{f:AdaMultiplexing}.
 Go also provides a timeout via a channel and a @default@ clause like Ada @else@ for asynchronous multiplexing.
 …
 In following example, either channel @C1@ or @C2@ must be satisfied but nothing can be done for at least 1 or 3 seconds after the channel read, respectively.
 \begin{cfa}[deletekeywords={timeout}]
 waituntil( i << C1 ); and waituntil( timeout( 1`s ) );
 or waituntil( i << C2 ); and waituntil( timeout( 3`s ) );
+waituntil( i << C1 ){} and waituntil( timeout( 1`s ) ){}
+or waituntil( i << C2 ){} and waituntil( timeout( 3`s ) ){}
 \end{cfa}
 If only @C2@ is satisfied, \emph{both} timeout code-blocks trigger because 1 second occurs before 3 seconds.
 …
 Now the unblocked WUT is guaranteed to have a satisfied resource and its code block can safely executed.
 The insertion circumvents the channel buffer via the wait-morphing in the \CFA channel implementation \see{Section~\ref{s:chan_impl}}, allowing @waituntil@ channel unblocking to not be special-cased.
+Note that all channel operations are fair and no preference is given between @waituntil@ and direct channel operations when unblocking.
 Furthermore, if both @and@ and @or@ operators are used, the @or@ operations stop behaving like exclusive-or due to the race among channel operations, \eg:

libcfa/prelude/extras.c

rc1e66d9	rdeda7e6
3	3	#include <uchar.h> // char16_t, char32_t
4	4	#include <wchar.h> // wchar_t
5		#include <stdlib.h> // malloc, free, ~~exit, atexit, abort~~
	5	#include <stdlib.h> // malloc, free, getenv, exit, atexit, abort, printf
6	6	#include <stdio.h> // printf
	7	#include <string.h> // strlen, strcmp, strncmp

libcfa/prelude/extras.regx2

-                      rc1e66d9
+                      rdeda7e6
 extern void \*malloc[^;]*;
 extern void free[^;]*;
+extern char \*getenv[^;]*;
 extern void exit[^;]*;
 extern int atexit[^;]*;
 extern void abort[^;]*;
 extern int printf[^;]*;
+int strcmp[^;]*;
+int strncmp[^;]*;
+size_t strlen[^;]*;

libcfa/src/Makefile.am

-                      rc1e66d9
+                      rdeda7e6
 ## Created On       : Sun May 31 08:54:01 2015
 ## Last Modified By : Peter A. Buhr
 ## Last Modified On : Wed Aug 30 21:22:45 2023
 ## Update Count     : 263
+## Last Modified On : Mon Sep 18 17:06:56 2023
+## Update Count     : 264
 ###############################################################################
 …
         concurrency/mutex_stmt.hfa \
         concurrency/channel.hfa \
         concurrency/actor.hfa
+        concurrency/actor.hfa
 inst_thread_headers_src = \
 …
         concurrency/mutex.hfa \
         concurrency/select.hfa \
+        concurrency/thread.hfa
+        concurrency/thread.hfa \
+        concurrency/cofor.hfa
 thread_libsrc = ${inst_thread_headers_src} ${inst_thread_headers_src:.hfa=.cfa} \

libcfa/src/clock.hfa

-                      rc1e66d9
+                      rdeda7e6
 // Created On       : Thu Apr 12 14:36:06 2018
 // Last Modified By : Peter A. Buhr
 // Last Modified On : Sun Apr 18 08:12:16 2021
 // Update Count     : 28
+// Last Modified On : Sat Sep  9 14:07:17 2023
+// Update Count     : 30
 //
 …
         // discontinuous jumps when the OS is not running the kernal thread. A duration is returned because the value is
         // relative and cannot be converted to real-time (wall-clock) time.
         Duration processor() {                                                          // non-monotonic duration of kernel thread
+        Duration processor_cpu() {                                                      // non-monotonic duration of kernel thread
                 timespec ts;
                 clock_gettime( CLOCK_THREAD_CPUTIME_ID, &ts );
                 return (Duration){ ts };
         } // processor
+        } // processor_cpu
         // Program CPU-time watch measures CPU time consumed by all processors (kernel threads) in the UNIX process.  This
         // watch is affected by discontinuous jumps when the OS is not running the kernel threads. A duration is returned
         // because the value is relative and cannot be converted to real-time (wall-clock) time.
         Duration program() {                                                            // non-monotonic duration of program CPU
+        Duration program_cpu() {                                                        // non-monotonic duration of program CPU
                 timespec ts;
                 clock_gettime( CLOCK_PROCESS_CPUTIME_ID, &ts );
                 return (Duration){ ts };
         } // program
+        } // program_cpu
         // Monotonic duration from machine boot and including system suspension. This watch is unaffected by discontinuous

libcfa/src/collections/string.cfa

-                      rc1e66d9
+                      rdeda7e6
 // Comparison
+bool ?==?(const string & s, const string & other) {
+    return *s.inner == *other.inner;
+}
+bool ?!=?(const string & s, const string & other) {
+    return *s.inner != *other.inner;
+}
+bool ?==?(const string & s, const char * other) {
+    return *s.inner == other;
+}
+bool ?!=?(const string & s, const char * other) {
+    return *s.inner != other;
+}
+int  cmp (const string &s1, const string &s2) { return cmp(*s1.inner ,  *s2.inner); }
+bool ?==?(const string &s1, const string &s2) { return     *s1.inner == *s2.inner ; }
+bool ?!=?(const string &s1, const string &s2) { return     *s1.inner != *s2.inner ; }
+bool ?>? (const string &s1, const string &s2) { return     *s1.inner >  *s2.inner ; }
+bool ?>=?(const string &s1, const string &s2) { return     *s1.inner >= *s2.inner ; }
+bool ?<=?(const string &s1, const string &s2) { return     *s1.inner <= *s2.inner ; }
+bool ?<? (const string &s1, const string &s2) { return     *s1.inner <  *s2.inner ; }
+int  cmp (const string &s1, const char*   s2) { return cmp(*s1.inner ,   s2      ); }
+bool ?==?(const string &s1, const char*   s2) { return     *s1.inner ==  s2       ; }
+bool ?!=?(const string &s1, const char*   s2) { return     *s1.inner !=  s2       ; }
+bool ?>? (const string &s1, const char*   s2) { return     *s1.inner >   s2       ; }
+bool ?>=?(const string &s1, const char*   s2) { return     *s1.inner >=  s2       ; }
+bool ?<=?(const string &s1, const char*   s2) { return     *s1.inner <=  s2       ; }
+bool ?<? (const string &s1, const char*   s2) { return     *s1.inner <   s2       ; }
+int  cmp (const char*   s1, const string &s2) { return cmp( s1       ,  *s2.inner); }
+bool ?==?(const char*   s1, const string &s2) { return      s1       == *s2.inner ; }
+bool ?!=?(const char*   s1, const string &s2) { return      s1       != *s2.inner ; }
+bool ?>? (const char*   s1, const string &s2) { return      s1       >  *s2.inner ; }
+bool ?>=?(const char*   s1, const string &s2) { return      s1       >= *s2.inner ; }
+bool ?<=?(const char*   s1, const string &s2) { return      s1       <= *s2.inner ; }
+bool ?<? (const char*   s1, const string &s2) { return      s1       <  *s2.inner ; }
 ////////////////////////////////////////////////////////

libcfa/src/collections/string.hfa

-                      rc1e66d9
+                      rdeda7e6
 // Comparisons
+bool ?==?(const string & s, const string & other);
+bool ?!=?(const string & s, const string & other);
+bool ?==?(const string & s, const char * other);
+bool ?!=?(const string & s, const char * other);
+int  cmp (const string &, const string &);
+bool ?==?(const string &, const string &);
+bool ?!=?(const string &, const string &);
+bool ?>? (const string &, const string &);
+bool ?>=?(const string &, const string &);
+bool ?<=?(const string &, const string &);
+bool ?<? (const string &, const string &);
+int  cmp (const string &, const char*);
+bool ?==?(const string &, const char*);
+bool ?!=?(const string &, const char*);
+bool ?>? (const string &, const char*);
+bool ?>=?(const string &, const char*);
+bool ?<=?(const string &, const char*);
+bool ?<? (const string &, const char*);
+int  cmp (const char*, const string &);
+bool ?==?(const char*, const string &);
+bool ?!=?(const char*, const string &);
+bool ?>? (const char*, const string &);
+bool ?>=?(const char*, const string &);
+bool ?<=?(const char*, const string &);
+bool ?<? (const char*, const string &);
 // Slicing

