source: doc/papers/concurrency/Paper.tex @ 1e5d0f0c

arm-ehcleanup-dtorsjacob/cs343-translationjenkins-sandboxnew-astnew-ast-unique-expr
Last change on this file since 1e5d0f0c was 1e5d0f0c, checked in by Peter A. Buhr <pabuhr@…>, 3 years ago

start rewrite of coroutine section

  • Property mode set to 100644
File size: 135.6 KB
Line 
1\documentclass[AMA,STIX1COL]{WileyNJD-v2}
2
3\articletype{RESEARCH ARTICLE}%
4
5\received{XXXXX}
6\revised{XXXXX}
7\accepted{XXXXX}
8
9\raggedbottom
10
11%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
12
13% Latex packages used in the document.
14
15\usepackage{epic,eepic}
16\usepackage{xspace}
17\usepackage{comment}
18\usepackage{upquote}                                            % switch curled `'" to straight
19\usepackage{listings}                                           % format program code
20\usepackage[labelformat=simple,aboveskip=0pt,farskip=0pt]{subfig}
21\renewcommand{\thesubfigure}{(\Alph{subfigure})}
22\captionsetup{justification=raggedright,singlelinecheck=false}
23\usepackage{dcolumn}                                            % align decimal points in tables
24\usepackage{capt-of}
25
26\hypersetup{breaklinks=true}
27\definecolor{OliveGreen}{cmyk}{0.64 0 0.95 0.40}
28\definecolor{Mahogany}{cmyk}{0 0.85 0.87 0.35}
29\definecolor{Plum}{cmyk}{0.50 1 0 0}
30
31\usepackage[pagewise]{lineno}
32\renewcommand{\linenumberfont}{\scriptsize\sffamily}
33
34\renewcommand{\topfraction}{0.8}                        % float must be greater than X of the page before it is forced onto its own page
35\renewcommand{\bottomfraction}{0.8}                     % float must be greater than X of the page before it is forced onto its own page
36\renewcommand{\floatpagefraction}{0.8}          % float must be greater than X of the page before it is forced onto its own page
37\renewcommand{\textfraction}{0.0}                       % the entire page maybe devoted to floats with no text on the page at all
38
39\lefthyphenmin=3                                                        % hyphen only after 4 characters
40\righthyphenmin=3
41
42%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
43
44% Names used in the document.
45
46\newcommand{\CFAIcon}{\textsf{C}\raisebox{\depth}{\rotatebox{180}{\textsf{A}}}\xspace} % Cforall symbolic name
47\newcommand{\CFA}{\protect\CFAIcon}             % safe for section/caption
48\newcommand{\CFL}{\textrm{Cforall}\xspace}      % Cforall symbolic name
49\newcommand{\Celeven}{\textrm{C11}\xspace}      % C11 symbolic name
50\newcommand{\CC}{\textrm{C}\kern-.1em\hbox{+\kern-.25em+}\xspace} % C++ symbolic name
51\newcommand{\CCeleven}{\textrm{C}\kern-.1em\hbox{+\kern-.25em+}11\xspace} % C++11 symbolic name
52\newcommand{\CCfourteen}{\textrm{C}\kern-.1em\hbox{+\kern-.25em+}14\xspace} % C++14 symbolic name
53\newcommand{\CCseventeen}{\textrm{C}\kern-.1em\hbox{+\kern-.25em+}17\xspace} % C++17 symbolic name
54\newcommand{\CCtwenty}{\textrm{C}\kern-.1em\hbox{+\kern-.25em+}20\xspace} % C++20 symbolic name
55\newcommand{\Csharp}{C\raisebox{-0.7ex}{\Large$^\sharp$}\xspace} % C# symbolic name
56
57%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
58
59\newcommand{\Textbf}[2][red]{{\color{#1}{\textbf{#2}}}}
60\newcommand{\Emph}[2][red]{{\color{#1}\textbf{\emph{#2}}}}
61\newcommand{\uC}{$\mu$\CC}
62\newcommand{\TODO}[1]{{\Textbf{#1}}}
63
64%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
65
66% Default underscore is too low and wide. Cannot use lstlisting "literate" as replacing underscore
67% removes it as a variable-name character so keywords in variables are highlighted. MUST APPEAR
68% AFTER HYPERREF.
69%\DeclareTextCommandDefault{\textunderscore}{\leavevmode\makebox[1.2ex][c]{\rule{1ex}{0.1ex}}}
70\renewcommand{\textunderscore}{\leavevmode\makebox[1.2ex][c]{\rule{1ex}{0.075ex}}}
71
72\renewcommand*{\thefootnote}{\Alph{footnote}} % hack because fnsymbol does not work
73%\renewcommand*{\thefootnote}{\fnsymbol{footnote}}
74
75\makeatletter
76% parindent is relative, i.e., toggled on/off in environments like itemize, so store the value for
77% use rather than use \parident directly.
78\newlength{\parindentlnth}
79\setlength{\parindentlnth}{\parindent}
80
81\newcommand{\LstBasicStyle}[1]{{\lst@basicstyle{\lst@basicstyle{#1}}}}
82\newcommand{\LstKeywordStyle}[1]{{\lst@basicstyle{\lst@keywordstyle{#1}}}}
83\newcommand{\LstCommentStyle}[1]{{\lst@basicstyle{\lst@commentstyle{#1}}}}
84
85\newlength{\gcolumnposn}                                        % temporary hack because lstlisting does not handle tabs correctly
86\newlength{\columnposn}
87\setlength{\gcolumnposn}{3.5in}
88\setlength{\columnposn}{\gcolumnposn}
89
90\newcommand{\C}[2][\@empty]{\ifx#1\@empty\else\global\setlength{\columnposn}{#1}\global\columnposn=\columnposn\fi\hfill\makebox[\textwidth-\columnposn][l]{\lst@basicstyle{\LstCommentStyle{#2}}}}
91\newcommand{\CRT}{\global\columnposn=\gcolumnposn}
92
93% Denote newterms in particular font and index them without particular font and in lowercase, e.g., \newterm{abc}.
94% The option parameter provides an index term different from the new term, e.g., \newterm[\texttt{abc}]{abc}
95% The star version does not lowercase the index information, e.g., \newterm*{IBM}.
96\newcommand{\newtermFontInline}{\emph}
97\newcommand{\newterm}{\@ifstar\@snewterm\@newterm}
98\newcommand{\@newterm}[2][\@empty]{\lowercase{\def\temp{#2}}{\newtermFontInline{#2}}\ifx#1\@empty\index{\temp}\else\index{#1@{\protect#2}}\fi}
99\newcommand{\@snewterm}[2][\@empty]{{\newtermFontInline{#2}}\ifx#1\@empty\index{#2}\else\index{#1@{\protect#2}}\fi}
100
101% Latin abbreviation
102\newcommand{\abbrevFont}{\textit}                       % set empty for no italics
103\@ifundefined{eg}{
104\newcommand{\EG}{\abbrevFont{e}\abbrevFont{g}}
105\newcommand*{\eg}{%
106        \@ifnextchar{,}{\EG}%
107                {\@ifnextchar{:}{\EG}%
108                        {\EG,\xspace}}%
109}}{}%
110\@ifundefined{ie}{
111\newcommand{\IE}{\abbrevFont{i}\abbrevFont{e}}
112\newcommand*{\ie}{%
113        \@ifnextchar{,}{\IE}%
114                {\@ifnextchar{:}{\IE}%
115                        {\IE,\xspace}}%
116}}{}%
117\@ifundefined{etc}{
118\newcommand{\ETC}{\abbrevFont{etc}}
119\newcommand*{\etc}{%
120        \@ifnextchar{.}{\ETC}%
121        {\ETC.\xspace}%
122}}{}%
123\@ifundefined{etal}{
124\newcommand{\ETAL}{\abbrevFont{et}~\abbrevFont{al}}
125\newcommand*{\etal}{%
126        \@ifnextchar{.}{\protect\ETAL}%
127                {\protect\ETAL.\xspace}%
128}}{}%
129\@ifundefined{viz}{
130\newcommand{\VIZ}{\abbrevFont{viz}}
131\newcommand*{\viz}{%
132        \@ifnextchar{.}{\VIZ}%
133                {\VIZ.\xspace}%
134}}{}%
135\makeatother
136
137\newenvironment{cquote}
138               {\list{}{\lstset{resetmargins=true,aboveskip=0pt,belowskip=0pt}\topsep=3pt\parsep=0pt\leftmargin=\parindentlnth\rightmargin\leftmargin}%
139                \item\relax}
140               {\endlist}
141
142%\newenvironment{cquote}{%
143%\list{}{\lstset{resetmargins=true,aboveskip=0pt,belowskip=0pt}\topsep=3pt\parsep=0pt\leftmargin=\parindentlnth\rightmargin\leftmargin}%
144%\item\relax%
145%}{%
146%\endlist%
147%}% cquote
148
149% CFA programming language, based on ANSI C (with some gcc additions)
150\lstdefinelanguage{CFA}[ANSI]{C}{
151        morekeywords={
152                _Alignas, _Alignof, __alignof, __alignof__, asm, __asm, __asm__, __attribute, __attribute__,
153                auto, _Bool, catch, catchResume, choose, _Complex, __complex, __complex__, __const, __const__,
154                coroutine, disable, dtype, enable, exception, __extension__, fallthrough, fallthru, finally,
155                __float80, float80, __float128, float128, forall, ftype, _Generic, _Imaginary, __imag, __imag__,
156                inline, __inline, __inline__, __int128, int128, __label__, monitor, mutex, _Noreturn, one_t, or,
157                otype, restrict, __restrict, __restrict__, __signed, __signed__, _Static_assert, thread,
158                _Thread_local, throw, throwResume, timeout, trait, try, ttype, typeof, __typeof, __typeof__,
159                virtual, __volatile, __volatile__, waitfor, when, with, zero_t},
160        moredirectives={defined,include_next}%
161}
162
163\lstset{
164language=CFA,
165columns=fullflexible,
166basicstyle=\linespread{0.9}\sf,                                                 % reduce line spacing and use sanserif font
167stringstyle=\tt,                                                                                % use typewriter font
168tabsize=5,                                                                                              % N space tabbing
169xleftmargin=\parindentlnth,                                                             % indent code to paragraph indentation
170%mathescape=true,                                                                               % LaTeX math escape in CFA code $...$
171escapechar=\$,                                                                                  % LaTeX escape in CFA code
172keepspaces=true,                                                                                %
173showstringspaces=false,                                                                 % do not show spaces with cup
174showlines=true,                                                                                 % show blank lines at end of code
175aboveskip=4pt,                                                                                  % spacing above/below code block
176belowskip=3pt,
177% replace/adjust listing characters that look bad in sanserif
178literate={-}{\makebox[1ex][c]{\raisebox{0.4ex}{\rule{0.8ex}{0.1ex}}}}1 {^}{\raisebox{0.6ex}{$\scriptstyle\land\,$}}1
179        {~}{\raisebox{0.3ex}{$\scriptstyle\sim\,$}}1 % {`}{\ttfamily\upshape\hspace*{-0.1ex}`}1
180        {<}{\textrm{\textless}}1 {>}{\textrm{\textgreater}}1
181        {<-}{$\leftarrow$}2 {=>}{$\Rightarrow$}2 {->}{\makebox[1ex][c]{\raisebox{0.5ex}{\rule{0.8ex}{0.075ex}}}\kern-0.2ex{\textrm{\textgreater}}}2,
182moredelim=**[is][\color{red}]{`}{`},
183}% lstset
184
185% uC++ programming language, based on ANSI C++
186\lstdefinelanguage{uC++}[ANSI]{C++}{
187        morekeywords={
188                _Accept, _AcceptReturn, _AcceptWait, _Actor, _At, _CatchResume, _Cormonitor, _Coroutine, _Disable,
189                _Else, _Enable, _Event, _Finally, _Monitor, _Mutex, _Nomutex, _PeriodicTask, _RealTimeTask,
190                _Resume, _Select, _SporadicTask, _Task, _Timeout, _When, _With, _Throw},
191}
192\lstdefinelanguage{Golang}{
193        morekeywords=[1]{package,import,func,type,struct,return,defer,panic,recover,select,var,const,iota,},
194        morekeywords=[2]{string,uint,uint8,uint16,uint32,uint64,int,int8,int16,int32,int64,
195                bool,float32,float64,complex64,complex128,byte,rune,uintptr, error,interface},
196        morekeywords=[3]{map,slice,make,new,nil,len,cap,copy,close,true,false,delete,append,real,imag,complex,chan,},
197        morekeywords=[4]{for,break,continue,range,goto,switch,case,fallthrough,if,else,default,},
198        morekeywords=[5]{Println,Printf,Error,},
199        sensitive=true,
200        morecomment=[l]{//},
201        morecomment=[s]{/*}{*/},
202        morestring=[b]',
203        morestring=[b]",
204        morestring=[s]{`}{`},
205}
206
207\lstnewenvironment{cfa}[1][]
208{\lstset{#1}}
209{}
210\lstnewenvironment{C++}[1][]                            % use C++ style
211{\lstset{language=C++,moredelim=**[is][\protect\color{red}]{`}{`},#1}\lstset{#1}}
212{}
213\lstnewenvironment{uC++}[1][]
214{\lstset{#1}}
215{}
216\lstnewenvironment{Go}[1][]
217{\lstset{language=go,moredelim=**[is][\protect\color{red}]{`}{`},#1}\lstset{#1}}
218{}
219\lstnewenvironment{python}[1][]
220{\lstset{language=python,moredelim=**[is][\protect\color{red}]{`}{`},#1}\lstset{#1}}
221{}
222
223% inline code @...@
224\lstMakeShortInline@%
225
226\let\OLDthebibliography\thebibliography
227\renewcommand\thebibliography[1]{
228  \OLDthebibliography{#1}
229  \setlength{\parskip}{0pt}
230  \setlength{\itemsep}{4pt plus 0.3ex}
231}
232
233\title{\texorpdfstring{Advanced Control-flow and Concurrency in \protect\CFA}{Advanced Control-flow in Cforall}}
234
235\author[1]{Thierry Delisle}
236\author[1]{Peter A. Buhr*}
237\authormark{DELISLE \textsc{et al.}}
238
239\address[1]{\orgdiv{Cheriton School of Computer Science}, \orgname{University of Waterloo}, \orgaddress{\state{Waterloo, ON}, \country{Canada}}}
240
241\corres{*Peter A. Buhr, Cheriton School of Computer Science, University of Waterloo, 200 University Avenue West, Waterloo, ON, N2L 3G1, Canada. \email{pabuhr{\char`\@}uwaterloo.ca}}
242
243\fundingInfo{Natural Sciences and Engineering Research Council of Canada}
244
245\abstract[Summary]{
246\CFA is a modern, polymorphic, non-object-oriented, backwards-compatible extension of the C programming language.
247This paper discusses some advanced control-flow and concurrency/parallelism features in \CFA, along with the supporting runtime.
248These features are created from scratch because they do not exist in ISO C, or are low-level and/or unimplemented, so C programmers continue to rely on library features, like C pthreads.
249\CFA introduces language-level control-flow mechanisms, like coroutines, user-level threading, and monitors for mutual exclusion and synchronization.
250A unique contribution of this work is allowing multiple monitors to be safely acquired \emph{simultaneously} (deadlock free), while integrating this capability with monitor synchronization mechanisms.
251These features also integrate with the \CFA polymorphic type-system and exception handling, while respecting the expectations and style of C programmers.
252Experimental results show comparable performance of the new features with similar mechanisms in other concurrent programming-languages.
253}%
254
255\keywords{coroutines, concurrency, parallelism, threads, monitors, runtime, C, \CFA (Cforall)}
256
257
258\begin{document}
259\linenumbers                                            % comment out to turn off line numbering
260
261\maketitle
262
263
264\section{Introduction}
265
266This paper discusses the design of language-level control-flow and concurrency/parallelism extensions in \CFA and its runtime.
267\CFA is a modern, polymorphic, non-object-oriented\footnote{
268\CFA has features often associated with object-oriented programming languages, such as constructors, destructors, virtuals and simple inheritance.
269However, functions \emph{cannot} be nested in structures, so there is no lexical binding between a structure and set of functions (member/method) implemented by an implicit \lstinline@this@ (receiver) parameter.},
270backwards-compatible extension of the C programming language~\cite{Moss18}.
271Within the \CFA framework, new control-flow features are created from scratch.
272ISO \Celeven defines only a subset of the \CFA extensions, where the overlapping features are concurrency~\cite[\S~7.26]{C11}.
273However, \Celeven concurrency is largely wrappers for a subset of the pthreads library~\cite{Butenhof97,Pthreads}.
274Furthermore, \Celeven and pthreads concurrency is simple, based on thread fork/join in a function and a few locks, which is low-level and error prone;
275no high-level language concurrency features are defined.
276Interestingly, almost a decade after publication of the \Celeven standard, neither gcc-8, clang-8 nor msvc-19 (most recent versions) support the \Celeven include @threads.h@, indicating little interest in the C11 concurrency approach.
277Finally, while the \Celeven standard does not state a concurrent threading-model, the historical association with pthreads suggests implementations would adopt kernel-level threading (1:1)~\cite{ThreadModel}.
278
279In contrast, there has been a renewed interest during the past decade in user-level (M:N, green) threading in old and new programming languages.
280As multi-core hardware became available in the 1980/90s, both user and kernel threading were examined.
281Kernel threading was chosen, largely because of its simplicity and fit with the simpler operating systems and hardware architectures at the time, which gave it a performance advantage~\cite{Drepper03}.
282Libraries like pthreads were developed for C, and the Solaris operating-system switched from user (JDK 1.1~\cite{JDK1.1}) to kernel threads.
283As a result, languages like Java, Scala~\cite{Scala}, Objective-C~\cite{obj-c-book}, \CCeleven~\cite{C11}, and C\#~\cite{Csharp} adopted the 1:1 kernel-threading model, with a variety of presentation mechanisms.
284From 2000 onwards, languages like Go~\cite{Go}, Erlang~\cite{Erlang}, Haskell~\cite{Haskell}, D~\cite{D}, and \uC~\cite{uC++,uC++book} have championed the M:N user-threading model, and many user-threading libraries have appeared~\cite{Qthreads,MPC,BoostThreads}, including putting green threads back into Java~\cite{Quasar}.
285The main argument for user-level threading is that they are lighter weight than kernel threads (locking and context switching do not cross the kernel boundary), so there is less restriction on programming styles that encourage large numbers of threads performing smaller work-units to facilitate load balancing by the runtime~\cite{Verch12}.
286As well, user-threading facilitates a simpler concurrency approach using thread objects that leverage sequential patterns versus events with call-backs~\cite{vonBehren03}.
287Finally, performant user-threading implementations (both time and space) are largely competitive with direct kernel-threading implementations, while achieving the programming advantages of high concurrency levels and safety.
288
289A further effort over the past decade is the development of language memory-models to deal with the conflict between certain language features and compiler/hardware optimizations.