libcfa/src/collections/string_res.cfa

-                      rc1e66d9
+                      rdeda7e6
 // Comparisons
+bool ?==?(const string_res &s1, const string_res &s2) {
+    return ByteCmp( s1.Handle.s, 0, s1.Handle.lnth, s2.Handle.s, 0, s2.Handle.lnth) == 0;
+}
+bool ?!=?(const string_res &s1, const string_res &s2) {
+    return !(s1 == s2);
+}
+bool ?==?(const string_res &s, const char* other) {
+    string_res sother = other;
+    return s == sother;
+}
+bool ?!=?(const string_res &s, const char* other) {
+    return !(s == other);
+}
+int cmp(const string_res &s1, const string_res &s2) {
+    // return 0;
+    int ans1 = memcmp(s1.Handle.s, s2.Handle.s, min(s1.Handle.lnth, s2.Handle.lnth));
+    if (ans1 != 0) return ans1;
+    return s1.Handle.lnth - s2.Handle.lnth;
+}
+bool ?==?(const string_res &s1, const string_res &s2) { return cmp(s1, s2) == 0; }
+bool ?!=?(const string_res &s1, const string_res &s2) { return cmp(s1, s2) != 0; }
+bool ?>? (const string_res &s1, const string_res &s2) { return cmp(s1, s2) >  0; }
+bool ?>=?(const string_res &s1, const string_res &s2) { return cmp(s1, s2) >= 0; }
+bool ?<=?(const string_res &s1, const string_res &s2) { return cmp(s1, s2) <= 0; }
+bool ?<? (const string_res &s1, const string_res &s2) { return cmp(s1, s2) <  0; }
+int cmp (const string_res &s1, const char* s2) {
+    string_res s2x = s2;
+    return cmp(s1, s2x);
+}
+bool ?==?(const string_res &s1, const char* s2) { return cmp(s1, s2) == 0; }
+bool ?!=?(const string_res &s1, const char* s2) { return cmp(s1, s2) != 0; }
+bool ?>? (const string_res &s1, const char* s2) { return cmp(s1, s2) >  0; }
+bool ?>=?(const string_res &s1, const char* s2) { return cmp(s1, s2) >= 0; }
+bool ?<=?(const string_res &s1, const char* s2) { return cmp(s1, s2) <= 0; }
+bool ?<? (const string_res &s1, const char* s2) { return cmp(s1, s2) <  0; }
+int cmp (const char* s1, const string_res & s2) {
+    string_res s1x = s1;
+    return cmp(s1x, s2);
+}
+bool ?==?(const char* s1, const string_res &s2) { return cmp(s1, s2) == 0; }
+bool ?!=?(const char* s1, const string_res &s2) { return cmp(s1, s2) != 0; }
+bool ?>? (const char* s1, const string_res &s2) { return cmp(s1, s2) >  0; }
+bool ?>=?(const char* s1, const string_res &s2) { return cmp(s1, s2) >= 0; }
+bool ?<=?(const char* s1, const string_res &s2) { return cmp(s1, s2) <= 0; }
+bool ?<? (const char* s1, const string_res &s2) { return cmp(s1, s2) <  0; }

libcfa/src/collections/string_res.hfa

-                      rc1e66d9
+                      rdeda7e6
 // Comparisons
+bool ?==?(const string_res &s, const string_res &other);
+bool ?!=?(const string_res &s, const string_res &other);
+bool ?==?(const string_res &s, const char* other);
+bool ?!=?(const string_res &s, const char* other);
+int  cmp (const string_res &, const string_res &);
+bool ?==?(const string_res &, const string_res &);
+bool ?!=?(const string_res &, const string_res &);
+bool ?>? (const string_res &, const string_res &);
+bool ?>=?(const string_res &, const string_res &);
+bool ?<=?(const string_res &, const string_res &);
+bool ?<? (const string_res &, const string_res &);
+int  cmp (const string_res &, const char*);
+bool ?==?(const string_res &, const char*);
+bool ?!=?(const string_res &, const char*);
+bool ?>? (const string_res &, const char*);
+bool ?>=?(const string_res &, const char*);
+bool ?<=?(const string_res &, const char*);
+bool ?<? (const string_res &, const char*);
+int  cmp (const char*, const string_res &);
+bool ?==?(const char*, const string_res &);
+bool ?!=?(const char*, const string_res &);
+bool ?>? (const char*, const string_res &);
+bool ?>=?(const char*, const string_res &);
+bool ?<=?(const char*, const string_res &);
+bool ?<? (const char*, const string_res &);
 // String search

libcfa/src/common.hfa

-                      rc1e66d9
+                      rdeda7e6
         T min( T v1, T v2 ) { return v1 < v2 ? v1 : v2; }
         forall( T, Ts... | { T min( T, T ); T min( T, Ts ); } )
         T min( T v1, T v2, Ts vs ) { return min( min( v1, v2 ), vs ); }
+        forall( T, Ts... | { T min( T, T ); T min( T, T, Ts ); } )
+        T min( T v1, T v2, T v3, Ts vs ) { return min( min( v1, v2 ), v3, vs ); }
         forall( T | { int ?>?( T, T ); } )
         T max( T v1, T v2 ) { return v1 > v2 ? v1 : v2; }
         forall( T, Ts... | { T max( T, T ); T max( T, Ts ); } )
         T max( T v1, T v2, Ts vs ) { return max( max( v1, v2 ), vs ); }
+        forall( T, Ts... | { T max( T, T ); T max( T, T, Ts ); } )
+        T max( T v1, T v2, T v3, Ts vs ) { return max( max( v1, v2 ), v3, vs ); }
         forall( T | { T min( T, T ); T max( T, T ); } )

libcfa/src/concurrency/coroutine.cfa

-                      rc1e66d9
+                      rdeda7e6
 // Created On       : Mon Nov 28 12:27:26 2016
 // Last Modified By : Peter A. Buhr
 // Last Modified On : Thu Feb 16 15:34:46 2023
 // Update Count     : 24
+// Last Modified On : Mon Sep 18 21:47:12 2023
+// Update Count     : 25
 //
 …
 // resume non local exception at receiver (i.e. enqueue in ehm buffer)
 forall(exceptT *, T & | ehm_resume_at( exceptT, T ))
 void resumeAt( T & receiver, exceptT & ex )  libcfa_public {
+void resumeAt( T & receiver, exceptT & ex ) libcfa_public {
     coroutine$ * cor = get_coroutine( receiver );
     nonlocal_exception * nl_ex = alloc();

libcfa/src/concurrency/kernel/cluster.hfa

-                      rc1e66d9
+                      rdeda7e6
 // warn normally all ints
 #define warn_large_before warnf( !strict || old_avg < 33_000_000_000, "Suspiciously large previous average: %'llu (%llx), %'" PRId64 "ms \n", old_avg, old_avg, program()`ms )
 #define warn_large_after warnf( !strict || ret < 33_000_000_000, "Suspiciously large new average after %'" PRId64 "ms cputime: %'llu (%llx) from %'llu-%'llu (%'llu, %'llu) and %'llu\n", program()`ms, ret, ret, currtsc, intsc, new_val, new_val / 1000000, old_avg )
+#define warn_large_before warnf( !strict || old_avg < 33_000_000_000, "Suspiciously large previous average: %'llu (%llx), %'" PRId64 "ms \n", old_avg, old_avg, program_cpu()`ms )
+#define warn_large_after warnf( !strict || ret < 33_000_000_000, "Suspiciously large new average after %'" PRId64 "ms cputime: %'llu (%llx) from %'llu-%'llu (%'llu, %'llu) and %'llu\n", program_cpu()`ms, ret, ret, currtsc, intsc, new_val, new_val / 1000000, old_avg )
 // 8X linear factor is just 8 * x
 …
 static inline __readyQ_avg_t __to_readyQ_avg(unsigned long long intsc) { if(unlikely(0 == intsc)) return 0.0; else return log2((__readyQ_avg_t)intsc); }
 #define warn_large_before warnf( !strict || old_avg < 35.0, "Suspiciously large previous average: %'lf, %'" PRId64 "ms \n", old_avg, program()`ms )
 #define warn_large_after warnf( !strict || ret < 35.3, "Suspiciously large new average after %'" PRId64 "ms cputime: %'lf from %'llu-%'llu (%'llu, %'llu) and %'lf\n", program()`ms, ret, currtsc, intsc, new_val, new_val / 1000000, old_avg ); \
+#define warn_large_before warnf( !strict || old_avg < 35.0, "Suspiciously large previous average: %'lf, %'" PRId64 "ms \n", old_avg, program_cpu()`ms )
+#define warn_large_after warnf( !strict || ret < 35.3, "Suspiciously large new average after %'" PRId64 "ms cputime: %'lf from %'llu-%'llu (%'llu, %'llu) and %'lf\n", program_cpu()`ms, ret, currtsc, intsc, new_val, new_val / 1000000, old_avg ); \
 verify(ret >= 0)

libcfa/src/heap.cfa

-                      rc1e66d9
+                      rdeda7e6
 // Created On       : Tue Dec 19 21:58:35 2017
 // Last Modified By : Peter A. Buhr
 // Last Modified On : Wed Aug  2 18:48:30 2023
 // Update Count     : 1614
+// Last Modified On : Mon Sep 11 11:21:10 2023
+// Update Count     : 1615
 //
 …
         return stats;
 } // collectStats
+static inline void clearStats() {
+        lock( mgrLock );
+        // Zero the heap master and all active thread heaps.
+        HeapStatisticsCtor( heapMaster.stats );
+        for ( Heap * heap = heapMaster.heapManagersList; heap; heap = heap->nextHeapManager ) {
+                HeapStatisticsCtor( heap->stats );
+        } // for
+        unlock( mgrLock );
+} // clearStats
 #endif // __STATISTICS__
 …
+        // Zero the heap master and all active thread heaps.
+        void malloc_stats_clear() {
+                #ifdef __STATISTICS__
+                clearStats();
+                #else
+                #define MALLOC_STATS_MSG "malloc_stats statistics disabled.\n"
+                if ( write( STDERR_FILENO, MALLOC_STATS_MSG, sizeof( MALLOC_STATS_MSG ) - 1 /* size includes '\0' */ ) == -1 ) {
+                        abort( "**** Error **** write failed in malloc_stats" );
+                } // if
+                #endif // __STATISTICS__
+        } // malloc_stats_clear
         // Changes the file descriptor where malloc_stats() writes statistics.
         int malloc_stats_fd( int fd __attribute__(( unused )) ) libcfa_public {

libcfa/src/heap.hfa

-                      rc1e66d9
+                      rdeda7e6
 // Created On       : Tue May 26 11:23:55 2020
 // Last Modified By : Peter A. Buhr
 // Last Modified On : Tue Oct  4 19:08:55 2022
 // Update Count     : 23
+// Last Modified On : Mon Sep 11 11:18:18 2023
+// Update Count     : 24
 //
 …
         size_t malloc_mmap_start();                                                     // crossover allocation size from sbrk to mmap
         size_t malloc_unfreed();                                                        // heap unfreed size (bytes)
+        void malloc_stats_clear();                                                      // clear heap statistics
 } // extern "C"

libcfa/src/iostream.cfa

-                      rc1e66d9
+                      rdeda7e6
 //
 // Cforall Version 1.0.0 Copyright (C) 2015 University of Waterloo
 …
                 if ( f.flags.ignore ) { fmtstr[1] = '*'; start += 1; }
                 // no maximum width necessary because text ignored => width is read width
+                if ( f.wd != -1 ) { start += sprintf( &fmtstr[start], "%d", f.wd ); }
+                if ( f.wd != -1 ) {
+                        // wd is buffer bytes available (for input chars + null terminator)
+                        // rwd is count of input chars
+                        int rwd = f.flags.rwd ? f.wd : (f.wd - 1);
+                        start += sprintf( &fmtstr[start], "%d", rwd );
+                }
                 if ( ! scanset ) {
 …
                 } // if
                 int check = f.wd - 1;
+                int check = f.wd - 2;
                 if ( ! f.flags.rwd ) f.s[check] = '\0';                 // insert sentinel
                 len = fmt( is, fmtstr, f.s );

src/AST/Util.cpp

-                      rc1e66d9
+                      rdeda7e6
+        }
         assertf( false, "Member not found." );
+}
+template<typename node_t>
+void oneOfExprOrType( const node_t * node ) {
+        if ( node->expr ) {
+                assertf( node->expr && !node->type, "Exactly one of expr or type should be set." );
+        } else {
+                assertf( !node->expr && node->type, "Exactly one of expr or type should be set." );
+        }
+}
 …
+        }
+        void previsit( const SizeofExpr * node ) {
+                previsit( (const ParseNode *)node );
+                oneOfExprOrType( node );
+        }
+        void previsit( const AlignofExpr * node ) {
+                previsit( (const ParseNode *)node );
+                oneOfExprOrType( node );
+        }
         void previsit( const StructInstType * node ) {
                 previsit( (const Node *)node );
 …
 /// referring to is in scope by the structural rules of code.
 // Any escapes marked with a bug should be removed once the bug is fixed.
+// This is a separate pass because of it changes the visit pattern and
+// must always be run on the entire translation unit.
 struct InScopeCore : public ast::WithShortCircuiting {
         ScopedSet<DeclWithType const *> typedDecls;

src/ControlStruct/MultiLevelExit.cpp

-                      rc1e66d9
+                      rdeda7e6
 // Created On       : Mon Nov  1 13:48:00 2021
 // Last Modified By : Andrew Beach
 // Last Modified On : Wed Sep  6 12:00:00 2023
 // Update Count     : 35
+// Last Modified On : Fri Sep  8 17:04:00 2023
+// Update Count     : 36
 //
 …
 namespace {
+/// The return context is used to remember if returns are allowed and if
+/// not, why not. It is the nearest local control flow blocking construct.
+enum ReturnContext {
+        MayReturn,
+        InTryWithHandler,
+        InResumeHandler,
+        InTerminateHandler,
+        InFinally,
+};
 class Entry {
 …
         void previsit( const TryStmt * );
         void postvisit( const TryStmt * );
+        void previsit( const CatchClause * );
         void previsit( const FinallyClause * );
 …
         vector<Entry> enclosing_control_structures;
         Label break_label;
         bool inFinally;
+        ReturnContext ret_context;
         template<typename LoopNode>
 …
                 const list<ptr<Stmt>> & kids, bool caseClause );
+        void enterSealedContext( ReturnContext );
         template<typename UnaryPredicate>
         auto findEnclosingControlStructure( UnaryPredicate pred ) {
 …
 MultiLevelExitCore::MultiLevelExitCore( const LabelToStmt & lt ) :
         target_table( lt ), break_label( CodeLocation(), "" ),
         inFinally( false )
+        ret_context( ReturnContext::MayReturn )
 {}
 …
 void MultiLevelExitCore::previsit( const ReturnStmt * stmt ) {
+        if ( inFinally ) {
+                SemanticError( stmt->location, "'return' may not appear in a finally clause" );
+        }
+        char const * context;
+        switch ( ret_context ) {
+        case ReturnContext::MayReturn:
+                return;
+        case ReturnContext::InTryWithHandler:
+                context = "try statement with a catch clause";
+                break;
+        case ReturnContext::InResumeHandler:
+                context = "catchResume clause";
+                break;
+        case ReturnContext::InTerminateHandler:
+                context = "catch clause";
+                break;
+        case ReturnContext::InFinally:
+                context = "finally clause";
+                break;
+        default:
+                assert(0);
+        }
+        SemanticError( stmt->location, toString( "'return' may not appear in a ", context ) );
+}
 …
                 GuardAction([this](){ enclosing_control_structures.pop_back(); } );
+        }
+        // Try statements/try blocks are only sealed with a termination handler.
+        for ( auto clause : stmt->handlers ) {
+                if ( ast::Terminate == clause->kind ) {
+                        return enterSealedContext( ReturnContext::InTryWithHandler );
+                }
+        }
+}
 …
+}
+void MultiLevelExitCore::previsit( const CatchClause * clause ) {
+        ReturnContext context = ( ast::Terminate == clause->kind )
+                ? ReturnContext::InTerminateHandler : ReturnContext::InResumeHandler;
+        enterSealedContext( context );
+}
 void MultiLevelExitCore::previsit( const FinallyClause * ) {
+        GuardAction([this, old = std::move( enclosing_control_structures)](){ enclosing_control_structures = std::move(old); });
+        enclosing_control_structures = vector<Entry>();
+        GuardValue( inFinally ) = true;
+        enterSealedContext( ReturnContext::InFinally );
+}
 …
+}
+void MultiLevelExitCore::enterSealedContext( ReturnContext enter_context ) {
+        GuardAction([this, old = std::move(enclosing_control_structures)](){ enclosing_control_structures = std::move(old); });
+        enclosing_control_structures = vector<Entry>();
+        GuardValue( ret_context ) = enter_context;
+}
 } // namespace