290This issue can be rephrased as: some language features are pervasive (language and runtime) and cannot be safely added via a library to prevent invalidation by sequential optimizations~\cite{Buhr95a,Boehm05}.
291The consequence is that a language must be cognizant of these features and provide sufficient tools to program around any safety issues.
292For example, C created the @volatile@ qualifier to provide correct execution for @setjmp@/@logjmp@ (concurrency came later).
293The common solution is to provide a handful of complex qualifiers and functions (e.g., @volatile@ and atomics) allowing programmers to write consistent/race-free programs, often in the sequentially-consistent memory-model~\cite{Boehm12}.
294
295While having a sufficient memory-model allows sound libraries to be constructed, writing these libraries can quickly become awkward and error prone, and using these low-level libraries has the same issues.
296Essentially, using low-level explicit locks is the concurrent equivalent of assembler programming.
297Just as most assembler programming is replaced with high-level programming, explicit locks can be replaced with high-level concurrency in a programming language.
298Then the goal is for the compiler to check for correct usage and follow any complex coding conventions implicitly.
299The drawback is that language constructs may preclude certain specialized techniques, therefore introducing inefficiency or inhibiting concurrency.
300For most concurrent programs, these drawbacks are insignificant in comparison to the speed of composition, and subsequent reliability and maintainability of the high-level concurrent program.
301(The same is true for high-level programming versus assembler programming.)
302Only very rarely should it be necessary to drop down to races and/or explicit locks to apply a specialized technique to achieve maximum speed or concurrency.
303As stated, this observation applies to non-concurrent forms of complex control-flow, like exception handling and coroutines.
304
305Adapting the programming language to these features also allows matching the control-flow model with the programming-language style, versus adopting one general (sound) library/paradigm.
306For example, it is possible to provide exceptions, coroutines, monitors, and tasks as specialized types in an object-oriented language, integrating these constructs to allow leveraging the type-system (static type-checking) and all other object-oriented capabilities~\cite{uC++}.
307It is also possible to leverage call/return for blocking communication via new control structures, versus switching to alternative communication paradigms, like channels or message passing.
308As well, user threading is often a complementary feature, allowing light-weight threading to match with low-cost objects, while hiding the application/kernel boundary.
309User threading also allows layering of implicit concurrency models (no explicit thread creation), such executors, data-flow, actors, into a single language, so programmers can chose the model that best fits an algorithm.\footnote{
310All implicit concurrency models have explicit threading in their implementation, and hence, can be build from explicit threading;
311however, the reverse is seldom true, i.e., given implicit concurrency, e.g., actors, it is virtually impossible to create explicit concurrency, e.g., blocking thread objects.}
312Finally, with extended language features and user-level threading it is possible to discretely fold locking and non-blocking I/O multiplexing into the language's I/O libraries, so threading implicitly dovetails with the I/O subsystem.
313\CFA embraces language extensions and user-level threading to provide advanced control-flow (exception handling\footnote{
314\CFA exception handling will be presented in a separate paper.
315The key feature that dovetails with this paper is non-local exceptions allowing exceptions to be raised across stacks, with synchronous exceptions raised among coroutines and asynchronous exceptions raised among threads, similar to that in \uC~\cite[\S~5]{uC++}
316} and coroutines) and concurrency.
317
318Most augmented traditional (Fortran 18~\cite{Fortran18}, Cobol 14~\cite{Cobol14}, Ada 12~\cite{Ada12}, Java 11~\cite{Java11}) and new languages (Go~\cite{Go}, Rust~\cite{Rust}, and D~\cite{D}), except \CC, diverge from C with different syntax and semantics, only interoperate indirectly with C, and are not systems languages, for those with managed memory.
319As a result, there is a significant learning curve to move to these languages, and C legacy-code must be rewritten.
320While \CC, like \CFA, takes an evolutionary approach to extend C, \CC's constantly growing complex and interdependent features-set (e.g., objects, inheritance, templates, etc.) mean idiomatic \CC code is difficult to use from C, and C programmers must expend significant effort learning \CC.
321Hence, rewriting and retraining costs for these languages, even \CC, are prohibitive for companies with a large C software-base.
322\CFA with its orthogonal feature-set, its high-performance runtime, and direct access to all existing C libraries circumvents these problems.
323
324We present comparative examples so the reader can judge if the \CFA control-flow extensions are equivalent or better than those in or proposed for \Celeven, \CC and other concurrent, imperative programming languages, and perform experiments to show the \CFA runtime is competitive with other similar mechanisms.
325The detailed contributions of this work are:
326\begin{itemize}
327\item
328allowing multiple monitors to be safely acquired \emph{simultaneously} (deadlock free), while seamlessly integrating this capability with all monitor synchronization mechanisms.
329\item
330all control-flow features respect the expectations of C programmers, with statically type-safe interfaces that integrate with the \CFA polymorphic type-system and other language features.
331\item
332experimental results show comparable performance of the new features with similar mechanisms in other concurrent programming-languages.
333\end{itemize}
334
335\begin{comment}
336This paper provides a minimal concurrency \newterm{Application Program Interface} (API) that is simple, efficient and can be used to build other concurrency features.
337While the simplest concurrency system is a thread and a lock, this low-level approach is hard to master.
338An easier approach for programmers is to support higher-level constructs as the basis of concurrency.
339Indeed, for highly-productive concurrent-programming, high-level approaches are much more popular~\cite{Hochstein05}.
340Examples of high-level approaches are jobs (thread pool)~\cite{TBB}, implicit threading~\cite{OpenMP}, monitors~\cite{Java}, channels~\cite{CSP,Go}, and message passing~\cite{Erlang,MPI}.
341
342The following terminology is used.
343A \newterm{thread} is a fundamental unit of execution that runs a sequence of code and requires a stack to maintain state.
344Multiple simultaneous threads give rise to \newterm{concurrency}, which requires locking to ensure access to shared data and safe communication.
345\newterm{Locking}, and by extension \newterm{locks}, are defined as a mechanism to prevent progress of threads to provide safety.
346\newterm{Parallelism} is running multiple threads simultaneously.
347Parallelism implies \emph{actual} simultaneous execution, where concurrency only requires \emph{apparent} simultaneous execution.
348As such, parallelism only affects performance, which is observed through differences in space and/or time at runtime.
349Hence, there are two problems to be solved: concurrency and parallelism.
350While these two concepts are often combined, they are distinct, requiring different tools~\cite[\S~2]{Buhr05a}.
351Concurrency tools handle mutual exclusion and synchronization, while parallelism tools handle performance, cost, and resource utilization.
352
353The proposed concurrency API is implemented in a dialect of C, called \CFA (pronounced C-for-all).
354The paper discusses how the language features are added to the \CFA translator with respect to parsing, semantics, and type checking, and the corresponding high-performance runtime-library to implement the concurrent features.
355\end{comment}
356
357
358\begin{comment}
359\section{\CFA Overview}
360
361The following is a quick introduction to the \CFA language, specifically tailored to the features needed to support concurrency.
362Extended versions and explanation of the following code examples are available at the \CFA website~\cite{Cforall} or in Moss~\etal~\cite{Moss18}.
363
364\CFA is a non-object-oriented extension of ISO-C, and hence, supports all C paradigms.
365Like C, the building blocks of \CFA are structures and routines.
366Virtually all of the code generated by the \CFA translator respects C memory layouts and calling conventions.
367While \CFA is not object oriented, lacking the concept of a receiver (\eg @this@) and nominal inheritance-relationships, C has a notion of objects: ``region of data storage in the execution environment, the contents of which can represent values''~\cite[3.15]{C11}.
368While some object-oriented features appear in \CFA, they are independent capabilities, allowing \CFA to adopt them while maintaining a procedural paradigm.
369
370
371\subsection{References}
372
373\CFA provides multi-level rebindable references, as an alternative to pointers, which significantly reduces syntactic noise.
374\begin{cfa}
375int x = 1, y = 2, z = 3;
376int * p1 = &x, ** p2 = &p1,  *** p3 = &p2,      $\C{// pointers to x}$
377    `&` r1 = x,   `&&` r2 = r1,   `&&&` r3 = r2;        $\C{// references to x}$
378int * p4 = &z, `&` r4 = z;
379
380*p1 = 3; **p2 = 3; ***p3 = 3;       // change x
381r1 =  3;     r2 = 3;      r3 = 3;        // change x: implicit dereferences *r1, **r2, ***r3
382**p3 = &y; *p3 = &p4;                // change p1, p2
383`&`r3 = &y; `&&`r3 = &`&`r4;             // change r1, r2: cancel implicit dereferences (&*)**r3, (&(&*)*)*r3, &(&*)r4
384\end{cfa}
385A reference is a handle to an object, like a pointer, but is automatically dereferenced the specified number of levels.
386Referencing (address-of @&@) a reference variable cancels one of the implicit dereferences, until there are no more implicit references, after which normal expression behaviour applies.
387
388
389\subsection{\texorpdfstring{\protect\lstinline{with} Statement}{with Statement}}
390\label{s:WithStatement}
391
392Heterogeneous data is aggregated into a structure/union.
393To reduce syntactic noise, \CFA provides a @with@ statement (see Pascal~\cite[\S~4.F]{Pascal}) to elide aggregate field-qualification by opening a scope containing the field identifiers.
394\begin{cquote}
395\vspace*{-\baselineskip}%???
396\lstDeleteShortInline@%
397\begin{cfa}
398struct S { char c; int i; double d; };
399struct T { double m, n; };
400// multiple aggregate parameters
401\end{cfa}
402\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}|@{\hspace{2\parindentlnth}}l@{}}
403\begin{cfa}
404void f( S & s, T & t ) {
405        `s.`c; `s.`i; `s.`d;
406        `t.`m; `t.`n;
407}
408\end{cfa}
409&
410\begin{cfa}
411void f( S & s, T & t ) `with ( s, t )` {
412        c; i; d;                // no qualification
413        m; n;
414}
415\end{cfa}
416\end{tabular}
417\lstMakeShortInline@%
418\end{cquote}
419Object-oriented programming languages only provide implicit qualification for the receiver.
420
421In detail, the @with@-statement syntax is:
422\begin{cfa}
423$\emph{with-statement}$:
424        'with' '(' $\emph{expression-list}$ ')' $\emph{compound-statement}$
425\end{cfa}
426and may appear as the body of a routine or nested within a routine body.
427Each expression in the expression-list provides a type and object.
428The type must be an aggregate type.
429(Enumerations are already opened.)
430The object is the implicit qualifier for the open structure-fields.
431All expressions in the expression list are opened in parallel within the compound statement, which is different from Pascal, which nests the openings from left to right.
432
433
434\subsection{Overloading}
435
436\CFA maximizes the ability to reuse names via overloading to aggressively address the naming problem.
437Both variables and routines may be overloaded, where selection is based on number and types of returns and arguments (as in Ada~\cite{Ada}).
438\newpage
439\vspace*{-2\baselineskip}%???
440\begin{cquote}
441\begin{cfa}
442// selection based on type
443\end{cfa}
444\lstDeleteShortInline@%
445\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}|@{\hspace{2\parindentlnth}}l@{}}
446\begin{cfa}
447const short int `MIN` = -32768;
448const int `MIN` = -2147483648;
449const long int `MIN` = -9223372036854775808L;
450\end{cfa}
451&
452\begin{cfa}
453short int si = `MIN`;
454int i = `MIN`;
455long int li = `MIN`;
456\end{cfa}
457\end{tabular}
458\begin{cfa}
459// selection based on type and number of parameters
460\end{cfa}
461\begin{tabular}{@{}l@{\hspace{2.7\parindentlnth}}|@{\hspace{2\parindentlnth}}l@{}}
462\begin{cfa}
463void `f`( void );
464void `f`( char );
465void `f`( int, double );
466\end{cfa}
467&
468\begin{cfa}
469`f`();
470`f`( 'a' );
471`f`( 3, 5.2 );
472\end{cfa}
473\end{tabular}
474\begin{cfa}
475// selection based on type and number of returns
476\end{cfa}
477\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}|@{\hspace{2\parindentlnth}}l@{}}
478\begin{cfa}
479char `f`( int );
480double `f`( int );
481[char, double] `f`( int );
482\end{cfa}
483&
484\begin{cfa}
485char c = `f`( 3 );
486double d = `f`( 3 );
487[d, c] = `f`( 3 );
488\end{cfa}
489\end{tabular}
490\lstMakeShortInline@%
491\end{cquote}
492Overloading is important for \CFA concurrency since the runtime system relies on creating different types to represent concurrency objects.
493Therefore, overloading eliminates long prefixes and other naming conventions to prevent name clashes.
494As seen in Section~\ref{s:Concurrency}, routine @main@ is heavily overloaded.
495As another example, variable overloading is useful in the parallel semantics of the @with@ statement for fields with the same name:
496\begin{cfa}
497struct S { int `i`; int j; double m; } s;
498struct T { int `i`; int k; int m; } t;
499with ( s, t ) {
500        j + k;                                                                  $\C{// unambiguous, s.j + t.k}$
501        m = 5.0;                                                                $\C{// unambiguous, s.m = 5.0}$
502        m = 1;                                                                  $\C{// unambiguous, t.m = 1}$
503        int a = m;                                                              $\C{// unambiguous, a = t.m }$
504        double b = m;                                                   $\C{// unambiguous, b = s.m}$
505        int c = `s.i` + `t.i`;                                  $\C{// unambiguous, qualification}$
506        (double)m;                                                              $\C{// unambiguous, cast s.m}$
507}
508\end{cfa}
509For parallel semantics, both @s.i@ and @t.i@ are visible with the same type, so only @i@ is ambiguous without qualification.
510
511
512\subsection{Operators}
513
514Overloading also extends to operators.
515Operator-overloading syntax creates a routine name with an operator symbol and question marks for the operands:
516\begin{cquote}
517\lstDeleteShortInline@%
518\begin{tabular}{@{}ll@{\hspace{\parindentlnth}}|@{\hspace{\parindentlnth}}l@{}}
519\begin{cfa}
520int ++?(int op);
521int ?++(int op);
522int `?+?`(int op1, int op2);
523int ?<=?(int op1, int op2);
524int ?=? (int & op1, int op2);
525int ?+=?(int & op1, int op2);
526\end{cfa}
527&
528\begin{cfa}
529// unary prefix increment
530// unary postfix increment
531// binary plus
532// binary less than
533// binary assignment
534// binary plus-assignment
535\end{cfa}
536&
537\begin{cfa}
538struct S { int i, j; };
539S `?+?`( S op1, S op2) { // add two structures
540        return (S){op1.i + op2.i, op1.j + op2.j};
541}
542S s1 = {1, 2}, s2 = {2, 3}, s3;
543s3 = s1 `+` s2;         // compute sum: s3 == {2, 5}
544\end{cfa}
545\end{tabular}
546\lstMakeShortInline@%
547\end{cquote}
548
549
550\subsection{Constructors / Destructors}
551
552Object lifetime is a challenge in non-managed programming languages.
553\CFA responds with \CC-like constructors and destructors, using a different operator-overloading syntax.
554\begin{cfa}
555struct VLA { int len, * data; };                        $\C{// variable length array of integers}$
556void ?{}( VLA & vla ) with ( vla ) { len = 10;  data = alloc( len ); }  $\C{// default constructor}$
557void ?{}( VLA & vla, int size, char fill ) with ( vla ) { len = size;  data = alloc( len, fill ); } // initialization
558void ?{}( VLA & vla, VLA other ) { vla.len = other.len;  vla.data = other.data; } $\C{// copy, shallow}$
559void ^?{}( VLA & vla ) with ( vla ) { free( data ); } $\C{// destructor}$
560{
561        VLA  x,            y = { 20, 0x01 },     z = y; $\C{// z points to y}$
562        // $\LstCommentStyle{\color{red}\ \ \ x\{\};\ \ \ \ \ \ \ \ \ y\{ 20, 0x01 \};\ \ \ \ \ \ \ \ \ \ z\{ z, y \};\ \ \ \ \ \ \ implicit calls}$
563        ^x{};                                                                   $\C{// deallocate x}$
564        x{};                                                                    $\C{// reallocate x}$
565        z{ 5, 0xff };                                                   $\C{// reallocate z, not pointing to y}$
566        ^y{};                                                                   $\C{// deallocate y}$
567        y{ x };                                                                 $\C{// reallocate y, points to x}$
568        x{};                                                                    $\C{// reallocate x, not pointing to y}$
569}       //  $\LstCommentStyle{\color{red}\^{}z\{\};\ \ \^{}y\{\};\ \ \^{}x\{\};\ \ \ implicit calls}$
570\end{cfa}
571Like \CC, construction is implicit on allocation (stack/heap) and destruction is implicit on deallocation.
572The object and all their fields are constructed/destructed.
573\CFA also provides @new@ and @delete@ as library routines, which behave like @malloc@ and @free@, in addition to constructing and destructing objects:
574\begin{cfa}
575{
576        ... struct S s = {10}; ...                              $\C{// allocation, call constructor}$
577}                                                                                       $\C{// deallocation, call destructor}$
578struct S * s = new();                                           $\C{// allocation, call constructor}$
579...
580delete( s );                                                            $\C{// deallocation, call destructor}$
581\end{cfa}
582\CFA concurrency uses object lifetime as a means of mutual exclusion and/or synchronization.
583
584
585\subsection{Parametric Polymorphism}
586\label{s:ParametricPolymorphism}
587
588The signature feature of \CFA is parametric-polymorphic routines~\cite{Cforall} with routines generalized using a @forall@ clause (giving the language its name), which allow separately compiled routines to support generic usage over multiple types.
589For example, the following sum routine works for any type that supports construction from 0 and addition:
590\begin{cfa}
591forall( otype T | { void `?{}`( T *, zero_t ); T `?+?`( T, T ); } ) // constraint type, 0 and +
592T sum( T a[$\,$], size_t size ) {
593        `T` total = { `0` };                                    $\C{// initialize by 0 constructor}$
594        for ( size_t i = 0; i < size; i += 1 )
595                total = total `+` a[i];                         $\C{// select appropriate +}$
596        return total;
597}
598S sa[5];
599int i = sum( sa, 5 );                                           $\C{// use S's 0 construction and +}$
600\end{cfa}
601Type variables can be @otype@ or @dtype@.
602@otype@ refers to a \emph{complete type}, \ie, a type with size, alignment, default constructor, copy constructor, destructor, and assignment operator.
603@dtype@ refers to an \emph{incomplete type}, \eg, void or a forward-declared type.
604The builtin types @zero_t@ and @one_t@ overload constant 0 and 1 for a new types, where both 0 and 1 have special meaning in C.
605
606\CFA provides \newterm{traits} to name a group of type assertions, where the trait name allows specifying the same set of assertions in multiple locations, preventing repetition mistakes at each routine declaration:
607\begin{cfa}
608trait `sumable`( otype T ) {
609        void `?{}`( T &, zero_t );                              $\C{// 0 literal constructor}$
610        T `?+?`( T, T );                                                $\C{// assortment of additions}$
611        T ?+=?( T &, T );
612        T ++?( T & );
613        T ?++( T & );
614};
615forall( otype T `| sumable( T )` )                      $\C{// use trait}$
616T sum( T a[$\,$], size_t size );
617\end{cfa}
618
619Using the return type for overload discrimination, it is possible to write a type-safe @alloc@ based on the C @malloc@:
620\begin{cfa}
621forall( dtype T | sized(T) ) T * alloc( void ) { return (T *)malloc( sizeof(T) ); }
622int * ip = alloc();                                                     $\C{// select type and size from left-hand side}$
623double * dp = alloc();
624struct S {...} * sp = alloc();
625\end{cfa}
626where the return type supplies the type/size of the allocation, which is impossible in most type systems.
627\end{comment}
628
629
630\section{Coroutines: Stepping Stone}
631\label{coroutine}
632
633Coroutines are generalized routines allowing execution to be temporarily suspended and later resumed.
634Hence, unlike a normal routine, a coroutine may not terminate when it returns to its caller, allowing it to be restarted with the values and execution location present at the point of suspension.
635This capability is accomplished via the coroutine's stack, where suspend/resume context switch among stacks.
636Because threading design-challenges are present in coroutines, their design effort is relevant, and this effort can be easily exposed to programmers giving them a useful new programming paradigm because a coroutine handles the class of problems that need to retain state between calls, \eg plugins, device drivers, and finite-state machines.
637Therefore, the two fundamental features of the core \CFA coroutine-API are independent call-stacks and @suspend@/@resume@ operations.
638
639For example, a problem made easier with coroutines is unbounded generators, \eg generating an infinite sequence of Fibonacci numbers
640\begin{displaymath}
641\mathsf{fib}(n) = \left \{
642\begin{array}{ll}
6430                                       & n = 0         \\
6441                                       & n = 1         \\
645\mathsf{fib}(n-1) + \mathsf{fib}(n-2)   & n \ge 2       \\
646\end{array}
647\right.
648\end{displaymath}
649where Figure~\ref{f:C-fibonacci} shows conventional approaches for writing a Fibonacci generator in C.
650Figure~\ref{f:GlobalVariables} illustrates the following problems: unique unencapsulated global variables necessary to retain state between calls, only one Fibonacci generator, and execution state must be explicitly retained via explicit state variables.
651Figure~\ref{f:ExternalState} addresses these issues: unencapsulated program global variables become encapsulated structure variables, unique global variables are replaced by multiple Fibonacci objects, and explicit execution state is removed by precomputing the first two Fibonacci numbers and returning $\mathsf{fib}(n-2)$.
652
653\begin{figure}
654\centering
655\newbox\myboxA
656% \begin{lrbox}{\myboxA}
657% \begin{cfa}[aboveskip=0pt,belowskip=0pt]
658% `int fn1, fn2, state = 1;`   // single global variables
659% int fib() {
660%       int fn;
661%       `switch ( state )` {  // explicit execution state
662%         case 1: fn = 0;  fn1 = fn;  state = 2;  break;
663%         case 2: fn = 1;  fn2 = fn1;  fn1 = fn;  state = 3;  break;
664%         case 3: fn = fn1 + fn2;  fn2 = fn1;  fn1 = fn;  break;
665%       }
666%       return fn;
667% }
668% int main() {
669%
670%       for ( int i = 0; i < 10; i += 1 ) {
671%               printf( "%d\n", fib() );
672%       }
673% }
674% \end{cfa}
675% \end{lrbox}
676\begin{lrbox}{\myboxA}
677\begin{cfa}[aboveskip=0pt,belowskip=0pt]
678#define FIB_INIT { 0, 1 }
679typedef struct { int fn1, fn; } Fib;
680int fib( Fib * f ) {
681
682        int ret = f->fn1;
683        f->fn1 = f->fn;
684        f->fn = ret + f->fn;
685        return ret;
686}
687
688
689
690int main() {
691        Fib f1 = FIB_INIT, f2 = FIB_INIT;
692        for ( int i = 0; i < 10; i += 1 ) {
693                printf( "%d %d\n",
694                                fib( &f1 ), fib( &f2 ) );
695        }
696}
697\end{cfa}
698\end{lrbox}
699
700\newbox\myboxB
701\begin{lrbox}{\myboxB}
702\begin{cfa}[aboveskip=0pt,belowskip=0pt]
703`coroutine` Fib { int fn1; };
704void main( Fib & fib ) with( fib ) {
705        int fn;
706        [fn1, fn] = [0, 1];
707        for () {
708                `suspend();`
709                [fn1, fn] = [fn, fn1 + fn];
710        }
711}
712int ?()( Fib & fib ) with( fib ) {
713        `resume( fib );`  return fn1;
714}
715int main() {
716        Fib f1, f2;
717        for ( 10 ) {
718                sout | f1() | f2();
719}
720
721
722\end{cfa}
723\end{lrbox}
724
725\newbox\myboxC
726\begin{lrbox}{\myboxC}
727\begin{python}[aboveskip=0pt,belowskip=0pt]
728
729def Fib():
730
731    fn1, fn = 0, 1
732    while True:
733        `yield fn1`
734        fn1, fn = fn, fn1 + fn
735
736
737// next prewritten
738
739
740f1 = Fib()
741f2 = Fib()
742for i in range( 10 ):
743        print( next( f1 ), next( f2 ) )
744
745
746
747\end{python}
748\end{lrbox}
749
750\subfloat[C]{\label{f:GlobalVariables}\usebox\myboxA}
751\hspace{3pt}
752\vrule
753\hspace{3pt}
754\subfloat[\CFA]{\label{f:ExternalState}\usebox\myboxB}
755\hspace{3pt}
756\vrule
757\hspace{3pt}
758\subfloat[Python]{\label{f:ExternalState}\usebox\myboxC}
759\caption{Fibonacci Generator}
760\label{f:C-fibonacci}
761
762% \bigskip
763%
764% \newbox\myboxA
765% \begin{lrbox}{\myboxA}
766% \begin{cfa}[aboveskip=0pt,belowskip=0pt]
767% `coroutine` Fib { int fn; };
768% void main( Fib & fib ) with( fib ) {
769%       fn = 0;  int fn1 = fn; `suspend()`;
770%       fn = 1;  int fn2 = fn1;  fn1 = fn; `suspend()`;
771%       for () {
772%               fn = fn1 + fn2; fn2 = fn1; fn1 = fn; `suspend()`; }
773% }
774% int next( Fib & fib ) with( fib ) { `resume( fib );` return fn; }
775% int main() {
776%       Fib f1, f2;
777%       for ( 10 )
778%               sout | next( f1 ) | next( f2 );
779% }
780% \end{cfa}
781% \end{lrbox}
782% \newbox\myboxB
783% \begin{lrbox}{\myboxB}
784% \begin{python}[aboveskip=0pt,belowskip=0pt]
785%
786% def Fibonacci():
787%       fn = 0; fn1 = fn; `yield fn`  # suspend
788%       fn = 1; fn2 = fn1; fn1 = fn; `yield fn`
789%       while True:
790%               fn = fn1 + fn2; fn2 = fn1; fn1 = fn; `yield fn`
791%
792%
793% f1 = Fibonacci()
794% f2 = Fibonacci()
795% for i in range( 10 ):
796%       print( `next( f1 )`, `next( f2 )` ) # resume
797%
798% \end{python}
799% \end{lrbox}
800% \subfloat[\CFA]{\label{f:Coroutine3States}\usebox\myboxA}
801% \qquad
802% \subfloat[Python]{\label{f:Coroutine1State}\usebox\myboxB}
803% \caption{Fibonacci input coroutine, 3 states, internal variables}
804% \label{f:cfa-fibonacci}
805\end{figure}
806
807Using a coroutine, it is possible to express the Fibonacci formula directly without any of the C problems.
808Figure~\ref{f:Coroutine3States} creates a @coroutine@ type, @`coroutine` Fib { int fn; }@, which provides communication, @fn@, for the \newterm{coroutine main}, @main@, which runs on the coroutine stack, and possibly multiple interface routines, \eg @next@.
809Like the structure in Figure~\ref{f:ExternalState}, the coroutine type allows multiple instances, where instances of this type are passed to the (overloaded) coroutine main.
810The coroutine main's stack holds the state for the next generation, @f1@ and @f2@, and the code represents the three states in the Fibonacci formula via the three suspend points, to context switch back to the caller's @resume@.
811The interface routine @next@, takes a Fibonacci instance and context switches to it using @resume@;
812on restart, the Fibonacci field, @fn@, contains the next value in the sequence, which is returned.
813The first @resume@ is special because it allocates the coroutine stack and cocalls its coroutine main on that stack;
814when the coroutine main returns, its stack is deallocated.
815Hence, @Fib@ is an object at creation, transitions to a coroutine on its first resume, and transitions back to an object when the coroutine main finishes.
816Figure~\ref{f:Coroutine1State} shows the coroutine version of the C version in Figure~\ref{f:ExternalState}.
817Coroutine generators are called \newterm{output coroutines} because values are only returned.
818
819Figure~\ref{f:CFAFmt} shows an \newterm{input coroutine}, @Format@, for restructuring text into groups of characters of fixed-size blocks.
820For example, the input of the left is reformatted into the output on the right.
821\begin{quote}
822\tt
823\begin{tabular}{@{}l|l@{}}
824\multicolumn{1}{c|}{\textbf{\textrm{input}}} & \multicolumn{1}{c}{\textbf{\textrm{output}}} \\
825abcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyz
826&
827\begin{tabular}[t]{@{}lllll@{}}
828abcd    & efgh  & ijkl  & mnop  & qrst  \\
829uvwx    & yzab  & cdef  & ghij  & klmn  \\
830opqr    & stuv  & wxyz  &               &
831\end{tabular}
832\end{tabular}
833\end{quote}
834The example takes advantage of resuming a coroutine in the constructor to prime the loops so the first character sent for formatting appears inside the nested loops.
835The destructor provides a newline, if formatted text ends with a full line.
836Figure~\ref{f:CFmt} shows the C equivalent formatter, where the loops of the coroutine are flattened (linearized) and rechecked on each call because execution location is not retained between calls.
837(Linearized code is the bane of device drivers.)
838
839\begin{figure}
840\centering
841\newbox\myboxA
842\begin{lrbox}{\myboxA}
843\begin{cfa}[aboveskip=0pt,belowskip=0pt]
844`coroutine` Fmt {
845        char ch;   // communication variables
846        int g, b;   // needed in destructor
847};
848void main( Fmt & fmt ) with( fmt ) {
849        for () {
850                for ( g = 0; g < 5; g += 1 ) { // groups
851                        for ( b = 0; b < 4; b += 1 ) { // blocks
852                                `suspend();`
853                                sout | ch; } // print character
854                        sout | "  "; } // block separator
855                sout | nl; }  // group separator
856}
857void ?{}( Fmt & fmt ) { `resume( fmt );` } // prime
858void ^?{}( Fmt & fmt ) with( fmt ) { // destructor
859        if ( g != 0 || b != 0 ) // special case
860                sout | nl; }
861void send( Fmt & fmt, char c ) { fmt.ch = c; `resume( fmt )`; }
862int main() {
863        Fmt fmt;
864        sout | nlOff;   // turn off auto newline
865        for ( 41 )
866                send( fmt, 'a' );
867}
868\end{cfa}
869\end{lrbox}
870
871\newbox\myboxB
872\begin{lrbox}{\myboxB}
873\begin{python}[aboveskip=0pt,belowskip=0pt]
874
875
876
877def Fmt():
878        try:
879                while True:
880                        for g in range( 5 ):
881                                for b in range( 4 ):
882
883                                        print( `(yield)`, end='' )
884                                print( '  ', end='' )
885                        print()
886
887
888        except GeneratorExit:
889                if g != 0 | b != 0:
890                        print()
891
892
893fmt = Fmt()
894`next( fmt )`                    # prime
895for i in range( 41 ):
896        `fmt.send( 'a' );`      # send to yield
897
898\end{python}
899\end{lrbox}
900\subfloat[\CFA]{\label{f:CFAFmt}\usebox\myboxA}
901\qquad
902\subfloat[Python]{\label{f:CFmt}\usebox\myboxB}
903\caption{Output formatting text}
904\label{f:fmt-line}
905\end{figure}
906
907The previous examples are \newterm{asymmetric (semi) coroutine}s because one coroutine always calls a resuming routine for another coroutine, and the resumed coroutine always suspends back to its last resumer, similar to call/return for normal routines.
908However, @resume@ and @suspend@ context switch among existing stack-frames, rather than create new ones so there is no stack growth.
909\newterm{Symmetric (full) coroutine}s have a coroutine call to a resuming routine for another coroutine, and its coroutine main calls another resuming routine, which eventually forms a resuming-call cycle.
910(The trivial cycle is a coroutine resuming itself.)
911This control flow is similar to recursion for normal routines, but again there is no stack growth from the context switch.
912
913\begin{figure}
914\centering
915\lstset{language=CFA,escapechar={},moredelim=**[is][\protect\color{red}]{`}{`}}% allow $
916\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}l@{}}
917\begin{cfa}
918`coroutine` Prod {
919        Cons & c;
920        int N, money, receipt;
921};
922void main( Prod & prod ) with( prod ) {
923        // 1st resume starts here
924        for ( i; N ) {
925                int p1 = random( 100 ), p2 = random( 100 );
926                sout | p1 | " " | p2;
927                int status = delivery( c, p1, p2 );
928                sout | " $" | money | nl | status;
929                receipt += 1;
930        }
931        stop( c );
932        sout | "prod stops";
933}
934int payment( Prod & prod, int money ) {
935        prod.money = money;
936        `resume( prod );`
937        return prod.receipt;
938}
939void start( Prod & prod, int N, Cons &c ) {
940        &prod.c = &c; // reassignable reference
941        prod.[N, receipt] = [N, 0];
942        `resume( prod );`
943}
944int main() {
945        Prod prod;
946        Cons cons = { prod };
947        start( prod, 5, cons );
948}
949\end{cfa}
950&
951\begin{cfa}
952`coroutine` Cons {
953        Prod & p;
954        int p1, p2, status;
955        bool done;
956};
957void ?{}( Cons & cons, Prod & p ) {
958        &cons.p = &p; // reassignable reference
959        cons.[status, done ] = [0, false];
960}
961void ^?{}( Cons & cons ) {}
962void main( Cons & cons ) with( cons ) {
963        // 1st resume starts here
964        int money = 1, receipt;
965        for ( ; ! done; ) {
966                sout | p1 | " " | p2 | nl | " $" | money;
967                status += 1;
968                receipt = payment( p, money );
969                sout | " #" | receipt;
970                money += 1;
971        }
972        sout | "cons stops";
973}
974int delivery( Cons & cons, int p1, int p2 ) {
975        cons.[p1, p2] = [p1, p2];
976        `resume( cons );`
977        return cons.status;
978}
979void stop( Cons & cons ) {
980        cons.done = true;
981        `resume( cons );`
982}
983\end{cfa}
984\end{tabular}
985\caption{Producer / consumer: resume-resume cycle, bi-directional communication}
986\label{f:ProdCons}
987\end{figure}
988
989Figure~\ref{f:ProdCons} shows a producer/consumer symmetric-coroutine performing bi-directional communication.
990Since the solution involves a full-coroutining cycle, the program main creates one coroutine in isolation, passes this coroutine to its partner, and closes the cycle at the call to @start@.
991The @start@ routine communicates both the number of elements to be produced and the consumer into the producer's coroutine-structure.
992Then the @resume@ to @prod@ creates @prod@'s stack with a frame for @prod@'s coroutine main at the top, and context switches to it.
993@prod@'s coroutine main starts, creates local variables that are retained between coroutine activations, and executes $N$ iterations, each generating two random values, calling the consumer to deliver the values, and printing the status returned from the consumer.
994
995The producer call to @delivery@ transfers values into the consumer's communication variables, resumes the consumer, and returns the consumer status.
996For the first resume, @cons@'s stack is initialized, creating local variables retained between subsequent activations of the coroutine.
997The consumer iterates until the @done@ flag is set, prints the values delivered by the producer, increments status, and calls back to the producer via @payment@, and on return from @payment@, prints the receipt from the producer and increments @money@ (inflation).
998The call from the consumer to @payment@ introduces the cycle between producer and consumer.
999When @payment@ is called, the consumer copies values into the producer's communication variable and a resume is executed.
1000The context switch restarts the producer at the point where it last context switched, so it continues in @delivery@ after the resume.
1001
1002@delivery@ returns the status value in @prod@'s coroutine main, where the status is printed.
1003The loop then repeats calling @delivery@, where each call resumes the consumer coroutine.
1004The context switch to the consumer continues in @payment@.
1005The consumer increments and returns the receipt to the call in @cons@'s coroutine main.
1006The loop then repeats calling @payment@, where each call resumes the producer coroutine.
1007
1008After iterating $N$ times, the producer calls @stop@.
1009The @done@ flag is set to stop the consumer's execution and a resume is executed.
1010The context switch restarts @cons@ in @payment@ and it returns with the last receipt.
1011The consumer terminates its loops because @done@ is true, its @main@ terminates, so @cons@ transitions from a coroutine back to an object, and @prod@ reactivates after the resume in @stop@.
1012@stop@ returns and @prod@'s coroutine main terminates.
1013The program main restarts after the resume in @start@.
1014@start@ returns and the program main terminates.
1015
1016One \emph{killer} application for a coroutine is device drivers, which at one time caused 70\%-85\% of failures in Windows/Linux~\cite{Swift05}.
1017Many device drivers are a finite state-machine parsing a protocol, e.g.:
1018\begin{tabbing}
1019\ldots STX \= \ldots message \ldots \= ESC \= ETX \= \ldots message \ldots  \= ETX \= 2-byte crc \= \ldots      \kill
1020\ldots STX \> \ldots message \ldots \> ESC \> ETX \> \ldots message \ldots  \> ETX \> 2-byte crc \> \ldots
1021\end{tabbing}
1022where a network message begins with the control character STX and ends with an ETX, followed by a 2-byte cyclic-redundancy check.
1023Control characters may appear in a message if preceded by an ESC.
1024Because FSMs can be complex and occur frequently in important domains, direct support of the coroutine is crucial in a systems programminglanguage.