src/GenPoly/GenPoly.cc

-                      rc1e66d9
+                      rdeda7e6
+                }
+                bool hasPolyParams( const std::vector<ast::ptr<ast::Expr>> & params, const ast::TypeSubstitution * env) {
+                        for (auto &param : params) {
+                                auto paramType = param.strict_as<ast::TypeExpr>();
+                                if (isPolyType(paramType->type, env)) return true;
+                bool hasPolyParams( const std::vector<ast::ptr<ast::Expr>> & params, const ast::TypeSubstitution * env ) {
+                        for ( auto &param : params ) {
+                                auto paramType = param.as<ast::TypeExpr>();
+                                assertf( paramType, "Aggregate parameters should be type expressions" );
+                                if ( isPolyType( paramType->type, env ) ) return true;
+                        }
                         return false;
 …
                                 assertf(paramType, "Aggregate parameters should be type expressions");
                                 if ( isPolyType( paramType->get_type(), tyVars, env ) ) return true;
+                        }
+                        return false;
+                }
+                bool hasPolyParams( const std::vector<ast::ptr<ast::Expr>> & params, const TypeVarMap & typeVars, const ast::TypeSubstitution * env ) {
+                        for ( auto & param : params ) {
+                                auto paramType = param.as<ast::TypeExpr>();
+                                assertf( paramType, "Aggregate parameters should be type expressions" );
+                                if ( isPolyType( paramType->type, typeVars, env ) ) return true;
+                        }
                         return false;
 …
+        }
-        const ast::Type * isPolyType(const ast::Type * type, const TyVarMap & tyVars, const ast::TypeSubstitution * env) {
-                type = replaceTypeInst( type, env );
-                if ( auto typeInst = dynamic_cast< const ast::TypeInstType * >( type ) ) {
-                        if ( tyVars.contains( typeInst->typeString() ) ) return type;
-                } else if ( auto arrayType = dynamic_cast< const ast::ArrayType * >( type ) ) {
-                        return isPolyType( arrayType->base, env );
-                } else if ( auto structType = dynamic_cast< const ast::StructInstType* >( type ) ) {
-                        if ( hasPolyParams( structType->params, env ) ) return type;
-                } else if ( auto unionType = dynamic_cast< const ast::UnionInstType* >( type ) ) {
-                        if ( hasPolyParams( unionType->params, env ) ) return type;
+                }
-                return nullptr;
+        }
 const ast::Type * isPolyType( const ast::Type * type,
                 const TypeVarMap & typeVars, const ast::TypeSubstitution * subst ) {
 …
                 if ( typeVars.contains( *inst ) ) return type;
         } else if ( auto array = dynamic_cast< const ast::ArrayType * >( type ) ) {
                 return isPolyType( array->base, subst );
+                return isPolyType( array->base, typeVars, subst );
         } else if ( auto sue = dynamic_cast< const ast::StructInstType * >( type ) ) {
                 if ( hasPolyParams( sue->params, subst ) ) return type;
+                if ( hasPolyParams( sue->params, typeVars, subst ) ) return type;
         } else if ( auto sue = dynamic_cast< const ast::UnionInstType * >( type ) ) {
                 if ( hasPolyParams( sue->params, subst ) ) return type;
+                if ( hasPolyParams( sue->params, typeVars, subst ) ) return type;
+        }
         return nullptr;

tests/.expect/minmax.txt

-                      rc1e66d9
+                      rdeda7e6
 double                  4. 3.1  max 4.
 long double             4. 3.1  max 4.
+arguments
+3 4   min 2   max 4
+2 3   min 2   max 4
+4 2   min 2   max 4
+arguments
+2 5 4 min 2   max 5
+3 4 2 min 2   max 5

tests/concurrency/actors/dynamic.cfa

-                      rc1e66d9
+                      rdeda7e6
 struct derived_actor { inline actor; };
 struct derived_msg {
     inline message;
     int cnt;
+        inline message;
+        int cnt;
 };
 void ?{}( derived_msg & this, int cnt ) {
     ((message &) this){ Delete };
     this.cnt = cnt;
+        set_allocation( this, Delete );
+        this.cnt = cnt;
+}
 void ?{}( derived_msg & this ) { ((derived_msg &)this){ 0 }; }
 allocation receive( derived_actor & receiver, derived_msg & msg ) {
     if ( msg.cnt >= Times ) {
         sout | "Done";
         return Delete;
+    }
     derived_msg * d_msg = alloc();
     (*d_msg){ msg.cnt + 1 };
     derived_actor * d_actor = alloc();
     (*d_actor){};
     *d_actor | *d_msg;
     return Delete;
+        if ( msg.cnt >= Times ) {
+                sout | "Done";
+                return Delete;
+        }
+        derived_msg * d_msg = alloc();
+        (*d_msg){ msg.cnt + 1 };
+        derived_actor * d_actor = alloc();
+        (*d_actor){};
+        *d_actor | *d_msg;
+        return Delete;
+}
 int main( int argc, char * argv[] ) {
     switch ( argc ) {
+        switch ( argc ) {
           case 2:
                 if ( strcmp( argv[1], "d" ) != 0 ) {                    // default ?
                         Times = atoi( argv[1] );
                         if ( Times < 1 ) goto Usage;
+                        Times = ato( argv[1] );
+                        if ( Times < 1 ) fallthru default;
                 } // if
           case 1:                                                                                       // use defaults
                 break;
           default:
+          Usage:
+                sout | "Usage: " | argv[0] | " [ times (> 0) ]";
+                exit( EXIT_FAILURE );
+                exit | "Usage: " | argv[0] | " [ times (> 0) ]";
         } // switch
     printf("starting\n");
+        sout | "starting";
     executor e{ 0, 1, 1, false };
     start_actor_system( e );
+        executor e{ 0, 1, 1, false };
+        start_actor_system( e );
     printf("started\n");
+        sout | "started";
     derived_msg * d_msg = alloc();
     (*d_msg){};
     derived_actor * d_actor = alloc();
     (*d_actor){};
     *d_actor | *d_msg;
+        derived_msg * d_msg = alloc();
+        (*d_msg){};
+        derived_actor * d_actor = alloc();
+        (*d_actor){};
+        *d_actor | *d_msg;
     printf("stopping\n");
+        sout | "stopping";
     stop_actor_system();
+        stop_actor_system();
+    printf("stopped\n");
+    return 0;
+        sout | "stopped";
+}