1025
1026\begin{figure}
1027\begin{cfa}
1028enum Status { CONT, MSG, ESTX, ELNTH, ECRC };
1029`coroutine` Driver {
1030        Status status;
1031        char * msg, byte;
1032};
1033void ?{}( Driver & d, char * m ) { d.msg = m; }         $\C[3.0in]{// constructor}$
1034Status next( Driver & d, char b ) with( d ) {           $\C{// 'with' opens scope}$
1035        byte = b; `resume( d );` return status;
1036}
1037void main( Driver & d ) with( d ) {
1038        enum { STX = '\002', ESC = '\033', ETX = '\003', MaxMsg = 64 };
1039        unsigned short int crc;                                                 $\C{// error checking}$
1040  msg: for () {                                                                         $\C{// parse message}$
1041                status = CONT;
1042                unsigned int lnth = 0, sum = 0;
1043                while ( byte != STX ) `suspend();`
1044          emsg: for () {
1045                        `suspend();`                                                    $\C{// process byte}$
1046                        choose ( byte ) {                                               $\C{// switch with default break}$
1047                          case STX:
1048                                status = ESTX; `suspend();` continue msg;
1049                          case ETX:
1050                                break emsg;
1051                          case ESC:
1052                                suspend();
1053                        } // choose
1054                        if ( lnth >= MaxMsg ) {                                 $\C{// buffer full ?}$
1055                                status = ELNTH; `suspend();` continue msg; }
1056                        msg[lnth++] = byte;
1057                        sum += byte;
1058                } // for
1059                msg[lnth] = '\0';                                                       $\C{// terminate string}\CRT$
1060                `suspend();`
1061                crc = (unsigned char)byte << 8; // prevent sign extension for signed char
1062                `suspend();`
1063                status = (crc | (unsigned char)byte) == sum ? MSG : ECRC;
1064                `suspend();`
1065        } // for
1066}
1067\end{cfa}
1068\caption{Device driver for simple communication protocol}
1069\end{figure}
1070
1071
1072\subsection{Coroutine Implementation}
1073
1074A significant implementation challenge for coroutines (and threads, see Section~\ref{threads}) is adding extra fields and executing code after/before the coroutine constructor/destructor and coroutine main to create/initialize/de-initialize/destroy extra fields and the stack.
1075There are several solutions to this problem and the chosen option forced the \CFA coroutine design.
1076
1077Object-oriented inheritance provides extra fields and code in a restricted context, but it requires programmers to explicitly perform the inheritance:
1078\begin{cfa}[morekeywords={class,inherits}]
1079class mycoroutine inherits baseCoroutine { ... }
1080\end{cfa}
1081and the programming language (and possibly its tool set, \eg debugger) may need to understand @baseCoroutine@ because of the stack.
1082Furthermore, the execution of constructors/destructors is in the wrong order for certain operations.
1083For example, for threads if the thread is implicitly started, it must start \emph{after} all constructors, because the thread relies on a completely initialized object, but the inherited constructor runs \emph{before} the derived.
1084
1085An alternative is composition:
1086\begin{cfa}
1087struct mycoroutine {
1088        ... // declarations
1089        baseCoroutine dummy; // composition, last declaration
1090}
1091\end{cfa}
1092which also requires an explicit declaration that must be the last one to ensure correct initialization order.
1093However, there is nothing preventing wrong placement or multiple declarations.
1094
1095For coroutines as for threads, many implementations are based on routine pointers or routine objects~\cite{Butenhof97, C++14, MS:VisualC++, BoostCoroutines15}.
1096For example, Boost implements coroutines in terms of four functor object-types:
1097\begin{cfa}
1098asymmetric_coroutine<>::pull_type
1099asymmetric_coroutine<>::push_type
1100symmetric_coroutine<>::call_type
1101symmetric_coroutine<>::yield_type
1102\end{cfa}
1103Similarly, the canonical threading paradigm is often based on routine pointers, \eg @pthreads@~\cite{Butenhof97}, \Csharp~\cite{Csharp}, Go~\cite{Go}, and Scala~\cite{Scala}.
1104However, the generic thread-handle (identifier) is limited (few operations), unless it is wrapped in a custom type.
1105\begin{cfa}
1106void mycor( coroutine_t cid, void * arg ) {
1107        int * value = (int *)arg;                               $\C{// type unsafe, pointer-size only}$
1108        // Coroutine body
1109}
1110int main() {
1111        int input = 0, output;
1112        coroutine_t cid = coroutine_create( &mycor, (void *)&input ); $\C{// type unsafe, pointer-size only}$
1113        coroutine_resume( cid, (void *)input, (void **)&output ); $\C{// type unsafe, pointer-size only}$
1114}
1115\end{cfa}
1116Since the custom type is simple to write in \CFA and solves several issues, added support for routine/lambda-based coroutines adds very little.
1117
1118Note, the type @coroutine_t@ must be an abstract handle to the coroutine, because the coroutine descriptor and its stack are non-copyable.
1119Copying the coroutine descriptor results in copies being out of date with the current state of the stack.
1120Correspondingly, copying the stack results is copies being out of date with the coroutine descriptor, and pointers in the stack being out of date to data on the stack.
1121(There is no mechanism in C to find all stack-specific pointers and update them as part of a copy.)
1122
1123The selected approach is to use language support by introducing a new kind of aggregate (structure):
1124\begin{cfa}
1125coroutine Fibonacci {
1126        int fn; // communication variables
1127};
1128\end{cfa}
1129The @coroutine@ keyword means the compiler (and tool set) can find and inject code where needed.
1130The downside of this approach is that it makes coroutine a special case in the language.
1131Users wanting to extend coroutines or build their own for various reasons can only do so in ways offered by the language.
1132Furthermore, implementing coroutines without language supports also displays the power of a programming language.
1133While this is ultimately the option used for idiomatic \CFA code, coroutines and threads can still be constructed without language support.
1134The reserved keyword simply eases use for the common case.
1135
1136Part of the mechanism to generalize coroutines is using a \CFA trait, which defines a coroutine as anything satisfying the trait @is_coroutine@, and this trait restricts the available set of coroutine-manipulation routines:
1137\begin{cfa}
1138trait is_coroutine( `dtype` T ) {
1139        void main( T & );
1140        coroutine_desc * get_coroutine( T & );
1141};
1142forall( `dtype` T | is_coroutine(T) ) void suspend( T & );
1143forall( `dtype` T | is_coroutine(T) ) void resume( T & );
1144\end{cfa}
1145The @dtype@ property provides no implicit copying operations and the @is_coroutine@ trait provides no explicit copying operations, so all coroutines must be passed by reference (pointer).
1146The routine definitions ensures there is a statically-typed @main@ routine that is the starting point (first stack frame) of a coroutine, and a mechanism to get (read) the currently executing coroutine handle.
1147The @main@ routine has no return value or additional parameters because the coroutine type allows an arbitrary number of interface routines with corresponding arbitrary typed input/output values versus fixed ones.
1148The advantage of this approach is that users can easily create different types of coroutines, \eg changing the memory layout of a coroutine is trivial when implementing the @get_coroutine@ routine, and possibly redefining @suspend@ and @resume@.
1149The \CFA keyword @coroutine@ implicitly implements the getter and forward declarations required for implementing the coroutine main:
1150\begin{cquote}
1151\begin{tabular}{@{}ccc@{}}
1152\begin{cfa}
1153coroutine MyCor {
1154        int value;
1155
1156};
1157\end{cfa}
1158&
1159{\Large $\Rightarrow$}
1160&
1161\begin{tabular}{@{}ccc@{}}
1162\begin{cfa}
1163struct MyCor {
1164        int value;
1165        coroutine_desc cor;
1166};
1167\end{cfa}
1168&
1169\begin{cfa}
1170static inline coroutine_desc *
1171get_coroutine( MyCor & this ) {
1172        return &this.cor;
1173}
1174\end{cfa}
1175&
1176\begin{cfa}
1177void main( MyCor * this );
1178
1179
1180
1181\end{cfa}
1182\end{tabular}
1183\end{tabular}
1184\end{cquote}
1185The combination of these two approaches allows an easy and concise specification to coroutining (and concurrency) for normal users, while more advanced users have tighter control on memory layout and initialization.
1186
1187
1188\section{Concurrency}
1189\label{s:Concurrency}
1190
1191At its core, concurrency is based on multiple call-stacks and scheduling threads executing on these stacks.
1192Multiple call stacks (or contexts) and a single thread of execution, called \newterm{coroutining}~\cite{Conway63,Marlin80}, does \emph{not} imply concurrency~\cite[\S~2]{Buhr05a}.
1193In coroutining, the single thread is self-scheduling across the stacks, so execution is deterministic, \ie the execution path from input to output is fixed and predictable.
1194A \newterm{stackless} coroutine executes on the caller's stack~\cite{Python} but this approach is restrictive, \eg preventing modularization and supporting only iterator/generator-style programming;
1195a \newterm{stackful} coroutine executes on its own stack, allowing full generality.
1196Only stackful coroutines are a stepping stone to concurrency.
1197
1198The transition to concurrency, even for execution with a single thread and multiple stacks, occurs when coroutines also context switch to a \newterm{scheduling oracle}, introducing non-determinism from the coroutine perspective~\cite[\S~3]{Buhr05a}.
1199Therefore, a minimal concurrency system is possible using coroutines (see Section \ref{coroutine}) in conjunction with a scheduler to decide where to context switch next.
1200The resulting execution system now follows a cooperative threading-model, called \newterm{non-preemptive scheduling}.
1201
1202Because the scheduler is special, it can either be a stackless or stackful coroutine.
1203For stackless, the scheduler performs scheduling on the stack of the current coroutine and switches directly to the next coroutine, so there is one context switch.
1204For stackful, the current coroutine switches to the scheduler, which performs scheduling, and it then switches to the next coroutine, so there are two context switches.
1205A stackful scheduler is often used for simplicity and security.
1206
1207Regardless of the approach used, a subset of concurrency related challenges start to appear.
1208For the complete set of concurrency challenges to occur, the missing feature is \newterm{preemption}, where context switching occurs randomly between any two instructions, often based on a timer interrupt, called \newterm{preemptive scheduling}.
1209While a scheduler introduces uncertainty in the order of execution, preemption introduces uncertainty about where context switches occur.
1210Interestingly, uncertainty is necessary for the runtime (operating) system to give the illusion of parallelism on a single processor and increase performance on multiple processors.
1211The reason is that only the runtime has complete knowledge about resources and how to best utilized them.
1212However, the introduction of unrestricted non-determinism results in the need for \newterm{mutual exclusion} and \newterm{synchronization} to restrict non-determinism for correctness;
1213otherwise, it is impossible to write meaningful programs.
1214Optimal performance in concurrent applications is often obtained by having as much non-determinism as correctness allows.
1215
1216An important missing feature in C is threading\footnote{While the C11 standard defines a \protect\lstinline@threads.h@ header, it is minimal and defined as optional.
1217As such, library support for threading is far from widespread.
1218At the time of writing the paper, neither \protect\lstinline@gcc@ nor \protect\lstinline@clang@ support \protect\lstinline@threads.h@ in their standard libraries.}.
1219In modern programming languages, a lack of threading is unacceptable~\cite{Sutter05, Sutter05b}, and therefore existing and new programming languages must have tools for writing efficient concurrent programs to take advantage of parallelism.
1220As an extension of C, \CFA needs to express these concepts in a way that is as natural as possible to programmers familiar with imperative languages.
1221Furthermore, because C is a system-level language, programmers expect to choose precisely which features they need and which cost they are willing to pay.
1222Hence, concurrent programs should be written using high-level mechanisms, and only step down to lower-level mechanisms when performance bottlenecks are encountered.
1223
1224
1225\subsection{Thread Interface}
1226\label{threads}
1227
1228Both user and kernel threads are supported, where user threads provide concurrency and kernel threads provide parallelism.
1229Like coroutines and for the same design reasons, the selected approach for user threads is to use language support by introducing a new kind of aggregate (structure) and a \CFA trait:
1230\begin{cquote}
1231\begin{tabular}{@{}c@{\hspace{3\parindentlnth}}c@{}}
1232\begin{cfa}
1233thread myThread {
1234        // communication variables
1235};
1236
1237
1238\end{cfa}
1239&
1240\begin{cfa}
1241trait is_thread( `dtype` T ) {
1242      void main( T & );
1243      thread_desc * get_thread( T & );
1244      void ^?{}( T & `mutex` );
1245};
1246\end{cfa}
1247\end{tabular}
1248\end{cquote}
1249(The qualifier @mutex@ for the destructor parameter is discussed in Section~\ref{s:Monitor}.)
1250Like a coroutine, the statically-typed @main@ routine is the starting point (first stack frame) of a user thread.
1251The difference is that a coroutine borrows a thread from its caller, so the first thread resuming a coroutine creates an instance of @main@;
1252whereas, a user thread receives its own thread from the runtime system, which starts in @main@ as some point after the thread constructor is run.\footnote{
1253The \lstinline@main@ routine is already a special routine in C, \ie where the program's initial thread begins, so it is a natural extension of this semantics to use overloading to declare \lstinline@main@s for user coroutines and threads.}
1254No return value or additional parameters are necessary for this routine because the task type allows an arbitrary number of interface routines with corresponding arbitrary typed input/output values.
1255
1256\begin{comment} % put in appendix with coroutine version ???
1257As such the @main@ routine of a thread can be defined as
1258\begin{cfa}
1259thread foo {};
1260
1261void main(foo & this) {
1262        sout | "Hello World!";
1263}
1264\end{cfa}
1265
1266In this example, threads of type @foo@ start execution in the @void main(foo &)@ routine, which prints @"Hello World!".@ While this paper encourages this approach to enforce strongly typed programming, users may prefer to use the routine-based thread semantics for the sake of simplicity.
1267With the static semantics it is trivial to write a thread type that takes a routine pointer as a parameter and executes it on its stack asynchronously.
1268\begin{cfa}
1269typedef void (*voidRtn)(int);
1270
1271thread RtnRunner {
1272        voidRtn func;
1273        int arg;
1274};
1275
1276void ?{}(RtnRunner & this, voidRtn inRtn, int arg) {
1277        this.func = inRtn;
1278        this.arg  = arg;
1279}
1280
1281void main(RtnRunner & this) {
1282        // thread starts here and runs the routine
1283        this.func( this.arg );
1284}
1285
1286void hello(/*unused*/ int) {
1287        sout | "Hello World!";
1288}
1289
1290int main() {
1291        RtnRunner f = {hello, 42};
1292        return 0?
1293}
1294\end{cfa}
1295A consequence of the strongly typed approach to main is that memory layout of parameters and return values to/from a thread are now explicitly specified in the \textbf{API}.
1296\end{comment}
1297
1298For user threads to be useful, it must be possible to start and stop the underlying thread, and wait for it to complete execution.
1299While using an API such as @fork@ and @join@ is relatively common, such an interface is awkward and unnecessary.
1300A simple approach is to use allocation/deallocation principles, and have threads implicitly @fork@ after construction and @join@ before destruction.
1301\begin{cfa}
1302thread World {};
1303void main( World & this ) {
1304        sout | "World!";
1305}
1306int main() {
1307        World w`[10]`;                                                  $\C{// implicit forks after creation}$
1308        sout | "Hello ";                                        $\C{// "Hello " and 10 "World!" printed concurrently}$
1309}                                                                                       $\C{// implicit joins before destruction}$
1310\end{cfa}
1311This semantics ensures a thread is started and stopped exactly once, eliminating some programming error, and scales to multiple threads for basic (termination) synchronization.
1312This tree-structure (lattice) create/delete from C block-structure is generalized by using dynamic allocation, so threads can outlive the scope in which they are created, much like dynamically allocating memory lets objects outlive the scope in which they are created.
1313\begin{cfa}
1314int main() {
1315        MyThread * heapLive;
1316        {
1317                MyThread blockLive;                                     $\C{// fork block-based thread}$
1318                heapLive = `new`( MyThread );           $\C{// fork heap-based thread}$
1319                ...
1320        }                                                                               $\C{// join block-based thread}$
1321        ...
1322        `delete`( heapLive );                                   $\C{// join heap-based thread}$
1323}
1324\end{cfa}
1325The heap-based approach allows arbitrary thread-creation topologies, with respect to fork/join-style concurrency.
1326
1327Figure~\ref{s:ConcurrentMatrixSummation} shows concurrently adding the rows of a matrix and then totalling the subtotals sequentially, after all the row threads have terminated.
1328The program uses heap-based threads because each thread needs different constructor values.
1329(Python provides a simple iteration mechanism to initialize array elements to different values allowing stack allocation.)
1330The allocation/deallocation pattern appears unusual because allocated objects are immediately deallocated without any intervening code.
1331However, for threads, the deletion provides implicit synchronization, which is the intervening code.
1332While the subtotals are added in linear order rather than completion order, which slightly inhibits concurrency, the computation is restricted by the critical-path thread (\ie the thread that takes the longest), and so any inhibited concurrency is very small as totalling the subtotals is trivial.
1333
1334\begin{figure}
1335\begin{cfa}
1336`thread` Adder { int * row, cols, & subtotal; } $\C{// communication variables}$
1337void ?{}( Adder & adder, int row[], int cols, int & subtotal ) {
1338    adder.[ row, cols, &subtotal ] = [ row, cols, &subtotal ];
1339}
1340void main( Adder & adder ) with( adder ) {
1341    subtotal = 0;
1342    for ( int c = 0; c < cols; c += 1 ) { subtotal += row[c]; }
1343}
1344int main() {
1345    const int rows = 10, cols = 1000;
1346    int matrix[rows][cols], subtotals[rows], total = 0;
1347    // read matrix
1348    Adder * adders[rows];
1349    for ( int r = 0; r < rows; r += 1 ) {       $\C{// start threads to sum rows}$
1350                adders[r] = `new( matrix[r], cols, &subtotals[r] );`
1351    }
1352    for ( int r = 0; r < rows; r += 1 ) {       $\C{// wait for threads to finish}$
1353                `delete( adders[r] );`                          $\C{// termination join}$
1354                total += subtotals[r];                          $\C{// total subtotal}$
1355    }
1356    sout | total;
1357}
1358\end{cfa}
1359\caption{Concurrent Matrix Summation}
1360\label{s:ConcurrentMatrixSummation}
1361\end{figure}
1362
1363
1364\section{Mutual Exclusion / Synchronization}
1365
1366Uncontrolled non-deterministic execution is meaningless.
1367To reestablish meaningful execution requires mechanisms to reintroduce determinism, \ie restrict non-determinism, called mutual exclusion and synchronization, where mutual exclusion is an access-control mechanism on data shared by threads, and synchronization is a timing relationship among threads~\cite[\S~4]{Buhr05a}.
1368Since many deterministic challenges appear with the use of mutable shared state, some languages/libraries disallow it, \eg Erlang~\cite{Erlang}, Haskell~\cite{Haskell}, Akka~\cite{Akka} (Scala).
1369In these paradigms, interaction among concurrent objects is performed by stateless message-passing~\cite{Thoth,Harmony,V-Kernel} or other paradigms closely related to networking concepts, \eg channels~\cite{CSP,Go}.
1370However, in call/return-based languages, these approaches force a clear distinction, \ie introduce a new programming paradigm between regular and concurrent computation, \eg routine call versus message passing.
1371Hence, a programmer must learn and manipulate two sets of design patterns.
1372While this distinction can be hidden away in library code, effective use of the library still has to take both paradigms into account.
1373In contrast, approaches based on stateful models more closely resemble the standard call/return programming-model, resulting in a single programming paradigm.