tests/concurrency/actors/executor.cfa

-                      rc1e66d9
+                      rdeda7e6
 static int ids = 0;
 struct d_actor {
     inline actor;
     d_actor * gstart;
     int id, rounds, recs, sends;
+        inline actor;
+        d_actor * gstart;
+        int id, rounds, recs, sends;
 };
 void ?{}( d_actor & this ) with(this) {
     id = ids++;
     gstart = (&this + (id / Set * Set - id)); // remember group-start array-element
     rounds = Set * Rounds;      // send at least one message to each group member
     recs = 0;
     sends = 0;
+        id = ids++;
+        gstart = (&this + (id / Set * Set - id)); // remember group-start array-element
+        rounds = Set * Rounds;  // send at least one message to each group member
+        recs = 0;
+        sends = 0;
+}
 …
 allocation receive( d_actor & this, d_msg & msg ) with( this ) {
     if ( recs == rounds ) return Finished;
     if ( recs % Batch == 0 ) {
         for ( i; Batch ) {
             gstart[sends % Set] | shared_msg;
             sends += 1;
+        }
+    }
     recs += 1;
     return Nodelete;
+        if ( recs == rounds ) return Finished;
+        if ( recs % Batch == 0 ) {
+                for ( i; Batch ) {
+                        gstart[sends % Set] | shared_msg;
+                        sends += 1;
+                }
+        }
+        recs += 1;
+        return Nodelete;
+}
 int main( int argc, char * argv[] ) {
     switch ( argc ) {
+        switch ( argc ) {
           case 7:
                 if ( strcmp( argv[6], "d" ) != 0 ) {                    // default ?
                         BufSize = atoi( argv[6] );
                         if ( BufSize < 0 ) goto Usage;
+                        BufSize = ato( argv[6] );
+                        if ( BufSize < 0 ) fallthru default;
                 } // if
           case 6:
                 if ( strcmp( argv[5], "d" ) != 0 ) {                    // default ?
                         Batch = atoi( argv[5] );
                         if ( Batch < 1 ) goto Usage;
+                        Batch = ato( argv[5] );
+                        if ( Batch < 1 ) fallthru default;
                 } // if
           case 5:
                 if ( strcmp( argv[4], "d" ) != 0 ) {                    // default ?
                         Processors = atoi( argv[4] );
                         if ( Processors < 1 ) goto Usage;
+                        Processors = ato( argv[4] );
+                        if ( Processors < 1 ) fallthru default;
                 } // if
           case 4:
                 if ( strcmp( argv[3], "d" ) != 0 ) {                    // default ?
                         Rounds = atoi( argv[3] );
                         if ( Rounds < 1 ) goto Usage;
+                        Rounds = ato( argv[3] );
+                        if ( Rounds < 1 ) fallthru default;
                 } // if
           case 3:
                 if ( strcmp( argv[2], "d" ) != 0 ) {                    // default ?
                         Set = atoi( argv[2] );
                         if ( Set < 1 ) goto Usage;
+                        Set = ato( argv[2] );
+                        if ( Set < 1 ) fallthru default;
                 } // if
           case 2:
                 if ( strcmp( argv[1], "d" ) != 0 ) {                    // default ?
                         Actors = atoi( argv[1] );
                         if ( Actors < 1 || Actors <= Set || Actors % Set != 0 ) goto Usage;
+                        Actors = ato( argv[1] );
+                        if ( Actors < 1 || Actors <= Set || Actors % Set != 0 ) fallthru default;
                 } // if
           case 1:                                                                                       // use defaults
                 break;
           default:
+          Usage:
+                sout | "Usage: " | argv[0]
+             | " [ actors (> 0 && > set && actors % set == 0 ) | 'd' (default " | Actors
+                exit | "Usage: " | argv[0]
+                         | " [ actors (> 0 && > set && actors % set == 0 ) | 'd' (default " | Actors
                          | ") ] [ set (> 0) | 'd' (default " | Set
                          | ") ] [ rounds (> 0) | 'd' (default " | Rounds
 …
                          | ") ] [ buffer size (>= 0) | 'd' (default " | BufSize
                          | ") ]" ;
-                exit( EXIT_FAILURE );
         } // switch
     executor e{ Processors, Processors, Processors == 1 ? 1 : Processors * 512, true };
+        executor e{ Processors, Processors, Processors == 1 ? 1 : Processors * 512, true };
     printf("starting\n");
+        sout | "starting";
     start_actor_system( e );
+        start_actor_system( e );
     printf("started\n");
+        sout | "started";
     d_actor actors[ Actors ];
+        d_actor actors[ Actors ];
         for ( i; Actors ) {
 …
         } // for
     printf("stopping\n");
+        sout | "stopping";
     stop_actor_system();
+        stop_actor_system();
+    printf("stopped\n");
+    return 0;
+        sout | "stopped";
+}

tests/concurrency/actors/inherit.cfa

-                      rc1e66d9
+                      rdeda7e6
 allocation handle() {
     return Finished;
+        return Finished;
+}
 …
 int main() {
     sout | "Start";
+    {
         start_actor_system();
         D_msg * dm = alloc();
         (*dm){};
         D_msg2 * dm2 = alloc();
         (*dm2){};
         Server2 * s = alloc();
         (*s){};
         Server2 * s2 = alloc();
         (*s2){};
         *s | *dm;
         *s2 | *dm2;
         stop_actor_system();
+    }
+    {
         start_actor_system();
         Server s[2];
         D_msg * dm = alloc();
         (*dm){};
         D_msg2 * dm2 = alloc();
         (*dm2){};
         s[0] | *dm;
         s[1] | *dm2;
         stop_actor_system();
+    }
     sout | "Finished";
+        sout | "Start";
+        {
+                start_actor_system();
+                D_msg * dm = alloc();
+                (*dm){};
+                D_msg2 * dm2 = alloc();
+                (*dm2){};
+                Server2 * s = alloc();
+                (*s){};
+                Server2 * s2 = alloc();
+                (*s2){};
+                *s | *dm;
+                *s2 | *dm2;
+                stop_actor_system();
+        }
+        {
+                start_actor_system();
+                Server s[2];
+                D_msg * dm = alloc();
+                (*dm){};
+                D_msg2 * dm2 = alloc();
+                (*dm2){};
+                s[0] | *dm;
+                s[1] | *dm2;
+                stop_actor_system();
+        }
+        sout | "Finished";
+}

tests/concurrency/actors/inline.cfa

-                      rc1e66d9
+                      rdeda7e6
 struct d_actor {
     inline actor;
+        inline actor;
 };
 struct msg_wrapper {
     int b;
     inline message;
+        int b;
+        inline message;
 };
 void ^?{}( msg_wrapper & this ) { sout | "msg_wrapper dtor"; }
 struct d_msg {
     int m;
     inline msg_wrapper;
+        int m;
+        inline msg_wrapper;
 };
 void ?{}( d_msg & this, int m, int b ) { this.m = m; this.b = b; set_allocation( this, Delete ); }
 …
 allocation receive( d_actor &, d_msg & msg ) {
     sout | msg.m;
     sout | msg.b;
     return Finished;
+        sout | msg.m;
+        sout | msg.b;
+        return Finished;
+}
 struct d_msg2 {
     int m;
     inline msg_wrapper;
+        int m;
+        inline msg_wrapper;
 };
 void ^?{}( d_msg2 & this ) { sout | "d_msg2 dtor";}
 allocation receive( d_actor &, d_msg2 & msg ) {
     sout | msg.m;
     return Finished;
+        sout | msg.m;
+        return Finished;
+}
 int main() {
     processor p;
+    {
         start_actor_system();                                // sets up executor
         d_actor da;
         d_msg * dm = alloc();
         (*dm){ 42, 2423 };
         da | *dm;
         stop_actor_system();                                // waits until actors finish
+    }
+    {
         start_actor_system();                                // sets up executor
         d_actor da;
         d_msg2 dm{ 29079 };
         set_allocation( dm, Nodelete );
         msg_wrapper * mw = &dm;
         message * mg = &dm;
         virtual_dtor * v = &dm;
         da | dm;
         stop_actor_system();                                // waits until actors finish
+    }
+        processor p;
+        {
+                start_actor_system();                                                           // sets up executor
+                d_actor da;
+                d_msg * dm = alloc();
+                (*dm){ 42, 2423 };
+                da | *dm;
+                stop_actor_system();                                                            // waits until actors finish
+        }
+        {
+                start_actor_system();                                                           // sets up executor
+                d_actor da;
+                d_msg2 dm{ 29079 };
+                set_allocation( dm, Nodelete );
+                msg_wrapper * mw = &dm;
+                message * mg = &dm;
+                virtual_dtor * v = &dm;
+                da | dm;
+                stop_actor_system();                                                            // waits until actors finish
+        }
+}

tests/concurrency/actors/pingpong.cfa

-                      rc1e66d9
+                      rdeda7e6
 struct p_msg {
     inline message;
     size_t count;
+        inline message;
+        size_t count;
 };
+static inline void ?{}( p_msg & this ) { ((message &)this){}; this.count = 0; }
+//static inline void ?{}( p_msg & this ) { ((message &)this){}; this.count = 0; }
+static inline void ?{}( p_msg & this ) { this.count = 0; }
 ping * pi;
 …
 allocation receive( ping & receiver, p_msg & msg ) {
     msg.count++;
     if ( msg.count > times ) return Finished;
+        msg.count++;
+        if ( msg.count > times ) return Finished;
     allocation retval = Nodelete;
     if ( msg.count == times ) retval = Finished;
     *po | msg;
     return retval;
+        allocation retval = Nodelete;
+        if ( msg.count == times ) retval = Finished;
+        *po | msg;
+        return retval;
+}
 allocation receive( pong & receiver, p_msg & msg ) {
     msg.count++;
     if ( msg.count > times ) return Finished;
     allocation retval = Nodelete;
     if ( msg.count == times ) retval = Finished;
     *pi | msg;
     return retval;
+        msg.count++;
+        if ( msg.count > times ) return Finished;
+        allocation retval = Nodelete;
+        if ( msg.count == times ) retval = Finished;
+        *pi | msg;
+        return retval;
+}
 …
 int main( int argc, char * argv[] ) {
     printf("start\n");
+        sout | "start";
     processor p[Processors - 1];
+        processor p[Processors - 1];
+    start_actor_system( Processors ); // test passing number of processors
+        start_actor_system( Processors ); // test passing number of processors
+        ping pi_actor;
+        pong po_actor;
+        po = &po_actor;
+        pi = &pi_actor;
+        p_msg m;
+        pi_actor | m;
+        stop_actor_system();
+    ping pi_actor;
+    pong po_actor;
+    po = &po_actor;
+    pi = &pi_actor;
+    p_msg m;
+    pi_actor | m;
+    stop_actor_system();
+    printf("end\n");
+    return 0;
+        sout | "end";
+}