1374
1375At the lowest level, concurrent control is implemented by atomic operations, upon which different kinds of locking mechanisms are constructed, \eg semaphores~\cite{Dijkstra68b}, barriers, and path expressions~\cite{Campbell74}.
1376However, for productivity it is always desirable to use the highest-level construct that provides the necessary efficiency~\cite{Hochstein05}.
1377A newer approach for restricting non-determinism is transactional memory~\cite{Herlihy93}.
1378While this approach is pursued in hardware~\cite{Nakaike15} and system languages, like \CC~\cite{Cpp-Transactions}, the performance and feature set is still too restrictive to be the main concurrency paradigm for system languages, which is why it is rejected as the core paradigm for concurrency in \CFA.
1379
1380One of the most natural, elegant, and efficient mechanisms for mutual exclusion and synchronization for shared-memory systems is the \emph{monitor}.
1381First proposed by Brinch Hansen~\cite{Hansen73} and later described and extended by C.A.R.~Hoare~\cite{Hoare74}, many concurrent programming-languages provide monitors as an explicit language construct: \eg Concurrent Pascal~\cite{ConcurrentPascal}, Mesa~\cite{Mesa}, Modula~\cite{Modula-2}, Turing~\cite{Turing:old}, Modula-3~\cite{Modula-3}, NeWS~\cite{NeWS}, Emerald~\cite{Emerald}, \uC~\cite{Buhr92a} and Java~\cite{Java}.
1382In addition, operating-system kernels and device drivers have a monitor-like structure, although they often use lower-level primitives such as mutex locks or semaphores to simulate monitors.
1383For these reasons, \CFA selected monitors as the core high-level concurrency-construct, upon which higher-level approaches can be easily constructed.
1384
1385
1386\subsection{Mutual Exclusion}
1387
1388A group of instructions manipulating a specific instance of shared data that must be performed atomically is called an (individual) \newterm{critical-section}~\cite{Dijkstra65}.
1389The generalization is called a \newterm{group critical-section}~\cite{Joung00}, where multiple tasks with the same session may use the resource simultaneously, but different sessions may not use the resource simultaneously.
1390The readers/writer problem~\cite{Courtois71} is an instance of a group critical-section, where readers have the same session and all writers have a unique session.
1391\newterm{Mutual exclusion} enforces that the correct kind and number of threads are using a critical section.
1392
1393However, many solutions exist for mutual exclusion, which vary in terms of performance, flexibility and ease of use.
1394Methods range from low-level locks, which are fast and flexible but require significant attention for correctness, to higher-level concurrency techniques, which sacrifice some performance to improve ease of use.
1395Ease of use comes by either guaranteeing some problems cannot occur, \eg deadlock free, or by offering a more explicit coupling between shared data and critical section.
1396For example, the \CC @std::atomic<T>@ offers an easy way to express mutual-exclusion on a restricted set of operations, \eg reading/writing, for numerical types.
1397However, a significant challenge with locks is composability because it takes careful organization for multiple locks to be used while preventing deadlock.
1398Easing composability is another feature higher-level mutual-exclusion mechanisms can offer.
1399
1400
1401\subsection{Synchronization}
1402
1403Synchronization enforces relative ordering of execution, and synchronization tools provide numerous mechanisms to establish these timing relationships.
1404Low-level synchronization primitives offer good performance and flexibility at the cost of ease of use;
1405higher-level mechanisms often simplify usage by adding better coupling between synchronization and data, \eg message passing, or offering a simpler solution to otherwise involved challenges, \eg barrier lock.
1406Often synchronization is used to order access to a critical section, \eg ensuring a reader thread is the next kind of thread to enter a critical section.
1407If a writer thread is scheduled for next access, but another reader thread acquires the critical section first, that reader \newterm{barged}.
1408Barging can result in staleness/freshness problems, where a reader barges ahead of a writer and reads temporally stale data, or a writer barges ahead of another writer overwriting data with a fresh value preventing the previous value from ever being read (lost computation).
1409Preventing or detecting barging is an involved challenge with low-level locks, which can be made much easier by higher-level constructs.
1410This challenge is often split into two different approaches: barging avoidance and barging prevention.
1411Algorithms that allow a barger, but divert it until later using current synchronization state (flags), are avoiding the barger;
1412algorithms that preclude a barger from entering during synchronization in the critical section prevent barging completely.
1413Techniques like baton-passing locks~\cite{Andrews89} between threads instead of unconditionally releasing locks is an example of barging prevention.
1414
1415
1416\section{Monitor}
1417\label{s:Monitor}
1418
1419A \textbf{monitor} is a set of routines that ensure mutual exclusion when accessing shared state.
1420More precisely, a monitor is a programming technique that binds mutual exclusion to routine scope, as opposed to locks, where mutual-exclusion is defined by acquire/release calls, independent of lexical context (analogous to block and heap storage allocation).
1421The strong association with the call/return paradigm eases programmability, readability and maintainability, at a slight cost in flexibility and efficiency.
1422
1423Note, like coroutines/threads, both locks and monitors require an abstract handle to reference them, because at their core, both mechanisms are manipulating non-copyable shared-state.
1424Copying a lock is insecure because it is possible to copy an open lock and then use the open copy when the original lock is closed to simultaneously access the shared data.
1425Copying a monitor is secure because both the lock and shared data are copies, but copying the shared data is meaningless because it no longer represents a unique entity.
1426As for coroutines/tasks, the @dtype@ property provides no implicit copying operations and the @is_monitor@ trait provides no explicit copying operations, so all locks/monitors must be passed by reference (pointer).
1427\begin{cfa}
1428trait is_monitor( `dtype` T ) {
1429        monitor_desc * get_monitor( T & );
1430        void ^?{}( T & mutex );
1431};
1432\end{cfa}
1433
1434
1435\subsection{Mutex Acquisition}
1436\label{s:MutexAcquisition}
1437
1438While correctness implies a monitor's mutual exclusion is acquired and released, there are implementation options about when and where the locking/unlocking occurs.
1439(Much of this discussion also applies to basic locks.)
1440For example, a monitor may need to be passed through multiple helper routines before it becomes necessary to acquire the monitor mutual-exclusion.
1441\begin{cfa}[morekeywords=nomutex]
1442monitor Aint { int cnt; };                                      $\C{// atomic integer counter}$
1443void ?{}( Aint & `nomutex` this ) with( this ) { cnt = 0; } $\C{// constructor}$
1444int ?=?( Aint & `mutex`$\(_{opt}\)$ lhs, int rhs ) with( lhs ) { cnt = rhs; } $\C{// conversions}$
1445void ?{}( int & this, Aint & `mutex`$\(_{opt}\)$ v ) { this = v.cnt; }
1446int ?=?( int & lhs, Aint & `mutex`$\(_{opt}\)$ rhs ) with( rhs ) { lhs = cnt; }
1447int ++?( Aint & `mutex`$\(_{opt}\)$ this ) with( this ) { return ++cnt; } $\C{// increment}$
1448\end{cfa}
1449The @Aint@ constructor, @?{}@, uses the \lstinline[morekeywords=nomutex]@nomutex@ qualifier indicating mutual exclusion is unnecessary during construction because an object is inaccessible (private) until after it is initialized.
1450(While a constructor may publish its address into a global variable, doing so generates a race-condition.)
1451The conversion operators for initializing and assigning with a normal integer only need @mutex@, if reading/writing the implementation type is not atomic.
1452Finally, the prefix increment operato, @++?@, is normally @mutex@ to protect the incrementing from race conditions, unless there is an atomic increment instruction for the implementation type.
1453
1454The atomic counter is used without any explicit mutual-exclusion and provides thread-safe semantics, which is similar to the \CC template @std::atomic@.
1455\begin{cfa}
1456Aint x, y, z;
1457++x; ++y; ++z;                                                          $\C{// safe increment by multiple threads}$
1458x = 2; y = 2; z = 2;                                            $\C{// conversions}$
1459int i = x, j = y, k = z;
1460i = x; j = y; k = z;
1461\end{cfa}
1462
1463For maximum usability, monitors have \newterm{multi-acquire} semantics allowing a thread to acquire it multiple times without deadlock.
1464\begin{cfa}
1465monitor M { ... } m;
1466void foo( M & mutex m ) { ... }                         $\C{// acquire mutual exclusion}$
1467void bar( M & mutex m ) {                                       $\C{// acquire mutual exclusion}$
1468        ... `foo( m );` ...                                             $\C{// reacquire mutual exclusion}$
1469}
1470`bar( m );`                                                                     $\C{// nested monitor call}$
1471\end{cfa}
1472
1473The benefit of mandatory monitor qualifiers is self-documentation, but requiring both @mutex@ and \lstinline[morekeywords=nomutex]@nomutex@ for all monitor parameters is redundant.
1474Instead, the semantics have one qualifier as the default, and the other required.
1475For example, make the safe @mutex@ qualifier the default because assuming \lstinline[morekeywords=nomutex]@nomutex@ may cause subtle errors.
1476Alternatively, make the unsafe \lstinline[morekeywords=nomutex]@nomutex@ qualifier the default because it is the \emph{normal} parameter semantics while @mutex@ parameters are rare.
1477Providing a default qualifier implies knowing whether a parameter is a monitor.
1478Since \CFA relies heavily on traits as an abstraction mechanism, the distinction between a type that is a monitor and a type that looks like a monitor can become blurred.
1479For this reason, \CFA requires programmers to identify the kind of parameter with the @mutex@ keyword and uses no keyword to mean \lstinline[morekeywords=nomutex]@nomutex@.
1480
1481The next semantic decision is establishing which parameter \emph{types} may be qualified with @mutex@.
1482Given:
1483\begin{cfa}
1484monitor M { ... }
1485int f1( M & mutex m );
1486int f2( M * mutex m );
1487int f3( M * mutex m[] );
1488int f4( stack( M * ) & mutex m );
1489\end{cfa}
1490the issue is that some of these parameter types are composed of multiple objects.
1491For @f1@, there is only a single parameter object.
1492Adding indirection in @f2@ still identifies a single object.
1493However, the matrix in @f3@ introduces multiple objects.
1494While shown shortly, multiple acquisition is possible;
1495however array lengths are often unknown in C.
1496This issue is exacerbated in @f4@, where the data structure must be safely traversed to acquire all of its elements.
1497
1498To make the issue tractable, \CFA only acquires one monitor per parameter with at most one level of indirection.
1499However, there is an ambiguity in the C type-system with respects to arrays.
1500Is the argument for @f2@ a single object or an array of objects?
1501If it is an array, only the first element of the array is acquired, which seems unsafe;
1502hence, @mutex@ is disallowed for array parameters.
1503\begin{cfa}
1504int f1( M & mutex m );                                          $\C{// allowed: recommended case}$
1505int f2( M * mutex m );                                          $\C{// disallowed: could be an array}$
1506int f3( M mutex m[$\,$] );                                      $\C{// disallowed: array length unknown}$
1507int f4( M ** mutex m );                                         $\C{// disallowed: could be an array}$
1508int f5( M * mutex m[$\,$] );                            $\C{// disallowed: array length unknown}$
1509\end{cfa}
1510% Note, not all array routines have distinct types: @f2@ and @f3@ have the same type, as do @f4@ and @f5@.
1511% However, even if the code generation could tell the difference, the extra information is still not sufficient to extend meaningfully the monitor call semantic.
1512
1513For object-oriented monitors, calling a mutex member \emph{implicitly} acquires mutual exclusion of the receiver object, @`rec`.foo(...)@.
1514\CFA has no receiver, and hence, must use an explicit mechanism to specify which object acquires mutual exclusion.
1515A positive consequence of this design decision is the ability to support multi-monitor routines.
1516\begin{cfa}
1517int f( M & mutex x, M & mutex y );              $\C{// multiple monitor parameter of any type}$
1518M m1, m2;
1519f( m1, m2 );
1520\end{cfa}
1521(While object-oriented monitors can be extended with a mutex qualifier for multiple-monitor members, no prior example of this feature could be found.)
1522In practice, writing multi-locking routines that do not deadlock is tricky.
1523Having language support for such a feature is therefore a significant asset for \CFA.
1524
1525The capability to acquire multiple locks before entering a critical section is called \newterm{bulk acquire} (see Section~\ref{s:Implementation} for implementation details).
1526In the previous example, \CFA guarantees the order of acquisition is consistent across calls to different routines using the same monitors as arguments.
1527This consistent ordering means acquiring multiple monitors is safe from deadlock.
1528However, users can force the acquiring order.
1529For example, notice the use of @mutex@/\lstinline[morekeywords=nomutex]@nomutex@ and how this affects the acquiring order:
1530\begin{cfa}
1531void foo( M & mutex m1, M & mutex m2 );         $\C{// acquire m1 and m2}$
1532void bar( M & mutex m1, M & /* nomutex */ m2 ) { $\C{// acquire m1}$
1533        ... foo( m1, m2 ); ...                                  $\C{// acquire m2}$
1534}
1535void baz( M & /* nomutex */ m1, M & mutex m2 ) { $\C{// acquire m2}$
1536        ... foo( m1, m2 ); ...                                  $\C{// acquire m1}$
1537}
1538\end{cfa}
1539The multi-acquire semantics allows @bar@ or @baz@ to acquire a monitor lock and reacquire it in @foo@.
1540In the calls to @bar@ and @baz@, the monitors are acquired in opposite order.
1541
1542However, such use leads to lock acquiring order problems resulting in deadlock~\cite{Lister77}, where detecting it requires dynamic tracking of monitor calls, and dealing with it requires rollback semantics~\cite{Dice10}.
1543In \CFA, a safety aid is provided by using bulk acquire of all monitors to shared objects, whereas other monitor systems provide no aid.
1544While \CFA provides only a partial solution, it handles many useful cases, \eg:
1545\begin{cfa}
1546monitor BankAccount { ... };
1547void deposit( BankAccount & `mutex` b, int deposit );
1548void transfer( BankAccount & `mutex` my, BankAccount & `mutex` your, int me2you ) {
1549        deposit( my, `-`me2you );                               $\C{// debit}$
1550        deposit( your, me2you );                                $\C{// credit}$
1551}
1552\end{cfa}
1553This example shows a trivial solution to the bank-account transfer problem.
1554Without multi- and bulk acquire, the solution to this problem requires careful engineering.
1555
1556
1557\subsection{\protect\lstinline@mutex@ statement}
1558\label{mutex-stmt}
1559
1560The monitor call-semantics associate all locking semantics to routines.
1561Like Java, \CFA offers an alternative @mutex@ statement to reduce refactoring and naming.
1562\begin{cquote}
1563\begin{tabular}{@{}l@{\hspace{3\parindentlnth}}l@{}}
1564\begin{cfa}
1565monitor M { ... };
1566void foo( M & mutex m1, M & mutex m2 ) {
1567        // critical section
1568}
1569void bar( M & m1, M & m2 ) {
1570        foo( m1, m2 );
1571}
1572\end{cfa}
1573&
1574\begin{cfa}
1575
1576void bar( M & m1, M & m2 ) {
1577        mutex( m1, m2 ) {       // remove refactoring and naming
1578                // critical section
1579        }
1580}
1581
1582\end{cfa}
1583\\
1584\multicolumn{1}{c}{\textbf{routine call}} & \multicolumn{1}{c}{\lstinline@mutex@ \textbf{statement}}
1585\end{tabular}
1586\end{cquote}
1587
1588
1589\section{Scheduling}
1590\label{s:Scheduling}
1591
1592While monitor mutual-exclusion provides safe access to shared data, the monitor data may indicate that a thread accessing it cannot proceed.
1593For example, Figure~\ref{f:GenericBoundedBuffer} shows a bounded buffer that may be full/empty so produce/consumer threads must block.
1594Leaving the monitor and trying again (busy waiting) is impractical for high-level programming.
1595Monitors eliminate busy waiting by providing synchronization to schedule threads needing access to the shared data, where threads block versus spinning.
1596Synchronization is generally achieved with internal~\cite{Hoare74} or external~\cite[\S~2.9.2]{uC++} scheduling, where \newterm{scheduling} defines which thread acquires the critical section next.
1597\newterm{Internal scheduling} is characterized by each thread entering the monitor and making an individual decision about proceeding or blocking, while \newterm{external scheduling} is characterized by an entering thread making a decision about proceeding for itself and on behalf of other threads attempting entry.
1598
1599Figure~\ref{f:BBInt} shows a \CFA generic bounded-buffer with internal scheduling, where producers/consumers enter the monitor, see the buffer is full/empty, and block on an appropriate condition lock, @full@/@empty@.
1600The @wait@ routine atomically blocks the calling thread and implicitly releases the monitor lock(s) for all monitors in the routine's parameter list.
1601The appropriate condition lock is signalled to unblock an opposite kind of thread after an element is inserted/removed from the buffer.
1602Signalling is unconditional, because signalling an empty condition lock does nothing.
1603
1604Signalling semantics cannot have the signaller and signalled thread in the monitor simultaneously, which means:
1605\begin{enumerate}
1606\item
1607The signalling thread returns immediately, and the signalled thread continues.
1608\item
1609The signalling thread continues and the signalled thread is marked for urgent unblocking at the next scheduling point (exit/wait).
1610\item
1611The signalling thread blocks but is marked for urgrent unblocking at the next scheduling point and the signalled thread continues.
1612\end{enumerate}
1613The first approach is too restrictive, as it precludes solving a reasonable class of problems, \eg dating service (see Figure~\ref{f:DatingService}).
1614\CFA supports the next two semantics as both are useful.
1615Finally, while it is common to store a @condition@ as a field of the monitor, in \CFA, a @condition@ variable can be created/stored independently.
1616Furthermore, a condition variable is tied to a \emph{group} of monitors on first use, called \newterm{branding}, which means that using internal scheduling with distinct sets of monitors requires one condition variable per set of monitors.
1617
1618\begin{figure}
1619\centering
1620\newbox\myboxA
1621\begin{lrbox}{\myboxA}
1622\begin{cfa}[aboveskip=0pt,belowskip=0pt]
1623forall( otype T ) { // distribute forall
1624        monitor Buffer {
1625                `condition` full, empty;
1626                int front, back, count;
1627                T elements[10];
1628        };
1629        void ?{}( Buffer(T) & buffer ) with(buffer) {
1630                [front, back, count] = 0;
1631        }
1632
1633        void insert( Buffer(T) & mutex buffer, T elem )
1634                                with(buffer) {
1635                if ( count == 10 ) `wait( empty )`;
1636                // insert elem into buffer
1637                `signal( full )`;
1638        }
1639        T remove( Buffer(T) & mutex buffer ) with(buffer) {
1640                if ( count == 0 ) `wait( full )`;
1641                // remove elem from buffer
1642                `signal( empty )`;
1643                return elem;
1644        }
1645}
1646\end{cfa}
1647\end{lrbox}
1648
1649\newbox\myboxB
1650\begin{lrbox}{\myboxB}
1651\begin{cfa}[aboveskip=0pt,belowskip=0pt]
1652forall( otype T ) { // distribute forall
1653        monitor Buffer {
1654
1655                int front, back, count;
1656                T elements[10];
1657        };
1658        void ?{}( Buffer(T) & buffer ) with(buffer) {
1659                [front, back, count] = 0;
1660        }
1661        T remove( Buffer(T) & mutex buffer ); // forward
1662        void insert( Buffer(T) & mutex buffer, T elem )
1663                                with(buffer) {
1664                if ( count == 10 ) `waitfor( remove, buffer )`;
1665                // insert elem into buffer
1666
1667        }
1668        T remove( Buffer(T) & mutex buffer ) with(buffer) {
1669                if ( count == 0 ) `waitfor( insert, buffer )`;
1670                // remove elem from buffer
1671
1672                return elem;
1673        }
1674}
1675\end{cfa}
1676\end{lrbox}
1677
1678\subfloat[Internal Scheduling]{\label{f:BBInt}\usebox\myboxA}
1679%\qquad
1680\subfloat[External Scheduling]{\label{f:BBExt}\usebox\myboxB}
1681\caption{Generic Bounded-Buffer}
1682\label{f:GenericBoundedBuffer}
1683\end{figure}
1684
1685Figure~\ref{f:BBExt} shows a \CFA generic bounded-buffer with external scheduling, where producers/consumers detecting a full/empty buffer block and prevent more producers/consumers from entering the monitor until there is a free/empty slot in the buffer.
1686External scheduling is controlled by the @waitfor@ statement, which atomically blocks the calling thread, releases the monitor lock, and restricts the routine calls that can next acquire mutual exclusion.
1687If the buffer is full, only calls to @remove@ can acquire the buffer, and if the buffer is empty, only calls to @insert@ can acquire the buffer.
1688Threads making calls to routines that are currently excluded, block outside of (external to) the monitor on a calling queue, versus blocking on condition queues inside of (internal to) the monitor.
1689External scheduling allows users to wait for events from other threads without concern of unrelated events occurring.
1690The mechnaism can be done in terms of control flow, \eg Ada @accept@ or \uC @_Accept@, or in terms of data, \eg Go channels.
1691While both mechanisms have strengths and weaknesses, this project uses a control-flow mechanism to stay consistent with other language semantics.
1692Two challenges specific to \CFA for external scheduling are loose object-definitions (see Section~\ref{s:LooseObjectDefinitions}) and multiple-monitor routines (see Section~\ref{s:Multi-MonitorScheduling}).
1693
1694For internal scheduling, non-blocking signalling (as in the producer/consumer example) is used when the signaller is providing the cooperation for a waiting thread;
1695the signaller enters the monitor and changes state, detects a waiting threads that can use the state, performs a non-blocking signal on the condition queue for the waiting thread, and exits the monitor to run concurrently.
1696The waiter unblocks next from the urgent queue, uses/takes the state, and exits the monitor.
1697Blocking signalling is the reverse, where the waiter is providing the cooperation for the signalling thread;
1698the signaller enters the monitor, detects a waiting thread providing the necessary state, performs a blocking signal to place it on the urgent queue and unblock the waiter.
1699The waiter changes state and exits the monitor, and the signaller unblocks next from the urgent queue to use/take the state.
1700
1701Figure~\ref{f:DatingService} shows a dating service demonstrating non-blocking and blocking signalling.
1702The dating service matches girl and boy threads with matching compatibility codes so they can exchange phone numbers.
1703A thread blocks until an appropriate partner arrives.
1704The complexity is exchanging phone numbers in the monitor because of the mutual-exclusion property.
1705For signal scheduling, the @exchange@ condition is necessary to block the thread finding the match, while the matcher unblocks to take the opposite number, post its phone number, and unblock the partner.
1706For signal-block scheduling, the implicit urgent-queue replaces the explict @exchange@-condition and @signal_block@ puts the finding thread on the urgent condition and unblocks the matcher.
1707
1708The dating service is an example of a monitor that cannot be written using external scheduling because it requires knowledge of calling parameters to make scheduling decisions, and parameters of waiting threads are unavailable;
1709as well, an arriving thread may not find a partner and must wait, which requires a condition variable, and condition variables imply internal scheduling.
1710
1711\begin{figure}
1712\centering
1713\newbox\myboxA
1714\begin{lrbox}{\myboxA}
1715\begin{cfa}[aboveskip=0pt,belowskip=0pt]
1716enum { CCodes = 20 };
1717monitor DS {
1718        int GirlPhNo, BoyPhNo;
1719        condition Girls[CCodes], Boys[CCodes];
1720        condition exchange;
1721};
1722int girl( DS & mutex ds, int phNo, int ccode ) {
1723        if ( is_empty( Boys[ccode] ) ) {
1724                wait( Girls[ccode] );
1725                GirlPhNo = phNo;
1726                `signal( exchange );`
1727        } else {
1728                GirlPhNo = phNo;
1729                `signal( Boys[ccode] );`
1730                `wait( exchange );`
1731        } // if
1732        return BoyPhNo;
1733}
1734int boy( DS & mutex ds, int phNo, int ccode ) {
1735        // as above with boy/girl interchanged
1736}
1737\end{cfa}
1738\end{lrbox}
1739
1740\newbox\myboxB
1741\begin{lrbox}{\myboxB}
1742\begin{cfa}[aboveskip=0pt,belowskip=0pt]
1743
1744monitor DS {
1745        int GirlPhNo, BoyPhNo;
1746        condition Girls[CCodes], Boys[CCodes];
1747
1748};
1749int girl( DS & mutex ds, int phNo, int ccode ) {
1750        if ( is_empty( Boys[ccode] ) ) { // no compatible
1751                wait( Girls[ccode] ); // wait for boy
1752                GirlPhNo = phNo; // make phone number available
1753
1754        } else {
1755                GirlPhNo = phNo; // make phone number available
1756                `signal_block( Boys[ccode] );` // restart boy
1757
1758        } // if
1759        return BoyPhNo;
1760}
1761int boy( DS & mutex ds, int phNo, int ccode ) {
1762        // as above with boy/girl interchanged
1763}
1764\end{cfa}
1765\end{lrbox}
1766
1767\subfloat[\lstinline@signal@]{\label{f:DatingSignal}\usebox\myboxA}
1768\qquad
1769\subfloat[\lstinline@signal_block@]{\label{f:DatingSignalBlock}\usebox\myboxB}
1770\caption{Dating service. }
1771\label{f:DatingService}
1772\end{figure}
1773
1774Both internal and external scheduling extend to multiple monitors in a natural way.
1775\begin{cquote}
1776\begin{tabular}{@{}l@{\hspace{3\parindentlnth}}l@{}}
1777\begin{cfa}
1778monitor M { `condition e`; ... };
1779void foo( M & mutex m1, M & mutex m2 ) {
1780        ... wait( `e` ); ...   // wait( e, m1, m2 )
1781        ... wait( `e, m1` ); ...
1782        ... wait( `e, m2` ); ...
1783}
1784\end{cfa}
1785&
1786\begin{cfa}
1787void rtn$\(_1\)$( M & mutex m1, M & mutex m2 );
1788void rtn$\(_2\)$( M & mutex m1 );
1789void bar( M & mutex m1, M & mutex m2 ) {
1790        ... waitfor( `rtn` ); ...       // $\LstCommentStyle{waitfor( rtn\(_1\), m1, m2 )}$
1791        ... waitfor( `rtn, m1` ); ... // $\LstCommentStyle{waitfor( rtn\(_2\), m1 )}$
1792}
1793\end{cfa}
1794\end{tabular}
1795\end{cquote}
1796For @wait( e )@, the default semantics is to atomically block the signaller and release all acquired mutex types in the parameter list, \ie @wait( e, m1, m2 )@.
1797To override the implicit multi-monitor wait, specific mutex parameter(s) can be specified, \eg @wait( e, m1 )@.
1798Wait statically verifies the released monitors are the acquired mutex-parameters so unconditional release is safe.
1799Finally, a signaller,
1800\begin{cfa}
1801void baz( M & mutex m1, M & mutex m2 ) {
1802        ... signal( e ); ...
1803}
1804\end{cfa}
1805must have acquired at least the same locks as the waiting thread signalled from the condition queue.
1806
1807Similarly, for @waitfor( rtn )@, the default semantics is to atomically block the acceptor and release all acquired mutex types in the parameter list, \ie @waitfor( rtn, m1, m2 )@.
1808To override the implicit multi-monitor wait, specific mutex parameter(s) can be specified, \eg @waitfor( rtn, m1 )@.
1809@waitfor@ statically verifies the released monitors are the same as the acquired mutex-parameters of the given routine or routine pointer.
1810To statically verify the released monitors match with the accepted routine's mutex parameters, the routine (pointer) prototype must be accessible.
1811% When an overloaded routine appears in an @waitfor@ statement, calls to any routine with that name are accepted.
1812% The rationale is that members with the same name should perform a similar function, and therefore, all should be eligible to accept a call.
1813Overloaded routines can be disambiguated using a cast:
1814\begin{cfa}
1815void rtn( M & mutex m );
1816`int` rtn( M & mutex m );
1817waitfor( (`int` (*)( M & mutex ))rtn, m );
1818\end{cfa}
1819
1820The ability to release a subset of acquired monitors can result in a \newterm{nested monitor}~\cite{Lister77} deadlock.
1821\begin{cfa}
1822void foo( M & mutex m1, M & mutex m2 ) {
1823        ... wait( `e, m1` ); ...                                $\C{// release m1, keeping m2 acquired )}$
1824void bar( M & mutex m1, M & mutex m2 ) {        $\C{// must acquire m1 and m2 )}$
1825        ... signal( `e` ); ...
1826\end{cfa}
1827The @wait@ only releases @m1@ so the signalling thread cannot acquire both @m1@ and @m2@ to  enter @bar@ to get to the @signal@.
1828While deadlock issues can occur with multiple/nesting acquisition, this issue results from the fact that locks, and by extension monitors, are not perfectly composable.
1829
1830Finally, an important aspect of monitor implementation is barging, \ie can calling threads barge ahead of signalled threads?
1831If barging is allowed, synchronization between a signaller and signallee is difficult, often requiring multiple unblock/block cycles (looping around a wait rechecking if a condition is met).
1832In fact, signals-as-hints is completely opposite from that proposed by Hoare in the seminal paper on monitors:
1833\begin{quote}
1834However, we decree that a signal operation be followed immediately by resumption of a waiting program, without possibility of an intervening procedure call from yet a third program.
1835It is only in this way that a waiting program has an absolute guarantee that it can acquire the resource just released by the signalling program without any danger that a third program will interpose a monitor entry and seize the resource instead.~\cite[p.~550]{Hoare74}
1836\end{quote}
1837\CFA scheduling \emph{precludes} barging, which simplifies synchronization among threads in the monitor and increases correctness.
1838Furthermore, \CFA concurrency has no spurious wakeup~\cite[\S~9]{Buhr05a}, which eliminates an implict form of barging.
1839For example, there are no loops in either bounded buffer solution in Figure~\ref{f:GenericBoundedBuffer}.
1840Supporting barging prevention as well as extending internal scheduling to multiple monitors is the main source of complexity in the design and implementation of \CFA concurrency.
1841
1842
1843\subsection{Barging Prevention}
1844
1845Figure~\ref{f:BargingPrevention} shows \CFA code where bulk acquire adds complexity to the internal-signalling semantics.
1846The complexity begins at the end of the inner @mutex@ statement, where the semantics of internal scheduling need to be extended for multiple monitors.
1847The problem is that bulk acquire is used in the inner @mutex@ statement where one of the monitors is already acquired.
1848When the signalling thread reaches the end of the inner @mutex@ statement, it should transfer ownership of @m1@ and @m2@ to the waiting threads to prevent barging into the outer @mutex@ statement by another thread.
1849However, both the signalling and waiting thread W1 still need monitor @m1@.
1850
1851\begin{figure}
1852\newbox\myboxA
1853\begin{lrbox}{\myboxA}
1854\begin{cfa}[aboveskip=0pt,belowskip=0pt]
1855monitor M m1, m2;
1856condition c;
1857mutex( m1 ) { // $\LstCommentStyle{\color{red}outer}$
1858        ...
1859        mutex( m1, m2 ) { // $\LstCommentStyle{\color{red}inner}$
1860                ... `signal( c )`; ...
1861                // m1, m2 acquired
1862        } // $\LstCommentStyle{\color{red}release m2}$
1863        // m1 acquired
1864} // release m1
1865\end{cfa}
1866\end{lrbox}
1867
1868\newbox\myboxB
1869\begin{lrbox}{\myboxB}
1870\begin{cfa}[aboveskip=0pt,belowskip=0pt]
1871
1872
1873mutex( m1 ) {
1874        ...
1875        mutex( m1, m2 ) {
1876                ... `wait( c )`; // block and release m1, m2
1877                // m1, m2 acquired
1878        } // $\LstCommentStyle{\color{red}release m2}$
1879        // m1 acquired
1880} // release m1
1881\end{cfa}
1882\end{lrbox}
1883
1884\newbox\myboxC
1885\begin{lrbox}{\myboxC}
1886\begin{cfa}[aboveskip=0pt,belowskip=0pt]
1887
1888
1889mutex( m2 ) {
1890        ... `wait( c )`; ...
1891        // m2 acquired
1892} // $\LstCommentStyle{\color{red}release m2}$
1893
1894
1895
1896
1897\end{cfa}
1898\end{lrbox}
1899
1900\begin{cquote}
1901\subfloat[Signalling Thread]{\label{f:SignallingThread}\usebox\myboxA}
1902\hspace{2\parindentlnth}
1903\subfloat[Waiting Thread (W1)]{\label{f:WaitingThread}\usebox\myboxB}
1904\hspace{2\parindentlnth}
1905\subfloat[Waiting Thread (W2)]{\label{f:OtherWaitingThread}\usebox\myboxC}
1906\end{cquote}
1907\caption{Barging Prevention}
1908\label{f:BargingPrevention}
1909\end{figure}
1910
1911One scheduling solution is for the signaller to keep ownership of all locks until the last lock is ready to be transferred, because this semantics fits most closely to the behaviour of single-monitor scheduling.
1912However, Figure~\ref{f:OtherWaitingThread} shows this solution is complex depending on other waiters, resulting in options when the signaller finishes the inner mutex-statement.
1913The signaller can retain @m2@ until completion of the outer mutex statement and pass the locks to waiter W1, or it can pass @m2@ to waiter W2 after completing the inner mutex-statement, while continuing to hold @m1@.
1914In the latter case, waiter W2 must eventually pass @m2@ to waiter W1, which is complex because W1 may have waited before W2, so W2 is unaware of it.
1915Furthermore, there is an execution sequence where the signaller always finds waiter W2, and hence, waiter W1 starves.
1916
1917While a number of approaches were examined~\cite[\S~4.3]{Delisle18}, the solution chosen for \CFA is a novel techique called \newterm{partial signalling}.
1918Signalled threads are moved to the urgent queue and the waiter at the front defines the set of monitors necessary for it to unblock.
1919Partial signalling transfers ownership of monitors to the front waiter.
1920When the signaller thread exits or waits in the monitor, the front waiter is unblocked if all its monitors are released.
1921The benefit is encapsulating complexity into only two actions: passing monitors to the next owner when they should be released and conditionally waking threads if all conditions are met.
1922
1923
1924\subsection{Loose Object Definitions}
1925\label{s:LooseObjectDefinitions}
1926
1927In an object-oriented programming-language, a class includes an exhaustive list of operations.
1928However, new members can be added via static inheritance or dynamic members, \eg JavaScript~\cite{JavaScript}.
1929Similarly, monitor routines can be added at any time in \CFA, making it less clear for programmers and more difficult to implement.
1930\begin{cfa}
1931monitor M { ... };
1932void `f`( M & mutex m );
1933void g( M & mutex m ) { waitfor( `f` ); }       $\C{// clear which f}$
1934void `f`( M & mutex m, int );                           $\C{// different f}$
1935void h( M & mutex m ) { waitfor( `f` ); }       $\C{// unclear which f}$
1936\end{cfa}
1937Hence, the cfa-code for entering a monitor looks like:
1938\begin{cfa}
1939if ( $\textrm{\textit{monitor is free}}$ ) $\LstCommentStyle{// \color{red}enter}$
1940else if ( $\textrm{\textit{already own monitor}}$ ) $\LstCommentStyle{// \color{red}continue}$
1941else if ( $\textrm{\textit{monitor accepts me}}$ ) $\LstCommentStyle{// \color{red}enter}$
1942else $\LstCommentStyle{// \color{red}block}$
1943\end{cfa}
1944For the first two conditions, it is easy to implement a check that can evaluate the condition in a few instructions.
1945However, a fast check for \emph{monitor accepts me} is much harder to implement depending on the constraints put on the monitors.
1946Figure~\ref{fig:ClassicalMonitor} shows monitors are often expressed as an entry (calling) queue, some acceptor queues, and an urgent stack/queue.
1947
1948\begin{figure}
1949\centering
1950\subfloat[Classical monitor] {
1951\label{fig:ClassicalMonitor}
1952{\resizebox{0.45\textwidth}{!}{\input{monitor.pstex_t}}}
1953}% subfloat
1954\quad
1955\subfloat[Bulk acquire monitor] {
1956\label{fig:BulkMonitor}
1957{\resizebox{0.45\textwidth}{!}{\input{ext_monitor.pstex_t}}}
1958}% subfloat
1959\caption{Monitor Implementation}
1960\label{f:MonitorImplementation}
1961\end{figure}
1962
1963For a fixed (small) number of mutex routines (\eg 128), the accept check reduces to a bitmask of allowed callers, which can be checked with a single instruction.
1964This approach requires a unique dense ordering of routines with a small upper-bound and the ordering must be consistent across translation units.
1965For object-oriented languages these constraints are common, but \CFA mutex routines can be added in any scope and are only visible in certain translation unit, precluding program-wide dense-ordering among mutex routines.
1966
1967Figure~\ref{fig:BulkMonitor} shows the \CFA monitor implementation.
1968The mutex routine called is associated with each thread on the entry queue, while a list of acceptable routines is kept separately.
1969The accepted list is a variable-sized array of accepted routine pointers, so the single instruction bitmask comparison is replaced by dereferencing a pointer followed by a (usually short) linear search.
1970
1971
1972\subsection{Multi-Monitor Scheduling}
1973\label{s:Multi-MonitorScheduling}
1974
1975External scheduling, like internal scheduling, becomes significantly more complex for multi-monitor semantics.
1976Even in the simplest case, new semantics needs to be established.
1977\newpage
1978\begin{cfa}
1979monitor M { ... };
1980void f( M & mutex m1 );
1981void g( M & mutex m1, M & mutex m2 ) {
1982        waitfor( f );                                                   $\C{\color{red}// pass m1 or m2 to f?}$
1983}
1984\end{cfa}
1985The solution is for the programmer to disambiguate:
1986\begin{cfa}
1987        waitfor( f, m2 );                                               $\C{\color{red}// wait for call to f with argument m2}$
1988\end{cfa}
1989Both locks are acquired by routine @g@, so when routine @f@ is called, the lock for monitor @m2@ is passed from @g@ to @f@, while @g@ still holds lock @m1@.