tests/concurrency/actors/poison.cfa

-                      rc1e66d9
+                      rdeda7e6
 int main() {
     sout | "Start";
+        sout | "Start";
     sout | "Finished";
+    {
         start_actor_system();
         Server s[10];
         for ( i; 10 ) {
             s[i] | finished_msg;
+        }
         stop_actor_system();
+    }
+        sout | "Finished";
+        {
+                start_actor_system();
+                Server s[10];
+                for ( i; 10 ) {
+                        s[i] | finished_msg;
+                }
+                stop_actor_system();
+        }
     sout | "Delete";
+    {
         start_actor_system();
         for ( i; 10 ) {
             Server * s = alloc();
             (*s){};
             (*s) | delete_msg;
+        }
         stop_actor_system();
+    }
+        sout | "Delete";
+        {
+                start_actor_system();
+                for ( i; 10 ) {
+                        Server * s = alloc();
+                        (*s){};
+                        (*s) | delete_msg;
+                }
+                stop_actor_system();
+        }
     sout | "Destroy";
+    {
         start_actor_system();
         Server s[10];
         for ( i; 10 )
             s[i] | destroy_msg;
         stop_actor_system();
         for ( i; 10 )
             if (s[i].val != 777)
                 sout | "Error: dtor not called correctly.";
+    }
+        sout | "Destroy";
+        {
+                start_actor_system();
+                Server s[10];
+                for ( i; 10 )
+                        s[i] | destroy_msg;
+                stop_actor_system();
+                for ( i; 10 )
+                        if (s[i].val != 777)
+                                sout | "Error: dtor not called correctly.";
+        }
+    sout | "Done";
+    return 0;
+        sout | "Done";
+}

tests/concurrency/actors/static.cfa

-                      rc1e66d9
+                      rdeda7e6
 struct derived_actor { inline actor; };
 struct derived_msg {
     inline message;
     int cnt;
+        inline message;
+        int cnt;
 };
 void ?{}( derived_msg & this, int cnt ) {
     ((message &) this){ Nodelete };
     this.cnt = cnt;
+        set_allocation( this, Nodelete );
+        this.cnt = cnt;
+}
 void ?{}( derived_msg & this ) { ((derived_msg &)this){ 0 }; }
 allocation receive( derived_actor & receiver, derived_msg & msg ) {
     if ( msg.cnt >= Times ) {
         sout | "Done";
         return Finished;
+    }
     msg.cnt++;
     receiver | msg;
     return Nodelete;
+        if ( msg.cnt >= Times ) {
+                sout | "Done";
+                return Finished;
+        }
+        msg.cnt++;
+        receiver | msg;
+        return Nodelete;
+}
 int main( int argc, char * argv[] ) {
     switch ( argc ) {
+        switch ( argc ) {
           case 2:
                 if ( strcmp( argv[1], "d" ) != 0 ) {                    // default ?
                         Times = atoi( argv[1] );
                         if ( Times < 1 ) goto Usage;
+                        Times = ato( argv[1] );
+                        if ( Times < 1 ) fallthru default;
                 } // if
           case 1:                                                                                       // use defaults
                 break;
           default:
+          Usage:
+                sout | "Usage: " | argv[0] | " [ times (> 0) ]";
+                exit( EXIT_FAILURE );
+                exit | "Usage: " | argv[0] | " [ times (> 0) ]";
         } // switch
     printf("starting\n");
+        sout | "starting";
     executor e{ 0, 1, 1, false };
     start_actor_system( e );
+        executor e{ 0, 1, 1, false };
+        start_actor_system( e );
     printf("started\n");
+        sout | "started";
     derived_msg msg;
+        derived_msg msg;
     derived_actor actor;
+        derived_actor actor;
     actor | msg;
+        actor | msg;
     printf("stopping\n");
+        sout | "stopping";
     stop_actor_system();
+        stop_actor_system();
+    printf("stopped\n");
+    return 0;
+        sout | "stopped";
+}

tests/concurrency/actors/types.cfa

-                      rc1e66d9
+                      rdeda7e6
 struct derived_actor {
     inline actor;
     int counter;
+        inline actor;
+        int counter;
 };
 static inline void ?{}( derived_actor & this ) { ((actor &)this){}; this.counter = 0; }
 struct d_msg {
     inline message;
     int num;
+        inline message;
+        int num;
 };
 // this isn't a valid receive routine since int is not a message type
 allocation receive( derived_actor & receiver, int i ) with( receiver ) {
     mutex(sout) sout | i;
     counter++;
     if ( counter == 2 ) return Finished;
     return Nodelete;
+        mutex(sout) sout | i;
+        counter++;
+        if ( counter == 2 ) return Finished;
+        return Nodelete;
+}
 allocation receive( derived_actor & receiver, d_msg & msg ) {
     return receive( receiver, msg.num );
+        return receive( receiver, msg.num );
+}
 struct derived_actor2 {
     struct nested { int i; }; // testing nested before inline
     inline actor;
+        struct nested { int i; }; // testing nested before inline
+        inline actor;
 };
 allocation receive( derived_actor2 & receiver, d_msg & msg ) {
     mutex(sout) sout | msg.num;
     return Finished;
+        mutex(sout) sout | msg.num;
+        return Finished;
+}
 …
 struct derived_actor4 { inline derived_actor3; };
 struct d_msg2 {
     inline message;
     int num;
+        inline message;
+        int num;
 };
 allocation receive( derived_actor3 & receiver, d_msg & msg ) {
     mutex(sout) sout | msg.num;
     if ( msg.num == -1 ) return Nodelete;
     return Finished;
+        mutex(sout) sout | msg.num;
+        if ( msg.num == -1 ) return Nodelete;
+        return Finished;
+}
 allocation receive( derived_actor3 & receiver, d_msg2 & msg ) {
     mutex(sout) sout | msg.num;
     return Finished;
+        mutex(sout) sout | msg.num;
+        return Finished;
+}
 …
 int main( int argc, char * argv[] ) {
     printf("start\n");
+        sout | "start";
     processor p[Processors - 1];
+        processor p[Processors - 1];
     printf("basic test\n");
     start_actor_system( Processors ); // test passing number of processors
     derived_actor a;
     d_msg b, c;
     b.num = 1;
     c.num = 2;
     a | b | c;
     stop_actor_system();
+        sout | "basic test";
+        start_actor_system( Processors ); // test passing number of processors
+        derived_actor a;
+        d_msg b, c;
+        b.num = 1;
+        c.num = 2;
+        a | b | c;
+        stop_actor_system();
     printf("same message and different actors test\n");
     start_actor_system(); // let system detect # of processors
     derived_actor2 d_ac2_0, d_ac2_1;
     d_msg d_ac2_msg;
     d_ac2_msg.num = 3;
     d_ac2_0 | d_ac2_msg;
     d_ac2_1 | d_ac2_msg;
     stop_actor_system();
+        sout | "same message and different actors test";
+        start_actor_system(); // let system detect # of processors
+        derived_actor2 d_ac2_0, d_ac2_1;
+        d_msg d_ac2_msg;
+        d_ac2_msg.num = 3;
+        d_ac2_0 | d_ac2_msg;
+        d_ac2_1 | d_ac2_msg;
+        stop_actor_system();
+    {
         printf("same message and different actor types test\n");
         executor e{ 0, Processors, Processors == 1 ? 1 : Processors * 4, false };
         start_actor_system( e ); // pass an explicit executor
         derived_actor2 d_ac2_2;
         derived_actor3 d_ac3_0;
         d_msg d_ac23_msg;
         d_ac23_msg.num = 4;
         d_ac3_0 | d_ac23_msg;
         d_ac2_2 | d_ac23_msg;
         stop_actor_system();
     } // RAII to clean up executor
+        {
+                sout | "same message and different actor types test";
+                executor e{ 0, Processors, Processors == 1 ? 1 : Processors * 4, false };
+                start_actor_system( e ); // pass an explicit executor
+                derived_actor2 d_ac2_2;
+                derived_actor3 d_ac3_0;
+                d_msg d_ac23_msg;
+                d_ac23_msg.num = 4;
+                d_ac3_0 | d_ac23_msg;
+                d_ac2_2 | d_ac23_msg;
+                stop_actor_system();
+        } // RAII to clean up executor
+    {
         printf("different message types, one actor test\n");
         executor e{ 1, Processors, Processors == 1 ? 1 : Processors * 4, true };
         start_actor_system( Processors );
         derived_actor3 a3;
         d_msg b1;
         d_msg2 c2;
         b1.num = -1;
         c2.num = 5;
         a3 | b1 | c2;
         stop_actor_system();
     } // RAII to clean up executor
+        {
+                sout | "different message types, one actor test";
+                executor e{ 1, Processors, Processors == 1 ? 1 : Processors * 4, true };
+                start_actor_system( Processors );
+                derived_actor3 a3;
+                d_msg b1;
+                d_msg2 c2;
+                b1.num = -1;
+                c2.num = 5;
+                a3 | b1 | c2;
+                stop_actor_system();
+        } // RAII to clean up executor
+    {
         printf("nested inheritance actor test\n");
         executor e{ 1, Processors, Processors == 1 ? 1 : Processors * 4, true };
         start_actor_system( Processors );
         derived_actor4 a4;
         d_msg b1;
         d_msg2 c2;
         b1.num = -1;
         c2.num = 5;
         a4 | b1 | c2;
         stop_actor_system();
     } // RAII to clean up executor
+        {
+                sout | "nested inheritance actor test";
+                executor e{ 1, Processors, Processors == 1 ? 1 : Processors * 4, true };
+                start_actor_system( Processors );
+                derived_actor4 a4;
+                d_msg b1;
+                d_msg2 c2;
+                b1.num = -1;
+                c2.num = 5;
+                a4 | b1 | c2;
+                stop_actor_system();
+        } // RAII to clean up executor
+    printf("end\n");
+    return 0;
+        sout | "end";
+}