1990This behaviour can be extended to the multi-monitor @waitfor@ statement.
1991\begin{cfa}
1992monitor M { ... };
1993void f( M & mutex m1, M & mutex m2 );
1994void g( M & mutex m1, M & mutex m2 ) {
1995        waitfor( f, m1, m2 );                                   $\C{\color{red}// wait for call to f with arguments m1 and m2}$
1996}
1997\end{cfa}
1998Again, the set of monitors passed to the @waitfor@ statement must be entirely contained in the set of monitors already acquired by the accepting routine.
1999Also, the order of the monitors in a @waitfor@ statement is unimportant.
2000
2001Figure~\ref{f:UnmatchedMutexSets} shows an example where, for internal and external scheduling with multiple monitors, a signalling or accepting thread must match exactly, \ie partial matching results in waiting.
2002For both examples, the set of monitors is disjoint so unblocking is impossible.
2003
2004\begin{figure}
2005\lstDeleteShortInline@%
2006\begin{tabular}{@{}l@{\hspace{\parindentlnth}}|@{\hspace{\parindentlnth}}l@{}}
2007\begin{cfa}
2008monitor M1 {} m11, m12;
2009monitor M2 {} m2;
2010condition c;
2011void f( M1 & mutex m1, M2 & mutex m2 ) {
2012        signal( c );
2013}
2014void g( M1 & mutex m1, M2 & mutex m2 ) {
2015        wait( c );
2016}
2017g( `m11`, m2 ); // block on wait
2018f( `m12`, m2 ); // cannot fulfil
2019\end{cfa}
2020&
2021\begin{cfa}
2022monitor M1 {} m11, m12;
2023monitor M2 {} m2;
2024
2025void f( M1 & mutex m1, M2 & mutex m2 ) {
2026
2027}
2028void g( M1 & mutex m1, M2 & mutex m2 ) {
2029        waitfor( f, m1, m2 );
2030}
2031g( `m11`, m2 ); // block on accept
2032f( `m12`, m2 ); // cannot fulfil
2033\end{cfa}
2034\end{tabular}
2035\lstMakeShortInline@%
2036\caption{Unmatched \protect\lstinline@mutex@ sets}
2037\label{f:UnmatchedMutexSets}
2038\end{figure}
2039
2040
2041\subsection{Extended \protect\lstinline@waitfor@}
2042
2043Figure~\ref{f:ExtendedWaitfor} show the extended form of the @waitfor@ statement to conditionally accept one of a group of mutex routines, with a specific action to be performed \emph{after} the mutex routine finishes.
2044For a @waitfor@ clause to be executed, its @when@ must be true and an outstanding call to its corresponding member(s) must exist.
2045The \emph{conditional-expression} of a @when@ may call a routine, but the routine must not block or context switch.
2046If there are multiple acceptable mutex calls, selection occurs top-to-bottom (prioritized) in the @waitfor@ clauses, whereas some programming languages with similar mechanisms accept non-deterministically for this case, \eg Go \lstinline[morekeywords=select]@select@.
2047If some accept guards are true and there are no outstanding calls to these members, the acceptor is accept-blocked until a call to one of these members is made.
2048If all the accept guards are false, the statement does nothing, unless there is a terminating @else@ clause with a true guard, which is executed instead.
2049Hence, the terminating @else@ clause allows a conditional attempt to accept a call without blocking.
2050If there is a @timeout@ clause, it provides an upper bound on waiting.
2051If both a @timeout@ clause and an @else@ clause are present, the @else@ must be conditional, or the @timeout@ is never triggered.
2052In all cases, the statement following is executed \emph{after} a clause is executed to know which of the clauses executed.
2053
2054\begin{figure}
2055\begin{cfa}
2056`when` ( $\emph{conditional-expression}$ )      $\C{// optional guard}$
2057        waitfor( $\emph{mutex-member-name}$ )
2058                $\emph{statement}$                                      $\C{// action after call}$
2059`or` `when` ( $\emph{conditional-expression}$ ) $\C{// optional guard}$
2060        waitfor( $\emph{mutex-member-name}$ )
2061                $\emph{statement}$                                      $\C{// action after call}$
2062`or`    ...                                                                     $\C{// list of waitfor clauses}$
2063`when` ( $\emph{conditional-expression}$ )      $\C{// optional guard}$
2064        `timeout`                                                               $\C{// optional terminating timeout clause}$
2065                $\emph{statement}$                                      $\C{// action after timeout}$
2066`when` ( $\emph{conditional-expression}$ )      $\C{// optional guard}$
2067        `else`                                                                  $\C{// optional terminating clause}$
2068                $\emph{statement}$                                      $\C{// action when no immediate calls}$
2069\end{cfa}
2070\caption{Extended \protect\lstinline@waitfor@}
2071\label{f:ExtendedWaitfor}
2072\end{figure}
2073
2074Note, a group of conditional @waitfor@ clauses is \emph{not} the same as a group of @if@ statements, e.g.:
2075\begin{cfa}
2076if ( C1 ) waitfor( mem1 );                       when ( C1 ) waitfor( mem1 );
2077else if ( C2 ) waitfor( mem2 );         or when ( C2 ) waitfor( mem2 );
2078\end{cfa}
2079The left example accepts only @mem1@ if @C1@ is true or only @mem2@ if @C2@ is true.
2080The right example accepts either @mem1@ or @mem2@ if @C1@ and @C2@ are true.
2081
2082An interesting use of @waitfor@ is accepting the @mutex@ destructor to know when an object is deallocated.
2083\begin{cfa}
2084void insert( Buffer(T) & mutex buffer, T elem ) with( buffer ) {
2085        if ( count == 10 )
2086                waitfor( remove, buffer ) {
2087                        // insert elem into buffer
2088                } or `waitfor( ^?{}, buffer )` throw insertFail;
2089}
2090\end{cfa}
2091When the buffer is deallocated, the current waiter is unblocked and informed, so it can perform an appropriate action.
2092However, the basic @waitfor@ semantics do not support this functionality, since using an object after its destructor is called is undefined.
2093Therefore, to make this useful capability work, the semantics for accepting the destructor is the same as @signal@, \ie the call to the destructor is placed on the urgent queue and the acceptor continues execution, which throws an exception to the acceptor and then the caller is unblocked from the urgent queue to deallocate the object.
2094Accepting the destructor is an idiomatic way to terminate a thread in \CFA.
2095
2096
2097\subsection{\protect\lstinline@mutex@ Threads}
2098
2099Threads in \CFA are monitors to allow direct communication among threads, \ie threads can have mutex routines that are called by other threads.
2100Hence, all monitor features are available when using threads.
2101The following shows an example of two threads directly calling each other and accepting calls from each other in a cycle.
2102\begin{cfa}
2103thread Ping {} pi;
2104thread Pong {} po;
2105void ping( Ping & mutex ) {}
2106void pong( Pong & mutex ) {}
2107int main() {}
2108\end{cfa}
2109\vspace{-0.8\baselineskip}
2110\begin{cquote}
2111\begin{tabular}{@{}l@{\hspace{3\parindentlnth}}l@{}}
2112\begin{cfa}
2113void main( Ping & pi ) {
2114        for ( int i = 0; i < 10; i += 1 ) {
2115                `waitfor( ping, pi );`
2116                `pong( po );`
2117        }
2118}
2119\end{cfa}
2120&
2121\begin{cfa}
2122void main( Pong & po ) {
2123        for ( int i = 0; i < 10; i += 1 ) {
2124                `ping( pi );`
2125                `waitfor( pong, po );`
2126        }
2127}
2128\end{cfa}
2129\end{tabular}
2130\end{cquote}
2131% \lstMakeShortInline@%
2132% \caption{Threads ping/pong using external scheduling}
2133% \label{f:pingpong}
2134% \end{figure}
2135Note, the ping/pong threads are globally declared, @pi@/@po@, and hence, start (and possibly complete) before the program main starts.
2136
2137
2138\subsection{Low-level Locks}
2139
2140For completeness and efficiency, \CFA provides a standard set of low-level locks: recursive mutex, condition, semaphore, barrier, \etc, and atomic instructions: @fetchAssign@, @fetchAdd@, @testSet@, @compareSet@, \etc.
2141
2142
2143\section{Parallelism}
2144
2145Historically, computer performance was about processor speeds.
2146However, with heat dissipation being a direct consequence of speed increase, parallelism is the new source for increased performance~\cite{Sutter05, Sutter05b}.
2147Now, high-performance applications must care about parallelism, which requires concurrency.
2148The lowest-level approach of parallelism is to use \newterm{kernel threads} in combination with semantics like @fork@, @join@, \etc.
2149However, kernel threads are better as an implementation tool because of complexity and higher cost.
2150Therefore, different abstractions are often layered onto kernel threads to simplify them, \eg pthreads.
2151
2152
2153\subsection{User Threads with Preemption}
2154
2155A direct improvement on kernel threads is user threads, \eg Erlang~\cite{Erlang} and \uC~\cite{uC++book}.
2156This approach provides an interface that matches the language paradigms, more control over concurrency by the language runtime, and an abstract (and portable) interface to the underlying kernel threads across operating systems.
2157In many cases, user threads can be used on a much larger scale (100,000 threads).
2158Like kernel threads, user threads support preemption, which maximizes nondeterminism, but introduces the same concurrency errors: race, livelock, starvation, and deadlock.
2159\CFA adopts user-threads as they represent the truest realization of concurrency and can build any other concurrency approach, \eg thread pools and actors~\cite{Actors}.
2160
2161
2162\subsection{User Threads without Preemption (Fiber)}
2163\label{s:fibers}
2164
2165A variant of user thread is \newterm{fibers}, which removes preemption, \eg Go~\cite{Go} @goroutine@s.
2166Like functional programming, which removes mutation and its associated problems, removing preemption from concurrency reduces nondeterminism, making race and deadlock errors more difficult to generate.
2167However, preemption is necessary for concurrency that relies on spinning, so there are a class of problems that cannot be programmed without preemption.
2168
2169
2170\subsection{Thread Pools}
2171
2172In contrast to direct threading is indirect \newterm{thread pools}, where small jobs (work units) are inserted into a work pool for execution.
2173If the jobs are dependent, \ie interact, there is an implicit/explicit dependency graph that ties them together.
2174While removing direct concurrency, and hence the amount of context switching, thread pools significantly limit the interaction that can occur among jobs.
2175Indeed, jobs should not block because that also blocks the underlying thread, which effectively means the CPU utilization, and therefore throughput, suffers.
2176While it is possible to tune the thread pool with sufficient threads, it becomes difficult to obtain high throughput and good core utilization as job interaction increases.
2177As well, concurrency errors return, which threads pools are suppose to mitigate.
2178
2179
2180\section{\protect\CFA Runtime Structure}
2181
2182Figure~\ref{f:RunTimeStructure} illustrates the runtime structure of a \CFA program.
2183In addition to the new kinds of objects introduced by \CFA, there are two more runtime entities used to control parallel execution: cluster and (virtual) processor.
2184An executing thread is illustrated by its containment in a processor.
2185
2186\begin{figure}
2187\centering
2188\input{RunTimeStructure}
2189\caption{\CFA Runtime Structure}
2190\label{f:RunTimeStructure}
2191\end{figure}
2192
2193
2194\subsection{Cluster}
2195\label{s:RuntimeStructureCluster}
2196
2197A \newterm{cluster} is a collection of threads and virtual processors (abstract kernel-thread) that execute the threads (like a virtual machine).
2198The purpose of a cluster is to control the amount of parallelism that is possible among threads, plus scheduling and other execution defaults.
2199The default cluster-scheduler is single-queue multi-server, which provides automatic load-balancing of threads on processors.
2200However, the scheduler is pluggable, supporting alternative schedulers.
2201If several clusters exist, both threads and virtual processors, can be explicitly migrated from one cluster to another.
2202No automatic load balancing among clusters is performed by \CFA.
2203
2204When a \CFA program begins execution, it creates a user cluster with a single processor and a special processor to handle preemption that does not execute user threads.
2205The user cluster is created to contain the application user-threads.
2206Having all threads execute on the one cluster often maximizes utilization of processors, which minimizes runtime.
2207However, because of limitations of the underlying operating system, heterogeneous hardware, or scheduling requirements (real-time), multiple clusters are sometimes necessary.
2208
2209
2210\subsection{Virtual Processor}
2211\label{s:RuntimeStructureProcessor}
2212
2213A virtual processor is implemented by a kernel thread (\eg UNIX process), which is subsequently scheduled for execution on a hardware processor by the underlying operating system.
2214Programs may use more virtual processors than hardware processors.
2215On a multiprocessor, kernel threads are distributed across the hardware processors resulting in virtual processors executing in parallel.
2216(It is possible to use affinity to lock a virtual processor onto a particular hardware processor~\cite{affinityLinux, affinityWindows, affinityFreebsd, affinityNetbsd, affinityMacosx}, which is used when caching issues occur or for heterogeneous hardware processors.)
2217The \CFA runtime attempts to block unused processors and unblock processors as the system load increases;
2218balancing the workload with processors is difficult.
2219Preemption occurs on virtual processors rather than user threads, via operating-system interrupts.
2220Thus virtual processors execute user threads, where preemption frequency applies to a virtual processor, so preemption occurs randomly across the executed user threads.
2221Turning off preemption transforms user threads into fibers.
2222
2223
2224\subsection{Debug Kernel}
2225
2226There are two versions of the \CFA runtime kernel: debug and non-debug.
2227The debugging version has many runtime checks and internal assertions, \eg stack (non-writable) guard page, and checks for stack overflow whenever context switches occur among coroutines and threads, which catches most stack overflows.
2228After a program is debugged, the non-debugging version can be used to decrease space and increase performance.
2229
2230
2231\section{Implementation}
2232\label{s:Implementation}
2233
2234Currently, \CFA has fixed-sized stacks, where the stack size can be set at coroutine/thread creation but with no subsequent growth.
2235Schemes exist for dynamic stack-growth, such as stack copying and chained stacks.
2236However, stack copying requires pointer adjustment to items on the stack, which is impossible without some form of garbage collection.
2237As well, chained stacks require all modules be recompiled to use this feature, which breaks backward compatibility with existing C libraries.
2238In the long term, it is likely C libraries will migrate to stack chaining to support concurrency, at only a minimal cost to sequential programs.
2239Nevertheless, experience teaching \uC~\cite{CS343} shows fixed-sized stacks are rarely an issue in most concurrent programs.
2240
2241A primary implementation challenge is avoiding contention from dynamically allocating memory because of bulk acquire, \eg the internal-scheduling design is (almost) free of allocations.
2242All blocking operations are made by parking threads onto queues, therefore all queues are designed with intrusive nodes, where each node has preallocated link fields for chaining.
2243Furthermore, several bulk-acquire operations need a variable amount of memory.
2244This storage is allocated at the base of a thread's stack before blocking, which means programmers must add a small amount of extra space for stacks.
2245
2246In \CFA, ordering of monitor acquisition relies on memory ordering to prevent deadlock~\cite{Havender68}, because all objects have distinct non-overlapping memory layouts, and mutual-exclusion for a monitor is only defined for its lifetime.
2247When a mutex call is made, pointers to the concerned monitors are aggregated into a variable-length array and sorted.
2248This array persists for the entire duration of the mutual exclusion and is used extensively for synchronization operations.
2249
2250To improve performance and simplicity, context switching occurs inside a routine call, so only callee-saved registers are copied onto the stack and then the stack register is switched;
2251the corresponding registers are then restored for the other context.
2252Note, the instruction pointer is untouched since the context switch is always inside the same routine.
2253Unlike coroutines, threads do not context switch among each other;
2254they context switch to the cluster scheduler.
2255This method is a 2-step context-switch and provides a clear distinction between user and kernel code, where scheduling and other system operations happen.
2256The alternative 1-step context-switch uses the \emph{from} thread's stack to schedule and then context-switches directly to the \emph{to} thread's stack.
2257Experimental results (not presented) show the performance of these two approaches is virtually equivalent, because both approaches are dominated by locking to prevent a race condition.
2258
2259All kernel threads (@pthreads@) created a stack.
2260Each \CFA virtual processor is implemented as a coroutine and these coroutines run directly on the kernel-thread stack, effectively stealing this stack.
2261The exception to this rule is the program main, \ie the initial kernel thread that is given to any program.
2262In order to respect C expectations, the stack of the initial kernel thread is used by program main rather than the main processor, allowing it to grow dynamically as in a normal C program.
2263
2264Finally, an important aspect for a complete threading system is preemption, which introduces extra non-determinism via transparent interleaving, rather than cooperation among threads for proper scheduling and processor fairness from long-running threads.
2265Because preemption frequency is usually long (1 millisecond) performance cost is negligible.
2266Preemption is normally handled by setting a count-down timer on each virtual processor.
2267When the timer expires, an interrupt is delivered, and the interrupt handler resets the count-down timer, and if the virtual processor is executing in user code, the signal handler performs a user-level context-switch, or if executing in the language runtime-kernel, the preemption is ignored or rolled forward to the point where the runtime kernel context switches back to user code.
2268Multiple signal handlers may be pending.
2269When control eventually switches back to the signal handler, it returns normally, and execution continues in the interrupted user thread, even though the return from the signal handler may be on a different kernel thread than the one where the signal is delivered.
2270The only issue with this approach is that signal masks from one kernel thread may be restored on another as part of returning from the signal handler;
2271therefore, the same signal mask is required for all virtual processors in a cluster.
2272
2273However, on current UNIX systems:
2274\begin{quote}
2275A process-directed signal may be delivered to any one of the threads that does not currently have the signal blocked.
2276If more than one of the threads has the signal unblocked, then the kernel chooses an arbitrary thread to which to deliver the signal.
2277SIGNAL(7) - Linux Programmer's Manual
2278\end{quote}
2279Hence, the timer-expiry signal, which is generated \emph{externally} by the UNIX kernel to the UNIX process, is delivered to any of its UNIX subprocesses (kernel threads).
2280To ensure each virtual processor receives its own preemption signals, a discrete-event simulation is run on a special virtual processor, and only it sets and receives timer events.
2281Virtual processors register an expiration time with the discrete-event simulator, which is inserted in sorted order.
2282The simulation sets the count-down timer to the value at the head of the event list, and when the timer expires, all events less than or equal to the current time are processed.
2283Processing a preemption event sends an \emph{internal} @SIGUSR1@ signal to the registered virtual processor, which is always delivered to that processor.
2284
2285
2286\section{Performance}
2287\label{results}
2288
2289To verify the implementation of the \CFA runtime, a series of microbenchmarks are performed comparing \CFA with other widely used programming languages with concurrency.
2290Table~\ref{t:machine} shows the specifications of the computer used to run the benchmarks, and the versions of the software used in the comparison.
2291
2292\begin{table}
2293\centering
2294\caption{Experiment environment}
2295\label{t:machine}
2296
2297\begin{tabular}{|l|r||l|r|}
2298\hline
2299Architecture            & x86\_64                               & NUMA node(s)  & 8 \\
2300\hline
2301CPU op-mode(s)          & 32-bit, 64-bit                & Model name    & AMD Opteron\texttrademark\ Processor 6380 \\
2302\hline
2303Byte Order                      & Little Endian                 & CPU Freq              & 2.5 GHz \\
2304\hline
2305CPU(s)                          & 64                                    & L1d cache     & 16 KiB \\
2306\hline
2307Thread(s) per core      & 2                                     & L1i cache     & 64 KiB \\
2308\hline
2309Core(s) per socket      & 8                                     & L2 cache              & 2048 KiB \\
2310\hline
2311Socket(s)                       & 4                                     & L3 cache              & 6144 KiB \\
2312\hline
2313\hline
2314Operating system        & Ubuntu 16.04.3 LTS    & Kernel                & Linux 4.4-97-generic \\
2315\hline
2316gcc                                     & 6.3                                   & \CFA                  & 1.0.0 \\
2317\hline
2318Java                            & OpenJDK-9                     & Go                    & 1.9.2 \\
2319\hline
2320\end{tabular}
2321\end{table}
2322
2323All benchmarks are run using the following harness:
2324\begin{cfa}
2325unsigned int N = 10_000_000;
2326#define BENCH( run ) Time before = getTimeNsec(); run; Duration result = (getTimeNsec() - before) / N;
2327\end{cfa}
2328The method used to get time is @clock_gettime( CLOCK_REALTIME )@.
2329Each benchmark is performed @N@ times, where @N@ varies depending on the benchmark;
2330the total time is divided by @N@ to obtain the average time for a benchmark.
2331All omitted tests for other languages are functionally identical to the shown \CFA test.
2332
2333
2334\paragraph{Context-Switching}
2335
2336In procedural programming, the cost of a routine call is important as modularization (refactoring) increases.
2337(In many cases, a compiler inlines routine calls to eliminate this cost.)
2338Similarly, when modularization extends to coroutines/tasks, the time for a context switch becomes a relevant factor.
2339The coroutine context-switch is 2-step using resume/suspend, \ie from resumer to suspender and from suspender to resumer.
2340The thread context switch is 2-step using yield, \ie enter and return from the runtime kernel.
2341Figure~\ref{f:ctx-switch} shows the code for coroutines/threads with all results in Table~\ref{tab:ctx-switch}.
2342The difference in performance between coroutine and thread context-switch is the cost of scheduling for threads, whereas coroutines are self-scheduling.
2343
2344\begin{figure}
2345\lstset{language=CFA,moredelim=**[is][\color{red}]{@}{@},deletedelim=**[is][]{`}{`}}
2346
2347\newbox\myboxA
2348\begin{lrbox}{\myboxA}
2349\begin{cfa}[aboveskip=0pt,belowskip=0pt]
2350coroutine C {} c;
2351void main( C & ) { for ( ;; ) { @suspend();@ } }
2352int main() {
2353        BENCH(
2354                for ( size_t i = 0; i < N; i += 1 ) { @resume( c );@ } )
2355        sout | result`ns;
2356}
2357\end{cfa}
2358\end{lrbox}
2359
2360\newbox\myboxB
2361\begin{lrbox}{\myboxB}
2362\begin{cfa}[aboveskip=0pt,belowskip=0pt]
2363
2364
2365int main() {
2366        BENCH(
2367                for ( size_t i = 0; i < N; i += 1 ) { @yield();@ } )
2368        sout | result`ns;
2369}
2370\end{cfa}
2371\end{lrbox}
2372
2373\subfloat[Coroutine]{\usebox\myboxA}
2374\quad
2375\subfloat[Thread]{\usebox\myboxB}
2376\captionof{figure}{\CFA context-switch benchmark}
2377\label{f:ctx-switch}
2378
2379\centering
2380
2381\captionof{table}{Context switch comparison (nanoseconds)}
2382\label{tab:ctx-switch}
2383\bigskip
2384\begin{tabular}{|r|*{3}{D{.}{.}{3.2}|}}
2385\cline{2-4}
2386\multicolumn{1}{c|}{} & \multicolumn{1}{c|}{Median} &\multicolumn{1}{c|}{Average} & \multicolumn{1}{c|}{Std Dev} \\
2387\hline
2388Kernel Thread   & 333.5 & 332.96        & 4.1 \\
2389\CFA Coroutine  & 49            & 48.68 & 0.47    \\
2390\CFA Thread             & 105           & 105.57        & 1.37 \\
2391\uC Coroutine   & 44            & 44            & 0 \\
2392\uC Thread              & 100           & 99.29 & 0.96 \\
2393Goroutine               & 145           & 147.25        & 4.15 \\
2394Java Thread             & 373.5 & 375.14        & 8.72 \\
2395\hline
2396\end{tabular}
2397
2398\bigskip
2399\bigskip
2400
2401\lstset{language=CFA,moredelim=**[is][\color{red}]{@}{@},deletedelim=**[is][]{`}{`}}
2402\begin{cfa}
2403monitor M { ... } m1/*, m2, m3, m4*/;
2404void __attribute__((noinline)) do_call( M & mutex m/*, m2, m3, m4*/ ) {}
2405int main() {
2406        BENCH( for( size_t i = 0; i < N; i += 1 ) { @do_call( m1/*, m2, m3, m4*/ );@ } )
2407        sout | result`ns;
2408}
2409\end{cfa}
2410\captionof{figure}{\CFA acquire/release mutex benchmark}
2411\label{f:mutex}
2412
2413\centering
2414
2415\captionof{table}{Mutex comparison (nanoseconds)}
2416\label{tab:mutex}
2417\bigskip
2418
2419\begin{tabular}{|r|*{3}{D{.}{.}{3.2}|}}
2420\cline{2-4}
2421\multicolumn{1}{c|}{} & \multicolumn{1}{c|}{Median} &\multicolumn{1}{c|}{Average} & \multicolumn{1}{c|}{Std Dev} \\
2422\hline
2423C routine                                       & 2             & 2             & 0    \\
2424FetchAdd + FetchSub                     & 26            & 26            & 0    \\
2425Pthreads Mutex Lock                     & 31            & 31.71 & 0.97 \\
2426\uC @monitor@ member routine            & 31            & 31            & 0    \\
2427\CFA @mutex@ routine, 1 argument        & 46            & 46.68 & 0.93  \\
2428\CFA @mutex@ routine, 2 argument        & 84            & 85.36 & 1.99 \\
2429\CFA @mutex@ routine, 4 argument        & 158           & 161           & 4.22 \\
2430Java synchronized routine               & 27.5  & 29.79 & 2.93  \\
2431\hline
2432\end{tabular}
2433\end{figure}
2434
2435
2436\paragraph{Mutual-Exclusion}
2437
2438Mutual exclusion is measured by entering/leaving a critical section.
2439For monitors, entering and leaving a monitor routine is measured.
2440Figure~\ref{f:mutex} shows the code for \CFA with all results in Table~\ref{tab:mutex}.
2441To put the results in context, the cost of entering a non-inline routine and the cost of acquiring and releasing a @pthread_mutex@ lock is also measured.
2442Note, the incremental cost of bulk acquire for \CFA, which is largely a fixed cost for small numbers of mutex objects.
2443
2444
2445\paragraph{Internal Scheduling}
2446
2447Internal scheduling is measured by waiting on and signalling a condition variable.
2448Figure~\ref{f:int-sched} shows the code for \CFA, with results in Table~\ref{tab:int-sched}.
2449Note, the incremental cost of bulk acquire for \CFA, which is largely a fixed cost for small numbers of mutex objects.
2450Java scheduling is significantly greater because the benchmark explicitly creates multiple thread in order to prevent the JIT from making the program sequential, \ie removing all locking.
2451
2452\begin{figure}
2453\lstset{language=CFA,moredelim=**[is][\color{red}]{@}{@},deletedelim=**[is][]{`}{`}}
2454\begin{cfa}
2455volatile int go = 0;
2456condition c;
2457monitor M { ... } m;
2458void __attribute__((noinline)) do_call( M & mutex a1 ) { signal( c ); }
2459thread T {};
2460void main( T & this ) {
2461        while ( go == 0 ) { yield(); }  // wait for other thread to start
2462        while ( go == 1 ) { @do_call( m );@ }
2463}
2464int  __attribute__((noinline)) do_wait( M & mutex m ) {
2465        go = 1; // continue other thread
2466        BENCH( for ( size_t i = 0; i < N; i += 1 ) { @wait( c );@ } );
2467        go = 0; // stop other thread
2468        sout | result`ns;
2469}
2470int main() {
2471        T t;
2472        do_wait( m );
2473}
2474\end{cfa}
2475\captionof{figure}{\CFA Internal-scheduling benchmark}
2476\label{f:int-sched}
2477
2478\centering
2479\captionof{table}{Internal-scheduling comparison (nanoseconds)}
2480\label{tab:int-sched}
2481\bigskip
2482
2483\begin{tabular}{|r|*{3}{D{.}{.}{5.2}|}}
2484\cline{2-4}
2485\multicolumn{1}{c|}{} & \multicolumn{1}{c|}{Median} &\multicolumn{1}{c|}{Average} & \multicolumn{1}{c|}{Std Dev} \\
2486\hline
2487Pthreads Condition Variable             & 6005  & 5681.43       & 835.45 \\
2488\uC @signal@                                    & 324           & 325.54        & 3,02   \\
2489\CFA @signal@, 1 @monitor@              & 368.5         & 370.61        & 4.77   \\
2490\CFA @signal@, 2 @monitor@              & 467           & 470.5 & 6.79   \\
2491\CFA @signal@, 4 @monitor@              & 700.5         & 702.46        & 7.23  \\
2492Java @notify@                                   & 15471 & 172511        & 5689 \\
2493\hline
2494\end{tabular}
2495\end{figure}
2496
2497
2498\paragraph{External Scheduling}
2499
2500External scheduling is measured by accepting a call using the @waitfor@ statement (@_Accept@ in \uC).
2501Figure~\ref{f:ext-sched} shows the code for \CFA, with results in Table~\ref{tab:ext-sched}.
2502Note, the incremental cost of bulk acquire for \CFA, which is largely a fixed cost for small numbers of mutex objects.
2503
2504\begin{figure}
2505\lstset{language=CFA,moredelim=**[is][\color{red}]{@}{@},deletedelim=**[is][]{`}{`}}
2506\begin{cfa}
2507volatile int go = 0;
2508monitor M { ... } m;
2509thread T {};
2510void __attribute__((noinline)) do_call( M & mutex ) {}
2511void main( T & ) {
2512        while ( go == 0 ) { yield(); }  // wait for other thread to start
2513        while ( go == 1 ) { @do_call( m );@ }
2514}
2515int __attribute__((noinline)) do_wait( M & mutex m ) {
2516        go = 1; // continue other thread
2517        BENCH( for ( size_t i = 0; i < N; i += 1 ) { @waitfor( do_call, m );@ } )
2518        go = 0; // stop other thread
2519        sout | result`ns;
2520}
2521int main() {
2522        T t;
2523        do_wait( m );
2524}
2525\end{cfa}
2526\captionof{figure}{\CFA external-scheduling benchmark}
2527\label{f:ext-sched}
2528
2529\centering
2530
2531\captionof{table}{External-scheduling comparison (nanoseconds)}
2532\label{tab:ext-sched}
2533\bigskip
2534\begin{tabular}{|r|*{3}{D{.}{.}{3.2}|}}
2535\cline{2-4}
2536\multicolumn{1}{c|}{} & \multicolumn{1}{c|}{Median} &\multicolumn{1}{c|}{Average} & \multicolumn{1}{c|}{Std Dev} \\
2537\hline
2538\uC @_Accept@                           & 358           & 359.11        & 2.53  \\
2539\CFA @waitfor@, 1 @monitor@     & 359           & 360.93        & 4.07  \\
2540\CFA @waitfor@, 2 @monitor@     & 450           & 449.39        & 6.62  \\
2541\CFA @waitfor@, 4 @monitor@     & 652           & 655.64        & 7.73 \\
2542\hline
2543\end{tabular}
2544
2545\bigskip
2546\medskip
2547
2548\lstset{language=CFA,moredelim=**[is][\color{red}]{@}{@},deletedelim=**[is][]{`}{`}}
2549\begin{cfa}
2550thread MyThread {};
2551void main( MyThread & ) {}
2552int main() {
2553        BENCH( for ( size_t i = 0; i < N; i += 1 ) { @MyThread m;@ } )
2554        sout | result`ns;
2555}
2556\end{cfa}
2557\captionof{figure}{\CFA object-creation benchmark}
2558\label{f:creation}
2559
2560\centering
2561
2562\captionof{table}{Creation comparison (nanoseconds)}
2563\label{tab:creation}
2564\bigskip
2565
2566\begin{tabular}{|r|*{3}{D{.}{.}{5.2}|}}
2567\cline{2-4}
2568\multicolumn{1}{c|}{} & \multicolumn{1}{c|}{Median} & \multicolumn{1}{c|}{Average} & \multicolumn{1}{c|}{Std Dev} \\
2569\hline
2570Pthreads                                & 28091         & 28073.39      & 163.1  \\
2571\CFA Coroutine Lazy             & 6                     & 6.07          & 0.26   \\
2572\CFA Coroutine Eager    & 520           & 520.61        & 2.04   \\
2573\CFA Thread                             & 2032  & 2016.29       & 112.07  \\
2574\uC Coroutine                   & 106           & 107.36        & 1.47   \\
2575\uC Thread                              & 536.5 & 537.07        & 4.64   \\
2576Goroutine                               & 3103  & 3086.29       & 90.25  \\
2577Java Thread                             & 103416.5      & 103732.29     & 1137 \\
2578\hline
2579\end{tabular}
2580\end{figure}
2581
2582
2583\paragraph{Object Creation}
2584
2585Object creation is measured by creating/deleting the specific kind of concurrent object.
2586Figure~\ref{f:creation} shows the code for \CFA, with results in Table~\ref{tab:creation}.
2587The only note here is that the call stacks of \CFA coroutines are lazily created, therefore without priming the coroutine to force stack creation, the creation cost is artificially low.
2588
2589
2590\section{Conclusion}
2591
2592This paper demonstrates a concurrency API that is simple, efficient, and able to build higher-level concurrency features.
2593The approach provides concurrency based on a preemptive M:N user-level threading-system, executing in clusters, which encapsulate scheduling of work on multiple kernel threads providing parallelism.
2594The M:N model is judged to be efficient and provide greater flexibility than a 1:1 threading model.
2595High-level objects (monitor/task) are the core mechanism for mutual exclusion and synchronization.
2596A novel aspect is allowing multiple mutex-objects to be accessed simultaneously reducing the potential for deadlock for this complex scenario.
2597These concepts and the entire \CFA runtime-system are written in the \CFA language, demonstrating the expressiveness of the \CFA language.
2598Performance comparisons with other concurrent systems/languages show the \CFA approach is competitive across all low-level operations, which translates directly into good performance in well-written concurrent applications.
2599C programmers should feel comfortable using these mechanisms for developing concurrent applications, with the ability to obtain maximum available performance by mechanisms at the appropriate level.
2600
2601
2602\section{Future Work}
2603
2604While concurrency in \CFA has a strong start, development is still underway and there are missing features.
2605
2606\paragraph{Flexible Scheduling}
2607\label{futur:sched}
2608
2609An important part of concurrency is scheduling.
2610Different scheduling algorithms can affect performance (both in terms of average and variation).
2611However, no single scheduler is optimal for all workloads and therefore there is value in being able to change the scheduler for given programs.
2612One solution is to offer various tweaking options, allowing the scheduler to be adjusted to the requirements of the workload.
2613However, to be truly flexible, a pluggable scheduler is necessary.
2614Currently, the \CFA pluggable scheduler is too simple to handle complex scheduling, \eg quality of service and real-time, where the scheduler must interact with mutex objects to deal with issues like priority inversion.
2615
2616\paragraph{Non-Blocking I/O}
2617\label{futur:nbio}
2618
2619Many modern workloads are not bound by computation but IO operations, a common case being web servers and XaaS~\cite{XaaS} (anything as a service).
2620These types of workloads require significant engineering to amortizing costs of blocking IO-operations.
2621At its core, non-blocking I/O is an operating-system level feature queuing IO operations, \eg network operations, and registering for notifications instead of waiting for requests to complete.
2622Current trends use asynchronous programming like callbacks, futures, and/or promises, \eg Node.js~\cite{NodeJs} for JavaScript, Spring MVC~\cite{SpringMVC} for Java, and Django~\cite{Django} for Python.
2623However, these solutions lead to code that is hard to create, read and maintain.
2624A better approach is to tie non-blocking I/O into the concurrency system to provide ease of use with low overhead, \eg thread-per-connection web-services.
2625A non-blocking I/O library is currently under development for \CFA.
2626
2627\paragraph{Other Concurrency Tools}
2628\label{futur:tools}
2629
2630While monitors offer flexible and powerful concurrency for \CFA, other concurrency tools are also necessary for a complete multi-paradigm concurrency package.
2631Examples of such tools can include futures and promises~\cite{promises}, executors and actors.
2632These additional features are useful when monitors offer a level of abstraction that is inadequate for certain tasks.
2633As well, new \CFA extensions should make it possible to create a uniform interface for virtually all mutual exclusion, including monitors and low-level locks.
2634
2635\paragraph{Implicit Threading}
2636\label{futur:implcit}
2637
2638Basic concurrent (embarrassingly parallel) applications can benefit greatly from implicit concurrency, where sequential programs are converted to concurrent, possibly with some help from pragmas to guide the conversion.
2639This type of concurrency can be achieved both at the language level and at the library level.
2640The canonical example of implicit concurrency is concurrent nested @for@ loops, which are amenable to divide and conquer algorithms~\cite{uC++book}.
2641The \CFA language features should make it possible to develop a reasonable number of implicit concurrency mechanism to solve basic HPC data-concurrency problems.
2642However, implicit concurrency is a restrictive solution with significant limitations, so it can never replace explicit concurrent programming.
2643
2644
2645\section{Acknowledgements}
2646
2647The authors would like to recognize the design assistance of Aaron Moss, Rob Schluntz and Andrew Beach on the features described in this paper.
2648Funding for this project has been provided by Huawei Ltd.\ (\url{http://www.huawei.com}), and Peter Buhr is partially funded by the Natural Sciences and Engineering Research Council of Canada.
2649
2650{%
2651\fontsize{9bp}{12bp}\selectfont%
2652\bibliography{pl,local}
2653}%
2654
2655\end{document}
2656
2657% Local Variables: %
2658% tab-width: 4 %
2659% fill-column: 120 %
2660% compile-command: "make" %
2661% End: %
Note: See TracBrowser for help on using the repository browser.