tests/concurrency/channels/barrier.cfa

-                      rc1e66d9
+                      rdeda7e6
 size_t total_operations = 0;
+int Processors = 1, Tasks = 5, BarrierSize = 2;
+ssize_t Processors = 1, Tasks = 5, BarrierSize = 2;             // must be signed
 typedef channel( int ) Channel;
 …
           case 3:
                 if ( strcmp( argv[2], "d" ) != 0 ) {                    // default ?
                         BarrierSize = atoi( argv[2] );
             if ( Processors < 1 ) goto Usage;
+                        BarrierSize = ato( argv[2] );
+            if ( Processors < 1 ) fallthru default;
                 } // if
           case 2:
                 if ( strcmp( argv[1], "d" ) != 0 ) {                    // default ?
                         Processors = atoi( argv[1] );
                         if ( Processors < 1 ) goto Usage;
+                        Processors = ato( argv[1] );
+                        if ( Processors < 1 ) fallthru default;
                 } // if
           case 1:                                                                                       // use defaults
                 break;
           default:
+          Usage:
+                sout | "Usage: " | argv[0]
+                exit | "Usage: " | argv[0]
              | " [ processors (> 0) | 'd' (default " | Processors
                          | ") ] [ BarrierSize (> 0) | 'd' (default " | BarrierSize
                          | ") ]" ;
-                exit( EXIT_FAILURE );
         } // switch
     if ( Tasks < BarrierSize )

tests/concurrency/channels/big_elems.cfa

rc1e66d9	rdeda7e6
2	2	#include "parallel_harness.hfa"
3	3
4		size_t Processors = 10, Channels = 10, Producers = 40, Consumers = 40, ChannelSize = 128;
	4	ssize_t Processors = 10, Channels = 10, Producers = 40, Consumers = 40, ChannelSize = 128;
5	5
6	6	int main() {

tests/concurrency/channels/churn.cfa

-                      rc1e66d9
+                      rdeda7e6
 #include <time.hfa>
 size_t Processors = 1, Channels = 4, Producers = 2, Consumers = 2, ChannelSize = 128;
+ssize_t Processors = 1, Channels = 4, Producers = 2, Consumers = 2, ChannelSize = 128;
 owner_lock o;
 …
       case 4:
                 if ( strcmp( argv[3], "d" ) != 0 ) {                    // default ?
                         if ( atoi( argv[3] ) < 1 ) goto Usage;
                         ChannelSize = atoi( argv[3] );
+                        ChannelSize = ato( argv[3] );
+                        if ( ChannelSize < 1 ) fallthru default;
                 } // if
       case 3:
                 if ( strcmp( argv[2], "d" ) != 0 ) {                    // default ?
                         if ( atoi( argv[2] ) < 1 ) goto Usage;
                         Channels = atoi( argv[2] );
+                        Channels = ato( argv[2] );
+                        if ( Channels < 1 ) fallthru default;
                 } // if
       case 2:
                 if ( strcmp( argv[1], "d" ) != 0 ) {                    // default ?
                         if ( atoi( argv[1] ) < 1 ) goto Usage;
                         Processors = atoi( argv[1] );
+                        Processors = ato( argv[1] );
+                        if ( Processors < 1 ) fallthru default;
                 } // if
           case 1:                                                                                       // use defaults
                 break;
           default:
+          Usage:
+                sout | "Usage: " | argv[0]
+                exit | "Usage: " | argv[0]
              | " [ processors > 0 | d ]"
              | " [ producers > 0 | d ]"
              | " [ consumers > 0 | d ]"
              | " [ channels > 0 | d ]";
-                exit( EXIT_FAILURE );
+    }
     processor p[Processors - 1];

tests/concurrency/channels/contend.cfa

-                      rc1e66d9
+                      rdeda7e6
           case 3:
                 if ( strcmp( argv[2], "d" ) != 0 ) {                    // default ?
+                        ChannelSize = atoi( argv[2] );
+                        ChannelSize = ato( argv[2] );
+                        if ( ChannelSize < 1 ) fallthru default;
                 } // if
           case 2:
                 if ( strcmp( argv[1], "d" ) != 0 ) {                    // default ?
                         Processors = atoi( argv[1] );
                         if ( Processors < 1 ) goto Usage;
+                        Processors = ato( argv[1] );
+                        if ( Processors < 1 ) fallthru default;
                 } // if
           case 1:                                                                                       // use defaults
                 break;
           default:
+          Usage:
+                sout | "Usage: " | argv[0]
+                exit | "Usage: " | argv[0]
              | " [ processors (> 0) | 'd' (default " | Processors
                          | ") ] [ channel size (>= 0) | 'd' (default " | ChannelSize
                          | ") ]" ;
-                exit( EXIT_FAILURE );
         } // switch
     test(Processors, Channels, Producers, Consumers, ChannelSize);
+}

tests/concurrency/channels/daisy_chain.cfa

-                      rc1e66d9
+                      rdeda7e6
 size_t total_operations = 0;
 size_t Processors = 1, Tasks = 4;
+ssize_t Processors = 1, Tasks = 4;                                              // must be signed
 owner_lock o;
 …
           case 3:
                 if ( strcmp( argv[2], "d" ) != 0 ) {                    // default ?
                         Tasks = atoi( argv[2] );
             if ( Tasks < 1 ) goto Usage;
+                        Tasks = ato( argv[2] );
+            if ( Tasks < 1 ) fallthru default;
                 } // if
           case 2:
                 if ( strcmp( argv[1], "d" ) != 0 ) {                    // default ?
                         Processors = atoi( argv[1] );
                         if ( Processors < 1 ) goto Usage;
+                        Processors = ato( argv[1] );
+                        if ( Processors < 1 ) fallthru default;
                 } // if
           case 1:                                                                                       // use defaults
                 break;
           default:
+          Usage:
+                sout | "Usage: " | argv[0]
+                exit | "Usage: " | argv[0]
              | " [ processors (> 0) | 'd' (default " | Processors
                          | ") ] [ channel size (>= 0) | 'd' (default " | Tasks
                          | ") ]" ;
-                exit( EXIT_FAILURE );
         } // switch
     processor proc[Processors - 1];
 …
     // sout | total_operations;
     sout | "done";
-    return 0;
+}

tests/concurrency/channels/hot_potato.cfa

-                      rc1e66d9
+                      rdeda7e6
 size_t total_operations = 0;
 size_t Processors = 1, Tasks = 4;
+ssize_t Processors = 1, Tasks = 4;                                              // must be signed
 owner_lock o;
 …
+}
 int main( int argc, char * argv[] ) {
     switch ( argc ) {
           case 3:
                 if ( strcmp( argv[2], "d" ) != 0 ) {                    // default ?
                         Tasks = atoi( argv[2] );
             if ( Tasks < 1 ) goto Usage;
+                        Tasks = ato( argv[2] );
+            if ( Tasks < 1 ) fallthru default;
                 } // if
           case 2:
                 if ( strcmp( argv[1], "d" ) != 0 ) {                    // default ?
                         Processors = atoi( argv[1] );
                         if ( Processors < 1 ) goto Usage;
+                        Processors = ato( argv[1] );
+                        if ( Processors < 1 ) fallthru default;
                 } // if
           case 1:                                                                                       // use defaults
                 break;
           default:
+          Usage:
+                sout | "Usage: " | argv[0]
+                exit | "Usage: " | argv[0]
              | " [ processors (> 0) | 'd' (default " | Processors
                          | ") ] [ channel size (>= 0) | 'd' (default " | Tasks
                          | ") ]" ;
-                exit( EXIT_FAILURE );
         } // switch
     processor proc[Processors - 1];

tests/concurrency/channels/pub_sub.cfa

-                      rc1e66d9
+                      rdeda7e6
           case 3:
                 if ( strcmp( argv[2], "d" ) != 0 ) {                    // default ?
                         Tasks = atoi( argv[2] );
             if ( Tasks < 1 ) goto Usage;
+                        Tasks = ato( argv[2] );
+            if ( Tasks < 1 ) fallthru default;
                 } // if
           case 2:
                 if ( strcmp( argv[1], "d" ) != 0 ) {                    // default ?
                         Processors = atoi( argv[1] );
                         if ( Processors < 1 ) goto Usage;
+                        Processors = ato( argv[1] );
+                        if ( Processors < 1 ) fallthru default;
                 } // if
           case 1:                                                                                       // use defaults
                 break;
           default:
+          Usage:
+                sout | "Usage: " | argv[0]
+                exit | "Usage: " | argv[0]
              | " [ processors (> 0) | 'd' (default " | Processors
                          | ") ] [ Tasks (> 0) | 'd' (default " | Tasks
                          | ") ]" ;
-                exit( EXIT_FAILURE );
         } // switch
     BarrierSize = Tasks;

tests/concurrency/examples/matrixSum.cfa

-                      rc1e66d9
+                      rdeda7e6
 // Created On       : Mon Oct  9 08:29:28 2017
 // Last Modified By : Peter A. Buhr
 // Last Modified On : Wed Feb 20 08:37:53 2019
 // Update Count     : 16
+// Last Modified On : Fri Sep  8 19:05:34 2023
+// Update Count     : 19
 //
 #include <fstream.hfa>
-#include <kernel.hfa>
 #include <thread.hfa>
 …
 int main() {
         /* const */ int rows = 10, cols = 1000;
+        const int rows = 10, cols = 1000;
         int matrix[rows][cols], subtotals[rows], total = 0;
         processor p;                                                                            // add kernel thread
         for ( r; rows ) {
+        for ( r; rows ) {                                                                       // initialize
                 for ( c; cols ) {
                         matrix[r][c] = 1;
                 } // for
         } // for
         Adder * adders[rows];
         for ( r; rows ) {                                                                       // start threads to sum rows
                 adders[r] = &(*malloc()){ matrix[r], cols, subtotals[r] };
 //              adders[r] = new( matrix[r], cols, &subtotals[r] );
+                // adders[r] = new( matrix[r], cols, subtotals[r] );
         } // for
         for ( r; rows ) {                                                                       // wait for threads to finish
                 delete( adders[r] );
 …
 // Local Variables: //
-// tab-width: 4 //
 // compile-command: "cfa matrixSum.cfa" //
 // End: //

tests/concurrency/unified_locking/locks.cfa

rc1e66d9	rdeda7e6
1	1	#include <stdio.h>
2		#include ~~"locks.hfa"~~
	2	#include <locks.hfa>
3	3	#include <stdlib.hfa>
4	4	#include <thread.hfa>

tests/concurrency/unified_locking/pthread_locks.cfa

rc1e66d9	rdeda7e6
1	1	#include <stdio.h>
2		#include ~~"locks.hfa"~~
	2	#include <locks.hfa>
3	3	#include <stdlib.hfa>
4	4	#include <thread.hfa>

tests/concurrency/unified_locking/test_debug.cfa

rc1e66d9	rdeda7e6
1		#include ~~"locks.hfa"~~
	1	#include <locks.hfa>
2	2
3	3	fast_block_lock f;

tests/concurrency/unified_locking/thread_test.cfa

rc1e66d9	rdeda7e6
1	1	#include <stdio.h>
2		#include ~~"locks.hfa"~~
	2	#include <locks.hfa>
3	3	#include <stdlib.hfa>
4	4	#include <thread.hfa>

tests/concurrency/waituntil/locks.cfa

rc1e66d9	rdeda7e6
73	73	printf("done\n");
74	74	}
	75

tests/io/.expect/manipulatorsInput.arm64.txt

rc1e66d9	rdeda7e6
	1	pre1 "123456", canary ok
	2	pre2a "1234567", exception occurred, canary ok
	3	pre2b "89", canary ok
1	4	1 yyyyyyyyyyyyyyyyyyyy
2	5	2 abcxxx

tests/io/.expect/manipulatorsInput.x64.txt

rc1e66d9	rdeda7e6
	1	pre1 "123456", canary ok
	2	pre2a "1234567", exception occurred, canary ok
	3	pre2b "89", canary ok
1	4	1 yyyyyyyyyyyyyyyyyyyy
2	5	2 abcxxx

tests/io/.expect/manipulatorsInput.x86.txt

rc1e66d9	rdeda7e6
	1	pre1 "123456", canary ok
	2	pre2a "1234567", exception occurred, canary ok
	3	pre2b "89", canary ok
1	4	1 yyyyyyyyyyyyyyyyyyyy
2	5	2 abcxxx

tests/io/.in/manipulatorsInput.txt

rc1e66d9	rdeda7e6
	1	123456
	2	123456789
1	3	abc
2	4	abc

tests/io/manipulatorsInput.cfa

-                      rc1e66d9
+                      rdeda7e6
 int main() {
+        {
+                // Upfront checks to ensure buffer safety.  Once these pass, the simpler `wdi(sizeof(s),s)`
+                // usage, as in the scanf alignment cases below, is justified.
+                struct {
+                        char buf[8];
+                        char canary;
+                } data;
+                static_assert( sizeof(data.buf) == 8 );
+                static_assert( &data.buf[8] == &data.canary );  // canary comes right after buf
+                void rep(const char* casename) {
+                        data.canary = 42;
+                        bool caught = false;
+                        try {
+                                sin | wdi( sizeof(data.buf), data.buf );
+                        } catch (cstring_length*) {
+                                caught = true;
+                        }
+                        printf( "%s \"%s\"", casename, data.buf );
+                        if ( caught ) {
+                                printf(", exception occurred");
+                        }
+                        if ( data.canary == 42 ) {
+                                printf(", canary ok");
+                        } else {
+                                printf(", canary overwritten to %d", data.canary);
+                        }
+                        printf("\n");
+                }
+                rep("pre1");
+                rep("pre2a");
+                rep("pre2b");
+                scanf("\n");  // next test does not start with %s so does not tolerate leading whitespace
+        }
+        {
                 char s[] = "yyyyyyyyyyyyyyyyyyyy";

tests/minmax.cfa

-                      rc1e66d9
+                      rdeda7e6
         sout | "double\t\t\t"                           | 4.0 | 3.1 | "\tmax" | max( 4.0, 3.1 );
         sout | "long double\t\t"                        | 4.0l | 3.1l | "\tmax" | max( 4.0l, 3.1l );
+        sout | nl;
+        sout | "3 arguments";
+        sout | 2 | 3 | 4 | "\tmin" | min(2, 3, 4) | "\tmax" | max(2, 3, 4);
+        sout | 4 | 2 | 3 | "\tmin" | min(4, 2, 3) | "\tmax" | max(4, 2, 3);
+        sout | 3 | 4 | 2 | "\tmin" | min(3, 4, 2) | "\tmax" | max(3, 4, 2);
+        sout | "4 arguments";
+        sout | 3 | 2 | 5 | 4 | "\tmin" | min(3, 2, 5, 4) | "\tmax" | max(3, 2, 5, 4);
+        sout | 5 | 3 | 4 | 2 | "\tmin" | min(5, 3, 4, 2) | "\tmax" | max(5, 3, 4, 2);
 } // main

Context Navigation

Changes in / [c1e66d9:deda7e6]

Legend:

Download in other formats: