Changes in / [5452673:c886f4b]


Ignore:
Files:
1 added
9 deleted
21 edited

Legend:

Unmodified
Added
Removed

  • doc/bibliography/pl.bib

    r5452673 rc886f4b  
    99%    Predefined journal names:
    1010%  acmcs: Computing Surveys             acta: Acta Infomatica
     11@string{acta="Acta Infomatica"}
    1112%  cacm: Communications of the ACM
    1213%  ibmjrd: IBM J. Research & Development ibmsj: IBM Systems Journal
    … …  
    2122%  tcs: Theoretical Computer Science
    2223
    23 @string{acta="Acta Infomatica"}
    2424string{ieeepds="IEEE Transactions on Parallel and Distributed Systems"}
    2525@string{ieeepds="IEEE Trans. Parallel Distrib. Syst."}
    … …  
    124124    series      = {ACM Distinguished Dissertations},
    125125    year        = 1983,
    126 }
    127 
    128 @article{Zhang19,
    129     keywords    = {Algebraic effects, dynamic scoping, exceptions, parametricity, type systems},
    130     author      = {Zhang, Yizhou and Myers, Andrew C.},
    131     title       = {Abstraction-safe Effect Handlers via Tunneling},
    132     journal     = {Proc. ACM Program. Lang.},
    133     issue_date  = {January 2019},
    134     volume      = {3},
    135     number      = {POPL},
    136     month       = jan,
    137     year        = {2019},
    138     issn        = {2475-1421},
    139     pages       = {5:1--5:29},
    140     articleno   = {5},
    141     publisher   = {ACM},
    142     address     = {New York, NY, USA},
    143 }
    144 
    145 @inproceedings{Zhang16,
    146     keywords    = {Exception tunneling, Genus, exception handling},
    147     author      = {Zhang, Yizhou and Salvaneschi, Guido and Beightol, Quinn and Liskov, Barbara and Myers, Andrew C.},
    148     title       = {Accepting Blame for Safe Tunneled Exceptions},
    149     booktitle   = {Proceedings of the 37th ACM SIGPLAN Conference on Programming Language Design and Implementation},
    150     series      = {PLDI'16},
    151     year        = {2016},
    152     location    = {Santa Barbara, CA, USA},
    153     pages       = {281--295},
    154     publisher   = {ACM},
    155     address     = {New York, NY, USA},
    156126}
    157127
    … …  
    428398    journal     = sigplan,
    429399    year        = 1981,
    430     month       = feb,
    431     volume      = 16,
    432     number      = 2,
    433     pages       = {48-52},
     400    month       = feb, volume = 16, number = 2, pages = {48-52},
    434401    comment     = {
    435402        A one-pass, top-down algorithm for overload resolution.  Input is a
    … …  
    510477    title       = {An Alternative to Subclassing},
    511478    journal     = sigplan,
    512     volume      = {21},
    513     number      = {11},
     479    volume      = {21},    number = {11},
    514480    pages       = {424-428},
    515     month       = nov,
    516     year        = 1986,
     481    month       = nov, year = 1986,
    517482    comment     = {
    518483        The Smalltalk class hierarchy has three uses: factoring out code;
    … …  
    568533    isbn        = {3-540-66538-2},
    569534    location    = {Toulouse, France},
     535    doi         = {http://doi.acm.org/10.1145/318773.319251},
    570536    publisher   = {Springer},
    571537    address     = {London, UK},
    … …  
    665631    year        = 2010,
    666632    pages       = {39--50},
     633    numpages    = {12},
    667634    publisher   = {IEEE Computer Society},
    668635    address     = {Washington, DC, USA},
    … …  
    955922}
    956923
    957 @manual{C99,
    958     keywords    = {ISO/IEC C 9899},
    959     contributer = {pabuhr@plg},
    960     key         = {C99},
    961     title       = {C Programming Language {ISO/IEC} 9899:1999(E)},
    962     edition     = {2nd},
    963     publisher   = {International Standard Organization},
    964     address     = {\href{https://webstore.ansi.org/Standards/INCITS/INCITSISOIEC98991999R2005}{https://webstore.ansi.org/\-Standards/\-INCITS/\-INCITSISOIEC98991999R2005}},
    965     year        = 1999,
    966 }
    967 
    968924@manual{C11,
    969925    keywords    = {ISO/IEC C 11},
    … …  
    13491305    location    = {London, United Kingdom},
    13501306    pages       = {41--53},
     1307    numpages    = {13},
     1308    url         = {http://doi.acm.org/10.1145/360204.360207},
     1309    doi         = {10.1145/360204.360207},
     1310    acmid       = {360207},
    13511311    publisher   = {ACM},
    13521312    address     = {New York, NY, USA},
    … …  
    24482408    year        = 1993,
    24492409    pages       = {201--208},
     2410    url         = {http://doi.acm.org/10.1145/155360.155580},
    24502411    publisher   = {ACM},
    24512412    address     = {New York, NY, USA},
    … …  
    26452606    location    = {Boulder, Colorado, USA},
    26462607    pages       = {91--97},
     2608    numpages    = {7},
    26472609    publisher   = {ACM},
    26482610    address     = {New York, NY, USA},
    … …  
    26752637    issn        = {0004-5411},
    26762638    pages       = {215--225},
     2639    numpages    = {11},
     2640    url         = {http://doi.acm.org/10.1145/321879.321884},
     2641    doi         = {10.1145/321879.321884},
     2642    acmid       = {321884},
    26772643    publisher   = {ACM},
    26782644    address     = {New York, NY, USA},
    … …  
    27422708}
    27432709
    2744 @misc{Drepper13,
    2745     keywords    = {thread-local storage},
    2746     contributer = {pabuhr@plg},
    2747     author      = {Ulrich Drepper},
    2748     title       = {{ELF} Handling For Thread-Local Storage},
    2749     year        = 2013,
    2750     month       = aug,
    2751     note        = {WikipediA},
    2752     howpublished= {\href{http://www.akkadia.org/drepper/tls.pdf}
    2753                   {http://\-www.akkadia.org/\-drepper/\-tls.pdf}},
    2754 }
    2755 
    27562710@misc{Turley99,
    27572711    keywords    = {embedded system, micrprocessor},
    … …  
    27642718    howpublished= {\href{https://www.eetimes.com/author.asp?sectionid=36&doc_id=1287712}
    27652719                  {https://\-www.eetimes.com/\-author.asp?sectionid=\-36&doc_id=1287712}},
    2766 }
    2767 
    2768 @article{Xiao19,
    2769     keywords    = {bug classification, fault trigger, Linux operating system, regression bug},
    2770     contributer = {pabuhr@plg},
    2771     author      = {Guanping Xiao and Zheng Zheng and Beibei Yin and Kishor S. Trivedi and Xiaoting Du and Kai-Yuan Cai},
    2772     title       = {An Empirical Study of Fault Triggers in the Linux Operating System: An Evolutionary Perspective},
    2773     journal     = {IEEE Transactions on Reliability},
    2774     month       = dec,
    2775     year        = 2019,
    2776     volume      = 68,
    2777     number      = 4,
    2778     pages       = {1356-1383},
    27792720}
    27802721
    … …  
    31963137}
    31973138
    3198 @inproceedings{Palix11,
    3199     keywords    = {Linux, fault-finding tools},
    3200     contributer = {pabuhr@plg},
    3201     author      = {Nicolas Palix and Ga\"el Thomas and Suman Saha and Christophe Calv\`es and Julia Lawall and Gilles Muller},
    3202     title       = {Faults in Linux: Ten Years Later},
    3203     booktitle   = {Proc. of the 16 International Conf. on Arch. Support for Prog. Lang. and Oper. Sys.},
    3204     series      = {ASPLOS'11},
    3205     month       = mar,
    3206     year        = 2011,
    3207     location    = {Newport Beach, California, USA},
    3208     pages       = {305-318},
    3209     publisher   = {ACM},
    3210     address     = {New York, NY, USA},
    3211 }
    3212 
    32133139@article{Lamport87,
    32143140    keywords    = {software solutions, mutual exclusion, fast},
    … …  
    33323258    issn        = {0001-0782},
    33333259    pages       = {107--115},
     3260    numpages    = {9},
     3261    url         = {http://doi.acm.org/10.1145/1538788.1538814},
     3262    doi         = {10.1145/1538788.1538814},
     3263    acmid       = {1538814},
    33343264    publisher   = {ACM},
    33353265    address     = {New York, NY, USA},
    … …  
    37343664}
    37353665
    3736 @mastersthesis{Radhakrishnan19,
    3737     author      = {Srihari Radhakrishnan},
    3738     title       = {High Performance Web Servers: A Study In Concurrent Programming Models},
    3739     school      = {School of Computer Sc., University of Waterloo},
    3740     year        = 2019,
    3741     optaddress  = {Waterloo, Ontario, Canada, N2L 3G1},
    3742     note        = {\href{https://uwspace.uwaterloo.ca/handle/10012/14706}{https://\-uwspace.uwaterloo.ca/\-handle/\-10012/\-14706}},
    3743 }
    3744 
    37453666@article{katzenelson83b,
    37463667    contributer = {gjditchfield@plg},
    … …  
    37763697    pages       = {115-138},
    37773698    year        = 1971,
    3778 }
    3779 
    3780 @inproceedings{Hagersten03,
    3781     keywords    = {cache storage, parallel architectures, performance evaluation, shared memory systems},
    3782     author      = {Zoran Radovi\'{c} and Erik Hagersten},
    3783     title       = {Hierarchical backoff locks for nonuniform communication architectures},
    3784     booktitle   = {Proceedings of the Ninth International Symposium on High-Performance Computer Architecture},
    3785     year        = {2003},
    3786     location    = {Anaheim, CA, USA},
    3787     pages       = {241-252},
    3788     publisher   = {IEEE},
    37893699}
    37903700
    … …  
    44554365}
    44564366
    4457 @misc{gccValueLabels,
    4458     keywords    = {gcc extension, value labels},
    4459     contributer = {pabuhr@plg},
    4460     key         = {Labels as Values},
    4461     author      = {{gcc Extension}},
    4462     title       = {Labels as Values},
    4463     year        = {since gcc-3},
    4464     howpublished= {\href{https://gcc.gnu.org/onlinedocs/gcc/Labels-as-Values.html}
    4465                   {https:\-//gcc.gnu.org/\-onlinedocs/\-gcc/\-Labels-as-Values.html}},
    4466 }
    4467 
    44684367@mastersthesis{Clarke90,
    44694368    keywords    = {concurrency, postponing requests},
    … …  
    45584457
    45594458@article{Pierce00,
    4560     keywords    = {Scala, polymorphism, subtyping, type inference},
     4459    keywords    = {Scala},
    45614460    contributer = {a3moss@uwaterloo.ca},
    45624461    author      = {Pierce, Benjamin C. and Turner, David N.},
    … …  
    45704469    issn        = {0164-0925},
    45714470    pages       = {1--44},
     4471    numpages    = {44},
     4472    url         = {http://doi.acm.org/10.1145/345099.345100},
     4473    doi         = {10.1145/345099.345100},
     4474    acmid       = {345100},
    45724475    publisher   = {ACM},
    45734476    address     = {New York, NY, USA},
     4477    keywords    = {polymorphism, subtyping, type inference},
    45744478}
    4575 
    4576 @article{Dice15,
    4577     keywords    = {Concurrency, NUMA, hierarchical locks, locks, multicore, mutex, mutual exclusion, spin locks},
    4578     author      = {Dice, David and Marathe, Virendra J. and Shavit, Nir},
    4579     title       = {Lock Cohorting: A General Technique for Designing NUMA Locks},
    4580     journal     = {ACM Trans. Parallel Comput.},
    4581     issue_date  = {January 2015},
    4582     volume      = 1,
    4583     number      = 2,
    4584     month       = feb,
    4585     year        = 2015,
    4586     pages       = {13:1--13:42},
    4587     publisher   = {ACM},
    4588     address     = {New York, NY, USA},
    4589 }
    45904479
    45914480@article{Sundell08,
    … …  
    46654554    journal     = sigplan,
    46664555    year        = 1989,
    4667     month       = jun,
    4668     volume      = 24,
    4669     number      = 6,
    4670     pages       = {37-48},
     4556    month       = jun, volume = 24, number = 6, pages = {37-48},
    46714557    abstract    = {
    46724558        This paper describes a scheme we have used to manage a large
    … …  
    51094995    year        = 1986,
    51104996    pages       = {313--326},
     4997    numpages    = {14},
    51114998    publisher   = {ACM},
    51124999    address     = {New York, NY, USA},
    … …  
    51245011    year        = 1986,
    51255012    pages       = {327--348},
     5013    numpages    = {22},
    51265014    publisher   = {ACM},
    51275015    address     = {New York, NY, USA},
    … …  
    53205208    year        = 2005,
    53215209    pages       = {146-196},
     5210    numpages    = {51},
    53225211    publisher   = {ACM},
    53235212    address     = {New York, NY, USA},
    … …  
    54655354    year        = 2000,
    54665355    pages       = {29-46},
    5467     note        = {OOPSLA'00, Oct. 15--19, 2000, Minneapolis, Minn., U.S.A.},
     5356    note        = {OOPSLA'00, Oct. 15--19, 2000, Minneapolis, Minnesota, U.S.A.},
    54685357}
    54695358
    … …  
    55795468    location    = {San Diego, California, USA},
    55805469    pages       = {101--112},
     5470    numpages    = {12},
     5471    url         = {http://doi.acm.org/10.1145/2535838.2535878},
     5472    doi         = {10.1145/2535838.2535878},
     5473    acmid       = {2535878},
    55815474    publisher   = {ACM},
    55825475    address     = {New York, NY, USA},
    … …  
    56825575    issn        = {0362-1340},
    56835576    pages       = {30--42},
     5577    numpages    = {13},
     5578    url         = {http://doi.acm.org/10.1145/947586.947589},
     5579    doi         = {10.1145/947586.947589},
    56845580    publisher   = {ACM},
    56855581    address     = {New York, NY, USA}
    … …  
    62166112    month       = 9,
    62176113    year        = 2005,
    6218 }
    6219 
    6220 @article{Bauer15,
    6221     keywords    = {resumption exceptions, theory},
    6222     contributer = {pabuhr@plg},
    6223     author      = {Andrej Bauer and Matija Pretnar},
    6224     title       = {Programming with Algebraic Effects and Handlers},
    6225     journal     = {Journal of Logical and Algebraic Methods in Programming},
    6226     publisher   = {Elsevier BV},
    6227     volume      = 84,
    6228     number      = 1,
    6229     month       = jan,
    6230     year        = 2015,
    6231     pages       = {108-123},
    62326114}
    62336115
    … …  
    66176499    issn        = {0164-0925},
    66186500    pages       = {429-475},
     6501    url         = {http://doi.acm.org/10.1145/1133651.1133653},
     6502    doi         = {10.1145/1133651.1133653},
     6503    acmid       = {1133653},
    66196504    publisher   = {ACM},
    66206505    address     = {New York, NY, USA},
    … …  
    69946879    issn        = {0001-0782},
    69956880    pages       = {565--569},
     6881    numpages    = {5},
     6882    url         = {http://doi.acm.org/10.1145/359545.359566},
     6883    doi         = {10.1145/359545.359566},
     6884    acmid       = {359566},
    69966885    publisher   = {ACM},
    69976886    address     = {New York, NY, USA}
    … …  
    70116900    issn        = {0362-1340},
    70126901    pages       = {145--147},
     6902    numpages    = {3},
     6903    url         = {http://doi.acm.org/10.1145/122598.122614},
     6904    doi         = {10.1145/122598.122614},
     6905    acmid       = {122614},
    70136906    publisher   = {ACM},
    70146907    address     = {New York, NY, USA},
    … …  
    71137006    issn        = {0362-1340},
    71147007    pages       = {82--87},
     7008    numpages    = {6},
     7009    url         = {http://doi.acm.org/10.1145/947680.947688},
     7010    doi         = {10.1145/947680.947688},
    71157011    publisher   = {ACM},
    71167012    address     = {New York, NY, USA},
    … …  
    72577153}
    72587154
    7259 @article{Cascaval08,
    7260     author      = {Cascaval, Calin and Blundell, Colin and Michael, Maged and Cain, Harold W. and Wu, Peng and Chiras, Stefanie and Chatterjee, Siddhartha},
    7261     title       = {Software Transactional Memory: Why Is It Only a Research Toy?},
    7262     journal     = {Queue},
    7263     volume      = {6},
    7264     number      = {5},
    7265     month       = sep,
    7266     year        = {2008},
    7267     pages       = {40:46--40:58},
    7268     publisher   = {ACM},
    7269     address     = {New York, NY, USA},
    7270 }
    7271 
    72727155@article{Dijkstra65a,
    72737156    keywords    = {N-thread software-solution mutual exclusion},
    … …  
    74807363    year        = 1974,
    74817364    pages       = {261-301},
     7365    issn        = {0360-0300},
     7366    doi         = {http://doi.acm.org/10.1145/356635.356640},
    74827367    publisher   = {ACM},
    74837368    address     = {New York, NY, USA},
    … …  
    75697454    publisher   = {ACM Press},
    75707455    address     = {New York, NY, USA},
     7456    doi         = {http://doi.acm.org/10.1145/356586.356588},
    75717457}
    75727458
    … …  
    78697755    howpublished= {\href{https://projects.eclipse.org/proposals/trace-compass}{https://\-projects.eclipse.org/\-proposals/\-trace-compass}},
    78707756}
    7871 
    7872 @inproceedings{Boehm09,
    7873     author      = {Boehm, Hans-J.},
    7874     title       = {Transactional Memory Should Be an Implementation Technique, Not a Programming Interface},
    7875     booktitle   = {Proceedings of the First USENIX Conference on Hot Topics in Parallelism},
    7876     series      = {HotPar'09},
    7877     year        = {2009},
    7878     location    = {Berkeley, California},
    7879     publisher   = {USENIX Association},
    7880     address     = {Berkeley, CA, USA},
    7881 }
    7882 
     7757 
    78837758@article{Leroy00,
    78847759    keywords    = {type-systems, exceptions},
    … …  
    79307805    number      = {2},
    79317806    pages       = {204-214},
    7932     month       = apr,
    7933     year        = 1988,
     7807    month       = apr, year = 1988,
    79347808    comment     = {
    79357809        Extended record types add fields to their base record.  Assignment
    … …  
    82368110    issn        = {0004-5411},
    82378111    pages       = {245--281},
     8112    numpages    = {37},
     8113    url         = {http://doi.acm.org/10.1145/62.2160},
     8114    doi         = {10.1145/62.2160},
     8115    acmid       = {2160},
    82388116    publisher   = {ACM},
    82398117    address     = {New York, NY, USA},
    … …  
    82488126    contributer = {pabuhr@plg},
    82498127    author      = {Boehm, Hans-J. and Adve, Sarita V.},
    8250     title       = {You Don't Know Jack About Shared Variables or Memory Models},
     8128    title       = {You Don'T Know Jack About Shared Variables or Memory Models},
    82518129    journal     = cacm,
    82528130    volume      = 55,

  • doc/papers/concurrency/Paper.tex

    r5452673 rc886f4b  
    6161\newcommand{\CCseventeen}{\textrm{C}\kern-.1em\hbox{+\kern-.25em+}17\xspace} % C++17 symbolic name
    6262\newcommand{\CCtwenty}{\textrm{C}\kern-.1em\hbox{+\kern-.25em+}20\xspace} % C++20 symbolic name
    63 \newcommand{\Csharp}{C\raisebox{-0.7ex}{\large$^\sharp$}\xspace} % C# symbolic name
     63\newcommand{\Csharp}{C\raisebox{-0.7ex}{\Large$^\sharp$}\xspace} % C# symbolic name
    6464
    6565%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
    … …  
    127127\newcommand*{\etc}{%
    128128        \@ifnextchar{.}{\ETC}%
    129                 {\ETC.\xspace}%
     129        {\ETC.\xspace}%
    130130}}{}%
    131131\@ifundefined{etal}{
    132132\newcommand{\ETAL}{\abbrevFont{et}~\abbrevFont{al}}
    133133\newcommand*{\etal}{%
    134         \@ifnextchar{.}{\ETAL}%
    135                 {\ETAL.\xspace}%
     134        \@ifnextchar{.}{\protect\ETAL}%
     135                {\protect\ETAL.\xspace}%
    136136}}{}%
    137137\@ifundefined{viz}{
    … …  
    163163                __float80, float80, __float128, float128, forall, ftype, generator, _Generic, _Imaginary, __imag, __imag__,
    164164                inline, __inline, __inline__, __int128, int128, __label__, monitor, mutex, _Noreturn, one_t, or,
    165                 otype, restrict, resume, __restrict, __restrict__, __signed, __signed__, _Static_assert, suspend, thread,
     165                otype, restrict, __restrict, __restrict__, __signed, __signed__, _Static_assert, thread,
    166166                _Thread_local, throw, throwResume, timeout, trait, try, ttype, typeof, __typeof, __typeof__,
    167167                virtual, __volatile, __volatile__, waitfor, when, with, zero_t},
    168168        moredirectives={defined,include_next},
    169169        % replace/adjust listing characters that look bad in sanserif
    170         literate={-}{\makebox[1ex][c]{\raisebox{0.5ex}{\rule{0.8ex}{0.1ex}}}}1 {^}{\raisebox{0.6ex}{$\scriptstyle\land\,$}}1
     170        literate={-}{\makebox[1ex][c]{\raisebox{0.4ex}{\rule{0.8ex}{0.1ex}}}}1 {^}{\raisebox{0.6ex}{$\scriptstyle\land\,$}}1
    171171                {~}{\raisebox{0.3ex}{$\scriptstyle\sim\,$}}1 % {`}{\ttfamily\upshape\hspace*{-0.1ex}`}1
    172172                {<}{\textrm{\textless}}1 {>}{\textrm{\textgreater}}1
    … …  
    197197                _Else, _Enable, _Event, _Finally, _Monitor, _Mutex, _Nomutex, _PeriodicTask, _RealTimeTask,
    198198                _Resume, _Select, _SporadicTask, _Task, _Timeout, _When, _With, _Throw},
     199}
     200\lstdefinelanguage{Golang}{
     201        morekeywords=[1]{package,import,func,type,struct,return,defer,panic,recover,select,var,const,iota,},
     202        morekeywords=[2]{string,uint,uint8,uint16,uint32,uint64,int,int8,int16,int32,int64,
     203                bool,float32,float64,complex64,complex128,byte,rune,uintptr, error,interface},
     204        morekeywords=[3]{map,slice,make,new,nil,len,cap,copy,close,true,false,delete,append,real,imag,complex,chan,},
     205        morekeywords=[4]{for,break,continue,range,goto,switch,case,fallthrough,if,else,default,},
     206        morekeywords=[5]{Println,Printf,Error,},
     207        sensitive=true,
     208        morecomment=[l]{//},
     209        morecomment=[s]{/*}{*/},
     210        morestring=[b]',
     211        morestring=[b]",
     212        morestring=[s]{`}{`},
    199213}
    200214
    … …  
    227241{}
    228242\lstnewenvironment{uC++}[1][]
    229 {\lstset{language=uC++,moredelim=**[is][\protect\color{red}]{`}{`},#1}\lstset{#1}}
     243{\lstset{#1}}
    230244{}
    231245\lstnewenvironment{Go}[1][]
    … …  
    268282\CFA is a polymorphic, non-object-oriented, concurrent, backwards-compatible extension of the C programming language.
    269283This paper discusses the design philosophy and implementation of its advanced control-flow and concurrent/parallel features, along with the supporting runtime written in \CFA.
    270 These features are created from scratch as ISO C has only low-level and/or unimplemented concurrency, so C programmers continue to rely on library approaches like pthreads.
     284These features are created from scratch as ISO C has only low-level and/or unimplemented concurrency, so C programmers continue to rely on library features like pthreads.
    271285\CFA introduces modern language-level control-flow mechanisms, like generators, coroutines, user-level threading, and monitors for mutual exclusion and synchronization.
    272286% Library extension for executors, futures, and actors are built on these basic mechanisms.
    … …  
    281295
    282296\begin{document}
    283 \linenumbers                            % comment out to turn off line numbering
     297\linenumbers                                            % comment out to turn off line numbering
    284298
    285299\maketitle
    … …  
    288302\section{Introduction}
    289303
    290 \CFA~\cite{Moss18,Cforall} is a modern, polymorphic, non-object-oriented\footnote{
    291 \CFA has object-oriented features, such as constructors, destructors, virtuals and simple trait/interface inheritance.
    292 % Go interfaces, Rust traits, Swift Protocols, Haskell Type Classes and Java Interfaces.
    293 % "Trait inheritance" works for me. "Interface inheritance" might also be a good choice, and distinguish clearly from implementation inheritance.
    294 % You'll want to be a little bit careful with terms like "structural" and "nominal" inheritance as well. CFA has structural inheritance (I think Go as well) -- it's inferred based on the structure of the code. Java, Rust, and Haskell (not sure about Swift) have nominal inheritance, where there needs to be a specific statement that "this type inherits from this type".
     304This paper discusses the design philosophy and implementation of advanced language-level control-flow and concurrent/parallel features in \CFA~\cite{Moss18,Cforall} and its runtime, which is written entirely in \CFA.
     305\CFA is a modern, polymorphic, non-object-oriented\footnote{
     306\CFA has features often associated with object-oriented programming languages, such as constructors, destructors, virtuals and simple inheritance.
    295307However, 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.},
    296308backwards-compatible extension of the C programming language.
    297 In many ways, \CFA is to C as Scala~\cite{Scala} is to Java, providing a \emph{research vehicle} for new typing and control-flow capabilities on top of a highly popular programming language\footnote{
    298 The TIOBE index~\cite{TIOBE} for December 2019 ranks the top five \emph{popular} programming languages as Java 17\%, C 16\%, Python 10\%, and \CC 6\%, \Csharp 5\% = 54\%, and over the past 30 years, C has always ranked either first or second in popularity.}
    299 allowing immediate dissemination.
    300 This paper discusses the design philosophy and implementation of advanced language-level control-flow and concurrent/parallel features in \CFA and its runtime, which is written entirely in \CFA.
    301 The \CFA control-flow framework extends ISO \Celeven~\cite{C11} with new call/return and concurrent/parallel control-flow.
    302 
    303 % The call/return extensions retain state between callee and caller versus losing the callee's state on return;
    304 % the concurrency extensions allow high-level management of threads.
    305 
    306 Call/return control-flow with argument/parameter passing appeared in the first programming languages.
    307 Over the past 50 years, call/return has been augmented with features like static/dynamic call, exceptions (multi-level return) and generators/coroutines (retain state between calls).
    308 While \CFA has mechanisms for dynamic call (algebraic effects) and exceptions\footnote{
    309 \CFA exception handling will be presented in a separate paper.
    310 The key feature that dovetails with this paper is nonlocal 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++}}, this work only discusses retaining state between calls via generators/coroutines.
    311 \newterm{Coroutining} was introduced by Conway~\cite{Conway63} (1963), discussed by Knuth~\cite[\S~1.4.2]{Knuth73V1}, implemented in Simula67~\cite{Simula67}, formalized by Marlin~\cite{Marlin80}, and is now popular and appears in old and new programming languages: CLU~\cite{CLU}, \Csharp~\cite{Csharp}, Ruby~\cite{Ruby}, Python~\cite{Python}, JavaScript~\cite{JavaScript}, Lua~\cite{Lua}, \CCtwenty~\cite{C++20Coroutine19}.
    312 Coroutining is sequential execution requiring direct handoff among coroutines, \ie only the programmer is controlling execution order.
    313 If coroutines transfer to an internal event-engine for scheduling the next coroutines, the program transitions into the realm of concurrency~\cite[\S~3]{Buhr05a}.
    314 Coroutines are only a stepping stone towards concurrency where the commonality is that coroutines and threads retain state between calls.
    315 
    316 \Celeven/\CCeleven define concurrency~\cite[\S~7.26]{C11}, but it is largely wrappers for a subset of the pthreads library~\cite{Pthreads}.\footnote{Pthreads concurrency is based on simple thread fork/join in a function and mutex/condition locks, which is low-level and error-prone}
    317 Interestingly, almost a decade after the \Celeven standard, neither gcc-9, clang-9 nor msvc-19 (most recent versions) support the \Celeven include @threads.h@, indicating no interest in the C11 concurrency approach (possibly because of the recent effort to add concurrency to \CC).
    318 While the \Celeven standard does not state a threading model, the historical association with pthreads suggests implementations would adopt kernel-level threading (1:1)~\cite{ThreadModel}, as for \CC.
     309In many ways, \CFA is to C as Scala~\cite{Scala} is to Java, providing a \emph{research vehicle} for new typing and control-flow capabilities on top of a highly popular programming language allowing immediate dissemination.
     310Within the \CFA framework, new control-flow features are created from scratch because ISO \Celeven defines only a subset of the \CFA extensions, where the overlapping features are concurrency~\cite[\S~7.26]{C11}.
     311However, \Celeven concurrency is largely wrappers for a subset of the pthreads library~\cite{Butenhof97,Pthreads}, and \Celeven and pthreads concurrency is simple, based on thread fork/join in a function and mutex/condition locks, which is low-level and error-prone;
     312no high-level language concurrency features are defined.
     313Interestingly, almost a decade after publication of the \Celeven standard, neither gcc-8, clang-9 nor msvc-19 (most recent versions) support the \Celeven include @threads.h@, indicating little interest in the C11 concurrency approach (possibly because the effort to add concurrency to \CC).
     314Finally, while the \Celeven standard does not state a threading model, the historical association with pthreads suggests implementations would adopt kernel-level threading (1:1)~\cite{ThreadModel}.
     315
    319316In contrast, there has been a renewed interest during the past decade in user-level (M:N, green) threading in old and new programming languages.
    320317As multi-core hardware became available in the 1980/90s, both user and kernel threading were examined.
    321318Kernel 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}.
    322319Libraries like pthreads were developed for C, and the Solaris operating-system switched from user (JDK 1.1~\cite{JDK1.1}) to kernel threads.
    323 As a result, many current languages implementations adopt the 1:1 kernel-threading model, like Java (Scala), Objective-C~\cite{obj-c-book}, \CCeleven~\cite{C11}, C\#~\cite{Csharp} and Rust~\cite{Rust}, with a variety of presentation mechanisms.
    324 From 2000 onwards, several language implementations have championed the M:N user-threading model, like Go~\cite{Go}, Erlang~\cite{Erlang}, Haskell~\cite{Haskell}, D~\cite{D}, and \uC~\cite{uC++,uC++book}, including putting green threads back into Java~\cite{Quasar}, and many user-threading libraries have appeared~\cite{Qthreads,MPC,Marcel}.
    325 The main argument for user-level threading is that it is lighter weight than kernel threading (locking and context switching do not cross the kernel boundary), so there is less restriction on programming styles that encourages large numbers of threads performing medium-sized work to facilitate load balancing by the runtime~\cite{Verch12}.
     320As a result, languages like Java, Scala, Objective-C~\cite{obj-c-book}, \CCeleven~\cite{C11}, and C\#~\cite{Csharp} adopt the 1:1 kernel-threading model, with a variety of presentation mechanisms.
     321From 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,Marcel}, including putting green threads back into Java~\cite{Quasar}.
     322The main argument for user-level threading is that it is lighter weight than kernel threading (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 medium work units to facilitate load balancing by the runtime~\cite{Verch12}.
    326323As well, user-threading facilitates a simpler concurrency approach using thread objects that leverage sequential patterns versus events with call-backs~\cite{Adya02,vonBehren03}.
    327324Finally, performant user-threading implementations (both time and space) meet or exceed direct kernel-threading implementations, while achieving the programming advantages of high concurrency levels and safety.
    328325
    329 A further effort over the past two decades is the development of language memory models to deal with the conflict between language features and compiler/hardware optimizations, \eg some language features are unsafe in the presence of aggressive sequential optimizations~\cite{Buhr95a,Boehm05}.
     326A further effort over the past two decades is the development of language memory models to deal with the conflict between language features and compiler/hardware optimizations, \ie some language features are unsafe in the presence of aggressive sequential optimizations~\cite{Buhr95a,Boehm05}.
    330327The consequence is that a language must provide sufficient tools to program around safety issues, as inline and library code is all sequential to the compiler.
    331328One solution is low-level qualifiers and functions (\eg @volatile@ and atomics) allowing \emph{programmers} to explicitly write safe (race-free~\cite{Boehm12}) programs.
    332 A safer solution is high-level language constructs so the \emph{compiler} knows the concurrency boundaries (where mutual exclusion and synchronization are acquired/released) and provide implicit safety at and across these boundaries.
    333 While the optimization problem is best known with respect to concurrency, it applies to other complex control-flow, like exceptions and coroutines.
    334 As well, language solutions allow matching the language paradigm with the approach, \eg matching the functional paradigm with data-flow programming or the imperative paradigm with thread programming.
    335 
    336 Finally, it is important for a language to provide safety over performance \emph{as the default}, allowing careful reduction of safety (unsafe code) for performance when necessary.
    337 Two concurrency violations of this philosophy are \emph{spurious wakeup} (random wakeup~\cite[\S~9]{Buhr05a}) and \emph{barging}\footnote{
    338 Barging is competitive succession instead of direct handoff, \ie after a lock is released both arriving and preexisting waiter threads compete to acquire the lock.
    339 Hence, an arriving thread can temporally \emph{barge} ahead of threads already waiting for an event, which can repeat indefinitely leading to starvation of waiter threads.
     329A safer solution is high-level language constructs so the \emph{compiler} knows the optimization boundaries, and hence, provides implicit safety.
     330This problem is best known with respect to concurrency, but applies to other complex control-flow, like exceptions\footnote{
     331\CFA exception handling will be presented in a separate paper.
     332The key feature that dovetails with this paper is nonlocal 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++}
     333} and coroutines.
     334Finally, language solutions allow matching constructs with language paradigm, \ie imperative and functional languages often have different presentations of the same concept to fit their programming model.
     335
     336Finally, it is important for a language to provide safety over performance \emph{as the default}, allowing careful reduction of safety for performance when necessary.
     337Two concurrency violations of this philosophy are \emph{spurious wakeup} (random wakeup~\cite[\S~8]{Buhr05a}) and \emph{barging}\footnote{
     338The notion of competitive succession instead of direct handoff, \ie a lock owner releases the lock and an arriving thread acquires it ahead of preexisting waiter threads.
    340339} (signals-as-hints~\cite[\S~8]{Buhr05a}), where one is a consequence of the other, \ie once there is spurious wakeup, signals-as-hints follow.
    341 (Author experience teaching concurrency is that students are confused by these semantics.)
    342 However, spurious wakeup is \emph{not} a foundational concurrency property~\cite[\S~9]{Buhr05a};
    343 it is a performance design choice.
    344 We argue removing spurious wakeup and signals-as-hints make concurrent programming simpler and safer as there is less local non-determinism to manage.
    345 If barging acquisition is allowed, its specialized performance advantage should be available as an option not the default.
    346 
    347 \CFA embraces language extensions for advanced control-flow, user-level threading, and safety as the default.
    348 We present comparative examples to support our argument that the \CFA control-flow extensions are as expressive and safe as those in other concurrent imperative programming languages, and perform experiments to show the \CFA runtime is competitive with other similar mechanisms.
     340However, spurious wakeup is \emph{not} a foundational concurrency property~\cite[\S~8]{Buhr05a}, it is a performance design choice.
     341Similarly, signals-as-hints are often a performance decision.
     342We argue removing spurious wakeup and signals-as-hints make concurrent programming significantly safer because it removes local non-determinism and matches with programmer expectation.
     343(Author experience teaching concurrency is that students are highly confused by these semantics.)
     344Clawing back performance, when local non-determinism is unimportant, should be an option not the default.
     345
     346\begin{comment}
     347Most 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.
     348As a result, there is a significant learning curve to move to these languages, and C legacy-code must be rewritten.
     349While \CC, like \CFA, takes an evolutionary approach to extend C, \CC's constantly growing complex and interdependent features-set (\eg objects, inheritance, templates, etc.) mean idiomatic \CC code is difficult to use from C, and C programmers must expend significant effort learning \CC.
     350Hence, rewriting and retraining costs for these languages, even \CC, are prohibitive for companies with a large C software-base.
     351\CFA with its orthogonal feature-set, its high-performance runtime, and direct access to all existing C libraries circumvents these problems.
     352\end{comment}
     353
     354\CFA embraces user-level threading, language extensions for advanced control-flow, and safety as the default.
     355We present comparative examples so the reader can judge if the \CFA control-flow extensions are better and safer than those in other concurrent, imperative programming languages, and perform experiments to show the \CFA runtime is competitive with other similar mechanisms.
    349356The main contributions of this work are:
    350 \begin{itemize}[topsep=3pt,itemsep=0pt]
     357\begin{itemize}[topsep=3pt,itemsep=1pt]
    351358\item
    352 a set of fundamental execution properties that dictate which language-level control-flow features need to be supported,
    353 
     359language-level generators, coroutines and user-level threading, which respect the expectations of C programmers.
    354360\item
    355 integration of these language-level control-flow features, while respecting the style and expectations of C programmers,
    356 
     361monitor synchronization without barging, and the ability to safely acquiring multiple monitors \emph{simultaneously} (deadlock free), while seamlessly integrating these capabilities with all monitor synchronization mechanisms.
    357362\item
    358 monitor synchronization without barging, and the ability to safely acquiring multiple monitors \emph{simultaneously} (deadlock free), while seamlessly integrating these capabilities with all monitor synchronization mechanisms,
    359 
    360 \item
    361 providing statically type-safe interfaces that integrate with the \CFA polymorphic type-system and other language features,
    362 
     363providing statically type-safe interfaces that integrate with the \CFA polymorphic type-system and other language features.
    363364% \item
    364365% library extensions for executors, futures, and actors built on the basic mechanisms.
    365 
    366366\item
    367 a runtime system without spurious wake-up and no performance loss,
    368 
     367a runtime system with no spurious wakeup.
    369368\item
    370 a dynamic partitioning mechanism to segregate groups of executing user and kernel threads performing specialized work (\eg web-server or compute engine) or requiring different scheduling (\eg NUMA or real-time).
    371 
     369a dynamic partitioning mechanism to segregate the execution environment for specialized requirements.
    372370% \item
    373371% a non-blocking I/O library
    374 
    375372\item
    376 experimental results showing comparable performance of the \CFA features with similar mechanisms in other languages.
     373experimental results showing comparable performance of the new features with similar mechanisms in other programming languages.
    377374\end{itemize}
    378375
    379 Section~\ref{s:FundamentalExecutionProperties} presents the compositional hierarchy of execution properties directing the design of control-flow features in \CFA.
    380 Section~\ref{s:StatefulFunction} begins advanced control by introducing sequential functions that retain data and execution state between calls producing constructs @generator@ and @coroutine@.
    381 Section~\ref{s:Concurrency} begins concurrency, or how to create (fork) and destroy (join) a thread producing the @thread@ construct.
     376Section~\ref{s:StatefulFunction} begins advanced control by introducing sequential functions that retain data and execution state between calls, which produces constructs @generator@ and @coroutine@.
     377Section~\ref{s:Concurrency} begins concurrency, or how to create (fork) and destroy (join) a thread, which produces the @thread@ construct.
    382378Section~\ref{s:MutualExclusionSynchronization} discusses the two mechanisms to restricted nondeterminism when controlling shared access to resources (mutual exclusion) and timing relationships among threads (synchronization).
    383379Section~\ref{s:Monitor} shows how both mutual exclusion and synchronization are safely embedded in the @monitor@ and @thread@ constructs.
    384380Section~\ref{s:CFARuntimeStructure} describes the large-scale mechanism to structure (cluster) threads and virtual processors (kernel threads).
    385 Section~\ref{s:Performance} uses a series of microbenchmarks to compare \CFA threading with pthreads, Java 11.0.6, Go 1.12.6, Rust 1.37.0, Python 3.7.6, Node.js 12.14.1, and \uC 7.0.0.
    386 
    387 
    388 \section{Fundamental Execution Properties}
    389 \label{s:FundamentalExecutionProperties}
    390 
    391 The features in a programming language should be composed from a set of fundamental properties rather than an ad hoc collection chosen by the designers.
    392 To this end, the control-flow features created for \CFA are based on the fundamental properties of any language with function-stack control-flow (see also \uC~\cite[pp.~140-142]{uC++}).
    393 The fundamental properties are execution state, thread, and mutual-exclusion/synchronization (MES).
    394 These independent properties can be used alone, in pairs, or in triplets to compose different language features, forming a compositional hierarchy where the most advanced feature has all the properties (state/thread/MES).
    395 While it is possible for a language to only support the most advanced feature~\cite{Hermes90}, this unnecessarily complicates and makes inefficient solutions to certain classes of problems.
    396 As is shown, each of the (non-rejected) composed features solves a particular set of problems, and hence, has a defensible position in a programming language.
    397 If a compositional feature is missing, a programmer has too few/many fundamental properties resulting in a complex and/or is inefficient solution.
    398 
    399 In detail, the fundamental properties are:
    400 \begin{description}[leftmargin=\parindent,topsep=3pt,parsep=0pt]
    401 \item[\newterm{execution state}:]
    402 is the state information needed by a control-flow feature to initialize, manage compute data and execution location(s), and de-initialize.
    403 State is retained in fixed-sized aggregate structures and dynamic-sized stack(s), often allocated in the heap(s) managed by the runtime system.
    404 The lifetime of the state varies with the control-flow feature, where longer life-time and dynamic size provide greater power but also increase usage complexity and cost.
    405 Control-flow transfers among execution states occurs in multiple ways, such as function call, context switch, asynchronous await, etc.
    406 Because the programming language determines what constitutes an execution state, implicitly manages this state, and defines movement mechanisms among states, execution state is an elementary property of the semantics of a programming language.
    407 % An execution-state is related to the notion of a process continuation \cite{Hieb90}.
    408 
    409 \item[\newterm{threading}:]
    410 is execution of code that occurs independently of other execution, \ie the execution resulting from a thread is sequential.
    411 Multiple threads provide \emph{concurrent execution};
    412 concurrent execution becomes parallel when run on multiple processing units (hyper-threading, cores, sockets).
    413 There must be language mechanisms to create, block/unblock, and join with a thread.
    414 
    415 \item[\newterm{MES}:]
    416 is the concurrency mechanisms to perform an action without interruption and establish timing relationships among multiple threads.
    417 These two properties are independent, \ie mutual exclusion cannot provide synchronization and vice versa without introducing additional threads~\cite[\S~4]{Buhr05a}.
    418 Limiting MES, \eg no access to shared data, results in contrived solutions and inefficiency on multi-core von Neumann computers where shared memory is a foundational aspect of its design.
    419 \end{description}
    420 These properties are fundamental because they cannot be built from existing language features, \eg a basic programming language like C99~\cite{C99} cannot create new control-flow features, concurrency, or provide MES using atomic hardware mechanisms.
    421 
    422 
    423 \subsection{Execution Properties}
    424 
    425 Table~\ref{t:ExecutionPropertyComposition} shows how the three fundamental execution properties: state, thread, and mutual exclusion compose a hierarchy of control-flow features needed in a programming language.
    426 (When doing case analysis, not all combinations are meaningful.)
    427 Note, basic von Neumann execution requires at least one thread and an execution state providing some form of call stack.
    428 For table entries missing these minimal components, the property is borrowed from the invoker (caller).
    429 
    430 Case 1 is a function that borrows storage for its state (stack frame/activation) and a thread from its invoker and retains this state across \emph{callees}, \ie function local-variables are retained on the stack across calls.
    431 Case 2 is case 1 with access to shared state so callers are restricted during update (mutual exclusion) and scheduling for other threads (synchronization).
    432 Case 3 is a stateful function supporting resume/suspend along with call/return to retain state across \emph{callers}, but has some restrictions because the function's state is stackless.
    433 Note, stackless functions still borrow the caller's stack and thread, where the stack is used to preserve state across its callees.
    434 Case 4 is cases 2 and 3 with protection to shared state for stackless functions.
    435 Cases 5 and 6 are the same as 3 and 4 but only the thread is borrowed as the function state is stackful, so resume/suspend is a context switch from the caller's to the function's stack.
    436 Cases 7 and 8 are rejected because a function that is given a new thread must have its own stack where the thread begins and stack frames are stored for calls, \ie there is no stack to borrow.
    437 Cases 9 and 10 are rejected because a thread with a fixed state (no stack) cannot accept calls, make calls, block, or be preempted, all of which require an unknown amount of additional dynamic state.
    438 Hence, once started, this kind of thread must execute to completion, \ie computation only, which severely restricts runtime management.
    439 Cases 11 and 12 have a stackful thread with and without safe access to shared state.
    440 Execution properties increase the cost of creation and execution along with complexity of usage.
    441 
    442 \begin{table}
    443 \caption{Execution property composition}
    444 \centering
    445 \label{t:ExecutionPropertyComposition}
    446 \renewcommand{\arraystretch}{1.25}
    447 %\setlength{\tabcolsep}{5pt}
    448 \begin{tabular}{c|c||l|l}
    449 \multicolumn{2}{c||}{execution properties} & \multicolumn{2}{c}{mutual exclusion / synchronization} \\
    450 \hline
    451 stateful                        & thread        & \multicolumn{1}{c|}{No} & \multicolumn{1}{c}{Yes} \\
    452 \hline   
    453 \hline   
    454 No                                      & No            & \textbf{1}\ \ \ function                              & \textbf{2}\ \ \ @monitor@ function    \\
    455 \hline   
    456 Yes (stackless)         & No            & \textbf{3}\ \ \ @generator@                   & \textbf{4}\ \ \ @monitor@ @generator@ \\
    457 \hline   
    458 Yes (stackful)          & No            & \textbf{5}\ \ \ @coroutine@                   & \textbf{6}\ \ \ @monitor@ @coroutine@ \\
    459 \hline   
    460 No                                      & Yes           & \textbf{7}\ \ \ {\color{red}rejected} & \textbf{8}\ \ \ {\color{red}rejected} \\
    461 \hline   
    462 Yes (stackless)         & Yes           & \textbf{9}\ \ \ {\color{red}rejected} & \textbf{10}\ \ \ {\color{red}rejected} \\
    463 \hline   
    464 Yes (stackful)          & Yes           & \textbf{11}\ \ \ @thread@                             & \textbf{12}\ \ @monitor@ @thread@             \\
    465 \end{tabular}
    466 \end{table}
    467 
    468 Given the execution-properties taxonomy, programmers can now answer three basic questions: is state necessary across calls and how much, is a separate thread necessary, is access to shared state necessary.
    469 The answers define the optimal language feature need for implementing a programming problem.
    470 The next sections discusses how \CFA fills in the table with language features, while other programming languages may only provide a subset of the table.
    471 
    472 
    473 \subsection{Design Requirements}
    474 
    475 The following design requirements largely stem from building \CFA on top of C.
    476 \begin{itemize}[topsep=3pt,parsep=0pt]
    477 \item
    478 All communication must be statically type checkable for early detection of errors and efficient code generation.
    479 This requirement is consistent with the fact that C is a statically-typed programming-language.
    480 
    481 \item
    482 Direct interaction among language features must be possible allowing any feature to be selected without restricting comm\-unication.
    483 For example, many concurrent languages do not provide direct communication (calls) among threads, \ie threads only communicate indirectly through monitors, channels, messages, and/or futures.
    484 Indirect communication increases the number of objects, consuming more resources, and require additional synchronization and possibly data transfer.
    485 
    486 \item
    487 All communication is performed using function calls, \ie data is transmitted from argument to parameter and results are returned from function calls.
    488 Alternative forms of communication, such as call-backs, message passing, channels, or communication ports, step outside of C's normal form of communication.
    489 
    490 \item
    491 All stateful features must follow the same declaration scopes and lifetimes as other language data.
    492 For C that means at program startup, during block and function activation, and on demand using dynamic allocation.
    493 
    494 \item
    495 MES must be available implicitly in language constructs as well as explicitly for specialized requirements, because requiring programmers to build MES using low-level locks often leads to incorrect programs.
    496 Furthermore, reducing synchronization scope by encapsulating it within language constructs further reduces errors in concurrent programs.
    497 
    498 \item
    499 Both synchronous and asynchronous communication are needed.
    500 However, we believe the best way to provide asynchrony, such as call-buffering/chaining and/or returning futures~\cite{multilisp}, is building it from expressive synchronous features.
    501 
    502 \item
    503 Synchronization must be able to control the service order of requests including prioritizing selection from different kinds of outstanding requests, and postponing a request for an unspecified time while continuing to accept new requests.
    504 Otherwise, certain concurrency problems are difficult, e.g.\ web server, disk scheduling, and the amount of concurrency is inhibited~\cite{Gentleman81}.
    505 \end{itemize}
    506 We have satisfied these requirements in \CFA while maintaining backwards compatibility with the huge body of legacy C programs.
    507 % In contrast, other new programming languages must still access C programs (\eg operating-system service routines), but do so through fragile C interfaces.
    508 
    509 
    510 \subsection{Asynchronous Await / Call}
    511 
    512 Asynchronous await/call is a caller mechanism for structuring programs and/or increasing concurrency, where the caller (client) postpones an action into the future, which is subsequently executed by a callee (server).
    513 The caller detects the action's completion through a \newterm{future}/\newterm{promise}.
    514 The benefit is asynchronous caller execution with respect to the callee until future resolution.
    515 For single-threaded languages like JavaScript, an asynchronous call passes a callee action, which is queued in the event-engine, and continues execution with a promise.
    516 When the caller needs the promise to be fulfilled, it executes @await@.
    517 A promise-completion call-back can be part of the callee action or the caller is rescheduled;
    518 in either case, the call back is executed after the promise is fulfilled.
    519 While asynchronous calls generate new callee (server) events, we content this mechanism is insufficient for advanced control-flow mechanisms like generators or coroutines (which are discussed next).
    520 Specifically, control between caller and callee occurs indirectly through the event-engine precluding direct handoff and cycling among events, and requires complex resolution of a control promise and data.
    521 Note, @async-await@ is just syntactic-sugar over the event engine so it does not solve these deficiencies.
    522 For multi-threaded languages like Java, the asynchronous call queues a callee action with an executor (server), which subsequently executes the work by a thread in the executor thread-pool.
    523 The problem is when concurrent work-units need to interact and/or block as this effects the executor, \eg stops threads.
    524 While it is possible to extend this approach to support the necessary mechanisms, \eg message passing in Actors, we show monitors and threads provide an equally competitive approach that does not deviate from normal call communication and can be used to build asynchronous call, as is done in Java.
     381Section~\ref{s:Performance} uses a series of microbenchmarks to compare \CFA threading with pthreads, Java OpenJDK-9, Go 1.12.6 and \uC 7.0.0.
    525382
    526383
    … …  
    528385\label{s:StatefulFunction}
    529386
    530 A \emph{stateful function} has the ability to remember state between calls, where state can be either data or execution, \eg plugin, device driver, finite-state machine (FSM).
    531 A simple technique to retain data state between calls is @static@ declarations within a function, which is often implemented by hoisting the declarations to the global scope but hiding the names within the function using name mangling.
    532 However, each call starts the function at the top making it difficult to determine the last point of execution in an algorithm, and requiring multiple flag variables and testing to reestablish the continuation point.
    533 Hence, the next step of generalizing function state is implicitly remembering the return point between calls and reentering the function at this point rather than the top, called \emph{generators}\,/\,\emph{iterators} or \emph{stackless coroutines}.
    534 For example, a Fibonacci generator retains data and execution state allowing it to remember prior values needed to generate the next value and the location in the algorithm to compute that value.
    535 The next step of generalization is instantiating the function to allow multiple named instances, \eg multiple Fibonacci generators, where each instance has its own state, and hence, can generate an independent sequence of values.
    536 Note, a subset of generator state is a function \emph{closure}, \ie the technique of capturing lexical references when returning a nested function.
    537 A further generalization is adding a stack to a generator's state, called a \emph{coroutine}, so it can suspend outside of itself, \eg call helper functions to arbitrary depth before suspending back to its resumer without unwinding these calls.
    538 For example, a coroutine iterator for a binary tree can stop the traversal at the visit point (pre, infix, post traversal), return the node value to the caller, and then continue the recursive traversal from the current node on the next call.
    539 
    540 There are two styles of activating a stateful function, \emph{asymmetric} or \emph{symmetric}, identified by resume/suspend (no cycles) and resume/resume (cycles).
    541 These styles \emph{do not} cause incremental stack growth, \eg a million resume/suspend or resume/resume cycles do not remember each cycle just the last resumer for each cycle.
    542 Selecting between stackless/stackful semantics and asymmetric/symmetric style is a tradeoff between programming requirements, performance, and design, where stackless is faster and smaller (modified call/return between closures), stackful is more general but slower and larger (context switching between distinct stacks), and asymmetric is simpler control-flow than symmetric.
    543 Additionally, storage management for the closure/stack (especially in unmanaged languages, \ie no garbage collection) must be factored into design and performance.
    544 Note, creation cost (closure/stack) is amortized across usage, so activation cost (resume/suspend) is usually the dominant factor.
    545 
    546 % The stateful function is an old idea~\cite{Conway63,Marlin80} that is new again~\cite{C++20Coroutine19}, where execution is temporarily suspended and later resumed, \eg plugin, device driver, finite-state machine.
    547 % Hence, a stateful function may not end when it returns to its caller, allowing it to be restarted with the data and execution location present at the point of suspension.
    548 % If the closure is fixed size, we call it a \emph{generator} (or \emph{stackless}), and its control flow is restricted, \eg suspending outside the generator is prohibited.
    549 % If the closure is variable size, we call it a \emph{coroutine} (or \emph{stackful}), and as the names implies, often implemented with a separate stack with no programming restrictions.
    550 % Hence, refactoring a stackless coroutine may require changing it to stackful.
    551 % A foundational property of all \emph{stateful functions} is that resume/suspend \emph{do not} cause incremental stack growth, \ie resume/suspend operations are remembered through the closure not the stack.
    552 % As well, activating a stateful function is \emph{asymmetric} or \emph{symmetric}, identified by resume/suspend (no cycles) and resume/resume (cycles).
    553 % A fixed closure activated by modified call/return is faster than a variable closure activated by context switching.
    554 % Additionally, any storage management for the closure (especially in unmanaged languages, \ie no garbage collection) must also be factored into design and performance.
    555 % Therefore, selecting between stackless and stackful semantics is a tradeoff between programming requirements and performance, where stackless is faster and stackful is more general.
    556 % nppNote, creation cost is amortized across usage, so activation cost is usually the dominant factor.
    557 
    558 For example, Python presents asymmetric generators as a function object, \uC presents symmetric coroutines as a \lstinline[language=C++]|class|-like object, and many languages present threading using function pointers, @pthreads@~\cite{Butenhof97}, \Csharp~\cite{Csharp}, Go~\cite{Go}, and Scala~\cite{Scala}.
    559 \begin{center}
    560 \begin{tabular}{@{}l|l|l@{}}
    561 \multicolumn{1}{@{}c|}{Python asymmetric generator} & \multicolumn{1}{c|}{\uC symmetric coroutine} & \multicolumn{1}{c@{}}{Pthreads thread} \\
    562 \hline
    563 \begin{python}
    564 `def Gen():` $\LstCommentStyle{\color{red}// function}$
    565         ... yield val ...
    566 gen = Gen()
    567 for i in range( 10 ):
    568         print( next( gen ) )
    569 \end{python}
    570 &
    571 \begin{uC++}
    572 `_Coroutine Cycle {` $\LstCommentStyle{\color{red}// class}$
    573         Cycle * p;
    574         void main() { p->cycle(); }
    575         void cycle() { resume(); }  `};`
    576 Cycle c1, c2; c1.p=&c2; c2.p=&c1; c1.cycle();
    577 \end{uC++}
    578 &
    579 \begin{cfa}
    580 void * rtn( void * arg ) { ... }
    581 int i = 3, rc;
    582 pthread_t t; $\C{// thread id}$
    583 $\LstCommentStyle{\color{red}// function pointer}$
    584 rc=pthread_create(&t, `rtn`, (void *)i);
    585 \end{cfa}
    586 \end{tabular}
    587 \end{center}
    588 \CFA's preferred presentation model for generators/coroutines/threads is a hybrid of functions and classes, giving an object-oriented flavour.
    589 Essentially, the generator/coroutine/thread function is semantically coupled with a generator/coroutine/thread custom type via the type's name.
    590 The custom type solves several issues, while accessing the underlying mechanisms used by the custom types is still allowed for flexibility reasons.
    591 Each custom type is discussed in detail in the following sections.
    592 
    593 
    594 \subsection{Generator}
    595 
    596 Stackless generators (Table~\ref{t:ExecutionPropertyComposition} case 3) have the potential to be very small and fast, \ie as small and fast as function call/return for both creation and execution.
    597 The \CFA goal is to achieve this performance target, possibly at the cost of some semantic complexity.
    598 A series of different kinds of generators and their implementation demonstrate how this goal is accomplished.\footnote{
    599 The \CFA operator syntax uses \lstinline|?| to denote operands, which allows precise definitions for pre, post, and infix operators, \eg \lstinline|?++|, \lstinline|++?|, and \lstinline|?+?|, in addition \lstinline|?\{\}| denotes a constructor, as in \lstinline|foo `f` = `\{`...`\}`|, \lstinline|^?\{\}| denotes a destructor, and \lstinline|?()| is \CC function call \lstinline|operator()|.
    600 Operator \lstinline+|+ is overloaded for printing, like bit-shift \lstinline|<<| in \CC.
    601 The \CFA \lstinline|with| clause opens an aggregate scope making its fields directly accessible, like Pascal \lstinline|with|, but using parallel semantics;
    602 multiple aggregates may be opened.
    603 \CFA has rebindable references \lstinline|int i, & ip = i, j; `&ip = &j;`| and non-rebindable references \lstinline|int i, & `const` ip = i, j; `&ip = &j;` // disallowed|.
    604 }%
     387The stateful function is an old idea~\cite{Conway63,Marlin80} that is new again~\cite{C++20Coroutine19}, where execution is temporarily suspended and later resumed, \eg plugin, device driver, finite-state machine.
     388Hence, a stateful function may not end when it returns to its caller, allowing it to be restarted with the data and execution location present at the point of suspension.
     389This capability is accomplished by retaining a data/execution \emph{closure} between invocations.
     390If the closure is fixed size, we call it a \emph{generator} (or \emph{stackless}), and its control flow is restricted, \eg suspending outside the generator is prohibited.
     391If the closure is variable size, we call it a \emph{coroutine} (or \emph{stackful}), and as the names implies, often implemented with a separate stack with no programming restrictions.
     392Hence, refactoring a stackless coroutine may require changing it to stackful.
     393A foundational property of all \emph{stateful functions} is that resume/suspend \emph{do not} cause incremental stack growth, \ie resume/suspend operations are remembered through the closure not the stack.
     394As well, activating a stateful function is \emph{asymmetric} or \emph{symmetric}, identified by resume/suspend (no cycles) and resume/resume (cycles).
     395A fixed closure activated by modified call/return is faster than a variable closure activated by context switching.
     396Additionally, any storage management for the closure (especially in unmanaged languages, \ie no garbage collection) must also be factored into design and performance.
     397Therefore, selecting between stackless and stackful semantics is a tradeoff between programming requirements and performance, where stackless is faster and stackful is more general.
     398Note, creation cost is amortized across usage, so activation cost is usually the dominant factor.
    605399
    606400\begin{figure}
    … …  
    616410
    617411
    618 
    619 
    620412        int fn = f->fn; f->fn = f->fn1;
    621413                f->fn1 = f->fn + fn;
    622414        return fn;
     415
    623416}
    624417int main() {
    … …  
    639432void `main(Fib & fib)` with(fib) {
    640433
    641 
    642434        [fn1, fn] = [1, 0];
    643435        for () {
    … …  
    659451\begin{cfa}[aboveskip=0pt,belowskip=0pt]
    660452typedef struct {
    661         int `restart`, fn1, fn;
     453        int fn1, fn;  void * `next`;
    662454} Fib;
    663 #define FibCtor { `0`, 1, 0 }
     455#define FibCtor { 1, 0, NULL }
    664456Fib * comain( Fib * f ) {
    665         `static void * states[] = {&&s0, &&s1};`
    666         `goto *states[f->restart];`
    667   s0: f->`restart` = 1;
     457        if ( f->next ) goto *f->next;
     458        f->next = &&s1;
    668459        for ( ;; ) {
    669460                return f;
    670461          s1:; int fn = f->fn + f->fn1;
    671                 f->fn1 = f->fn; f->fn = fn;
     462                        f->fn1 = f->fn; f->fn = fn;
    672463        }
    673464}
    … …  
    681472\end{lrbox}
    682473
    683 \subfloat[C]{\label{f:CFibonacci}\usebox\myboxA}
     474\subfloat[C asymmetric generator]{\label{f:CFibonacci}\usebox\myboxA}
    684475\hspace{3pt}
    685476\vrule
    686477\hspace{3pt}
    687 \subfloat[\CFA]{\label{f:CFAFibonacciGen}\usebox\myboxB}
     478\subfloat[\CFA asymmetric generator]{\label{f:CFAFibonacciGen}\usebox\myboxB}
    688479\hspace{3pt}
    689480\vrule
    690481\hspace{3pt}
    691 \subfloat[C generated code for \CFA version]{\label{f:CFibonacciSim}\usebox\myboxC}
     482\subfloat[C generator implementation]{\label{f:CFibonacciSim}\usebox\myboxC}
    692483\caption{Fibonacci (output) asymmetric generator}
    693484\label{f:FibonacciAsymmetricGenerator}
    … …  
    702493};
    703494void ?{}( Fmt & fmt ) { `resume(fmt);` } // constructor
    704 void ^?{}( Fmt & f ) with(f) { $\C[2.25in]{// destructor}$
     495void ^?{}( Fmt & f ) with(f) { $\C[1.75in]{// destructor}$
    705496        if ( g != 0 || b != 0 ) sout | nl; }
    706497void `main( Fmt & f )` with(f) {
    … …  
    708499                for ( ; g < 5; g += 1 ) { $\C{// groups}$
    709500                        for ( ; b < 4; b += 1 ) { $\C{// blocks}$
    710                                 do { `suspend;` $\C{// wait for character}$
    711                                 while ( ch == '\n' ); // ignore newline
    712                                 sout | ch;                      $\C{// print character}$
    713                         } sout | " ";  $\C{// block separator}$
    714                 } sout | nl; $\C{// group separator}$
     501                                `suspend;` $\C{// wait for character}$
     502                                while ( ch == '\n' ) `suspend;` // ignore
     503                                sout | ch;                                              // newline
     504                        } sout | " ";  // block spacer
     505                } sout | nl; // group newline
    715506        }
    716507}
    … …  
    730521\begin{cfa}[aboveskip=0pt,belowskip=0pt]
    731522typedef struct {
    732         int `restart`, g, b;
     523        void * next;
    733524        char ch;
     525        int g, b;
    734526} Fmt;
    735527void comain( Fmt * f ) {
    736         `static void * states[] = {&&s0, &&s1};`
    737         `goto *states[f->restart];`
    738   s0: f->`restart` = 1;
     528        if ( f->next ) goto *f->next;
     529        f->next = &&s1;
    739530        for ( ;; ) {
    740531                for ( f->g = 0; f->g < 5; f->g += 1 ) {
    741532                        for ( f->b = 0; f->b < 4; f->b += 1 ) {
    742                                 do { return;  s1: ;
    743                                 } while ( f->ch == '\n' );
     533                                return;
     534                          s1:;  while ( f->ch == '\n' ) return;
    744535                                printf( "%c", f->ch );
    745536                        } printf( " " );
    … …  
    748539}
    749540int main() {
    750         Fmt fmt = { `0` };  comain( &fmt ); // prime
     541        Fmt fmt = { NULL };  comain( &fmt ); // prime
    751542        for ( ;; ) {
    752543                scanf( "%c", &fmt.ch );
    … …  
    759550\end{lrbox}
    760551
    761 \subfloat[\CFA]{\label{f:CFAFormatGen}\usebox\myboxA}
    762 \hspace{35pt}
     552\subfloat[\CFA asymmetric generator]{\label{f:CFAFormatGen}\usebox\myboxA}
     553\hspace{3pt}
    763554\vrule
    764555\hspace{3pt}
    765 \subfloat[C generated code for \CFA version]{\label{f:CFormatGenImpl}\usebox\myboxB}
     556\subfloat[C generator simulation]{\label{f:CFormatSim}\usebox\myboxB}
    766557\hspace{3pt}
    767558\caption{Formatter (input) asymmetric generator}
    … …  
    769560\end{figure}
    770561
    771 Figure~\ref{f:FibonacciAsymmetricGenerator} shows an unbounded asymmetric generator for an infinite sequence of Fibonacci numbers written (left to right) in C, \CFA, and showing the underlying C implementation for the \CFA version.
     562Stateful functions appear as generators, coroutines, and threads, where presentations are based on function objects or pointers~\cite{Butenhof97, C++14, MS:VisualC++, BoostCoroutines15}.
     563For example, Python presents generators as a function object:
     564\begin{python}
     565def Gen():
     566        ... `yield val` ...
     567gen = Gen()
     568for i in range( 10 ):
     569        print( next( gen ) )
     570\end{python}
     571Boost presents coroutines in terms of four functor object-types:
     572\begin{cfa}
     573asymmetric_coroutine<>::pull_type
     574asymmetric_coroutine<>::push_type
     575symmetric_coroutine<>::call_type
     576symmetric_coroutine<>::yield_type
     577\end{cfa}
     578and many languages present threading using function pointers, @pthreads@~\cite{Butenhof97}, \Csharp~\cite{Csharp}, Go~\cite{Go}, and Scala~\cite{Scala}, \eg pthreads:
     579\begin{cfa}
     580void * rtn( void * arg ) { ... }
     581int i = 3, rc;
     582pthread_t t; $\C{// thread id}$
     583`rc = pthread_create( &t, rtn, (void *)i );` $\C{// create and initialized task, type-unsafe input parameter}$
     584\end{cfa}
     585% void mycor( pthread_t cid, void * arg ) {
     586%       int * value = (int *)arg;                               $\C{// type unsafe, pointer-size only}$
     587%       // thread body
     588% }
     589% int main() {
     590%       int input = 0, output;
     591%       coroutine_t cid = coroutine_create( &mycor, (void *)&input ); $\C{// type unsafe, pointer-size only}$
     592%       coroutine_resume( cid, (void *)input, (void **)&output ); $\C{// type unsafe, pointer-size only}$
     593% }
     594\CFA's preferred presentation model for generators/coroutines/threads is a hybrid of objects and functions, with an object-oriented flavour.
     595Essentially, the generator/coroutine/thread function is semantically coupled with a generator/coroutine/thread custom type.
     596The custom type solves several issues, while accessing the underlying mechanisms used by the custom types is still allowed.
     597
     598
     599\subsection{Generator}
     600
     601Stackless generators have the potential to be very small and fast, \ie as small and fast as function call/return for both creation and execution.
     602The \CFA goal is to achieve this performance target, possibly at the cost of some semantic complexity.
     603A series of different kinds of generators and their implementation demonstrate how this goal is accomplished.
     604
     605Figure~\ref{f:FibonacciAsymmetricGenerator} shows an unbounded asymmetric generator for an infinite sequence of Fibonacci numbers written in C and \CFA, with a simple C implementation for the \CFA version.
    772606This generator is an \emph{output generator}, producing a new result on each resumption.
    773607To compute Fibonacci, the previous two values in the sequence are retained to generate the next value, \ie @fn1@ and @fn@, plus the execution location where control restarts when the generator is resumed, \ie top or middle.
    … …  
    777611The C version only has the middle execution state because the top execution state is declaration initialization.
    778612Figure~\ref{f:CFAFibonacciGen} shows the \CFA approach, which also has a manual closure, but replaces the structure with a custom \CFA @generator@ type.
    779 Each generator type must have a function named \lstinline|main|,
    780 % \footnote{
    781 % The name \lstinline|main| has special meaning in C, specifically the function where a program starts execution.
    782 % Leveraging starting semantics to this name for generator/coroutine/thread is a logical extension.}
    783 called a \emph{generator main} (leveraging the starting semantics for program @main@ in C), which is connected to the generator type via its single reference parameter.
     613This generator type is then connected to a function that \emph{must be named \lstinline|main|},\footnote{
     614The name \lstinline|main| has special meaning in C, specifically the function where a program starts execution.
     615Hence, overloading this name for other starting points (generator/coroutine/thread) is a logical extension.}
     616called a \emph{generator main},which takes as its only parameter a reference to the generator type.
    784617The generator main contains @suspend@ statements that suspend execution without ending the generator versus @return@.
    785 For the Fibonacci generator-main,
     618For the Fibonacci generator-main,\footnote{
     619The \CFA \lstinline|with| opens an aggregate scope making its fields directly accessible, like Pascal \lstinline|with|, but using parallel semantics.
     620Multiple aggregates may be opened.}
    786621the top initialization state appears at the start and the middle execution state is denoted by statement @suspend@.
    787622Any local variables in @main@ \emph{are not retained} between calls;
    … …  
    792627Resuming an ended (returned) generator is undefined.
    793628Function @resume@ returns its argument generator so it can be cascaded in an expression, in this case to print the next Fibonacci value @fn@ computed in the generator instance.
    794 Figure~\ref{f:CFibonacciSim} shows the C implementation of the \CFA asymmetric generator.
    795 Only one execution-state field, @restart@, is needed to subscript the suspension points in the generator.
    796 At the start of the generator main, the @static@ declaration, @states@, is initialized to the N suspend points in the generator (where operator @&&@ dereferences/references a label~\cite{gccValueLabels}).
    797 Next, the computed @goto@ selects the last suspend point and branches to it.
    798 The  cost of setting @restart@ and branching via the computed @goto@ adds very little cost to the suspend/resume calls.
    799 
    800 An advantage of the \CFA explicit generator type is the ability to allow multiple type-safe interface functions taking and returning arbitrary types.
     629Figure~\ref{f:CFibonacciSim} shows the C implementation of the \CFA generator only needs one additional field, @next@, to handle retention of execution state.
     630The computed @goto@ at the start of the generator main, which branches after the previous suspend, adds very little cost to the resume call.
     631Finally, an explicit generator type provides both design and performance benefits, such as multiple type-safe interface functions taking and returning arbitrary types.\footnote{
     632The \CFA operator syntax uses \lstinline|?| to denote operands, which allows precise definitions for pre, post, and infix operators, \eg \lstinline|++?|, \lstinline|?++|, and \lstinline|?+?|, in addition \lstinline|?\{\}| denotes a constructor, as in \lstinline|foo `f` = `\{`...`\}`|, \lstinline|^?\{\}| denotes a destructor, and \lstinline|?()| is \CC function call \lstinline|operator()|.
     633}%
    801634\begin{cfa}
    802635int ?()( Fib & fib ) { return `resume( fib )`.fn; } $\C[3.9in]{// function-call interface}$
    803 int ?()( Fib & fib, int N ) { for ( N - 1 ) `fib()`; return `fib()`; } $\C{// add parameter to skip N values}$
    804 double ?()( Fib & fib ) { return (int)`fib()` / 3.14159; } $\C{// different return type, cast prevents recursive call}$
    805 Fib f;  int i;  double d;
    806 i = f();  i = f( 2 );  d = f();                                         $\C{// alternative interfaces}\CRT$
     636int ?()( Fib & fib, int N ) { for ( N - 1 ) `fib()`; return `fib()`; } $\C{// use function-call interface to skip N values}$
     637double ?()( Fib & fib ) { return (int)`fib()` / 3.14159; } $\C{// different return type, cast prevents recursive call}\CRT$
     638sout | (int)f1() | (double)f1() | f2( 2 ); // alternative interface, cast selects call based on return type, step 2 values
    807639\end{cfa}
    808640Now, the generator can be a separately compiled opaque-type only accessed through its interface functions.
    809641For contrast, Figure~\ref{f:PythonFibonacci} shows the equivalent Python Fibonacci generator, which does not use a generator type, and hence only has a single interface, but an implicit closure.
    810642
    811 \begin{figure}
    812 %\centering
    813 \newbox\myboxA
    814 \begin{lrbox}{\myboxA}
    815 \begin{python}[aboveskip=0pt,belowskip=0pt]
    816 def Fib():
    817         fn1, fn = 0, 1
    818         while True:
    819                 `yield fn1`
    820                 fn1, fn = fn, fn1 + fn
    821 f1 = Fib()
    822 f2 = Fib()
    823 for i in range( 10 ):
    824         print( next( f1 ), next( f2 ) )
    825 
    826 
    827 
    828 
    829 
    830 
    831 
    832 
    833 
    834 
    835 \end{python}
    836 \end{lrbox}
    837 
    838 \newbox\myboxB
    839 \begin{lrbox}{\myboxB}
    840 \begin{python}[aboveskip=0pt,belowskip=0pt]
    841 def Fmt():
    842         try:
    843                 while True:                                             $\C[2.5in]{\# until destructor call}$
    844                         for g in range( 5 ):            $\C{\# groups}$
    845                                 for b in range( 4 ):    $\C{\# blocks}$
    846                                         while True:
    847                                                 ch = (yield)    $\C{\# receive from send}$
    848                                                 if '\n' not in ch: $\C{\# ignore newline}$
    849                                                         break
    850                                         print( ch, end='' )     $\C{\# print character}$
    851                                 print( '  ', end='' )   $\C{\# block separator}$
    852                         print()                                         $\C{\# group separator}$
    853         except GeneratorExit:                           $\C{\# destructor}$
    854                 if g != 0 | b != 0:                             $\C{\# special case}$
    855                         print()
    856 fmt = Fmt()
    857 `next( fmt )`                                                   $\C{\# prime, next prewritten}$
    858 for i in range( 41 ):
    859         `fmt.send( 'a' );`                                      $\C{\# send to yield}$
    860 \end{python}
    861 \end{lrbox}
    862 
    863 \hspace{30pt}
    864 \subfloat[Fibonacci]{\label{f:PythonFibonacci}\usebox\myboxA}
    865 \hspace{3pt}
    866 \vrule
    867 \hspace{3pt}
    868 \subfloat[Formatter]{\label{f:PythonFormatter}\usebox\myboxB}
    869 \caption{Python generator}
    870 \label{f:PythonGenerator}
    871 \end{figure}
    872 
    873 Having to manually create the generator closure by moving local-state variables into the generator type is an additional programmer burden (removed by the coroutine in Section~\ref{s:Coroutine}).
    874 This manual requirement follows from the generality of allowing variable-size local-state, \eg local state with a variable-length array requires dynamic allocation as the array size is unknown at compile time.
     643Having to manually create the generator closure by moving local-state variables into the generator type is an additional programmer burden.
     644(This restriction is removed by the coroutine in Section~\ref{s:Coroutine}.)
     645This requirement follows from the generality of variable-size local-state, \eg local state with a variable-length array requires dynamic allocation because the array size is unknown at compile time.
    875646However, dynamic allocation significantly increases the cost of generator creation/destruction and is a showstopper for embedded real-time programming.
    876647But more importantly, the size of the generator type is tied to the local state in the generator main, which precludes separate compilation of the generator main, \ie a generator must be inlined or local state must be dynamically allocated.
    877 With respect to safety, we believe static analysis can discriminate persistent generator state from temporary generator-main state and raise a compile-time error for temporary usage spanning suspend points.
    878 Our experience using generators is that the problems have simple data state, including local state, but complex execution state, so the burden of creating the generator type is small.
     648With respect to safety, we believe static analysis can discriminate local state from temporary variables in a generator, \ie variable usage spanning @suspend@, and generate a compile-time error.
     649Finally, our current experience is that most generator problems have simple data state, including local state, but complex execution state, so the burden of creating the generator type is small.
    879650As well, C programmers are not afraid of this kind of semantic programming requirement, if it results in very small, fast generators.
    880651
    … …  
    898669The example takes advantage of resuming a generator in the constructor to prime the loops so the first character sent for formatting appears inside the nested loops.
    899670The destructor provides a newline, if formatted text ends with a full line.
    900 Figure~\ref{f:CFormatGenImpl} shows the C implementation of the \CFA input generator with one additional field and the computed @goto@.
    901 For contrast, Figure~\ref{f:PythonFormatter} shows the equivalent Python format generator with the same properties as the format generator.
    902 
    903 % https://dl-acm-org.proxy.lib.uwaterloo.ca/
    904 
    905 Figure~\ref{f:DeviceDriverGen} shows an important application for an asymmetric generator, a device-driver, because device drivers are a significant source of operating-system errors: 85\% in Windows XP~\cite[p.~78]{Swift05} and 51.6\% in Linux~\cite[p.~1358,]{Xiao19}. %\cite{Palix11}
    906 Swift \etal~\cite[p.~86]{Swift05} restructure device drivers using the Extension Procedure Call (XPC) within the kernel via functions @nooks_driver_call@ and @nooks_kernel_call@, which have coroutine properties context switching to separate stacks with explicit hand-off calls;
    907 however, the calls do not retain execution state, and hence always start from the top.
    908 The alternative approach for implementing device drivers is using stack-ripping.
    909 However, Adya \etal~\cite{Adya02} argue against stack ripping in Section 3.2 and suggest a hybrid approach in Section 4 using cooperatively scheduled \emph{fibers}, which is coroutining.
    910 
    911 As an example, the following protocol:
     671Figure~\ref{f:CFormatSim} shows the C implementation of the \CFA input generator with one additional field and the computed @goto@.
     672For contrast, Figure~\ref{f:PythonFormatter} shows the equivalent Python format generator with the same properties as the Fibonacci generator.
     673
     674Figure~\ref{f:DeviceDriverGen} shows a \emph{killer} asymmetric generator, a device-driver, because device drivers caused 70\%-85\% of failures in Windows/Linux~\cite{Swift05}.
     675Device drives follow the pattern of simple data state but complex execution state, \ie finite state-machine (FSM) parsing a protocol.
     676For example, the following protocol:
    912677\begin{center}
    913678\ldots\, STX \ldots\, message \ldots\, ESC ETX \ldots\, message \ldots\, ETX 2-byte crc \ldots
    914679\end{center}
    915 is for a simple network message beginning with the control character STX, ending with an ETX, and followed by a 2-byte cyclic-redundancy check.
     680is a network message beginning with the control character STX, ending with an ETX, and followed by a 2-byte cyclic-redundancy check.
    916681Control characters may appear in a message if preceded by an ESC.
    917682When a message byte arrives, it triggers an interrupt, and the operating system services the interrupt by calling the device driver with the byte read from a hardware register.
    918 The device driver returns a status code of its current state, and when a complete message is obtained, the operating system read the message accumulated in the supplied buffer.
    919 Hence, the device driver is an input/output generator, where the cost of resuming the device-driver generator is the same as call/return, so performance in an operating-system kernel is excellent.
    920 The key benefits of using a generator are correctness, safety, and maintenance because the execution states are transcribed directly into the programming language rather than table lookup or stack ripping.
    921 The conclusion is that FSMs are complex and occur in important domains, so direct generator support is important in a system programming language.
     683The device driver returns a status code of its current state, and when a complete message is obtained, the operating system knows the message is in the message buffer.
     684Hence, the device driver is an input/output generator.
     685
     686Note, the cost of creating and resuming the device-driver generator, @Driver@, is virtually identical to call/return, so performance in an operating-system kernel is excellent.
     687As well, the data state is small, where variables @byte@ and @msg@ are communication variables for passing in message bytes and returning the message, and variables @lnth@, @crc@, and @sum@ are local variable that must be retained between calls and are manually hoisted into the generator type.
     688% Manually, detecting and hoisting local-state variables is easy when the number is small.
     689In contrast, the execution state is large, with one @resume@ and seven @suspend@s.
     690Hence, the key benefits of the generator are correctness, safety, and maintenance because the execution states are transcribed directly into the programming language rather than using a table-driven approach.
     691Because FSMs can be complex and frequently occur in important domains, direct generator support is important in a system programming language.
    922692
    923693\begin{figure}
    924694\centering
     695\newbox\myboxA
     696\begin{lrbox}{\myboxA}
     697\begin{python}[aboveskip=0pt,belowskip=0pt]
     698def Fib():
     699        fn1, fn = 0, 1
     700        while True:
     701                `yield fn1`
     702                fn1, fn = fn, fn1 + fn
     703f1 = Fib()
     704f2 = Fib()
     705for i in range( 10 ):
     706        print( next( f1 ), next( f2 ) )
     707
     708
     709
     710
     711
     712
     713\end{python}
     714\end{lrbox}
     715
     716\newbox\myboxB
     717\begin{lrbox}{\myboxB}
     718\begin{python}[aboveskip=0pt,belowskip=0pt]
     719def Fmt():
     720        try:
     721                while True:
     722                        for g in range( 5 ):
     723                                for b in range( 4 ):
     724                                        print( `(yield)`, end='' )
     725                                print( '  ', end='' )
     726                        print()
     727        except GeneratorExit:
     728                if g != 0 | b != 0:
     729                        print()
     730fmt = Fmt()
     731`next( fmt )`                    # prime, next prewritten
     732for i in range( 41 ):
     733        `fmt.send( 'a' );`      # send to yield
     734\end{python}
     735\end{lrbox}
     736\subfloat[Fibonacci]{\label{f:PythonFibonacci}\usebox\myboxA}
     737\hspace{3pt}
     738\vrule
     739\hspace{3pt}
     740\subfloat[Formatter]{\label{f:PythonFormatter}\usebox\myboxB}
     741\caption{Python generator}
     742\label{f:PythonGenerator}
     743
     744\bigskip
     745
    925746\begin{tabular}{@{}l|l@{}}
    926747\begin{cfa}[aboveskip=0pt,belowskip=0pt]
    … …  
    929750`generator` Driver {
    930751        Status status;
    931         char byte, * msg; // communication
    932         int lnth, sum;      // local state
    933         short int crc;
     752        unsigned char byte, * msg; // communication
     753        unsigned int lnth, sum;      // local state
     754        unsigned short int crc;
    934755};
    935756void ?{}( Driver & d, char * m ) { d.msg = m; }
    … …  
    979800(The trivial cycle is a generator resuming itself.)
    980801This control flow is similar to recursion for functions but without stack growth.
    981 Figure~\ref{f:PingPongFullCoroutineSteps} shows the steps for symmetric control-flow are creating, executing, and terminating the cycle.
     802The steps for symmetric control-flow are creating, executing, and terminating the cycle.
    982803Constructing the cycle must deal with definition-before-use to close the cycle, \ie, the first generator must know about the last generator, which is not within scope.
    983804(This issue occurs for any cyclic data structure.)
    984 The example creates the generators, @ping@/@pong@, and then assigns the partners that form the cycle.
    985 % (Alternatively, the constructor can assign the partners as they are declared, except the first, and the first-generator partner is set after the last generator declaration to close the cycle.)
    986 Once the cycle is formed, the program main resumes one of the generators, @ping@, and the generators can then traverse an arbitrary cycle using @resume@ to activate partner generator(s).
     805% The example creates all the generators and then assigns the partners that form the cycle.
     806% Alternatively, the constructor can assign the partners as they are declared, except the first, and the first-generator partner is set after the last generator declaration to close the cycle.
     807Once the cycle is formed, the program main resumes one of the generators, and the generators can then traverse an arbitrary cycle using @resume@ to activate partner generator(s).
    987808Terminating the cycle is accomplished by @suspend@ or @return@, both of which go back to the stack frame that started the cycle (program main in the example).
    988 Note, the creator and starter may be different, \eg if the creator calls another function that starts the cycle.
    989809The starting stack-frame is below the last active generator because the resume/resume cycle does not grow the stack.
    990 Also, since local variables are not retained in the generator function, there are no objects with destructors to be called, so the cost is the same as a function return.
    991 Destructor cost occurs when the generator instance is deallocated by the creator.
     810Also, since local variables are not retained in the generator function, it does not contain any objects with destructors that must be called, so the  cost is the same as a function return.
     811Destructor cost occurs when the generator instance is deallocated, which is easily controlled by the programmer.
     812
     813Figure~\ref{f:CPingPongSim} shows the implementation of the symmetric generator, where the complexity is the @resume@, which needs an extension to the calling convention to perform a forward rather than backward jump.
     814This jump-starts at the top of the next generator main to re-execute the normal calling convention to make space on the stack for its local variables.
     815However, before the jump, the caller must reset its stack (and any registers) equivalent to a @return@, but subsequently jump forward.
     816This semantics is basically a tail-call optimization, which compilers already perform.
     817The example shows the assembly code to undo the generator's entry code before the direct jump.
     818This assembly code depends on what entry code is generated, specifically if there are local variables and the level of optimization.
     819To provide this new calling convention requires a mechanism built into the compiler, which is beyond the scope of \CFA at this time.
     820Nevertheless, it is possible to hand generate any symmetric generators for proof of concept and performance testing.
     821A compiler could also eliminate other artifacts in the generator simulation to further increase performance, \eg LLVM has various coroutine support~\cite{CoroutineTS}, and \CFA can leverage this support should it fork @clang@.
    992822
    993823\begin{figure}
    … …  
    996826\begin{cfa}[aboveskip=0pt,belowskip=0pt]
    997827`generator PingPong` {
    998         int N, i;                               // local state
    999828        const char * name;
     829        int N;
     830        int i;                          // local state
    1000831        PingPong & partner; // rebindable reference
    1001832};
    1002833
    1003834void `main( PingPong & pp )` with(pp) {
    1004 
    1005 
    1006835        for ( ; i < N; i += 1 ) {
    1007836                sout | name | i;
    … …  
    1021850\begin{cfa}[escapechar={},aboveskip=0pt,belowskip=0pt]
    1022851typedef struct PingPong {
    1023         int restart, N, i;
    1024852        const char * name;
     853        int N, i;
    1025854        struct PingPong * partner;
     855        void * next;
    1026856} PingPong;
    1027 #define PPCtor(name, N) {0, N, 0, name, NULL}
     857#define PPCtor(name, N) {name,N,0,NULL,NULL}
    1028858void comain( PingPong * pp ) {
    1029         static void * states[] = {&&s0, &&s1};
    1030         goto *states[pp->restart];
    1031   s0: pp->restart = 1;
     859        if ( pp->next ) goto *pp->next;
     860        pp->next = &&cycle;
    1032861        for ( ; pp->i < pp->N; pp->i += 1 ) {
    1033862                printf( "%s %d\n", pp->name, pp->i );
    1034863                asm( "mov  %0,%%rdi" : "=m" (pp->partner) );
    1035864                asm( "mov  %rdi,%rax" );
    1036                 asm( "add  $16, %rsp" );
    1037                 asm( "popq %rbp" );
     865                asm( "popq %rbx" );
    1038866                asm( "jmp  comain" );
    1039           s1: ;
     867          cycle: ;
    1040868        }
    1041869}
    … …  
    1053881\end{figure}
    1054882
    1055 \begin{figure}
    1056 \centering
    1057 \input{FullCoroutinePhases.pstex_t}
    1058 \vspace*{-10pt}
    1059 \caption{Symmetric coroutine steps: Ping / Pong}
    1060 \label{f:PingPongFullCoroutineSteps}
    1061 \end{figure}
    1062 
    1063 Figure~\ref{f:CPingPongSim} shows the C implementation of the \CFA symmetric generator, where there is still only one additional field, @restart@, but @resume@ is more complex because it does a forward rather than backward jump.
    1064 Before the jump, the parameter for the next call @partner@ is placed into the register used for the first parameter, @rdi@, and the remaining registers are reset for a return.
    1065 The @jmp comain@ restarts the function but with a different parameter, so the new call's behaviour depends on the state of the coroutine type, i.e., branch to restart location with different data state.
    1066 While the semantics of call forward is a tail-call optimization, which compilers perform, the generator state is different on each call rather a common state for a tail-recursive function (i.e., the parameter to the function never changes during the forward calls.
    1067 However, this assembler code depends on what entry code is generated, specifically if there are local variables and the level of optimization.
    1068 Hence, internal compiler support is necessary for any forward call (or backwards return), \eg LLVM has various coroutine support~\cite{CoroutineTS}, and \CFA can leverage this support should it eventually fork @clang@.
    1069 For this reason, \CFA does not support general symmetric generators at this time, but, it is possible to hand generate any symmetric generators (as in Figure~\ref{f:CPingPongSim}) for proof of concept and performance testing.
    1070 
    1071 Finally, part of this generator work was inspired by the recent \CCtwenty coroutine proposal~\cite{C++20Coroutine19}, which uses the general term coroutine to mean generator.
     883Finally, part of this generator work was inspired by the recent \CCtwenty generator proposal~\cite{C++20Coroutine19} (which they call coroutines).
    1072884Our work provides the same high-performance asymmetric generators as \CCtwenty, and extends their work with symmetric generators.
    1073885An additional \CCtwenty generator feature allows @suspend@ and @resume@ to be followed by a restricted compound statement that is executed after the current generator has reset its stack but before calling the next generator, specified with \CFA syntax:
    … …  
    1084896\label{s:Coroutine}
    1085897
    1086 Stackful coroutines (Table~\ref{t:ExecutionPropertyComposition} case 5) extend generator semantics, \ie there is an implicit closure and @suspend@ may appear in a helper function called from the coroutine main.
     898Stackful coroutines extend generator semantics, \ie there is an implicit closure and @suspend@ may appear in a helper function called from the coroutine main.
    1087899A coroutine is specified by replacing @generator@ with @coroutine@ for the type.
    1088 Coroutine generality results in higher cost for creation, due to dynamic stack allocation, for execution, due to context switching among stacks, and for terminating, due to possible stack unwinding and dynamic stack deallocation.
     900Coroutine generality results in higher cost for creation, due to dynamic stack allocation, execution, due to context switching among stacks, and terminating, due to possible stack unwinding and dynamic stack deallocation.
    1089901A series of different kinds of coroutines and their implementations demonstrate how coroutines extend generators.
    1090902
    1091903First, the previous generator examples are converted to their coroutine counterparts, allowing local-state variables to be moved from the generator type into the coroutine main.
    1092 \begin{center}
    1093 \begin{tabular}{@{}l|l|l|l@{}}
    1094 \multicolumn{1}{c|}{Fibonacci} & \multicolumn{1}{c|}{Formatter} & \multicolumn{1}{c|}{Device Driver} & \multicolumn{1}{c}{PingPong} \\
    1095 \hline
     904\begin{description}
     905\item[Fibonacci]
     906Move the declaration of @fn1@ to the start of coroutine main.
    1096907\begin{cfa}[xleftmargin=0pt]
    1097 void main( Fib & fib ) ...
     908void main( Fib & fib ) with(fib) {
    1098909        `int fn1;`
    1099 
    1100 
    1101 \end{cfa}
    1102 &
     910\end{cfa}
     911\item[Formatter]
     912Move the declaration of @g@ and @b@ to the for loops in the coroutine main.
    1103913\begin{cfa}[xleftmargin=0pt]
    1104914for ( `g`; 5 ) {
    1105915        for ( `b`; 4 ) {
    1106 
    1107 
    1108 \end{cfa}
    1109 &
     916\end{cfa}
     917\item[Device Driver]
     918Move the declaration of @lnth@ and @sum@ to their points of initialization.
    1110919\begin{cfa}[xleftmargin=0pt]
    1111 status = CONT;
    1112 `int lnth = 0, sum = 0;`
    1113 ...
    1114 `short int crc = byte << 8;`
    1115 \end{cfa}
    1116 &
     920        status = CONT;
     921        `unsigned int lnth = 0, sum = 0;`
     922        ...
     923        `unsigned short int crc = byte << 8;`
     924\end{cfa}
     925\item[PingPong]
     926Move the declaration of @i@ to the for loop in the coroutine main.
    1117927\begin{cfa}[xleftmargin=0pt]
    1118 void main( PingPong & pp ) ...
     928void main( PingPong & pp ) with(pp) {
    1119929        for ( `i`; N ) {
    1120 
    1121 
    1122 \end{cfa}
    1123 \end{tabular}
    1124 \end{center}
     930\end{cfa}
     931\end{description}
    1125932It is also possible to refactor code containing local-state and @suspend@ statements into a helper function, like the computation of the CRC for the device driver.
    1126933\begin{cfa}
    1127 int Crc() {
     934unsigned int Crc() {
    1128935        `suspend;`
    1129         short int crc = byte << 8;
     936        unsigned short int crc = byte << 8;
    1130937        `suspend;`
    1131938        status = (crc | byte) == sum ? MSG : ECRC;
    … …  
    1138945
    1139946\begin{comment}
    1140 Figure~\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 functions, \eg @restart@.
     947Figure~\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 functions, \eg @next@.
    1141948Like 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.
    1142949The 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@.
    1143 The interface function @restart@, takes a Fibonacci instance and context switches to it using @resume@;
     950The interface function @next@, takes a Fibonacci instance and context switches to it using @resume@;
    1144951on restart, the Fibonacci field, @fn@, contains the next value in the sequence, which is returned.
    1145952The first @resume@ is special because it allocates the coroutine stack and cocalls its coroutine main on that stack;
    … …  
    13071114\begin{figure}
    13081115\centering
     1116\lstset{language=CFA,escapechar={},moredelim=**[is][\protect\color{red}]{`}{`}}% allow $
    13091117\begin{tabular}{@{}l@{\hspace{2\parindentlnth}}l@{}}
    13101118\begin{cfa}
    13111119`coroutine` Prod {
    1312         Cons & c;                       $\C[1.5in]{// communication}$
     1120        Cons & c;                       // communication
    13131121        int N, money, receipt;
    13141122};
    13151123void main( Prod & prod ) with( prod ) {
    1316         for ( i; N ) {          $\C{// 1st resume}\CRT$
     1124        // 1st resume starts here
     1125        for ( i; N ) {
    13171126                int p1 = random( 100 ), p2 = random( 100 );
     1127                sout | p1 | " " | p2;
    13181128                int status = delivery( c, p1, p2 );
     1129                sout | " $" | money | nl | status;
    13191130                receipt += 1;
    13201131        }
    13211132        stop( c );
     1133        sout | "prod stops";
    13221134}
    13231135int payment( Prod & prod, int money ) {
    … …  
    13401152\begin{cfa}
    13411153`coroutine` Cons {
    1342         Prod & p;                       $\C[1.5in]{// communication}$
     1154        Prod & p;                       // communication
    13431155        int p1, p2, status;
    13441156        bool done;
    13451157};
    13461158void ?{}( Cons & cons, Prod & p ) {
    1347         &cons.p = &p;           $\C{// reassignable reference}$
     1159        &cons.p = &p; // reassignable reference
    13481160        cons.[status, done ] = [0, false];
    13491161}
    13501162void main( Cons & cons ) with( cons ) {
    1351         int money = 1, receipt; $\C{// 1st resume}\CRT$
     1163        // 1st resume starts here
     1164        int money = 1, receipt;
    13521165        for ( ; ! done; ) {
     1166                sout | p1 | " " | p2 | nl | " $" | money;
    13531167                status += 1;
    13541168                receipt = payment( p, money );
     1169                sout | " #" | receipt;
    13551170                money += 1;
    13561171        }
     1172        sout | "cons stops";
    13571173}
    13581174int delivery( Cons & cons, int p1, int p2 ) {
    … …  
    13751191This example is illustrative because both producer/consumer have two interface functions with @resume@s that suspend execution in these interface (helper) functions.
    13761192The program main creates the producer coroutine, passes it to the consumer coroutine in its initialization, and closes the cycle at the call to @start@ along with the number of items to be produced.
    1377 The call to @start@ is the first @resume@ of @prod@, which remembers the program main as the starter and creates @prod@'s stack with a frame for @prod@'s coroutine main at the top, and context switches to it.
    1378 @prod@'s coroutine main starts, creates local-state variables that are retained between coroutine activations, and executes $N$ iterations, each generating two random values, calling the consumer's @deliver@ function to transfer the values, and printing the status returned from the consumer.
     1193The first @resume@ of @prod@ creates @prod@'s stack with a frame for @prod@'s coroutine main at the top, and context switches to it.
     1194@prod@'s coroutine main starts, creates local-state 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.
     1195
    13791196The producer call to @delivery@ transfers values into the consumer's communication variables, resumes the consumer, and returns the consumer status.
    1380 Similarly on the first resume, @cons@'s stack is created and initialized, holding local-state variables retained between subsequent activations of the coroutine.
    1381 The symmetric coroutine cycle forms when the consumer calls the producer's @payment@ function, which resumes the producer in the consumer's delivery function.
    1382 When the producer calls @delivery@ again, it resumes the consumer in the @payment@ function.
    1383 Both interface function than return to the their corresponding coroutine-main functions for the next cycle.
     1197On the first resume, @cons@'s stack is created and initialized, holding local-state variables retained between subsequent activations of the coroutine.
     1198The 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).
     1199The call from the consumer to @payment@ introduces the cycle between producer and consumer.
     1200When @payment@ is called, the consumer copies values into the producer's communication variable and a resume is executed.
     1201The context switch restarts the producer at the point where it last context switched, so it continues in @delivery@ after the resume.
     1202@delivery@ returns the status value in @prod@'s coroutine main, where the status is printed.
     1203The loop then repeats calling @delivery@, where each call resumes the consumer coroutine.
     1204The context switch to the consumer continues in @payment@.
     1205The consumer increments and returns the receipt to the call in @cons@'s coroutine main.
     1206The loop then repeats calling @payment@, where each call resumes the producer coroutine.
    13841207Figure~\ref{f:ProdConsRuntimeStacks} shows the runtime stacks of the program main, and the coroutine mains for @prod@ and @cons@ during the cycling.
    1385 As a consequence of a coroutine retaining its last resumer for suspending back, these reverse pointers allow @suspend@ to cycle \emph{backwards} around a symmetric coroutine cycle.
    13861208
    13871209\begin{figure}
    … …  
    13921214\caption{Producer / consumer runtime stacks}
    13931215\label{f:ProdConsRuntimeStacks}
     1216
     1217\medskip
     1218
     1219\begin{center}
     1220\input{FullCoroutinePhases.pstex_t}
     1221\end{center}
     1222\vspace*{-10pt}
     1223\caption{Ping / Pong coroutine steps}
     1224\label{f:PingPongFullCoroutineSteps}
    13941225\end{figure}
    13951226
    13961227Terminating a coroutine cycle is more complex than a generator cycle, because it requires context switching to the program main's \emph{stack} to shutdown the program, whereas generators started by the program main run on its stack.
    1397 Furthermore, each deallocated coroutine must execute all destructors for object allocated in the coroutine type \emph{and} allocated on the coroutine's stack at the point of suspension, which can be arbitrarily deep.
    1398 In the example, termination begins with the producer's loop stopping after N iterations and calling the consumer's @stop@ function, which sets the @done@ flag, resumes the consumer in function @payment@, terminating the call, and the consumer's loop in its coroutine main.
    1399 % (Not shown is having @prod@ raise a nonlocal @stop@ exception at @cons@ after it finishes generating values and suspend back to @cons@, which catches the @stop@ exception to terminate its loop.)
    1400 When the consumer's main ends, its stack is already unwound so any stack allocated objects with destructors are finalized.
    1401 The question now is where does control continue?
    1402 
     1228Furthermore, each deallocated coroutine must guarantee all destructors are run for object allocated in the coroutine type \emph{and} allocated on the coroutine's stack at the point of suspension, which can be arbitrarily deep.
     1229When a coroutine's main ends, its stack is already unwound so any stack allocated objects with destructors have been finalized.
    14031230The na\"{i}ve semantics for coroutine-cycle termination is to context switch to the last resumer, like executing a @suspend@/@return@ in a generator.
    14041231However, for coroutines, the last resumer is \emph{not} implicitly below the current stack frame, as for generators, because each coroutine's stack is independent.
    14051232Unfortunately, it is impossible to determine statically if a coroutine is in a cycle and unrealistic to check dynamically (graph-cycle problem).
    14061233Hence, a compromise solution is necessary that works for asymmetric (acyclic) and symmetric (cyclic) coroutines.
    1407 Our solution is to retain a coroutine's starter (first resumer), and context switch back to the starter when the coroutine ends.
    1408 Hence, the consumer restarts its first resumer, @prod@, in @stop@, and when the producer ends, it restarts its first resumer, program main, in @start@ (see dashed lines from the end of the coroutine mains in Figure~\ref{f:ProdConsRuntimeStacks}).
     1234
     1235Our solution is to context switch back to the first resumer (starter) once the coroutine ends.
    14091236This semantics works well for the most common asymmetric and symmetric coroutine usage patterns.
    1410 For asymmetric coroutines, it is common for the first resumer (starter) coroutine to be the only resumer;
    1411 for symmetric coroutines, it is common for the cycle creator to persist for the lifetime of the cycle.
     1237For asymmetric coroutines, it is common for the first resumer (starter) coroutine to be the only resumer.
     1238All previous generators converted to coroutines have this property.
     1239For symmetric coroutines, it is common for the cycle creator to persist for the lifetime of the cycle.
     1240Hence, the starter coroutine is remembered on the first resume and ending the coroutine resumes the starter.
     1241Figure~\ref{f:ProdConsRuntimeStacks} shows this semantic by the dashed lines from the end of the coroutine mains: @prod@ starts @cons@ so @cons@ resumes @prod@ at the end, and the program main starts @prod@ so @prod@ resumes the program main at the end.
    14121242For other scenarios, it is always possible to devise a solution with additional programming effort, such as forcing the cycle forward (backward) to a safe point before starting termination.
    14131243
    1414 Note, the producer/consumer example does not illustrate the full power of the starter semantics because @cons@ always ends first.
    1415 Assume generator @PingPong@ in Figure~\ref{f:PingPongSymmetricGenerator} is converted to a coroutine.
    1416 Unlike generators, coroutines have a starter structure with multiple levels, where the program main starts @ping@ and @ping@ starts @pong@.
    1417 By adjusting $N$ for either @ping@/@pong@, it is possible to have either finish first.
    1418 If @pong@ ends first, it resumes its starter @ping@ in its coroutine main, then @ping@ ends and resumes its starter the program main on return;
    1419 if @ping@ ends first, it resumes its starter the program main on return.
    1420 Regardless of the cycle complexity, the starter structure always leads back to the program main, but the path can be entered at an arbitrary point.
    1421 Once back at the program main (creator), coroutines @ping@ and @pong@ are deallocated, runnning any destructors for objects within the coroutine and possibly deallocating any coroutine stacks for non-terminated coroutines, where stack deallocation implies stack unwinding to find destructors for allocated objects on the stack.
    1422 Hence, the \CFA termination semantics for the generator and coroutine ensure correct deallocation semnatics, regardless of the coroutine's state (terminated or active), like any other aggregate object.
     1244The producer/consumer example does not illustrate the full power of the starter semantics because @cons@ always ends first.
     1245Assume generator @PingPong@ is converted to a coroutine.
     1246Figure~\ref{f:PingPongFullCoroutineSteps} shows the creation, starter, and cyclic execution steps of the coroutine version.
     1247The program main creates (declares) coroutine instances @ping@ and @pong@.
     1248Next, program main resumes @ping@, making it @ping@'s starter, and @ping@'s main resumes @pong@'s main, making it @pong@'s starter.
     1249Execution forms a cycle when @pong@ resumes @ping@, and cycles $N$ times.
     1250By adjusting $N$ for either @ping@/@pong@, it is possible to have either one finish first, instead of @pong@ always ending first.
     1251If @pong@ ends first, it resumes its starter @ping@ in its coroutine main, then @ping@ ends and resumes its starter the program main in function @start@.
     1252If @ping@ ends first, it resumes its starter the program main in function @start@.
     1253Regardless of the cycle complexity, the starter stack always leads back to the program main, but the stack can be entered at an arbitrary point.
     1254Once back at the program main, coroutines @ping@ and @pong@ are deallocated.
     1255For generators, deallocation runs the destructors for all objects in the generator type.
     1256For coroutines, deallocation deals with objects in the coroutine type and must also run the destructors for any objects pending on the coroutine's stack for any unterminated coroutine.
     1257Hence, if a coroutine's destructor detects the coroutine is not ended, it implicitly raises a cancellation exception (uncatchable exception) at the coroutine and resumes it so the cancellation exception can propagate to the root of the coroutine's stack destroying all local variable on the stack.
     1258So the \CFA semantics for the generator and coroutine, ensure both can be safely deallocated at any time, regardless of their current state, like any other aggregate object.
     1259Explicitly raising normal exceptions at another coroutine can replace flag variables, like @stop@, \eg @prod@ raises a @stop@ exception at @cons@ after it finishes generating values and resumes @cons@, which catches the @stop@ exception to terminate its loop.
     1260
     1261Finally, there is an interesting effect for @suspend@ with symmetric coroutines.
     1262A coroutine must retain its last resumer to suspend back because the resumer is on a different stack.
     1263These reverse pointers allow @suspend@ to cycle \emph{backwards}, which may be useful in certain cases.
     1264However, there is an anomaly if a coroutine resumes itself, because it overwrites its last resumer with itself, losing the ability to resume the last external resumer.
     1265To prevent losing this information, a self-resume does not overwrite the last resumer.
    14231266
    14241267
    … …  
    14511294Users wanting to extend custom types or build their own can only do so in ways offered by the language.
    14521295Furthermore, implementing custom types without language support may display the power of a programming language.
    1453 \CFA blends the two approaches, providing custom type for idiomatic \CFA code, while extending and building new custom types is still possible, similar to Java concurrency with builtin and library (@java.util.concurrent@) monitors.
     1296\CFA blends the two approaches, providing custom type for idiomatic \CFA code, while extending and building new custom types is still possible, similar to Java concurrency with builtin and library.
    14541297
    14551298Part of the mechanism to generalize custom types is the \CFA trait~\cite[\S~2.3]{Moss18}, \eg the definition for custom-type @coroutine@ is anything satisfying the trait @is_coroutine@, and this trait both enforces and restricts the coroutine-interface functions.
    … …  
    14611304forall( `dtype` T | is_coroutine(T) ) void $suspend$( T & ), resume( T & );
    14621305\end{cfa}
    1463 Note, copying generators/coroutines/threads is undefined because muliple objects cannot execute on a shared stack and stack copying does not work in unmanaged languages (no garbage collection), like C, because the stack may contain pointers to objects within it that require updating for the copy.
     1306Note, copying generators/coroutines/threads is not meaningful.
     1307For example, both the resumer and suspender descriptors can have bidirectional pointers;
     1308copying these coroutines does not update the internal pointers so behaviour of both copies would be difficult to understand.
     1309Furthermore, two coroutines cannot logically execute on the same stack.
     1310A deep coroutine copy, which copies the stack, is also meaningless in an unmanaged language (no garbage collection), like C, because the stack may contain pointers to object within it that require updating for the copy.
    14641311The \CFA @dtype@ property provides no \emph{implicit} copying operations and the @is_coroutine@ trait provides no \emph{explicit} copying operations, so all coroutines must be passed by reference (pointer).
    14651312The function definitions ensure there is a statically typed @main@ function that is the starting point (first stack frame) of a coroutine, and a mechanism to get (read) the coroutine descriptor from its handle.
    … …  
    15051352The combination of custom types and fundamental @trait@ description of these types allows a concise specification for programmers and tools, while more advanced programmers can have tighter control over memory layout and initialization.
    15061353
    1507 Figure~\ref{f:CoroutineMemoryLayout} shows different memory-layout options for a coroutine (where a thread is similar).
     1354Figure~\ref{f:CoroutineMemoryLayout} shows different memory-layout options for a coroutine (where a task is similar).
    15081355The coroutine handle is the @coroutine@ instance containing programmer specified type global/communication variables across interface functions.
    15091356The coroutine descriptor contains all implicit declarations needed by the runtime, \eg @suspend@/@resume@, and can be part of the coroutine handle or separate.
    15101357The coroutine stack can appear in a number of locations and be fixed or variable sized.
    1511 Hence, the coroutine's stack could be a variable-length structure (VLS)\footnote{
    1512 We are examining VLSs, where fields can be variable-sized structures or arrays.
     1358Hence, the coroutine's stack could be a VLS\footnote{
     1359We are examining variable-sized structures (VLS), where fields can be variable-sized structures or arrays.
    15131360Once allocated, a VLS is fixed sized.}
    15141361on the allocating stack, provided the allocating stack is large enough.
    15151362For a VLS stack allocation/deallocation is an inexpensive adjustment of the stack pointer, modulo any stack constructor costs (\eg initial frame setup).
    1516 For stack allocation in the heap, allocation/deallocation is an expensive allocation, where the heap can be a shared resource, modulo any stack constructor costs.
    1517 It is also possible to use a split (segmented) stack calling convention, available with gcc and clang, allowing a variable-sized stack via a set of connected blocks in the heap.
     1363For heap stack allocation, allocation/deallocation is an expensive heap allocation (where the heap can be a shared resource), modulo any stack constructor costs.
     1364With heap stack allocation, it is also possible to use a split (segmented) stack calling convention, available with gcc and clang, so the stack is variable sized.
    15181365Currently, \CFA supports stack/heap allocated descriptors but only fixed-sized heap allocated stacks.
    15191366In \CFA debug-mode, the fixed-sized stack is terminated with a write-only page, which catches most stack overflows.
    15201367Experience teaching concurrency with \uC~\cite{CS343} shows fixed-sized stacks are rarely an issue for students.
    1521 Split-stack allocation is under development but requires recompilation of legacy code, which is not always possible.
     1368Split-stack allocation is under development but requires recompilation of legacy code, which may be impossible.
    15221369
    15231370\begin{figure}
    … …  
    15331380
    15341381Concurrency is nondeterministic scheduling of independent sequential execution paths (threads), where each thread has its own stack.
    1535 A single thread with multiple stacks, \ie coroutining, does \emph{not} imply concurrency~\cite[\S~3]{Buhr05a}.
    1536 Coroutining self-schedule the thread across stacks so execution is deterministic.
     1382A single thread with multiple call stacks, \newterm{coroutining}~\cite{Conway63,Marlin80}, does \emph{not} imply concurrency~\cite[\S~2]{Buhr05a}.
     1383In coroutining, coroutines self-schedule the thread across stacks so execution is deterministic.
    15371384(It is \emph{impossible} to generate a concurrency error when coroutining.)
    1538 
    1539 The transition to concurrency, even for a single thread with multiple stacks, occurs when coroutines context switch to a \newterm{scheduling coroutine}, introducing non-determinism from the coroutine perspective~\cite[\S~3]{Buhr05a}.
     1385However, coroutines are a stepping stone towards concurrency.
     1386
     1387The transition to concurrency, even for a single thread with multiple stacks, occurs when coroutines context switch to a \newterm{scheduling coroutine}, introducing non-determinism from the coroutine perspective~\cite[\S~3,]{Buhr05a}.
    15401388Therefore, a minimal concurrency system requires coroutines \emph{in conjunction with a nondeterministic scheduler}.
    1541 The resulting execution system now follows a cooperative threading-model~\cite{Adya02,libdill} because context-switching points to the scheduler (blocking) are known, but the next unblocking point is unknown due to the scheduler.
    1542 Adding \newterm{preemption} introduces \newterm{non-cooperative} or \newterm{preemptive} scheduling, where context switching points to the scheduler are unknown as they can occur randomly between any two instructions often based on a timer interrupt.
     1389The resulting execution system now follows a cooperative threading model~\cite{Adya02,libdill}, called \newterm{non-preemptive scheduling}.
     1390Adding \newterm{preemption} introduces non-cooperative scheduling, where context switching occurs randomly between any two instructions often based on a timer interrupt, called \newterm{preemptive scheduling}.
     1391While a scheduler introduces uncertain execution among explicit context switches, preemption introduces uncertainty by introducing implicit context switches.
    15431392Uncertainty gives the illusion of parallelism on a single processor and provides a mechanism to access and increase performance on multiple processors.
    15441393The reason is that the scheduler/runtime have complete knowledge about resources and how to best utilized them.
    1545 However, the introduction of unrestricted nondeterminism results in the need for \newterm{mutual exclusion} and \newterm{synchronization}~\cite[\S~4]{Buhr05a}, which restrict nondeterminism for correctness;
     1394However, the introduction of unrestricted nondeterminism results in the need for \newterm{mutual exclusion} and \newterm{synchronization}, which restrict nondeterminism for correctness;
    15461395otherwise, it is impossible to write meaningful concurrent programs.
    15471396Optimal concurrent performance is often obtained by having as much nondeterminism as mutual exclusion and synchronization correctness allow.
    15481397
    1549 A scheduler can also be stackless or stackful.
     1398A scheduler can either be a stackless or stackful.
    15501399For 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.
    15511400For 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.
    … …  
    15561405\label{s:threads}
    15571406
    1558 Threading (Table~\ref{t:ExecutionPropertyComposition} case 11) needs the ability to start a thread and wait for its completion.
     1407Threading needs the ability to start a thread and wait for its completion.
    15591408A common API for this ability is @fork@ and @join@.
    1560 \vspace{4pt}
    1561 \par\noindent
    1562 \begin{tabular}{@{}l|l|l@{}}
    1563 \multicolumn{1}{c|}{\textbf{Java}} & \multicolumn{1}{c|}{\textbf{\Celeven}} & \multicolumn{1}{c}{\textbf{pthreads}} \\
    1564 \hline
    1565 \begin{cfa}
    1566 class MyThread extends Thread {...}
    1567 mythread t = new MyThread(...);
     1409\begin{cquote}
     1410\begin{tabular}{@{}lll@{}}
     1411\multicolumn{1}{c}{\textbf{Java}} & \multicolumn{1}{c}{\textbf{\Celeven}} & \multicolumn{1}{c}{\textbf{pthreads}} \\
     1412\begin{cfa}
     1413class MyTask extends Thread {...}
     1414mytask t = new MyTask(...);
    15681415`t.start();` // start
    15691416// concurrency
    … …  
    15721419&
    15731420\begin{cfa}
    1574 class MyThread { ... } // functor
    1575 MyThread mythread;
    1576 `thread t( mythread, ... );` // start
     1421class MyTask { ... } // functor
     1422MyTask mytask;
     1423`thread t( mytask, ... );` // start
    15771424// concurrency
    15781425`t.join();` // wait
    … …  
    15871434\end{cfa}
    15881435\end{tabular}
    1589 \vspace{1pt}
    1590 \par\noindent
     1436\end{cquote}
    15911437\CFA has a simpler approach using a custom @thread@ type and leveraging declaration semantics (allocation/deallocation), where threads implicitly @fork@ after construction and @join@ before destruction.
    15921438\begin{cfa}
    1593 thread MyThread {};
    1594 void main( MyThread & this ) { ... }
     1439thread MyTask {};
     1440void main( MyTask & this ) { ... }
    15951441int main() {
    1596         MyThread team`[10]`; $\C[2.5in]{// allocate stack-based threads, implicit start after construction}$
     1442        MyTask team`[10]`; $\C[2.5in]{// allocate stack-based threads, implicit start after construction}$
    15971443        // concurrency
    15981444} $\C{// deallocate stack-based threads, implicit joins before destruction}$
    … …  
    16021448Arbitrary topologies are possible using dynamic allocation, allowing threads to outlive their declaration scope, identical to normal dynamic allocation.
    16031449\begin{cfa}
    1604 MyThread * factory( int N ) { ... return `anew( N )`; } $\C{// allocate heap-based threads, implicit start after construction}$
     1450MyTask * factory( int N ) { ... return `anew( N )`; } $\C{// allocate heap-based threads, implicit start after construction}$
    16051451int main() {
    1606         MyThread * team = factory( 10 );
     1452        MyTask * team = factory( 10 );
    16071453        // concurrency
    16081454        `delete( team );` $\C{// deallocate heap-based threads, implicit joins before destruction}\CRT$
    … …  
    16501496
    16511497Threads in \CFA are user level run by runtime kernel threads (see Section~\ref{s:CFARuntimeStructure}), where user threads provide concurrency and kernel threads provide parallelism.
    1652 Like coroutines, and for the same design reasons, \CFA provides a custom @thread@ type and a @trait@ to enforce and restrict the thread-interface functions.
     1498Like coroutines, and for the same design reasons, \CFA provides a custom @thread@ type and a @trait@ to enforce and restrict the task-interface functions.
    16531499\begin{cquote}
    16541500\begin{tabular}{@{}c@{\hspace{3\parindentlnth}}c@{}}
    … …  
    16811527\label{s:MutualExclusionSynchronization}
    16821528
    1683 Unrestricted nondeterminism is meaningless as there is no way to know when a result is completed and safe to access.
     1529Unrestricted nondeterminism is meaningless as there is no way to know when the result is completed without synchronization.
    16841530To produce meaningful execution requires clawing back some determinism using mutual exclusion and synchronization, where mutual exclusion provides access control for threads using shared data, and synchronization is a timing relationship among threads~\cite[\S~4]{Buhr05a}.
    1685 The shared data protected by mutual exlusion is called a \newterm{critical section}~\cite{Dijkstra65}, and the protection can be simple (only 1 thread) or complex (only N kinds of threads, \eg group~\cite{Joung00} or readers/writer~\cite{Courtois71}).
    1686 Without synchronization control in a critical section, an arriving thread can barge ahead of preexisting waiter threads resulting in short/long-term starvation, staleness/freshness problems, and/or incorrect transfer of data.
    1687 Preventing or detecting barging is a challenge with low-level locks, but made easier through higher-level constructs.
    1688 This challenge is often split into two different approaches: barging \emph{avoidance} and \emph{prevention}.
    1689 Approaches that unconditionally releasing a lock for competing threads to acquire must use barging avoidance with flag/counter variable(s) to force barging threads to wait;
    1690 approaches that conditionally hold locks during synchronization, \eg baton-passing~\cite{Andrews89}, prevent barging completely.
    1691 
    1692 At the lowest level, concurrent control is provided by atomic operations, upon which different kinds of locking mechanisms are constructed, \eg spin locks, semaphores~\cite{Dijkstra68b}, barriers, and path expressions~\cite{Campbell74}.
    1693 However, for productivity it is always desirable to use the highest-level construct that provides the necessary efficiency~\cite{Hochstein05}.
    1694 A significant challenge with locks is composability because it takes careful organization for multiple locks to be used while preventing deadlock.
    1695 Easing composability is another feature higher-level mutual-exclusion mechanisms can offer.
    1696 Some concurrent systems eliminate mutable shared-state by switching to non-shared communication like message passing~\cite{Thoth,Harmony,V-Kernel,MPI} (Erlang, MPI), channels~\cite{CSP} (CSP,Go), actors~\cite{Akka} (Akka, Scala), or functional techniques (Haskell).
     1531Some concurrent systems eliminate mutable shared-state by switching to stateless communication like message passing~\cite{Thoth,Harmony,V-Kernel,MPI} (Erlang, MPI), channels~\cite{CSP} (CSP,Go), actors~\cite{Akka} (Akka, Scala), or functional techniques (Haskell).
    16971532However, these approaches introduce a new communication mechanism for concurrency different from the standard communication using function call/return.
    16981533Hence, a programmer must learn and manipulate two sets of design/programming patterns.
    16991534While this distinction can be hidden away in library code, effective use of the library still has to take both paradigms into account.
    1700 In contrast, approaches based on shared-state models more closely resemble the standard call/return programming model, resulting in a single programming paradigm.
    1701 Finally, a newer approach for restricting non-determinism is transactional memory~\cite{Herlihy93}.
    1702 While 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~\cite{Cascaval08,Boehm09} to be the main concurrency paradigm for system languages.
     1535In contrast, approaches based on stateful models more closely resemble the standard call/return programming model, resulting in a single programming paradigm.
     1536
     1537At 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}.
     1538However, for productivity it is always desirable to use the highest-level construct that provides the necessary efficiency~\cite{Hochstein05}.
     1539A newer approach for restricting non-determinism is transactional memory~\cite{Herlihy93}.
     1540While 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.
     1541
     1542One of the most natural, elegant, and efficient mechanisms for mutual exclusion and synchronization for shared-memory systems is the \emph{monitor}.
     1543First 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}.
     1544In 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.
     1545For these reasons, \CFA selected monitors as the core high-level concurrency construct, upon which higher-level approaches can be easily constructed.
     1546
     1547
     1548\subsection{Mutual Exclusion}
     1549
     1550A group of instructions manipulating a specific instance of shared data that must be performed atomically is called a \newterm{critical section}~\cite{Dijkstra65}, which is enforced by \newterm{simple mutual-exclusion}.
     1551The generalization is called a \newterm{group critical-section}~\cite{Joung00}, where multiple tasks with the same session use the resource simultaneously and different sessions are segregated, which is enforced by \newterm{complex mutual-exclusion} providing the correct kind and number of threads using a group critical-section.
     1552The readers/writer problem~\cite{Courtois71} is an instance of a group critical-section, where readers share a session but writers have a unique session.
     1553
     1554However, many solutions exist for mutual exclusion, which vary in terms of performance, flexibility and ease of use.
     1555Methods 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.
     1556Ease 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.
     1557For 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.
     1558However, a significant challenge with locks is composability because it takes careful organization for multiple locks to be used while preventing deadlock.
     1559Easing composability is another feature higher-level mutual-exclusion mechanisms can offer.
     1560
     1561
     1562\subsection{Synchronization}
     1563
     1564Synchronization enforces relative ordering of execution, and synchronization tools provide numerous mechanisms to establish these timing relationships.
     1565Low-level synchronization primitives offer good performance and flexibility at the cost of ease of use;
     1566higher-level mechanisms often simplify usage by adding better coupling between synchronization and data, \eg receive-specific versus receive-any thread in message passing or offering specialized solutions, \eg barrier lock.
     1567Often synchronization is used to order access to a critical section, \eg ensuring a waiting writer thread enters the critical section before a calling reader thread.
     1568If the calling reader is scheduled before the waiting writer, the reader has barged.
     1569Barging 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).
     1570Preventing or detecting barging is an involved challenge with low-level locks, which is made easier through higher-level constructs.
     1571This challenge is often split into two different approaches: barging avoidance and prevention.
     1572Algorithms that unconditionally releasing a lock for competing threads to acquire use barging avoidance during synchronization to force a barging thread to wait;
     1573algorithms that conditionally hold locks during synchronization, \eg baton-passing~\cite{Andrews89}, prevent barging completely.
    17031574
    17041575
    … …  
    17061577\label{s:Monitor}
    17071578
    1708 One of the most natural, elegant, efficient, high-level mechanisms for mutual exclusion and synchronization for shared-memory systems is the \emph{monitor} (Table~\ref{t:ExecutionPropertyComposition} case 2).
    1709 First 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}.
    1710 In 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 manually implement a monitor.
    1711 For these reasons, \CFA selected monitors as the core high-level concurrency construct, upon which higher-level approaches can be easily constructed.
    1712 
    1713 Specifically, a \textbf{monitor} is a set of functions that ensure mutual exclusion when accessing shared state.
    1714 More precisely, a monitor is a programming technique that implicitly binds mutual exclusion to static function scope by call/return, as opposed to locks, where mutual-exclusion is defined by acquire/release calls, independent of lexical context (analogous to block and heap storage allocation).
     1579A \textbf{monitor} is a set of functions that ensure mutual exclusion when accessing shared state.
     1580More precisely, a monitor is a programming technique that implicitly binds mutual exclusion to static function 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).
    17151581Restricting acquire/release points eases programming, comprehension, and maintenance, at a slight cost in flexibility and efficiency.
    17161582\CFA uses a custom @monitor@ type and leverages declaration semantics (deallocation) to protect active or waiting threads in a monitor.
    17171583
    17181584The following is a \CFA monitor implementation of an atomic counter.
    1719 \begin{cfa}
     1585\begin{cfa}[morekeywords=nomutex]
    17201586`monitor` Aint { int cnt; }; $\C[4.25in]{// atomic integer counter}$
    1721 int ++?( Aint & `mutex` this ) with( this ) { return ++cnt; } $\C{// increment}$
    1722 int ?=?( Aint & `mutex` lhs, int rhs ) with( lhs ) { cnt = rhs; } $\C{// conversions with int, mutex optional}\CRT$
    1723 int ?=?( int & lhs, Aint & `mutex` rhs ) with( rhs ) { lhs = cnt; }
    1724 \end{cfa}
    1725 The operators use the parameter-only declaration type-qualifier @mutex@ to mark which parameters require locking during function execution to protect from race conditions.
    1726 The assignment operators provide bidirectional conversion between an atomic and normal integer without accessing field @cnt@.
    1727 (These operations only need @mutex@, if reading/writing the implementation type is not atomic.)
    1728 The atomic counter is used without any explicit mutual-exclusion and provides thread-safe semantics.
     1587int ++?( Aint & `mutex`$\(_{opt}\)$ this ) with( this ) { return ++cnt; } $\C{// increment}$
     1588int ?=?( Aint & `mutex`$\(_{opt}\)$ lhs, int rhs ) with( lhs ) { cnt = rhs; } $\C{// conversions with int}\CRT$
     1589int ?=?( int & lhs, Aint & `mutex`$\(_{opt}\)$ rhs ) with( rhs ) { lhs = cnt; }
     1590\end{cfa}
     1591% The @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.
     1592% (While a constructor may publish its address into a global variable, doing so generates a race-condition.)
     1593The prefix increment operation, @++?@, is normally @mutex@, indicating mutual exclusion is necessary during function execution, to protect the incrementing from race conditions, unless there is an atomic increment instruction for the implementation type.
     1594The assignment operators provide bidirectional conversion between an atomic and normal integer without accessing field @cnt@;
     1595these operations only need @mutex@, if reading/writing the implementation type is not atomic.
     1596The atomic counter is used without any explicit mutual-exclusion and provides thread-safe semantics, which is similar to the \CC template @std::atomic@.
    17291597\begin{cfa}
    17301598int i = 0, j = 0, k = 5;
    … …  
    17341602i = x; j = y; k = z;
    17351603\end{cfa}
    1736 Note, like other concurrent programming languages, \CFA has specializations for the basic types using atomic instructions for performance and a general trait similar to the \CC template @std::atomic@.
    17371604
    17381605\CFA monitors have \newterm{multi-acquire} semantics so the thread in the monitor may acquire it multiple times without deadlock, allowing recursion and calling other interface functions.
    1739 \newpage
    17401606\begin{cfa}
    17411607monitor M { ... } m;
    … …  
    17461612\end{cfa}
    17471613\CFA monitors also ensure the monitor lock is released regardless of how an acquiring function ends (normal or exceptional), and returning a shared variable is safe via copying before the lock is released.
    1748 Similar safety is offered by \emph{explicit} opt-in disciplines like \CC RAII versus the monitor \emph{implicit} language-enforced safety guarantee ensuring no programmer usage errors.
     1614Similar safety is offered by \emph{explicit} mechanisms like \CC RAII;
     1615monitor \emph{implicit} safety ensures no programmer usage errors.
    17491616Furthermore, RAII mechanisms cannot handle complex synchronization within a monitor, where the monitor lock may not be released on function exit because it is passed to an unblocking thread;
    17501617RAII is purely a mutual-exclusion mechanism (see Section~\ref{s:Scheduling}).
    … …  
    17721639\end{cquote}
    17731640The @dtype@ property prevents \emph{implicit} copy operations and the @is_monitor@ trait provides no \emph{explicit} copy operations, so monitors must be passed by reference (pointer).
     1641% Copying 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.
     1642% Copying 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.
    17741643Similarly, the function definitions ensures there is a mechanism to get (read) the monitor descriptor from its handle, and a special destructor to prevent deallocation if a thread using the shared data.
    17751644The custom monitor type also inserts any locks needed to implement the mutual exclusion semantics.
    … …  
    17831652For example, a monitor may be passed through multiple helper functions before it is necessary to acquire the monitor's mutual exclusion.
    17841653
    1785 \CFA requires programmers to identify the kind of parameter with the @mutex@ keyword and uses no keyword to mean \lstinline[morekeywords=nomutex]@nomutex@, because @mutex@ parameters are rare and no keyword is the \emph{normal} parameter semantics.
    1786 Hence, @mutex@ parameters are documentation, at the function and its prototype, to both programmer and compiler, without other redundant keywords.
    1787 Furthermore, \CFA relies heavily on traits as an abstraction mechanism, so the @mutex@ qualifier prevents coincidentally matching of a monitor trait with a type that is not a monitor, similar to coincidental inheritance where a shape and playing card can both be drawable.
     1654The benefit of mandatory monitor qualifiers is self-documentation, but requiring both @mutex@ and \lstinline[morekeywords=nomutex]@nomutex@ for all monitor parameters is redundant.
     1655Instead, the semantics has one qualifier as the default and the other required.
     1656For example, make the safe @mutex@ qualifier the default because assuming \lstinline[morekeywords=nomutex]@nomutex@ may cause subtle errors.
     1657Alternatively, make the unsafe \lstinline[morekeywords=nomutex]@nomutex@ qualifier the default because it is the \emph{normal} parameter semantics while @mutex@ parameters are rare.
     1658Providing a default qualifier implies knowing whether a parameter is a monitor.
     1659Since \CFA relies heavily on traits as an abstraction mechanism, types can coincidentally match the monitor trait but not be a monitor, similar to inheritance where a shape and playing card can both be drawable.
     1660For 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@.
    17881661
    17891662The next semantic decision is establishing which parameter \emph{types} may be qualified with @mutex@.
    … …  
    17991672Function @f3@ has a multiple object matrix, and @f4@ a multiple object data structure.
    18001673While shown shortly, multiple object acquisition is possible, but the number of objects must be statically known.
    1801 Therefore, \CFA only acquires one monitor per parameter with exactly one level of indirection, and exclude pointer types to unknown sized arrays.
     1674Therefore, \CFA only acquires one monitor per parameter with at most one level of indirection, excluding pointers as it is impossible to statically determine the size.
    18021675
    18031676For object-oriented monitors, \eg Java, calling a mutex member \emph{implicitly} acquires mutual exclusion of the receiver object, @`rec`.foo(...)@.
    … …  
    18061679While object-oriented monitors can be extended with a mutex qualifier for multiple-monitor members, no prior example of this feature could be found.}
    18071680called \newterm{bulk acquire}.
    1808 \CFA guarantees bulk acquisition order is consistent across calls to @mutex@ functions using the same monitors as arguments, so acquiring multiple monitors in a bulk acquire is safe from deadlock.
     1681\CFA guarantees acquisition order is consistent across calls to @mutex@ functions using the same monitors as arguments, so acquiring multiple monitors is safe from deadlock.
    18091682Figure~\ref{f:BankTransfer} shows a trivial solution to the bank transfer problem~\cite{BankTransfer}, where two resources must be locked simultaneously, using \CFA monitors with implicit locking and \CC with explicit locking.
    18101683A \CFA programmer only has to manage when to acquire mutual exclusion;
    … …  
    18261699void transfer( BankAccount & `mutex` my,
    18271700        BankAccount & `mutex` your, int me2you ) {
    1828         // bulk acquire
     1701
    18291702        deposit( my, -me2you ); // debit
    18301703        deposit( your, me2you ); // credit
    … …  
    18561729void transfer( BankAccount & my,
    18571730                        BankAccount & your, int me2you ) {
    1858         `scoped_lock lock( my.m, your.m );` // bulk acquire
     1731        `scoped_lock lock( my.m, your.m );`
    18591732        deposit( my, -me2you ); // debit
    18601733        deposit( your, me2you ); // credit
    … …  
    18841757\end{figure}
    18851758
    1886 Users can still force the acquiring order by using or not using @mutex@.
     1759Users can still force the acquiring order by using @mutex@/\lstinline[morekeywords=nomutex]@nomutex@.
    18871760\begin{cfa}
    18881761void foo( M & mutex m1, M & mutex m2 ); $\C{// acquire m1 and m2}$
    1889 void bar( M & mutex m1, M & m2 ) { $\C{// only acquire m1}$
     1762void bar( M & mutex m1, M & /* nomutex */ m2 ) { $\C{// acquire m1}$
    18901763        ... foo( m1, m2 ); ... $\C{// acquire m2}$
    18911764}
    1892 void baz( M & m1, M & mutex m2 ) { $\C{// only acquire m2}$
     1765void baz( M & /* nomutex */ m1, M & mutex m2 ) { $\C{// acquire m2}$
    18931766        ... foo( m1, m2 ); ... $\C{// acquire m1}$
    18941767}
    … …  
    19331806% There are many aspects of scheduling in a concurrency system, all related to resource utilization by waiting threads, \ie which thread gets the resource next.
    19341807% Different forms of scheduling include access to processors by threads (see Section~\ref{s:RuntimeStructureCluster}), another is access to a shared resource by a lock or monitor.
    1935 This section discusses scheduling for waiting threads eligible for monitor entry, \ie which user thread gets the shared resource next. (See Section~\ref{s:RuntimeStructureCluster} for scheduling kernel threads on virtual processors.)
    1936 While monitor mutual-exclusion provides safe access to its shared data, the data may indicate a thread cannot proceed, \eg a bounded buffer may be full/\-empty so produce/consumer threads must block.
    1937 Leaving the monitor and retrying (busy waiting) is impractical for high-level programming.
    1938 
    1939 Monitors eliminate busy waiting by providing synchronization within the monitor critical-section to schedule threads needing access to the shared data, where threads block versus spin.
     1808This section discusses monitor scheduling for waiting threads eligible for entry, \ie which thread gets the shared resource next. (See Section~\ref{s:RuntimeStructureCluster} for scheduling threads on virtual processors.)
     1809While monitor mutual-exclusion provides safe access to shared data, the monitor data may indicate that a thread accessing it cannot proceed, \eg a bounded buffer may be full/empty so produce/consumer threads must block.
     1810Leaving the monitor and trying again (busy waiting) is impractical for high-level programming.
     1811Monitors eliminate busy waiting by providing synchronization to schedule threads needing access to the shared data, where threads block versus spinning.
    19401812Synchronization is generally achieved with internal~\cite{Hoare74} or external~\cite[\S~2.9.2]{uC++} scheduling.
    1941 \newterm{Internal} (largely) schedules threads located \emph{inside} the monitor and is accomplished using condition variables with signal and wait.
    1942 \newterm{External} (largely) schedules threads located \emph{outside} the monitor and is accomplished with the @waitfor@ statement.
    1943 Note, internal scheduling has a small amount of external scheduling and vice versus, so the naming denotes where the majority of the block threads reside (inside or outside) for scheduling.
    1944 For complex scheduling, the approaches can be combined, so there can be an equal number of threads waiting inside and outside.
    1945 
    1946 \CFA monitors do not allow calling threads to barge ahead of signalled threads (via barging prevention), which simplifies synchronization among threads in the monitor and increases correctness.
    1947 A direct consequence of this semantics is that unblocked waiting threads are not required to recheck the waiting condition, \ie waits are not in a starvation-prone busy-loop as required by the signals-as-hints style with barging.
    1948 Preventing barging comes directly from Hoare's semantics in the seminal paper on monitors~\cite[p.~550]{Hoare74}.
     1813\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.
     1814Finally, \CFA monitors do not allow calling threads to barge ahead of signalled threads, which simplifies synchronization among threads in the monitor and increases correctness.
     1815If barging is allowed, synchronization between a signaller and signallee is difficult, often requiring additional flags and multiple unblock/block cycles.
     1816In fact, signals-as-hints is completely opposite from that proposed by Hoare in the seminal paper on monitors~\cite[p.~550]{Hoare74}.
    19491817% \begin{cquote}
    19501818% However, 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.
    19511819% It 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}
    19521820% \end{cquote}
    1953 Furthermore, \CFA concurrency has no spurious wakeup~\cite[\S~9]{Buhr05a}, which eliminates an implicit self barging.
    1954 
    1955 Monitor mutual-exclusion means signalling cannot have the signaller and signalled thread in the monitor simultaneously, so only the signaller or signallee can proceed.
    1956 Figure~\ref{f:MonitorScheduling} shows internal/external scheduling for the bounded-buffer examples in Figure~\ref{f:GenericBoundedBuffer}.
    1957 For internal scheduling in Figure~\ref{f:BBInt}, the @signal@ moves the signallee (front thread of the specified condition queue) to urgent and the signaller continues (solid line).
    1958 Multiple signals move multiple signallees to urgent until the condition queue is empty.
    1959 When the signaller exits or waits, a thread is implicitly unblocked from urgent (if available) before unblocking a calling thread to prevent barging.
     1821Furthermore, \CFA concurrency has no spurious wakeup~\cite[\S~9]{Buhr05a}, which eliminates an implicit form of self barging.
     1822Hence, a \CFA @wait@ statement is not enclosed in a @while@ loop retesting a blocking predicate, which can cause thread starvation due to barging.
     1823
     1824Figure~\ref{f:MonitorScheduling} shows general internal/external scheduling (for the bounded-buffer example in Figure~\ref{f:InternalExternalScheduling}).
     1825External calling threads block on the calling queue, if the monitor is occupied, otherwise they enter in FIFO order.
     1826Internal threads block on condition queues via @wait@ and reenter from the condition in FIFO order.
     1827Alternatively, internal threads block on urgent from the @signal_block@ or @waitfor@, and reenter implicitly when the monitor becomes empty, \ie, the thread in the monitor exits or waits.
     1828
     1829There are three signalling mechanisms to unblock waiting threads to enter the monitor.
     1830Note, signalling cannot have the signaller and signalled thread in the monitor simultaneously because of the mutual exclusion, so either the signaller or signallee can proceed.
     1831For internal scheduling, threads are unblocked from condition queues using @signal@, where the signallee is moved to urgent and the signaller continues (solid line).
     1832Multiple signals move multiple signallees to urgent until the condition is empty.
     1833When the signaller exits or waits, a thread blocked on urgent is processed before calling threads to prevent barging.
    19601834(Java conceptually moves the signalled thread to the calling queue, and hence, allows barging.)
    1961 Signal is used when the signaller is providing the cooperation needed by the signallee (\eg creating an empty slot in a buffer for a producer) and the signaller immediately exits the monitor to run concurrently (consume the buffer element) and passes control of the monitor to the signalled thread, which can immediately take advantage of the state change.
    1962 Specifically, the @wait@ function atomically blocks the calling thread and implicitly releases the monitor lock(s) for all monitors in the function's parameter list.
    1963 Signalling is unconditional because signalling an empty condition queue does nothing.
    1964 It is common to declare condition queues as monitor fields to prevent shared access, hence no locking is required for access as the queues are protected by the monitor lock.
    1965 In \CFA, a condition queue can be created/stored independently.
     1835The alternative unblock is in the opposite order using @signal_block@, where the signaller is moved to urgent and the signallee continues (dashed line), and is implicitly unblocked from urgent when the signallee exits or waits.
     1836
     1837For external scheduling, the condition queues are not used;
     1838instead threads are unblocked directly from the calling queue using @waitfor@ based on function names requesting mutual exclusion.
     1839(The linear search through the calling queue to locate a particular call can be reduced to $O(1)$.)
     1840The @waitfor@ has the same semantics as @signal_block@, where the signalled thread executes before the signallee, which waits on urgent.
     1841Executing multiple @waitfor@s from different signalled functions causes the calling threads to move to urgent.
     1842External scheduling requires urgent to be a stack, because the signaller expects to execute immediately after the specified monitor call has exited or waited.
     1843Internal scheduling behaves the same for an urgent stack or queue, except for multiple signalling, where the threads unblock from urgent in reverse order from signalling.
     1844If the restart order is important, multiple signalling by a signal thread can be transformed into daisy-chain signalling among threads, where each thread signals the next thread.
     1845We tried both a stack for @waitfor@ and queue for signalling, but that resulted in complex semantics about which thread enters next.
     1846Hence, \CFA uses a single urgent stack to correctly handle @waitfor@ and adequately support both forms of signalling.
    19661847
    19671848\begin{figure}
    … …  
    19811862\end{figure}
    19821863
     1864Figure~\ref{f:BBInt} shows a \CFA generic bounded-buffer with internal scheduling, where producers/consumers enter the monitor, detect the buffer is full/empty, and block on an appropriate condition variable, @full@/@empty@.
     1865The @wait@ function atomically blocks the calling thread and implicitly releases the monitor lock(s) for all monitors in the function's parameter list.
     1866The appropriate condition variable is signalled to unblock an opposite kind of thread after an element is inserted/removed from the buffer.
     1867Signalling is unconditional, because signalling an empty condition variable does nothing.
     1868It is common to declare condition variables as monitor fields to prevent shared access, hence no locking is required for access as the conditions are protected by the monitor lock.
     1869In \CFA, a condition variable can be created/stored independently.
     1870% To still prevent expensive locking on access, a condition variable is tied to a \emph{group} of monitors on first use, called \newterm{branding}, resulting in a low-cost boolean test to detect sharing from other monitors.
     1871
     1872% Signalling semantics cannot have the signaller and signalled thread in the monitor simultaneously, which means:
     1873% \begin{enumerate}
     1874% \item
     1875% The signalling thread returns immediately and the signalled thread continues.
     1876% \item
     1877% The signalling thread continues and the signalled thread is marked for urgent unblocking at the next scheduling point (exit/wait).
     1878% \item
     1879% The signalling thread blocks but is marked for urgent unblocking at the next scheduling point and the signalled thread continues.
     1880% \end{enumerate}
     1881% The first approach is too restrictive, as it precludes solving a reasonable class of problems, \eg dating service (see Figure~\ref{f:DatingService}).
     1882% \CFA supports the next two semantics as both are useful.
     1883
    19831884\begin{figure}
    19841885\centering
    … …  
    19921893                T elements[10];
    19931894        };
    1994         void ?{}( Buffer(T) & buf ) with(buf) {
     1895        void ?{}( Buffer(T) & buffer ) with(buffer) {
    19951896                front = back = count = 0;
    19961897        }
    1997 
    1998         void insert(Buffer(T) & mutex buf, T elm) with(buf){
    1999                 if ( count == 10 ) `wait( empty )`; // full ?
    2000                 // insert elm into buf
     1898        void insert( Buffer(T) & mutex buffer, T elem )
     1899                                with(buffer) {
     1900                if ( count == 10 ) `wait( empty )`;
     1901                // insert elem into buffer
    20011902                `signal( full )`;
    20021903        }
    2003         T remove( Buffer(T) & mutex buf ) with(buf) {
    2004                 if ( count == 0 ) `wait( full )`; // empty ?
    2005                 // remove elm from buf
     1904        T remove( Buffer(T) & mutex buffer ) with(buffer) {
     1905                if ( count == 0 ) `wait( full )`;
     1906                // remove elem from buffer
    20061907                `signal( empty )`;
    2007                 return elm;
     1908                return elem;
    20081909        }
    20091910}
    20101911\end{cfa}
    20111912\end{lrbox}
     1913
     1914% \newbox\myboxB
     1915% \begin{lrbox}{\myboxB}
     1916% \begin{cfa}[aboveskip=0pt,belowskip=0pt]
     1917% forall( otype T ) { // distribute forall
     1918%       monitor Buffer {
     1919%
     1920%               int front, back, count;
     1921%               T elements[10];
     1922%       };
     1923%       void ?{}( Buffer(T) & buffer ) with(buffer) {
     1924%               [front, back, count] = 0;
     1925%       }
     1926%       T remove( Buffer(T) & mutex buffer ); // forward
     1927%       void insert( Buffer(T) & mutex buffer, T elem )
     1928%                               with(buffer) {
     1929%               if ( count == 10 ) `waitfor( remove, buffer )`;
     1930%               // insert elem into buffer
     1931%
     1932%       }
     1933%       T remove( Buffer(T) & mutex buffer ) with(buffer) {
     1934%               if ( count == 0 ) `waitfor( insert, buffer )`;
     1935%               // remove elem from buffer
     1936%
     1937%               return elem;
     1938%       }
     1939% }
     1940% \end{cfa}
     1941% \end{lrbox}
    20121942
    20131943\newbox\myboxB
    20141944\begin{lrbox}{\myboxB}
    20151945\begin{cfa}[aboveskip=0pt,belowskip=0pt]
    2016 forall( otype T ) { // distribute forall
    2017         monitor Buffer {
    2018 
    2019                 int front, back, count;
    2020                 T elements[10];
    2021         };
    2022         void ?{}( Buffer(T) & buf ) with(buf) {
    2023                 front = back = count = 0;
    2024         }
    2025         T remove( Buffer(T) & mutex buf ); // forward
    2026         void insert(Buffer(T) & mutex buf, T elm) with(buf){
    2027                 if ( count == 10 ) `waitfor( remove : buf )`;
    2028                 // insert elm into buf
    2029 
    2030         }
    2031         T remove( Buffer(T) & mutex buf ) with(buf) {
    2032                 if ( count == 0 ) `waitfor( insert : buf )`;
    2033                 // remove elm from buf
    2034 
    2035                 return elm;
    2036         }
    2037 }
    2038 \end{cfa}
    2039 \end{lrbox}
    2040 
    2041 \subfloat[Internal scheduling]{\label{f:BBInt}\usebox\myboxA}
    2042 \hspace{1pt}
    2043 \vrule
    2044 \hspace{3pt}
    2045 \subfloat[External scheduling]{\label{f:BBExt}\usebox\myboxB}
    2046 
    2047 \caption{Generic bounded buffer}
    2048 \label{f:GenericBoundedBuffer}
    2049 \end{figure}
    2050 
    2051 The @signal_block@ provides the opposite unblocking order, where the signaller is moved to urgent and the signallee continues and a thread is implicitly unblocked from urgent when the signallee exits or waits (dashed line).
    2052 Signal block is used when the signallee is providing the cooperation needed by the signaller (\eg if the buffer is removed and a producer hands off an item to a consumer, as in Figure~\ref{f:DatingSignalBlock}) so the signaller must wait until the signallee unblocks, provides the cooperation, exits the monitor to run concurrently, and passes control of the monitor to the signaller, which can immediately take advantage of the state change.
    2053 Using @signal@ or @signal_block@ can be a dynamic decision based on whether the thread providing the cooperation arrives before or after the thread needing the cooperation.
    2054 
    2055 External scheduling in Figure~\ref{f:BBExt} simplifies internal scheduling by eliminating condition queues and @signal@/@wait@ (cases where it cannot are discussed shortly), and has existed in the programming language Ada for almost 40 years with variants in other languages~\cite{SR,ConcurrentC++,uC++}.
    2056 While prior languages use external scheduling solely for thread interaction, \CFA generalizes it to both monitors and threads.
    2057 External scheduling allows waiting for events from other threads while restricting unrelated events, that would otherwise have to wait on condition queues in the monitor.
    2058 Scheduling is controlled by the @waitfor@ statement, which atomically blocks the calling thread, releases the monitor lock, and restricts the function calls that can next acquire mutual exclusion.
    2059 Specifically, a thread calling the monitor is unblocked directly from the calling queue based on function names that can fulfill the cooperation required by the signaller.
    2060 (The linear search through the calling queue to locate a particular call can be reduced to $O(1)$.)
    2061 Hence, the @waitfor@ has the same semantics as @signal_block@, where the signallee thread from the calling queue executes before the signaller, which waits on urgent.
    2062 Now when a producer/consumer detects a full/empty buffer, the necessary cooperation for continuation is specified by indicating the next function call that can occur.
    2063 For example, a producer detecting a full buffer must have cooperation from a consumer to remove an item so function @remove@ is accepted, which prevents producers from entering the monitor, and after a consumer calls @remove@, the producer waiting on urgent is \emph{implicitly} unblocked because it can now continue its insert operation.
    2064 Hence, this mechanism is done in terms of control flow, next call, versus in terms of data, channels, as in Go/Rust @select@.
    2065 While both mechanisms have strengths and weaknesses, \CFA uses the control-flow mechanism to be consistent with other language features.
    2066 
    2067 Figure~\ref{f:ReadersWriterLock} shows internal/external scheduling for a readers/writer lock with no barging and threads are serviced in FIFO order to eliminate staleness/freshness among the reader/writer threads.
    2068 For internal scheduling in Figure~\ref{f:RWInt}, the readers and writers wait on the same condition queue in FIFO order, making it impossible to tell if a waiting thread is a reader or writer.
    2069 To clawback the kind of thread, a \CFA condition can store user data in the node for a blocking thread at the @wait@, \ie whether the thread is a @READER@ or @WRITER@.
    2070 An unblocked reader thread checks if the thread at the front of the queue is a reader and unblock it, \ie the readers daisy-chain signal the next group of readers demarcated by the next writer or end of the queue.
    2071 For external scheduling in Figure~\ref{f:RWExt}, a waiting reader checks if a writer is using the resource, and if so, restricts further calls until the writer exits by calling @EndWrite@.
    2072 The writer does a similar action for each reader or writer using the resource.
    2073 Note, no new calls to @StartRead@/@StartWrite@ may occur when waiting for the call to @EndRead@/@EndWrite@.
    2074 
    2075 \begin{figure}
    2076 \centering
    2077 \newbox\myboxA
    2078 \begin{lrbox}{\myboxA}
    2079 \begin{cfa}[aboveskip=0pt,belowskip=0pt]
    2080 enum RW { READER, WRITER };
    20811946monitor ReadersWriter {
    2082         int rcnt, wcnt; // readers/writer using resource
    2083         `condition RWers;`
     1947        int rcnt, wcnt; // readers/writer using resource
    20841948};
    20851949void ?{}( ReadersWriter & rw ) with(rw) {
    … …  
    20881952void EndRead( ReadersWriter & mutex rw ) with(rw) {
    20891953        rcnt -= 1;
    2090         if ( rcnt == 0 ) `signal( RWers )`;
    20911954}
    20921955void EndWrite( ReadersWriter & mutex rw ) with(rw) {
    20931956        wcnt = 0;
    2094         `signal( RWers );`
    20951957}
    20961958void StartRead( ReadersWriter & mutex rw ) with(rw) {
    2097         if ( wcnt !=0 || ! empty( RWers ) )
    2098                 `wait( RWers, READER )`;
     1959        if ( wcnt > 0 ) `waitfor( EndWrite, rw );`
    20991960        rcnt += 1;
    2100         if ( ! empty(RWers) && `front(RWers) == READER` )
    2101                 `signal( RWers )`;  // daisy-chain signalling
    21021961}
    21031962void StartWrite( ReadersWriter & mutex rw ) with(rw) {
    2104         if ( wcnt != 0 || rcnt != 0 ) `wait( RWers, WRITER )`;
    2105 
     1963        if ( wcnt > 0 ) `waitfor( EndWrite, rw );`
     1964        else while ( rcnt > 0 ) `waitfor( EndRead, rw );`
    21061965        wcnt = 1;
    21071966}
     1967
    21081968\end{cfa}
    21091969\end{lrbox}
    21101970
    2111 \newbox\myboxB
    2112 \begin{lrbox}{\myboxB}
    2113 \begin{cfa}[aboveskip=0pt,belowskip=0pt]
    2114 
    2115 monitor ReadersWriter {
    2116         int rcnt, wcnt; // readers/writer using resource
    2117 
    2118 };
    2119 void ?{}( ReadersWriter & rw ) with(rw) {
    2120         rcnt = wcnt = 0;
    2121 }
    2122 void EndRead( ReadersWriter & mutex rw ) with(rw) {
    2123         rcnt -= 1;
    2124 
    2125 }
    2126 void EndWrite( ReadersWriter & mutex rw ) with(rw) {
    2127         wcnt = 0;
    2128 
    2129 }
    2130 void StartRead( ReadersWriter & mutex rw ) with(rw) {
    2131         if ( wcnt > 0 ) `waitfor( EndWrite : rw );`
    2132 
    2133         rcnt += 1;
    2134 
    2135 
    2136 }
    2137 void StartWrite( ReadersWriter & mutex rw ) with(rw) {
    2138         if ( wcnt > 0 ) `waitfor( EndWrite : rw );`
    2139         else while ( rcnt > 0 ) `waitfor( EndRead : rw );`
    2140         wcnt = 1;
    2141 }
    2142 \end{cfa}
    2143 \end{lrbox}
    2144 
    2145 \subfloat[Internal scheduling]{\label{f:RWInt}\usebox\myboxA}
    2146 \hspace{1pt}
     1971\subfloat[Generic bounded buffer, internal scheduling]{\label{f:BBInt}\usebox\myboxA}
     1972\hspace{3pt}
    21471973\vrule
    21481974\hspace{3pt}
    2149 \subfloat[External scheduling]{\label{f:RWExt}\usebox\myboxB}
    2150 
    2151 \caption{Readers / writer lock}
    2152 \label{f:ReadersWriterLock}
     1975\subfloat[Readers / writer lock, external scheduling]{\label{f:RWExt}\usebox\myboxB}
     1976
     1977\caption{Internal / external scheduling}
     1978\label{f:InternalExternalScheduling}
    21531979\end{figure}
    21541980
    2155 Finally, external scheduling requires urgent to be a stack, because the signaller expects to execute immediately after the specified monitor call has exited or waited.
    2156 Internal schedulling performing multiple signalling results in unblocking from urgent in the reverse order from signalling.
    2157 It is rare for the unblocking order to be important as an unblocked thread can be time-sliced immediately after leaving the monitor.
    2158 If the unblocking order is important, multiple signalling can be restructured into daisy-chain signalling, where each thread signals the next thread.
    2159 Hence, \CFA uses a single urgent stack to correctly handle @waitfor@ and adequately support both forms of signalling.
    2160 (Advanced @waitfor@ features are discussed in Section~\ref{s:ExtendedWaitfor}.)
     1981Figure~\ref{f:BBInt} can be transformed into external scheduling by removing the condition variables and signals/waits, and adding the following lines at the locations of the current @wait@s in @insert@/@remove@, respectively.
     1982\begin{cfa}[aboveskip=2pt,belowskip=1pt]
     1983if ( count == 10 ) `waitfor( remove, buffer )`;       |      if ( count == 0 ) `waitfor( insert, buffer )`;
     1984\end{cfa}
     1985Here, the producers/consumers detects a full/\-empty buffer and prevents more producers/consumers from entering the monitor until there is a free/empty slot in the buffer.
     1986External scheduling is controlled by the @waitfor@ statement, which atomically blocks the calling thread, releases the monitor lock, and restricts the function calls that can next acquire mutual exclusion.
     1987If 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.
     1988Threads calling excluded functions block outside of (external to) the monitor on the calling queue, versus blocking on condition queues inside of (internal to) the monitor.
     1989Figure~\ref{f:RWExt} shows a readers/writer lock written using external scheduling, where a waiting reader detects a writer using the resource and restricts further calls until the writer exits by calling @EndWrite@.
     1990The writer does a similar action for each reader or writer using the resource.
     1991Note, no new calls to @StarRead@/@StartWrite@ may occur when waiting for the call to @EndRead@/@EndWrite@.
     1992External scheduling allows waiting for events from other threads while restricting unrelated events, that would otherwise have to wait on conditions in the monitor.
     1993The mechnaism can be done in terms of control flow, \eg Ada @accept@ or \uC @_Accept@, or in terms of data, \eg Go @select@ on channels.
     1994While both mechanisms have strengths and weaknesses, this project uses the control-flow mechanism to be consistent with other language features.
     1995% Two challenges specific to \CFA for external scheduling are loose object-definitions (see Section~\ref{s:LooseObjectDefinitions}) and multiple-monitor functions (see Section~\ref{s:Multi-MonitorScheduling}).
     1996
     1997Figure~\ref{f:DatingService} shows a dating service demonstrating non-blocking and blocking signalling.
     1998The dating service matches girl and boy threads with matching compatibility codes so they can exchange phone numbers.
     1999A thread blocks until an appropriate partner arrives.
     2000The complexity is exchanging phone numbers in the monitor because of the mutual-exclusion property.
     2001For 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.
     2002For 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.
     2003The 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;
     2004as well, an arriving thread may not find a partner and must wait, which requires a condition variable, and condition variables imply internal scheduling.
     2005Furthermore, barging corrupts the dating service during an exchange because a barger may also match and change the phone numbers, invalidating the previous exchange phone number.
     2006Putting loops around the @wait@s does not correct the problem;
     2007the simple solution must be restructured to account for barging.
    21612008
    21622009\begin{figure}
    … …  
    21722019};
    21732020int girl( DS & mutex ds, int phNo, int ccode ) {
    2174         if ( empty( Boys[ccode] ) ) {
     2021        if ( is_empty( Boys[ccode] ) ) {
    21752022                wait( Girls[ccode] );
    21762023                GirlPhNo = phNo;
    … …  
    21992046};
    22002047int girl( DS & mutex ds, int phNo, int ccode ) {
    2201         if ( empty( Boys[ccode] ) ) { // no compatible
     2048        if ( is_empty( Boys[ccode] ) ) { // no compatible
    22022049                wait( Girls[ccode] ); // wait for boy
    22032050                GirlPhNo = phNo; // make phone number available
    … …  
    22192066\qquad
    22202067\subfloat[\lstinline@signal_block@]{\label{f:DatingSignalBlock}\usebox\myboxB}
    2221 \caption{Dating service Monitor}
    2222 \label{f:DatingServiceMonitor}
     2068\caption{Dating service}
     2069\label{f:DatingService}
    22232070\end{figure}
    22242071
    2225 Figure~\ref{f:DatingServiceMonitor} shows a dating service demonstrating non-blocking and blocking signalling.
    2226 The dating service matches girl and boy threads with matching compatibility codes so they can exchange phone numbers.
    2227 A thread blocks until an appropriate partner arrives.
    2228 The complexity is exchanging phone numbers in the monitor because of the mutual-exclusion property.
    2229 For 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.
    2230 For signal-block scheduling, the implicit urgent-queue replaces the explicit @exchange@-condition and @signal_block@ puts the finding thread on the urgent stack and unblocks the matcher.
    2231 
    2232 The dating service is an important example of a monitor that cannot be written using external scheduling.
    2233 First, because scheduling requires knowledge of calling parameters to make matching decisions, and parameters of calling threads are unavailable within the monitor.
    2234 For example, a girl thread within the monitor cannot examine the @ccode@ of boy threads waiting on the calling queue to determine if there is a matching partner.
    2235 Second, because a scheduling decision may be delayed when there is no immediate match, which requires a condition queue for waiting, and condition queues imply internal scheduling.
    2236 For example, if a girl thread could determine there is no calling boy with the same @ccode@, it must wait until a matching boy arrives.
    2237 Finally, barging corrupts the dating service during an exchange because a barger may also match and change the phone numbers, invalidating the previous exchange phone number.
    2238 This situation shows rechecking the waiting condition and waiting again (signals-as-hints) fails, requiring significant restructured to account for barging.
     2072In summation, for internal scheduling, non-blocking signalling (as in the producer/consumer example) is used when the signaller is providing the cooperation for a waiting thread;
     2073the 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.
     2074The waiter unblocks next from the urgent queue, uses/takes the state, and exits the monitor.
     2075Blocking signal is the reverse, where the waiter is providing the cooperation for the signalling thread;
     2076the 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.
     2077The waiter changes state and exits the monitor, and the signaller unblocks next from the urgent queue to use/take the state.
    22392078
    22402079Both internal and external scheduling extend to multiple monitors in a natural way.
    22412080\begin{cquote}
    2242 \begin{tabular}{@{}l@{\hspace{2\parindentlnth}}l@{}}
     2081\begin{tabular}{@{}l@{\hspace{3\parindentlnth}}l@{}}
    22432082\begin{cfa}
    22442083monitor M { `condition e`; ... };
    … …  
    22512090&
    22522091\begin{cfa}
    2253 void rtn$\(_1\)$( M & mutex m1, M & mutex m2 ); // overload rtn
     2092void rtn$\(_1\)$( M & mutex m1, M & mutex m2 );
    22542093void rtn$\(_2\)$( M & mutex m1 );
    22552094void bar( M & mutex m1, M & mutex m2 ) {
    2256         ... waitfor( `rtn`${\color{red}\(_1\)}$ ); ...       // $\LstCommentStyle{waitfor( rtn\(_1\) : m1, m2 )}$
    2257         ... waitfor( `rtn${\color{red}\(_2\)}$ : m1` ); ...
     2095        ... waitfor( `rtn` ); ...       // $\LstCommentStyle{waitfor( rtn\(_1\), m1, m2 )}$
     2096        ... waitfor( `rtn, m1` ); ... // $\LstCommentStyle{waitfor( rtn\(_2\), m1 )}$
    22582097}
    22592098\end{cfa}
    … …  
    22622101For @wait( e )@, the default semantics is to atomically block the signaller and release all acquired mutex parameters, \ie @wait( e, m1, m2 )@.
    22632102To override the implicit multi-monitor wait, specific mutex parameter(s) can be specified, \eg @wait( e, m1 )@.
    2264 Wait cannot statically verify the released monitors are the acquired mutex-parameters without disallowing separately compiled helper functions calling @wait@.
    2265 While \CC supports bulk locking, @wait@ only accepts a single lock for a condition queue, so bulk locking with condition queues is asymmetric.
     2103Wait cannot statically verifies the released monitors are the acquired mutex-parameters without disallowing separately compiled helper functions calling @wait@.
     2104While \CC supports bulk locking, @wait@ only accepts a single lock for a condition variable, so bulk locking with condition variables is asymmetric.
    22662105Finally, a signaller,
    22672106\begin{cfa}
    … …  
    22722111must have acquired at least the same locks as the waiting thread signalled from a condition queue to allow the locks to be passed, and hence, prevent barging.
    22732112
    2274 Similarly, for @waitfor( rtn )@, the default semantics is to atomically block the acceptor and release all acquired mutex parameters, \ie @waitfor( rtn : m1, m2 )@.
    2275 To override the implicit multi-monitor wait, specific mutex parameter(s) can be specified, \eg @waitfor( rtn : m1 )@.
     2113Similarly, for @waitfor( rtn )@, the default semantics is to atomically block the acceptor and release all acquired mutex parameters, \ie @waitfor( rtn, m1, m2 )@.
     2114To override the implicit multi-monitor wait, specific mutex parameter(s) can be specified, \eg @waitfor( rtn, m1 )@.
    22762115@waitfor@ does statically verify the monitor types passed are the same as the acquired mutex-parameters of the given function or function pointer, hence the function (pointer) prototype must be accessible.
    22772116% When an overloaded function appears in an @waitfor@ statement, calls to any function with that name are accepted.
    … …  
    22812120void rtn( M & mutex m );
    22822121`int` rtn( M & mutex m );
    2283 waitfor( (`int` (*)( M & mutex ))rtn : m );
    2284 \end{cfa}
    2285 
    2286 The ability to release a subset of acquired monitors can result in a \newterm{nested monitor}~\cite{Lister77} deadlock (see Section~\ref{s:MutexAcquisition}).
    2287 \newpage
     2122waitfor( (`int` (*)( M & mutex ))rtn, m );
     2123\end{cfa}
     2124
     2125The ability to release a subset of acquired monitors can result in a \newterm{nested monitor}~\cite{Lister77} deadlock.
    22882126\begin{cfa}
    22892127void foo( M & mutex m1, M & mutex m2 ) {
    2290         ... wait( `e, m1` ); ...                                $\C{// release m1, keeping m2 acquired}$
    2291 void bar( M & mutex m1, M & mutex m2 ) {        $\C{// must acquire m1 and m2}$
     2128        ... wait( `e, m1` ); ...                                $\C{// release m1, keeping m2 acquired )}$
     2129void bar( M & mutex m1, M & mutex m2 ) {        $\C{// must acquire m1 and m2 )}$
    22922130        ... signal( `e` ); ...
    22932131\end{cfa}
    22942132The @wait@ only releases @m1@ so the signalling thread cannot acquire @m1@ and @m2@ to enter @bar@ and @signal@ the condition.
    2295 While deadlock can occur with multiple/nesting acquisition, this is a consequence of locks, and by extension monitor locking is not perfectly composable.
     2133While deadlock can occur with multiple/nesting acquisition, this is a consequence of locks, and by extension monitors, not being perfectly composable.
     2134
    22962135
    22972136
    22982137\subsection{\texorpdfstring{Extended \protect\lstinline@waitfor@}{Extended waitfor}}
    2299 \label{s:ExtendedWaitfor}
    23002138
    23012139Figure~\ref{f:ExtendedWaitfor} shows the extended form of the @waitfor@ statement to conditionally accept one of a group of mutex functions, with an optional statement to be performed \emph{after} the mutex function finishes.
    … …  
    23082146Hence, the terminating @else@ clause allows a conditional attempt to accept a call without blocking.
    23092147If both @timeout@ and @else@ clause are present, the @else@ must be conditional, or the @timeout@ is never triggered.
    2310 There is also a traditional future wait queue (not shown) (\eg Microsoft @WaitForMultipleObjects@), to wait for a specified number of future elements in the queue.
    2311 Finally, there is a shorthand for specifying multiple functions using the same set of monitors: @waitfor( f, g, h : m1, m2, m3 )@.
     2148There is also a traditional future wait queue (not shown) (\eg Microsoft (@WaitForMultipleObjects@)), to wait for a specified number of future elements in the queue.
    23122149
    23132150\begin{figure}
    … …  
    23362173The right example accepts either @mem1@ or @mem2@ if @C1@ and @C2@ are true.
    23372174
    2338 An interesting use of @waitfor@ is accepting the @mutex@ destructor to know when an object is deallocated, \eg assume the bounded buffer is restructured from a monitor to a thread with the following @main@.
     2175An interesting use of @waitfor@ is accepting the @mutex@ destructor to know when an object is deallocated, \eg assume the bounded buffer is restructred from a monitor to a thread with the following @main@.
    23392176\begin{cfa}
    23402177void main( Buffer(T) & buffer ) with(buffer) {
    23412178        for () {
    2342                 `waitfor( ^?{} : buffer )` break;
    2343                 or when ( count != 20 ) waitfor( insert : buffer ) { ... }
    2344                 or when ( count != 0 ) waitfor( remove : buffer ) { ... }
     2179                `waitfor( ^?{}, buffer )` break;
     2180                or when ( count != 20 ) waitfor( insert, buffer ) { ... }
     2181                or when ( count != 0 ) waitfor( remove, buffer ) { ... }
    23452182        }
    23462183        // clean up
    … …  
    24342271To support this efficient semantics (and prevent barging), the implementation maintains a list of monitors acquired for each blocked thread.
    24352272When a signaller exits or waits in a monitor function/statement, the front waiter on urgent is unblocked if all its monitors are released.
    2436 Implementing a fast subset check for the necessary released monitors is important and discussed in the following sections.
     2273Implementing a fast subset check for the necessary released monitors is important.
    24372274% The 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.
    24382275
    24392276
    2440 \subsection{\texorpdfstring{\protect\lstinline@waitfor@ Implementation}{waitfor Implementation}}
    2441 \label{s:waitforImplementation}
    2442 
    2443 In a statically-typed object-oriented programming language, a class has an exhaustive list of members, even when members are added via static inheritance (see Figure~\ref{f:uCinheritance}).
    2444 Knowing all members at compilation (even separate compilation) allows uniquely numbered them so the accept-statement implementation can use a fast/compact bit mask with $O(1)$ compare.
    2445 
    2446 \begin{figure}
    2447 \centering
    2448 \begin{lrbox}{\myboxA}
    2449 \begin{uC++}[aboveskip=0pt,belowskip=0pt]
    2450 $\emph{translation unit 1}$
    2451 _Monitor B { // common type in .h file
    2452         _Mutex virtual void `f`( ... );
    2453         _Mutex virtual void `g`( ... );
    2454         _Mutex virtual void w1( ... ) { ... _Accept(`f`, `g`); ... }
    2455 };
    2456 $\emph{translation unit 2}$
    2457 // include B
    2458 _Monitor D : public B { // inherit
    2459         _Mutex void `h`( ... ); // add
    2460         _Mutex void w2( ... ) { ... _Accept(`f`, `h`); ... }
    2461 };
    2462 \end{uC++}
    2463 \end{lrbox}
    2464 
    2465 \begin{lrbox}{\myboxB}
    2466 \begin{cfa}[aboveskip=0pt,belowskip=0pt]
    2467 $\emph{translation unit 1}$
    2468 monitor M { ... }; // common type in .h file
    2469 void `f`( M & mutex m, ... );
    2470 void `g`( M & mutex m, ... );
    2471 void w1( M & mutex m, ... ) { ... waitfor(`f`, `g` : m); ... }
    2472 
    2473 $\emph{translation unit 2}$
    2474 // include M
    2475 extern void `f`( M & mutex m, ... ); // import f but not g
    2476 void `h`( M & mutex m ); // add
    2477 void w2( M & mutex m, ... ) { ... waitfor(`f`, `h` : m); ... }
    2478 
    2479 \end{cfa}
    2480 \end{lrbox}
    2481 
    2482 \subfloat[\uC]{\label{f:uCinheritance}\usebox\myboxA}
    2483 \hspace{3pt}
    2484 \vrule
    2485 \hspace{3pt}
    2486 \subfloat[\CFA]{\label{f:CFinheritance}\usebox\myboxB}
    2487 \caption{Member / Function visibility}
    2488 \label{f:MemberFunctionVisibility}
    2489 \end{figure}
    2490 
    2491 However, the @waitfor@ statement in translation unit 2 (see Figure~\ref{f:CFinheritance}) cannot see function @g@ in translation unit 1 precluding a unique numbering for a bit-mask because the monitor type only carries the protected shared-data.
     2277\subsection{Loose Object Definitions}
     2278\label{s:LooseObjectDefinitions}
     2279
     2280In an object-oriented programming language, a class includes an exhaustive list of operations.
     2281A new class can add members via static inheritance but the subclass still has an exhaustive list of operations.
     2282(Dynamic member adding, \eg JavaScript~\cite{JavaScript}, is not considered.)
     2283In the object-oriented scenario, the type and all its operators are always present at compilation (even separate compilation), so it is possible to number the operations in a bit mask and use an $O(1)$ compare with a similar bit mask created for the operations specified in a @waitfor@.
     2284
     2285However, in \CFA, monitor functions can be statically added/removed in translation units, making a fast subset check difficult.
     2286\begin{cfa}
     2287        monitor M { ... }; // common type, included in .h file
     2288translation unit 1
     2289        void `f`( M & mutex m );
     2290        void g( M & mutex m ) { waitfor( `f`, m ); }
     2291translation unit 2
     2292        void `f`( M & mutex m ); $\C{// replacing f and g for type M in this translation unit}$
     2293        void `g`( M & mutex m );
     2294        void h( M & mutex m ) { waitfor( `f`, m ) or waitfor( `g`, m ); } $\C{// extending type M in this translation unit}$
     2295\end{cfa}
     2296The @waitfor@ statements in each translation unit cannot form a unique bit-mask because the monitor type does not carry that information.
     2297Hence, function pointers are used to identify the functions listed in the @waitfor@ statement, stored in a variable-sized array.
     2298Then, the same implementation approach used for the urgent stack is used for the calling queue.
     2299Each caller has a list of monitors acquired, and the @waitfor@ statement performs a (usually short) linear search matching functions in the @waitfor@ list with called functions, and then verifying the associated mutex locks can be transfers.
    24922300(A possible way to construct a dense mapping is at link or load-time.)
    2493 Hence, function pointers are used to identify the functions listed in the @waitfor@ statement, stored in a variable-sized array.
    2494 Then, the same implementation approach used for the urgent stack (see Section~\ref{s:Scheduling}) is used for the calling queue.
    2495 Each caller has a list of monitors acquired, and the @waitfor@ statement performs a (short) linear search matching functions in the @waitfor@ list with called functions, and then verifying the associated mutex locks can be transfers.
    24962301
    24972302
    … …  
    25082313The solution is for the programmer to disambiguate:
    25092314\begin{cfa}
    2510 waitfor( f : `m2` ); $\C{// wait for call to f with argument m2}$
     2315waitfor( f, `m2` ); $\C{// wait for call to f with argument m2}$
    25112316\end{cfa}
    25122317Both locks are acquired by function @g@, so when function @f@ is called, the lock for monitor @m2@ is passed from @g@ to @f@, while @g@ still holds lock @m1@.
    … …  
    25152320monitor M { ... };
    25162321void f( M & mutex m1, M & mutex m2 );
    2517 void g( M & mutex m1, M & mutex m2 ) { waitfor( f : `m1, m2` ); $\C{// wait for call to f with arguments m1 and m2}$
     2322void g( M & mutex m1, M & mutex m2 ) { waitfor( f, `m1, m2` ); $\C{// wait for call to f with arguments m1 and m2}$
    25182323\end{cfa}
    25192324Again, the set of monitors passed to the @waitfor@ statement must be entirely contained in the set of monitors already acquired by the accepting function.
    2520 % Also, the order of the monitors in a @waitfor@ statement must match the order of the mutex parameters.
    2521 
    2522 Figure~\ref{f:UnmatchedMutexSets} shows internal and external scheduling with multiple monitors that must match exactly with a signalling or accepting thread, \ie partial matching results in waiting.
    2523 In both cases, the set of monitors is disjoint so unblocking is impossible.
     2325Also, the order of the monitors in a @waitfor@ statement is unimportant.
     2326
     2327Figure~\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.
     2328For both examples, the set of monitors is disjoint so unblocking is impossible.
    25242329
    25252330\begin{figure}
    … …  
    25502355}
    25512356void g( M1 & mutex m1, M2 & mutex m2 ) {
    2552         waitfor( f : m1, m2 );
     2357        waitfor( f, m1, m2 );
    25532358}
    25542359g( `m11`, m2 ); // block on accept
    … …  
    25652370\end{figure}
    25662371
     2372
     2373\subsection{\texorpdfstring{\protect\lstinline@mutex@ Threads}{mutex Threads}}
     2374
     2375Threads in \CFA can also be monitors to allow \emph{direct communication} among threads, \ie threads can have mutex functions that are called by other threads.
     2376Hence, all monitor features are available when using threads.
     2377Figure~\ref{f:DirectCommunication} shows a comparison of direct call communication in \CFA with direct channel communication in Go.
     2378(Ada provides a similar mechanism to the \CFA direct communication.)
     2379The program main in both programs communicates directly with the other thread versus indirect communication where two threads interact through a passive monitor.
     2380Both direct and indirection thread communication are valuable tools in structuring concurrent programs.
     2381
    25672382\begin{figure}
    25682383\centering
    … …  
    25712386
    25722387struct Msg { int i, j; };
    2573 monitor thread GoRtn { int i;  float f;  Msg m; };
     2388thread GoRtn { int i;  float f;  Msg m; };
    25742389void mem1( GoRtn & mutex gortn, int i ) { gortn.i = i; }
    25752390void mem2( GoRtn & mutex gortn, float f ) { gortn.f = f; }
    … …  
    25812396        for () {
    25822397
    2583                 `waitfor( mem1 : gortn )` sout | i;  // wait for calls
    2584                 or `waitfor( mem2 : gortn )` sout | f;
    2585                 or `waitfor( mem3 : gortn )` sout | m.i | m.j;
    2586                 or `waitfor( ^?{} : gortn )` break; // low priority
     2398                `waitfor( mem1, gortn )` sout | i;  // wait for calls
     2399                or `waitfor( mem2, gortn )` sout | f;
     2400                or `waitfor( mem3, gortn )` sout | m.i | m.j;
     2401                or `waitfor( ^?{}, gortn )` break;
    25872402
    25882403        }
    … …  
    26382453\hspace{3pt}
    26392454\subfloat[Go]{\label{f:Gochannel}\usebox\myboxB}
    2640 \caption{Direct versus indirect communication}
    2641 \label{f:DirectCommunicationComparison}
    2642 
    2643 \medskip
    2644 
    2645 \begin{cfa}
    2646 monitor thread DatingService {
    2647         condition Girls[CompCodes], Boys[CompCodes];
    2648         int girlPhoneNo, boyPhoneNo, ccode;
    2649 };
    2650 int girl( DatingService & mutex ds, int phoneno, int code ) with( ds ) {
    2651         girlPhoneNo = phoneno;  ccode = code;
    2652         `wait( Girls[ccode] );`                                                         $\C{// wait for boy}$
    2653         girlPhoneNo = phoneno;  return boyPhoneNo;
    2654 }
    2655 int boy( DatingService & mutex ds, int phoneno, int code ) with( ds ) {
    2656         boyPhoneNo = phoneno;  ccode = code;
    2657         `wait( Boys[ccode] );`                                                          $\C{// wait for girl}$
    2658         boyPhoneNo = phoneno;  return girlPhoneNo;
    2659 }
    2660 void main( DatingService & ds ) with( ds ) {                    $\C{// thread starts, ds defaults to mutex}$
    2661         for () {
    2662                 waitfor( ^?{} ) break;                                                  $\C{// high priority}$
    2663                 or waitfor( girl )                                                              $\C{// girl called, compatible boy ? restart boy then girl}$
    2664                         if ( ! is_empty( Boys[ccode] ) ) { `signal_block( Boys[ccode] );  signal_block( Girls[ccode] );` }
    2665                 or waitfor( boy ) {                                                             $\C{// boy called, compatible girl ? restart girl then boy}$
    2666                         if ( ! is_empty( Girls[ccode] ) ) { `signal_block( Girls[ccode] );  signal_block( Boys[ccode] );` }
    2667         }
    2668 }
    2669 \end{cfa}
    2670 \caption{Direct communication dating service}
    2671 \label{f:DirectCommunicationDatingService}
     2455\caption{Direct communication}
     2456\label{f:DirectCommunication}
    26722457\end{figure}
    26732458
    … …  
    26842469void main( Ping & pi ) {
    26852470        for ( 10 ) {
    2686                 `waitfor( ping : pi );`
     2471                `waitfor( ping, pi );`
    26872472                `pong( po );`
    26882473        }
    … …  
    26972482        for ( 10 ) {
    26982483                `ping( pi );`
    2699                 `waitfor( pong : po );`
     2484                `waitfor( pong, po );`
    27002485        }
    27012486}
    … …  
    27122497
    27132498
    2714 \subsection{\texorpdfstring{\protect\lstinline@monitor@ Generators / Coroutines / Threads}{monitor Generators / Coroutines / Threads}}
    2715 
    2716 \CFA generators, coroutines, and threads can also be monitors (Table~\ref{t:ExecutionPropertyComposition} cases 4, 6, 12) allowing safe \emph{direct communication} with threads, \ie the custom types can have mutex functions that are called by other threads.
    2717 All monitor features are available within these mutex functions.
    2718 For example, if the formatter generator (or coroutine equivalent) in Figure~\ref{f:CFAFormatGen} is extended with the monitor property and this interface function is used to communicate with the formatter:
    2719 \begin{cfa}
    2720 void fmt( Fmt & mutex fmt, char ch ) { fmt.ch = ch; resume( fmt ) }
    2721 \end{cfa}
    2722 multiple threads can safely pass characters for formatting.
    2723 
    2724 Figure~\ref{f:DirectCommunicationComparison} shows a comparison of direct call-communication in \CFA versus indirect channel-communication in Go.
    2725 (Ada has a similar mechanism to \CFA direct communication.)
    2726 The program thread in \CFA @main@ uses the call/return paradigm to directly communicate with the @GoRtn main@, whereas Go switches to the channel paradigm to indirectly communicate with the goroutine.
    2727 Communication by multiple threads is safe for the @gortn@ thread via mutex calls in \CFA or channel assignment in Go.
    2728 
    2729 Figure~\ref{f:DirectCommunicationDatingService} shows the dating-service problem in Figure~\ref{f:DatingServiceMonitor} extended from indirect monitor communication to direct thread communication.
    2730 When converting a monitor to a thread (server), the coding pattern is to move as much code as possible from the accepted members into the thread main so it does an much work as possible.
    2731 Notice, the dating server is postponing requests for an unspecified time while continuing to accept new requests.
    2732 For complex servers (web-servers), there can be hundreds of lines of code in the thread main and safe interaction with clients can be complex.
     2499\subsection{Execution Properties}
     2500
     2501Table~\ref{t:ObjectPropertyComposition} shows how the \CFA high-level constructs cover 3 fundamental execution properties: thread, stateful function, and mutual exclusion.
     2502Case 1 is a basic object, with none of the new execution properties.
     2503Case 2 allows @mutex@ calls to Case 1 to protect shared data.
     2504Case 3 allows stateful functions to suspend/resume but restricts operations because the state is stackless.
     2505Case 4 allows @mutex@ calls to Case 3 to protect shared data.
     2506Cases 5 and 6 are the same as 3 and 4 without restriction because the state is stackful.
     2507Cases 7 and 8 are rejected because a thread cannot execute without a stackful state in a preemptive environment when context switching from the signal handler.
     2508Cases 9 and 10 have a stackful thread without and with @mutex@ calls.
     2509For situations where threads do not require direct communication, case 9 provides faster creation/destruction by eliminating @mutex@ setup.
     2510
     2511\begin{table}
     2512\caption{Object property composition}
     2513\centering
     2514\label{t:ObjectPropertyComposition}
     2515\renewcommand{\arraystretch}{1.25}
     2516%\setlength{\tabcolsep}{5pt}
     2517\begin{tabular}{c|c||l|l}
     2518\multicolumn{2}{c||}{object properties} & \multicolumn{2}{c}{mutual exclusion} \\
     2519\hline
     2520thread  & stateful                              & \multicolumn{1}{c|}{No} & \multicolumn{1}{c}{Yes} \\
     2521\hline
     2522\hline
     2523No              & No                                    & \textbf{1}\ \ \ aggregate type                & \textbf{2}\ \ \ @monitor@ aggregate type \\
     2524\hline
     2525No              & Yes (stackless)               & \textbf{3}\ \ \ @generator@                   & \textbf{4}\ \ \ @monitor@ @generator@ \\
     2526\hline
     2527No              & Yes (stackful)                & \textbf{5}\ \ \ @coroutine@                   & \textbf{6}\ \ \ @monitor@ @coroutine@ \\
     2528\hline
     2529Yes             & No / Yes (stackless)  & \textbf{7}\ \ \ {\color{red}rejected} & \textbf{8}\ \ \ {\color{red}rejected} \\
     2530\hline
     2531Yes             & Yes (stackful)                & \textbf{9}\ \ \ @thread@                              & \textbf{10}\ \ @monitor@ @thread@ \\
     2532\end{tabular}
     2533\end{table}
    27332534
    27342535
    … …  
    27362537
    27372538For 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.
    2738 Some of these low-level mechanism are used to build the \CFA runtime, but we always advocate using high-level mechanisms whenever possible.
     2539Some of these low-level mechanism are used in the \CFA runtime, but we strongly advocate using high-level mechanisms whenever possible.
    27392540
    27402541
    … …  
    27792580\begin{cfa}
    27802581struct Adder {
    2781         int * row, cols;
     2582    int * row, cols;
    27822583};
    27832584int operator()() {
    … …  
    28382639\label{s:RuntimeStructureCluster}
    28392640
    2840 A \newterm{cluster} is a collection of user and kernel threads, where the kernel threads run the user threads from the cluster's ready queue, and the operating system runs the kernel threads on the processors from its ready queue.
    2841 The term \newterm{virtual processor} is introduced as a synonym for kernel thread to disambiguate between user and kernel thread.
    2842 From the language perspective, a virtual processor is an actual processor (core).
    2843 
     2641A \newterm{cluster} is a collection of threads and virtual processors (abstract kernel-thread) that execute the (user) threads from its own ready queue (like an OS executing kernel threads).
    28442642The purpose of a cluster is to control the amount of parallelism that is possible among threads, plus scheduling and other execution defaults.
    28452643The default cluster-scheduler is single-queue multi-server, which provides automatic load-balancing of threads on processors.
    … …  
    28602658Programs may use more virtual processors than hardware processors.
    28612659On a multiprocessor, kernel threads are distributed across the hardware processors resulting in virtual processors executing in parallel.
    2862 (It is possible to use affinity to lock a virtual processor onto a particular hardware processor~\cite{affinityLinux,affinityWindows}, which is used when caching issues occur or for heterogeneous hardware processors.) %, affinityFreebsd, affinityNetbsd, affinityMacosx
     2660(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.)
    28632661The \CFA runtime attempts to block unused processors and unblock processors as the system load increases;
    2864 balancing the workload with processors is difficult because it requires future knowledge, \ie what will the application workload do next.
     2662balancing the workload with processors is difficult because it requires future knowledge, \ie what will the applicaton workload do next.
    28652663Preemption occurs on virtual processors rather than user threads, via operating-system interrupts.
    28662664Thus virtual processors execute user threads, where preemption frequency applies to a virtual processor, so preemption occurs randomly across the executed user threads.
    … …  
    28972695Nondeterministic preemption provides fairness from long-running threads, and forces concurrent programmers to write more robust programs, rather than relying on code between cooperative scheduling to be atomic.
    28982696This atomic reliance can fail on multi-core machines, because execution across cores is nondeterministic.
    2899 A different reason for not supporting preemption is that it significantly complicates the runtime system, \eg Windows runtime does not support interrupts and on Linux systems, interrupts are complex (see below).
     2697A different reason for not supporting preemption is that it significantly complicates the runtime system, \eg Microsoft runtime does not support interrupts and on Linux systems, interrupts are complex (see below).
    29002698Preemption is normally handled by setting a countdown timer on each virtual processor.
    2901 When the timer expires, an interrupt is delivered, and its signal handler resets the countdown 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.
     2699When the timer expires, an interrupt is delivered, and the interrupt handler resets the countdown 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.
    29022700Multiple signal handlers may be pending.
    29032701When 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.
    29042702The 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;
    29052703therefore, the same signal mask is required for all virtual processors in a cluster.
    2906 Because preemption interval is usually long (1 millisecond) performance cost is negligible.
    2907 
    2908 Linux switched a decade ago from specific to arbitrary virtual-processor signal-delivery for applications with multiple kernel threads.
    2909 In the new semantics, a virtual-processor directed signal may be delivered to any virtual processor created by the application that does not have the signal blocked.
     2704Because preemption frequency is usually long (1 millisecond) performance cost is negligible.
     2705
     2706Linux switched a decade ago from specific to arbitrary process signal-delivery for applications with multiple kernel threads.
     2707\begin{cquote}
     2708A process-directed signal may be delivered to any one of the threads that does not currently have the signal blocked.
     2709If more than one of the threads has the signal unblocked, then the kernel chooses an arbitrary thread to which it will deliver the signal.
     2710SIGNAL(7) - Linux Programmer's Manual
     2711\end{cquote}
    29102712Hence, the timer-expiry signal, which is generated \emph{externally} by the Linux kernel to an application, is delivered to any of its Linux subprocesses (kernel threads).
    29112713To ensure each virtual processor receives a preemption signal, a discrete-event simulation is run on a special virtual processor, and only it sets and receives timer events.
    … …  
    29252727\label{s:Performance}
    29262728
    2927 To test the performance of the \CFA runtime, a series of microbenchmarks are used to compare \CFA with pthreads, Java 11.0.6, Go 1.12.6, Rust 1.37.0, Python 3.7.6, Node.js 12.14.1, and \uC 7.0.0.
     2729To verify the implementation of the \CFA runtime, a series of microbenchmarks are performed comparing \CFA with pthreads, Java OpenJDK-9, Go 1.12.6 and \uC 7.0.0.
    29282730For comparison, the package must be multi-processor (M:N), which excludes libdill/libmil~\cite{libdill} (M:1)), and use a shared-memory programming model, \eg not message passing.
    2929 The benchmark computer is an AMD Opteron\texttrademark\ 6380 NUMA 64-core, 8 socket, 2.5 GHz processor, running Ubuntu 16.04.6 LTS, and pthreads/\CFA/\uC are compiled with gcc 9.2.1.
     2731The benchmark computer is an AMD Opteron\texttrademark\ 6380 NUMA 64-core, 8 socket, 2.5 GHz processor, running Ubuntu 16.04.6 LTS, and \CFA/\uC are compiled with gcc 6.5.
    29302732
    29312733All benchmarks are run using the following harness. (The Java harness is augmented to circumvent JIT issues.)
    29322734\begin{cfa}
    2933 #define BENCH( `run` ) uint64_t start = cputime_ns();  `run;`  double result = (double)(cputime_ns() - start) / N;
    2934 \end{cfa}
    2935 where CPU time in nanoseconds is from the appropriate language clock.
    2936 Each benchmark is performed @N@ times, where @N@ is selected so the benchmark runs in the range of 2--20 seconds for the specific programming language.
    2937 The total time is divided by @N@ to obtain the average time for a benchmark.
    2938 Each benchmark experiment is run 13 times and the average appears in the table.
     2735unsigned int N = 10_000_000;
     2736#define BENCH( `run` ) Time before = getTimeNsec();  `run;`  Duration result = (getTimeNsec() - before) / N;
     2737\end{cfa}
     2738The method used to get time is @clock_gettime( CLOCK_REALTIME )@.
     2739Each benchmark is performed @N@ times, where @N@ varies depending on the benchmark;
     2740the total time is divided by @N@ to obtain the average time for a benchmark.
     2741Each benchmark experiment is run 31 times.
    29392742All omitted tests for other languages are functionally identical to the \CFA tests and available online~\cite{CforallBenchMarks}.
    2940 % tar --exclude-ignore=exclude -cvhf benchmark.tar benchmark
    2941 
    2942 \paragraph{Context Switching}
     2743% tar --exclude=.deps --exclude=Makefile --exclude=Makefile.in --exclude=c.c --exclude=cxx.cpp --exclude=fetch_add.c -cvhf benchmark.tar benchmark
     2744
     2745\paragraph{Object Creation}
     2746
     2747Object creation is measured by creating/deleting the specific kind of concurrent object.
     2748Figure~\ref{f:creation} shows the code for \CFA, with results in Table~\ref{tab:creation}.
     2749The 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.
     2750
     2751\begin{multicols}{2}
     2752\lstset{language=CFA,moredelim=**[is][\color{red}]{@}{@},deletedelim=**[is][]{`}{`}}
     2753\begin{cfa}
     2754@thread@ MyThread {};
     2755void @main@( MyThread & ) {}
     2756int main() {
     2757        BENCH( for ( N ) { @MyThread m;@ } )
     2758        sout | result`ns;
     2759}
     2760\end{cfa}
     2761\captionof{figure}{\CFA object-creation benchmark}
     2762\label{f:creation}
     2763
     2764\columnbreak
     2765
     2766\vspace*{-16pt}
     2767\captionof{table}{Object creation comparison (nanoseconds)}
     2768\label{tab:creation}
     2769
     2770\begin{tabular}[t]{@{}r*{3}{D{.}{.}{5.2}}@{}}
     2771\multicolumn{1}{@{}c}{} & \multicolumn{1}{c}{Median} & \multicolumn{1}{c}{Average} & \multicolumn{1}{c@{}}{Std Dev} \\
     2772\CFA Coroutine Lazy             & 13.2          & 13.1          & 0.44          \\
     2773\CFA Coroutine Eager    & 531.3         & 536.0         & 26.54         \\
     2774\CFA Thread                             & 2074.9        & 2066.5        & 170.76        \\
     2775\uC Coroutine                   & 89.6          & 90.5          & 1.83          \\
     2776\uC Thread                              & 528.2         & 528.5         & 4.94          \\
     2777Goroutine                               & 4068.0        & 4113.1        & 414.55        \\
     2778Java Thread                             & 103848.5      & 104295.4      & 2637.57       \\
     2779Pthreads                                & 33112.6       & 33127.1       & 165.90
     2780\end{tabular}
     2781\end{multicols}
     2782
     2783
     2784\paragraph{Context-Switching}
    29432785
    29442786In procedural programming, the cost of a function call is important as modularization (refactoring) increases.
    2945 (In many cases, a compiler inlines function calls to increase the size and number of basic blocks for optimizing.)
    2946 Similarly, when modularization extends to coroutines/threads, the time for a context switch becomes a relevant factor.
     2787(In many cases, a compiler inlines function calls to eliminate this cost.)
     2788Similarly, when modularization extends to coroutines/tasks, the time for a context switch becomes a relevant factor.
    29472789The coroutine test is from resumer to suspender and from suspender to resumer, which is two context switches.
    2948 %For async-await systems, the test is scheduling and fulfilling @N@ empty promises, where all promises are allocated before versus interleaved with fulfillment to avoid garbage collection.
    2949 For async-await systems, the test measures the cost of the @await@ expression entering the event engine by awaiting @N@ promises, where each created promise is resolved by an immediate event in the engine (using Node.js @setImmediate@).
    29502790The thread test is using yield to enter and return from the runtime kernel, which is two context switches.
    29512791The difference in performance between coroutine and thread context-switch is the cost of scheduling for threads, whereas coroutines are self-scheduling.
    2952 Figure~\ref{f:ctx-switch} shows the \CFA code for a coroutine/thread with results in Table~\ref{t:ctx-switch}.
    2953 
    2954 % From: Gregor Richards <gregor.richards@uwaterloo.ca>
    2955 % To: "Peter A. Buhr" <pabuhr@plg2.cs.uwaterloo.ca>
    2956 % Date: Fri, 24 Jan 2020 13:49:18 -0500
    2957 %
    2958 % I can also verify that the previous version, which just tied a bunch of promises together, *does not* go back to the
    2959 % event loop at all in the current version of Node. Presumably they're taking advantage of the fact that the ordering of
    2960 % events is intentionally undefined to just jump right to the next 'then' in the chain, bypassing event queueing
    2961 % entirely. That's perfectly correct behavior insofar as its difference from the specified behavior isn't observable, but
    2962 % it isn't typical or representative of much anything useful, because most programs wouldn't have whole chains of eager
    2963 % promises. Also, it's not representative of *anything* you can do with async/await, as there's no way to encode such an
    2964 % eager chain that way.
     2792Figure~\ref{f:ctx-switch} only shows the \CFA code for coroutines/threads (other systems are similar) with all results in Table~\ref{tab:ctx-switch}.
    29652793
    29662794\begin{multicols}{2}
    … …  
    29682796\begin{cfa}[aboveskip=0pt,belowskip=0pt]
    29692797@coroutine@ C {} c;
    2970 void main( C & ) { while () { @suspend;@ } }
     2798void main( C & ) { for ( ;; ) { @suspend;@ } }
    29712799int main() { // coroutine test
    29722800        BENCH( for ( N ) { @resume( c );@ } )
    2973         sout | result;
    2974 }
    2975 int main() { // thread test
     2801        sout | result`ns;
     2802}
     2803int main() { // task test
    29762804        BENCH( for ( N ) { @yield();@ } )
    2977         sout | result;
     2805        sout | result`ns;
    29782806}
    29792807\end{cfa}
    … …  
    29852813\vspace*{-16pt}
    29862814\captionof{table}{Context switch comparison (nanoseconds)}
    2987 \label{t:ctx-switch}
     2815\label{tab:ctx-switch}
    29882816\begin{tabular}{@{}r*{3}{D{.}{.}{3.2}}@{}}
    29892817\multicolumn{1}{@{}c}{} & \multicolumn{1}{c}{Median} &\multicolumn{1}{c}{Average} & \multicolumn{1}{c@{}}{Std Dev} \\
    2990 C function                      & 1.8           & 1.8           & 0.0   \\
    2991 \CFA generator          & 1.8           & 1.8           & 0.1   \\
    2992 \CFA coroutine          & 32.5          & 32.9          & 0.8   \\
    2993 \CFA thread                     & 93.8          & 93.6          & 2.2   \\
    2994 \uC coroutine           & 50.3          & 50.3          & 0.2   \\
    2995 \uC thread                      & 97.3          & 97.4          & 1.0   \\
    2996 Python generator        & 40.9          & 41.3          & 1.5   \\
    2997 Node.js generator       & 32.6          & 32.2          & 1.0   \\
    2998 Node.js await           & 1852.2        & 1854.7        & 16.4  \\
    2999 Goroutine thread        & 143.0         & 143.3         & 1.1   \\
    3000 Rust thread                     & 332.0         & 331.4         & 2.4   \\
    3001 Java thread                     & 405.0         & 415.0         & 17.6  \\
    3002 Pthreads thread         & 334.3         & 335.2         & 3.9
     2818C function              & 1.8   & 1.8   & 0.01  \\
     2819\CFA generator  & 2.4   & 2.2   & 0.25  \\
     2820\CFA Coroutine  & 36.2  & 36.2  & 0.25  \\
     2821\CFA Thread             & 93.2  & 93.5  & 2.09  \\
     2822\uC Coroutine   & 52.0  & 52.1  & 0.51  \\
     2823\uC Thread              & 96.2  & 96.3  & 0.58  \\
     2824Goroutine               & 141.0 & 141.3 & 3.39  \\
     2825Java Thread             & 374.0 & 375.8 & 10.38 \\
     2826Pthreads Thread & 361.0 & 365.3 & 13.19
    30032827\end{tabular}
    30042828\end{multicols}
    30052829
    3006 \paragraph{Internal Scheduling}
    3007 
    3008 Internal scheduling is measured using a cycle of two threads signalling and waiting.
    3009 Figure~\ref{f:schedint} shows the code for \CFA, with results in Table~\ref{t:schedint}.
     2830
     2831\paragraph{Mutual-Exclusion}
     2832
     2833Uncontented mutual exclusion, which frequently occurs, is measured by entering/leaving a critical section.
     2834For monitors, entering and leaving a monitor function is measured.
     2835To put the results in context, the cost of entering a non-inline function and the cost of acquiring and releasing a @pthread_mutex@ lock is also measured.
     2836Figure~\ref{f:mutex} shows the code for \CFA with all results in Table~\ref{tab:mutex}.
    30102837Note, the incremental cost of bulk acquire for \CFA, which is largely a fixed cost for small numbers of mutex objects.
    3011 Java 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.
    30122838
    30132839\begin{multicols}{2}
    30142840\lstset{language=CFA,moredelim=**[is][\color{red}]{@}{@},deletedelim=**[is][]{`}{`}}
    30152841\begin{cfa}
    3016 volatile int go = 0;
    3017 @condition c;@
    30182842@monitor@ M {} m1/*, m2, m3, m4*/;
    3019 void call( M & @mutex p1/*, p2, p3, p4*/@ ) {
    3020         @signal( c );@
    3021 }
    3022 void wait( M & @mutex p1/*, p2, p3, p4*/@ ) {
    3023         go = 1; // continue other thread
    3024         for ( N ) { @wait( c );@ } );
    3025 }
    3026 thread T {};
    3027 void main( T & ) {
    3028         while ( go == 0 ) { yield(); } // waiter must start first
    3029         BENCH( for ( N ) { call( m1/*, m2, m3, m4*/ ); } )
    3030         sout | result;
    3031 }
     2843void __attribute__((noinline))
     2844do_call( M & @mutex m/*, m2, m3, m4*/@ ) {}
    30322845int main() {
    3033         T t;
    3034         wait( m1/*, m2, m3, m4*/ );
    3035 }
    3036 \end{cfa}
    3037 \captionof{figure}{\CFA Internal-scheduling benchmark}
    3038 \label{f:schedint}
     2846        BENCH(
     2847                for( N ) do_call( m1/*, m2, m3, m4*/ );
     2848        )
     2849        sout | result`ns;
     2850}
     2851\end{cfa}
     2852\captionof{figure}{\CFA acquire/release mutex benchmark}
     2853\label{f:mutex}
    30392854
    30402855\columnbreak
    30412856
    30422857\vspace*{-16pt}
    3043 \captionof{table}{Internal-scheduling comparison (nanoseconds)}
    3044 \label{t:schedint}
    3045 \bigskip
    3046 
    3047 \begin{tabular}{@{}r*{3}{D{.}{.}{5.2}}@{}}
    3048 \multicolumn{1}{@{}c}{} & \multicolumn{1}{c}{Median} & \multicolumn{1}{c}{Average} & \multicolumn{1}{c@{}}{Std Dev} \\
    3049 \CFA @signal@, 1 monitor        & 364.4         & 364.2         & 4.4           \\
    3050 \CFA @signal@, 2 monitor        & 484.4         & 483.9         & 8.8           \\
    3051 \CFA @signal@, 4 monitor        & 709.1         & 707.7         & 15.0          \\
    3052 \uC @signal@ monitor            & 328.3         & 327.4         & 2.4           \\
    3053 Rust cond. variable                     & 7514.0        & 7437.4        & 397.2         \\
    3054 Java @notify@ monitor           & 9623.0        & 9654.6        & 236.2         \\
    3055 Pthreads cond. variable         & 5553.7        & 5576.1        & 345.6
     2858\captionof{table}{Mutex comparison (nanoseconds)}
     2859\label{tab:mutex}
     2860\begin{tabular}{@{}r*{3}{D{.}{.}{3.2}}@{}}
     2861\multicolumn{1}{@{}c}{} & \multicolumn{1}{c}{Median} &\multicolumn{1}{c}{Average} & \multicolumn{1}{c@{}}{Std Dev} \\
     2862test and test-and-test lock             & 19.1  & 18.9  & 0.40  \\
     2863\CFA @mutex@ function, 1 arg.   & 45.9  & 46.6  & 1.45  \\
     2864\CFA @mutex@ function, 2 arg.   & 105.0 & 104.7 & 3.08  \\
     2865\CFA @mutex@ function, 4 arg.   & 165.0 & 167.6 & 5.65  \\
     2866\uC @monitor@ member rtn.               & 54.0  & 53.7  & 0.82  \\
     2867Java synchronized method                & 31.0  & 31.1  & 0.50  \\
     2868Pthreads Mutex Lock                             & 33.6  & 32.6  & 1.14
    30562869\end{tabular}
    30572870\end{multicols}
    … …  
    30612874
    30622875External scheduling is measured using a cycle of two threads calling and accepting the call using the @waitfor@ statement.
    3063 Figure~\ref{f:schedext} shows the code for \CFA with results in Table~\ref{t:schedext}.
     2876Figure~\ref{f:ext-sched} shows the code for \CFA, with results in Table~\ref{tab:ext-sched}.
    30642877Note, the incremental cost of bulk acquire for \CFA, which is largely a fixed cost for small numbers of mutex objects.
    30652878
    … …  
    30682881\vspace*{-16pt}
    30692882\begin{cfa}
    3070 @monitor@ M {} m1/*, m2, m3, m4*/;
    3071 void call( M & @mutex p1/*, p2, p3, p4*/@ ) {}
    3072 void wait( M & @mutex p1/*, p2, p3, p4*/@ ) {
    3073         for ( N ) { @waitfor( call : p1/*, p2, p3, p4*/ );@ }
    3074 }
     2883volatile int go = 0;
     2884@monitor@ M {} m;
    30752885thread T {};
     2886void __attribute__((noinline))
     2887do_call( M & @mutex@ ) {}
    30762888void main( T & ) {
    3077         BENCH( for ( N ) { call( m1/*, m2, m3, m4*/ ); } )
    3078         sout | result;
     2889        while ( go == 0 ) { yield(); }
     2890        while ( go == 1 ) { do_call( m ); }
     2891}
     2892int __attribute__((noinline))
     2893do_wait( M & @mutex@ m ) {
     2894        go = 1; // continue other thread
     2895        BENCH( for ( N ) { @waitfor( do_call, m );@ } )
     2896        go = 0; // stop other thread
     2897        sout | result`ns;
    30792898}
    30802899int main() {
    30812900        T t;
    3082         wait( m1/*, m2, m3, m4*/ );
     2901        do_wait( m );
    30832902}
    30842903\end{cfa}
    30852904\captionof{figure}{\CFA external-scheduling benchmark}
    3086 \label{f:schedext}
     2905\label{f:ext-sched}
    30872906
    30882907\columnbreak
    … …  
    30902909\vspace*{-16pt}
    30912910\captionof{table}{External-scheduling comparison (nanoseconds)}
    3092 \label{t:schedext}
     2911\label{tab:ext-sched}
    30932912\begin{tabular}{@{}r*{3}{D{.}{.}{3.2}}@{}}
    30942913\multicolumn{1}{@{}c}{} & \multicolumn{1}{c}{Median} &\multicolumn{1}{c}{Average} & \multicolumn{1}{c@{}}{Std Dev} \\
    3095 \CFA @waitfor@, 1 monitor       & 367.1 & 365.3 & 5.0   \\
    3096 \CFA @waitfor@, 2 monitor       & 463.0 & 464.6 & 7.1   \\
    3097 \CFA @waitfor@, 4 monitor       & 689.6 & 696.2 & 21.5  \\
    3098 \uC \lstinline[language=uC++]|_Accept| monitor  & 328.2 & 329.1 & 3.4   \\
    3099 Go \lstinline[language=Golang]|select| channel  & 365.0 & 365.5 & 1.2
     2914\CFA @waitfor@, 1 @monitor@     & 376.4 & 376.8 & 7.63  \\
     2915\CFA @waitfor@, 2 @monitor@     & 491.4 & 492.0 & 13.31 \\
     2916\CFA @waitfor@, 4 @monitor@     & 681.0 & 681.7 & 19.10 \\
     2917\uC @_Accept@                           & 331.1 & 331.4 & 2.66
    31002918\end{tabular}
    31012919\end{multicols}
    31022920
    3103 \paragraph{Mutual-Exclusion}
    3104 
    3105 Uncontented mutual exclusion, which frequently occurs, is measured by entering/leaving a critical section.
    3106 For monitors, entering and leaving a monitor function is measured, otherwise the language-appropriate mutex-lock is measured.
    3107 For comparison, a spinning (versus blocking) test-and-test-set lock is presented.
    3108 Figure~\ref{f:mutex} shows the code for \CFA with results in Table~\ref{t:mutex}.
    3109 Note the incremental cost of bulk acquire for \CFA, which is largely a fixed cost for small numbers of mutex objects.
     2921
     2922\paragraph{Internal Scheduling}
     2923
     2924Internal scheduling is measured using a cycle of two threads signalling and waiting.
     2925Figure~\ref{f:int-sched} shows the code for \CFA, with results in Table~\ref{tab:int-sched}.
     2926Note, the incremental cost of bulk acquire for \CFA, which is largely a fixed cost for small numbers of mutex objects.
     2927Java 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.
    31102928
    31112929\begin{multicols}{2}
    31122930\lstset{language=CFA,moredelim=**[is][\color{red}]{@}{@},deletedelim=**[is][]{`}{`}}
    31132931\begin{cfa}
    3114 @monitor@ M {} m1/*, m2, m3, m4*/;
    3115 call( M & @mutex p1/*, p2, p3, p4*/@ ) {}
     2932volatile int go = 0;
     2933@monitor@ M { @condition c;@ } m;
     2934void __attribute__((noinline))
     2935do_call( M & @mutex@ a1 ) { @signal( c );@ }
     2936thread T {};
     2937void main( T & this ) {
     2938        while ( go == 0 ) { yield(); }
     2939        while ( go == 1 ) { do_call( m ); }
     2940}
     2941int  __attribute__((noinline))
     2942do_wait( M & mutex m ) with(m) {
     2943        go = 1; // continue other thread
     2944        BENCH( for ( N ) { @wait( c );@ } );
     2945        go = 0; // stop other thread
     2946        sout | result`ns;
     2947}
    31162948int main() {
    3117         BENCH( for( N ) call( m1/*, m2, m3, m4*/ ); )
    3118         sout | result;
    3119 }
    3120 \end{cfa}
    3121 \captionof{figure}{\CFA acquire/release mutex benchmark}
    3122 \label{f:mutex}
     2949        T t;
     2950        do_wait( m );
     2951}
     2952\end{cfa}
     2953\captionof{figure}{\CFA Internal-scheduling benchmark}
     2954\label{f:int-sched}
    31232955
    31242956\columnbreak
    31252957
    31262958\vspace*{-16pt}
    3127 \captionof{table}{Mutex comparison (nanoseconds)}
    3128 \label{t:mutex}
    3129 \begin{tabular}{@{}r*{3}{D{.}{.}{3.2}}@{}}
    3130 \multicolumn{1}{@{}c}{} & \multicolumn{1}{c}{Median} &\multicolumn{1}{c}{Average} & \multicolumn{1}{c@{}}{Std Dev} \\
    3131 test-and-test-set lock                  & 19.1  & 18.9  & 0.4   \\
    3132 \CFA @mutex@ function, 1 arg.   & 48.3  & 47.8  & 0.9   \\
    3133 \CFA @mutex@ function, 2 arg.   & 86.7  & 87.6  & 1.9   \\
    3134 \CFA @mutex@ function, 4 arg.   & 173.4 & 169.4 & 5.9   \\
    3135 \uC @monitor@ member rtn.               & 54.8  & 54.8  & 0.1   \\
    3136 Goroutine mutex lock                    & 34.0  & 34.0  & 0.0   \\
    3137 Rust mutex lock                                 & 33.0  & 33.2  & 0.8   \\
    3138 Java synchronized method                & 31.0  & 31.0  & 0.0   \\
    3139 Pthreads mutex Lock                             & 31.0  & 31.1  & 0.4
     2959\captionof{table}{Internal-scheduling comparison (nanoseconds)}
     2960\label{tab:int-sched}
     2961\bigskip
     2962
     2963\begin{tabular}{@{}r*{3}{D{.}{.}{5.2}}@{}}
     2964\multicolumn{1}{@{}c}{} & \multicolumn{1}{c}{Median} & \multicolumn{1}{c}{Average} & \multicolumn{1}{c@{}}{Std Dev} \\
     2965\CFA @signal@, 1 @monitor@      & 372.6         & 374.3         & 14.17         \\
     2966\CFA @signal@, 2 @monitor@      & 492.7         & 494.1         & 12.99         \\
     2967\CFA @signal@, 4 @monitor@      & 749.4         & 750.4         & 24.74         \\
     2968\uC @signal@                            & 320.5         & 321.0         & 3.36          \\
     2969Java @notify@                           & 10160.5       & 10169.4       & 267.71        \\
     2970Pthreads Cond. Variable         & 4949.6        & 5065.2        & 363
    31402971\end{tabular}
    31412972\end{multicols}
    31422973
    3143 \paragraph{Creation}
    3144 
    3145 Creation is measured by creating/deleting a specific kind of control-flow object.
    3146 Figure~\ref{f:creation} shows the code for \CFA with results in Table~\ref{t:creation}.
    3147 Note, the call stacks of \CFA coroutines are lazily created on the first resume, therefore the cost of creation with and without a stack are presented.
    3148 
    3149 \begin{multicols}{2}
    3150 \lstset{language=CFA,moredelim=**[is][\color{red}]{@}{@},deletedelim=**[is][]{`}{`}}
    3151 \begin{cfa}
    3152 @coroutine@ MyCoroutine {};
    3153 void ?{}( MyCoroutine & this ) {
    3154 #ifdef EAGER
    3155         resume( this );
    3156 #endif
    3157 }
    3158 void main( MyCoroutine & ) {}
    3159 int main() {
    3160         BENCH( for ( N ) { @MyCoroutine c;@ } )
    3161         sout | result;
    3162 }
    3163 \end{cfa}
    3164 \captionof{figure}{\CFA creation benchmark}
    3165 \label{f:creation}
    3166 
    3167 \columnbreak
    3168 
    3169 \vspace*{-16pt}
    3170 \captionof{table}{Creation comparison (nanoseconds)}
    3171 \label{t:creation}
    3172 
    3173 \begin{tabular}[t]{@{}r*{3}{D{.}{.}{5.2}}@{}}
    3174 \multicolumn{1}{@{}c}{} & \multicolumn{1}{c}{Median} & \multicolumn{1}{c}{Average} & \multicolumn{1}{c@{}}{Std Dev} \\
    3175 \CFA generator                  & 0.6           & 0.6           & 0.0           \\
    3176 \CFA coroutine lazy             & 13.4          & 13.1          & 0.5           \\
    3177 \CFA coroutine eager    & 144.7         & 143.9         & 1.5           \\
    3178 \CFA thread                             & 466.4         & 468.0         & 11.3          \\
    3179 \uC coroutine                   & 155.6         & 155.7         & 1.7           \\
    3180 \uC thread                              & 523.4         & 523.9         & 7.7           \\
    3181 Python generator                & 123.2         & 124.3         & 4.1           \\
    3182 Node.js generator               & 32.3          & 32.2          & 0.3           \\
    3183 Goroutine thread                & 751.0         & 750.5         & 3.1           \\
    3184 Rust thread                             & 53801.0       & 53896.8       & 274.9         \\
    3185 Java thread                             & 120274.0      & 120722.9      & 2356.7        \\
    3186 Pthreads thread                 & 31465.5       & 31419.5       & 140.4
    3187 \end{tabular}
    3188 \end{multicols}
    3189 
    3190 
    3191 \subsection{Discussion}
    3192 
    3193 Languages using 1:1 threading based on pthreads can at best meet or exceed (due to language overhead) the pthread results.
    3194 Note, pthreads has a fast zero-contention mutex lock checked in user space.
    3195 Languages with M:N threading have better performance than 1:1 because there is no operating-system interactions.
    3196 Languages with stackful coroutines have higher cost than stackless coroutines because of stack allocation and context switching;
    3197 however, stackful \uC and \CFA coroutines have approximately the same performance as stackless Python and Node.js generators.
    3198 The \CFA stackless generator is approximately 25 times faster for suspend/resume and 200 times faster for creation than stackless Python and Node.js generators.
    3199 
    32002974
    32012975\section{Conclusion}
    … …  
    32032977Advanced control-flow will always be difficult, especially when there is temporal ordering and nondeterminism.
    32042978However, many systems exacerbate the difficulty through their presentation mechanisms.
    3205 This paper shows it is possible to understand high-level control-flow using three properties: statefulness, thread, mutual-exclusion/synchronization.
    3206 Combining these properties creates a number of high-level, efficient, and maintainable control-flow types: generator, coroutine, thread, each of which can be a monitor.
    3207 Eliminated from \CFA are barging and spurious wakeup, which are nonintuitive and lead to errors, and having to work with a bewildering set of low-level locks and acquisition techniques.
    3208 \CFA high-level race-free monitors and threads provide the core mechanisms for mutual exclusion and synchronization, without having to resort to magic qualifiers like @volatile@/@atomic@.
     2979This paper shows it is possible to present a hierarchy of control-flow features, generator, coroutine, thread, and monitor, providing an integrated set of high-level, efficient, and maintainable control-flow features.
     2980Eliminated from \CFA are spurious wakeup and barging, which are nonintuitive and lead to errors, and having to work with a bewildering set of low-level locks and acquisition techniques.
     2981\CFA high-level race-free monitors and tasks provide the core mechanisms for mutual exclusion and synchronization, without having to resort to magic qualifiers like @volatile@/@atomic@.
    32092982Extending these mechanisms to handle high-level deadlock-free bulk acquire across both mutual exclusion and synchronization is a unique contribution.
    32102983The \CFA runtime 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.
    32112984The M:N model is judged to be efficient and provide greater flexibility than a 1:1 threading model.
    32122985These concepts and the \CFA runtime-system are written in the \CFA language, extensively leveraging the \CFA type-system, which demonstrates the expressiveness of the \CFA language.
    3213 Performance comparisons with other concurrent systems/languages show the \CFA approach is competitive across all basic operations, which translates directly into good performance in well-written applications with advanced control-flow.
    3214 C programmers should feel comfortable using these mechanisms for developing complex control-flow in applications, with the ability to obtain maximum available performance by selecting mechanisms at the appropriate level of need using only calling communication.
     2986Performance 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.
     2987C programmers should feel comfortable using these mechanisms for developing complex control-flow in applications, with the ability to obtain maximum available performance by selecting mechanisms at the appropriate level of need.
    32152988
    32162989
    … …  
    32323005\label{futur:nbio}
    32333006
    3234 Many modern workloads are not bound by computation but IO operations, common cases being web servers and XaaS~\cite{XaaS} (anything as a service).
     3007Many modern workloads are not bound by computation but IO operations, a common case being web servers and XaaS~\cite{XaaS} (anything as a service).
    32353008These types of workloads require significant engineering to amortizing costs of blocking IO-operations.
    32363009At 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.
    … …  
    32603033\section{Acknowledgements}
    32613034
    3262 The authors recognize the design assistance of Aaron Moss, Rob Schluntz, Andrew Beach, and Michael Brooks; David Dice for commenting and helping with the Java benchmarks; and Gregor Richards for helping with the Node.js benchmarks.
    3263 This research is funded by a grant from Waterloo-Huawei (\url{http://www.huawei.com}) Joint Innovation Lab. %, and Peter Buhr is partially funded by the Natural Sciences and Engineering Research Council of Canada.
     3035The authors would like to recognize the design assistance of Aaron Moss, Rob Schluntz, Andrew Beach and Michael Brooks on the features described in this paper.
     3036Funding 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.
    32643037
    32653038{%
    3266 \fontsize{9bp}{11.5bp}\selectfont%
     3039\fontsize{9bp}{12bp}\selectfont%
    32673040\bibliography{pl,local}
    32683041}%

  • doc/papers/concurrency/examples/Fib.py

    r5452673 rc886f4b  
    44        while True:
    55                fn = fn1 + fn2; fn2 = fn1; fn1 = fn; yield fn
     6
     7
    68
    79f1 = Fib()
    … …  
    1214# Local Variables: #
    1315# tab-width: 4 #
    14 # compile-command: "python3.7 Fib.py" #
     16# compile-command: "python3.5 Fib.py" #
    1517# End: #

  • doc/papers/concurrency/examples/Fib2.c

    r5452673 rc886f4b  
    11#include <stdio.h>
    22
     3void mary() {
     4        printf( "MARY\n" );
     5}
     6
    37#define FIB_INIT { 0 }
    4 typedef struct { int restart; int fn1, fn2; } Fib;
     8typedef struct { int next; int fn1, fn2; } Fib;
    59int fib( Fib * f ) {
    6         static void * states[] = { &&s0, &&s1, &&s2 };
    7         goto *states[f->restart];
    8   s0:
     10        static void * states[] = { &&s1, &&s2, &&s3 };
     11        goto *states[f->next];
     12  s1:
     13        mary();
    914        f->fn1 = 0;
    10         f->restart = 1;
     15        f->next = 1;
    1116        return f->fn1;
    12   s1:
     17  s2:
     18        mary();
    1319        f->fn2 = f->fn1;
    1420        f->fn1 = 1;
    15         f->restart = 2;
     21        f->next = 2;
    1622        return f->fn1;
    17   s2:;
     23  s3:;
     24        mary();
    1825        int fn = f->fn1 + f->fn2;
    1926        f->fn2 = f->fn1;

  • doc/papers/concurrency/examples/Fib2.py

    r5452673 rc886f4b  
    11def Fib():
    2     fn1, fn = 1, 0
     2    fn1, fn = 0, 1
    33    while True:
    4         yield fn
     4        yield fn1
    55        fn1, fn = fn, fn1 + fn
    66
    … …  
    1212# Local Variables: #
    1313# tab-width: 4 #
    14 # compile-command: "python3.7 Fib2.py" #
     14# compile-command: "python3.5 Fib2.py" #
    1515# End: #

  • doc/papers/concurrency/examples/Fib3.c

    r5452673 rc886f4b  
    22
    33typedef struct {
    4         int restart, fn1, fn;
     4        int fn1, fn;
     5        void * next;
    56} Fib;
    6 #define FibCtor { 0, 1, 0 }
     7#define FibCtor { 1, 0, NULL }
    78
    89Fib * comain( Fib * f ) {
    9         static void * states[] = {&&s0, &&s1};
    10         goto *states[f->restart];
    11   s0: f->restart = 1;
     10        if ( __builtin_expect(f->next != 0, 1) ) goto *f->next;
     11        f->next = &&s1;
    1212        for ( ;; ) {
    1313                return f;

  • doc/papers/concurrency/examples/FibRefactor.py

    r5452673 rc886f4b  
    2222# Local Variables: #
    2323# tab-width: 4 #
    24 # compile-command: "python3.7 FibRefactor.py" #
     24# compile-command: "python3.5 FibRefactor.py" #
    2525# End: #

  • doc/papers/concurrency/examples/Format.c

    r5452673 rc886f4b  
    22
    33typedef struct {
    4         int restart, g, b;
     4        void * next;
    55        char ch;
     6        int g, b;
    67} Fmt;
    78
    89void comain( Fmt * f ) {
    9         static void * states[] = {&&s0, &&s1};
    10         goto *states[f->restart];
    11   s0: f->restart = 1;
     10        if ( __builtin_expect(f->next != 0, 1) ) goto *f->next;
     11        f->next = &&s1;
    1212        for ( ;; ) {
    1313                for ( f->g = 0; f->g < 5; f->g += 1 ) {                 // groups
    1414                        for ( f->b = 0; f->b < 4; f->b += 1 ) {         // blocks
    15                                 do {
    16                                         return;  s1: ;
    17                                 } while ( f->ch == '\n' );                              // ignore
     15                                return;
     16                          s1:;  while ( f->ch == '\n' ) return;         // ignore
    1817                                printf( "%c", f->ch );                                  // print character
    1918                        }
    … …  
    2524
    2625int main() {
    27         Fmt fmt = { 0 };
     26        Fmt fmt = { NULL };
    2827        comain( &fmt );                                                                         // prime
    2928        for ( ;; ) {

  • doc/papers/concurrency/examples/Format.cc

    r5452673 rc886f4b  
    66                        for ( g = 0; g < 5; g += 1 ) { // groups of 5 blocks
    77                                for ( b = 0; b < 4; b += 1 ) { // blocks of 4 characters
    8                                         for ( ;; ) { // for newline characters
     8//                                      for ( ;; ) { // for newline characters
    99                                                suspend();
    10                                                 if ( ch != '\n' ) break; // ignore newline
    11                                         }
     10//                                              if ( ch != '\n' ) break; // ignore newline
     11//                                      }
    1212//                                      cout << ch; // print character
    1313                                }
    … …  
    3131// Local Variables: //
    3232// tab-width: 4 //
    33 // compile-command: "u++-work -O2 -nodebug Format.cc" //
     33// compile-command: "u++-work -O2 -nodebubg Format.cc" //
    3434// End: //

  • doc/papers/concurrency/examples/Format.cfa

    r5452673 rc886f4b  
    1111                for ( g = 0; g < 5; g += 1 ) {          // groups of 5 blocks
    1212                        for ( b = 0; b < 4; b += 1 ) {  // blocks of 4 characters
    13                                 do {
     13//                              do {
    1414                                        suspend();
    15                                 } while ( ch == '\n' || ch == '\t' );
     15//                              } while ( ch == '\n' || ch == '\t' );
    1616                                sout | ch;                                      // print character
    1717                        }

  • doc/papers/concurrency/examples/Format.data

    r5452673 rc886f4b  
    1 abcdefghijklmnop
    2 qrstuvwxyzx
    3 xxxxxxxxxxxxx
     1abcdefghijklmnopqrstuvwxyzxxxxxxxxxxxxxx

  • doc/papers/concurrency/examples/Format.py

    r5452673 rc886f4b  
    44                        for g in range( 5 ):    # groups of 5 blocks
    55                                for b in range( 4 ): # blocks of 4 characters
    6                                         while True:
    7                                                 ch = (yield) # receive from send
    8                                                 if '\n' not in ch:
    9                                                         break
    10                                         print( ch, end='' ) # receive from send
     6                                        print( (yield), end='' ) # receive from send
    117                                print( '  ', end='' ) # block separator
    128                        print()                                 # group separator
    … …  
    1511                        print()
    1612
    17 input = "abcdefghijklmnop\nqrstuvwx\nyzxxxxxxxxxxxxxx\n"
    18 
    1913fmt = Format()
    2014next( fmt )                                                     # prime generator
    21 for i in input:
    22         fmt.send( i );                          # send to yield
     15for i in range( 41 ):
     16        fmt.send( 'a' );                                # send to yield
    2317
    2418# Local Variables: #
    2519# tab-width: 4 #
    26 # compile-command: "python3.7 Format.py" #
     20# compile-command: "python3.5 Format.py" #
    2721# End: #

  • doc/papers/concurrency/examples/Format1.c

    r5452673 rc886f4b  
    22
    33typedef struct {
    4         int restart, g, b;
     4        void * next;
    55        char ch;
     6        int g, b;
    67} Fmt;
    78
    89void format( Fmt * f ) {
    9         static void * states[] = {&&s0, &&s1};
    10         goto *states[f->restart];
    11   s0: f->restart = 1;
     10        if ( __builtin_expect(f->next != 0, 1) ) goto *f->next;
     11        f->next = &&s1;
    1212        for ( ;; ) {
    1313                for ( f->g = 0; f->g < 5; f->g += 1 ) {                 // groups
    1414                        for ( f->b = 0; f->b < 4; f->b += 1 ) {         // blocks
    1515                                return;
    16                           s1: if ( f->ch == '\0' ) goto fini;           // EOF ?
     16                          s1: ;
     17                                if ( f->ch == '\0' ) goto fini;                 // EOF ?
    1718                                while ( f->ch == '\n' ) return;                 // ignore
    18 //                              printf( "%c", f->ch );                                  // print character
     19                                printf( "%c", f->ch );                                  // print character
    1920                        }
    20 //                      printf( " " );                                                          // block separator
     21                        printf( " " );                                                          // block separator
    2122                }
    22 //              printf( "\n" );                                                                 // group separator
     23                printf( "\n" );                                                                 // group separator
    2324        }
    24   fini:;
    25 //      if ( f->g != 0 || f->b != 0 ) printf( "\n" );
     25  fini:
     26        if ( f->g != 0 || f->b != 0 ) printf( "\n" );
    2627}
    2728
    2829int main() {
    29         Fmt fmt = { 0 };
     30        Fmt fmt = { NULL };
    3031        format( &fmt );                                                                         // prime
    31         fmt.ch = 'a';
    32         for ( long int i = 0; i < 1000000000; i += 1 ) {
    33 //              scanf( "%c", &fmt.ch );                                                 // direct read into communication variable
    34 //        if ( feof( stdin ) ) break;
     32        for ( ;; ) {
     33                scanf( "%c", &fmt.ch );                                                 // direct read into communication variable
     34          if ( feof( stdin ) ) break;
    3535                format( &fmt );
    3636        }
    37         fmt.ch = '\0';                                                                          // sentential (EOF)
     37        fmt.ch = '\0';
    3838        format( &fmt );
    3939}

  • doc/papers/concurrency/examples/PingPong.c

    r5452673 rc886f4b  
    22
    33typedef struct PingPong {
    4         int restart;                                                                            // style 1
     4        const char * name;
    55        int N, i;
    6         const char * name;
    76        struct PingPong * partner;
    8         void * next;                                                                            // style 2
     7        void * next;
    98} PingPong;
    10 #define PPCtor( name, N ) { 0, N, 0, name, NULL, NULL }
    11 
     9#define PPCtor( name, N ) { name, N, 0, NULL, NULL }
    1210void comain( PingPong * pp ) __attribute__(( noinline ));
    1311void comain( PingPong * pp ) {
     12        if ( __builtin_expect(pp->next != 0, 1) ) goto *pp->next;
    1413#if 0
    15         if ( __builtin_expect(pp->next != 0, 1) ) goto *pp->next;
     14        pp->next = &&here;
     15                asm( "mov  %0,%%rdi" : "=m" (pp) );
     16                asm( "mov  %rdi,%rax" );
     17#ifndef OPT
     18#ifdef PRINT
     19                asm( "add  $16, %rsp" );
     20#endif // PRINT
     21                asm( "popq %rbp" );
     22#endif // ! OPT
     23
     24#ifdef OPT
     25#ifdef PRINT
     26                asm( "popq %rbx" );
     27#endif // PRINT
     28#endif // OPT
     29                asm( "jmp  comain" );
     30  here: ;
     31#endif // 0
     32
    1633        pp->next = &&cycle;
    1734        for ( ; pp->i < pp->N; pp->i += 1 ) {
    … …  
    3653          cycle: ;
    3754        } // for
    38 #endif // 0
    39 
    40 #if 1
    41         static void * states[] = {&&s0, &&s1};
    42         goto *states[pp->restart];
    43   s0: pp->restart = 1;
    44         for ( ; pp->i < pp->N; pp->i += 1 ) {
    45 #ifdef PRINT
    46                 printf( "%s %d\n", pp->name, pp->i );
    47 #endif // PRINT
    48                 asm( "mov  %0,%%rdi" : "=m" (pp->partner) );
    49                 asm( "mov  %rdi,%rax" );
    50 #ifndef OPT
    51 #ifdef PRINT
    52                 asm( "add  $16, %rsp" );
    53 #endif // PRINT
    54                 asm( "popq %rbp" );
    55 #endif // ! OPT
    56 
    57 #ifdef OPT
    58 #ifdef PRINT
    59                 asm( "popq %rbx" );
    60 #endif // PRINT
    61 #endif // OPT
    62                 asm( "jmp  comain" );
    63           s1: ;
    64         } // for
    65 #endif // 0
    6655}
    6756
    … …  
    8170// Local Variables: //
    8271// tab-width: 4 //
    83 // compile-command: "gcc-9 -g -DPRINT PingPong.c" //
     72// compile-command: "gcc-8 -g -DPRINT PingPong.c" //
    8473// End: //

  • doc/papers/concurrency/examples/Pingpong.py

    r5452673 rc886f4b  
    11def PingPong( name, N ):
    2         partner = yield                         # get partner
    3         yield                                           # resume scheduler
     2        partner = (yield)           # get partner
     3        yield                       # resume scheduler
    44        for i in range( N ):
    55                print( name )
    6                 yield partner                   # execute next
     6                yield partner           # execute next
    77        print( "end", name )
    88
    99def Scheduler():
    10         n = yield                                       # starting coroutine
    11         try:
    12                 while True:
    13                         n = next( n )           # schedule coroutine
    14         except StopIteration:
    15                 pass
     10        n = (yield)                 # starting coroutine
     11        while True:
     12                n = next( n )           # schedule coroutine
    1613
    1714pi = PingPong( "ping", 5 )
    1815po = PingPong( "pong", 5 )
    19 next( pi )                                              # prime
    20 pi.send( po )                                   # send partner
    21 next( po )                                              # prime
    22 po.send( pi )                                   # send partner
     16next( pi )                      # prime
     17pi.send( po )                   # send partner
     18next( po )                      # prime
     19po.send( pi )                   # send partner
    2320
    2421s = Scheduler();
    25 next( s )                                               # prime
     22next( s )                       # prime
    2623try:
    2724        s.send( pi )                            # start cycle
    28 except StopIteration:                   # scheduler stopped
    29         pass
     25except StopIteration:
     26        print( "scheduler stop" )
    3027print( "stop" )
    3128
    3229# Local Variables: #
    3330# tab-width: 4 #
    34 # compile-command: "python3.7 Pingpong.py" #
     31# compile-command: "python3.5 Pingpong.py" #
    3532# End: #

  • doc/papers/concurrency/examples/ProdCons.py

    r5452673 rc886f4b  
    11def Prod( N ):
    2         cons = yield                            # get cons
    3         yield                                           # resume scheduler
     2        cons = (yield)              # get cons
     3        yield                       # resume scheduler
    44        for i in range( N ):
    55                print( "prod" )
    6                 yield cons                              # execute next
     6                yield cons              # execute next
    77        print( "end", "prod" )
    88
    99def Cons( N ):
    10         prod = yield                            # get prod
    11         yield                                           # resume scheduler
     10        prod = (yield)              # get prod
     11        yield                       # resume scheduler
    1212        for i in range( N ):
    1313                print( "cons" )
    14                 yield prod                              # execute next
     14                yield prod              # execute next
    1515        print( "end", "cons" )
    1616
    1717def Scheduler():
    18         n = yield                                       # starting coroutine
    19         try:
    20                 while True:
    21                         n = next( n )           # schedule coroutine
    22         except StopIteration:
    23                 pass
     18        n = (yield)                 # starting coroutine
     19        while True:
     20                n = next( n )           # schedule coroutine
    2421
    2522prod = Prod( 5 )
    2623cons = Cons( 5 )
    27 next( prod )                                    # prime
    28 prod.send( cons )                               # send cons
    29 next( cons )                                    # prime
    30 cons.send( prod )                               # send prod
     24next( prod )                    # prime
     25prod.send( cons )               # send cons
     26next( cons )                    # prime
     27cons.send( prod )               # send prod
    3128
    3229s = Scheduler();
    33 next( s )                                               # prime
     30next( s )                       # prime
    3431try:
    3532        s.send( prod )                          # start cycle
    36 except StopIteration:                   # scheduler stopped
    37         pass
     33except StopIteration:
     34        print( "scheduler stop" )
    3835print( "stop" )
    3936
    4037# Local Variables: #
    4138# tab-width: 4 #
    42 # compile-command: "python3.7 ProdCons.py" #
     39# compile-command: "python3.5 ProdCons.py" #
    4340# End: #

  • doc/papers/concurrency/examples/Refactor.py

    r5452673 rc886f4b  
    2626# Local Variables: #
    2727# tab-width: 4 #
    28 # compile-command: "python3.7 Refactor.py" #
     28# compile-command: "python3.5 Refactor.py" #
    2929# End: #

  • doc/papers/concurrency/figures/FullCoroutinePhases.fig

    r5452673 rc886f4b  
    88-2
    991200 2
    10 5 1 0 1 0 7 100 0 -1 0.000 0 0 1 0 5175.000 2437.500 4875 1875 5175 1800 5475 1875
     105 1 0 1 0 7 100 0 -1 0.000 0 0 1 0 4575.000 2437.500 4275 1875 4575 1800 4875 1875
    1111        1 1 1.00 45.00 90.00
    12 5 1 0 1 0 7 100 0 -1 0.000 0 0 1 0 5175.000 1537.500 5475 2100 5175 2175 4875 2100
     125 1 0 1 0 7 100 0 -1 0.000 0 0 1 0 4575.000 1537.500 4875 2100 4575 2175 4275 2100
    1313        1 1 1.00 45.00 90.00
    14 5 1 0 1 0 7 50 -1 -1 0.000 0 1 1 0 4807.500 1642.500 4725 1425 4575 1650 4800 1875
     145 1 0 1 0 7 50 -1 -1 0.000 0 1 1 0 4207.500 1642.500 4125 1425 3975 1650 4200 1875
    1515        1 1 1.00 45.00 90.00
    16 6 1575 1575 2700 2025
    17162 1 0 1 0 7 100 0 -1 0.000 0 0 -1 1 0 2
    1817        1 1 1.00 45.00 90.00
    … …  
    2120        1 1 1.00 45.00 90.00
    2221         2175 1575 2400 1800
     222 1 0 1 0 7 100 0 -1 0.000 0 0 -1 1 0 2
     23        1 1 1.00 45.00 90.00
     24         3300 1575 3300 1800
     252 1 0 1 0 7 100 0 -1 0.000 0 0 -1 1 0 2
     26        1 1 1.00 45.00 90.00
     27         3300 2025 3300 2250
     284 1 0 100 0 0 10 0.0000 2 105 555 2100 1200 creation\001
    23294 1 0 100 0 4 10 0.0000 2 165 300 1725 1950 ping\001
    24304 1 0 100 0 4 10 0.0000 2 135 360 2475 1950 pong\001
    25 -6
    26 6 3075 1575 4200 2025
    27 6 3075 1575 4200 2025
    28 2 1 0 1 0 7 100 0 -1 0.000 0 0 -1 1 0 2
    29         1 1 1.00 45.00 90.00
    30          3525 1575 3300 1800
    31 2 1 0 1 0 7 100 0 -1 0.000 0 0 -1 1 0 2
    32         1 1 1.00 45.00 90.00
    33          3675 1575 3900 1800
    34 4 1 0 100 0 4 10 0.0000 2 165 300 3225 1950 ping\001
    35 4 1 0 100 0 4 10 0.0000 2 135 360 3975 1950 pong\001
    36 -6
    37 -6
     314 1 0 100 0 4 10 0.0000 2 165 300 3300 1950 ping\001
     324 1 0 100 0 4 10 0.0000 2 135 360 3300 2400 pong\001
     334 1 0 100 0 0 10 0.0000 2 105 675 4575 1200 execution\001
     344 1 0 100 0 4 10 0.0000 2 165 300 4275 2025 ping\001
     354 1 0 100 0 4 10 0.0000 2 135 360 4875 2025 pong\001
     364 1 0 100 0 0 10 0.0000 2 90 420 3300 1200 starter\001
    38374 1 0 100 0 4 10 0.0000 2 165 705 2100 1500 pgm main\001
    39 4 1 0 100 0 4 10 0.0000 2 165 705 3600 1500 pgm main\001
    40 4 1 0 100 0 4 10 0.0000 2 165 300 4875 2025 ping\001
    41 4 1 0 100 0 4 10 0.0000 2 135 360 5475 2025 pong\001
    42 4 1 0 100 0 4 10 0.0000 2 165 705 5100 1500 pgm main\001
    43 4 1 0 100 0 2 10 0.0000 2 105 540 2100 1275 creator\001
    44 4 1 0 100 0 2 10 0.0000 2 105 495 3600 1275 starter\001
    45 4 1 0 100 0 2 10 0.0000 2 105 690 5175 1275 execution\001
     384 1 0 100 0 4 10 0.0000 2 165 705 3300 1500 pgm main\001
     394 1 0 100 0 4 10 0.0000 2 165 705 4500 1500 pgm main\001

  • doc/papers/concurrency/figures/RunTimeStructure.fig

    r5452673 rc886f4b  
    36361 3 0 1 -1 -1 0 0 20 0.000 1 0.0000 4500 3600 15 15 4500 3600 4515 3615
    3737-6
    38 6 3225 4125 4650 4425
    39 6 4350 4200 4650 4350
    40 1 3 0 1 -1 -1 0 0 20 0.000 1 0.0000 4425 4275 15 15 4425 4275 4440 4290
    41 1 3 0 1 -1 -1 0 0 20 0.000 1 0.0000 4500 4275 15 15 4500 4275 4515 4290
    42 1 3 0 1 -1 -1 0 0 20 0.000 1 0.0000 4575 4275 15 15 4575 4275 4590 4290
     386 2175 4650 7050 4950
     391 3 0 1 0 0 0 0 0 0.000 1 0.0000 2250 4830 30 30 2250 4830 2280 4860
     401 1 0 1 -1 -1 0 0 -1 0.000 1 0.0000 4200 4800 150 75 4200 4800 4350 4875
     411 3 0 1 -1 -1 0 0 -1 0.000 1 0.0000 3275 4800 100 100 3275 4800 3375 4800
     422 2 0 1 -1 -1 0 0 -1 0.000 0 0 0 0 0 5
     43         5400 4950 5400 4725 5175 4725 5175 4950 5400 4950
     442 2 1 1 -1 -1 0 0 -1 3.000 0 0 0 0 0 5
     45         6525 4950 6300 4950 6300 4725 6525 4725 6525 4950
     464 0 -1 0 0 0 10 0.0000 2 105 450 6600 4875 cluster\001
     474 0 -1 0 0 0 10 0.0000 2 105 660 5475 4875 processor\001
     484 0 -1 0 0 0 10 0.0000 2 105 555 4425 4875 monitor\001
     494 0 -1 0 0 0 10 0.0000 2 120 270 3450 4875 task\001
     504 0 -1 0 0 0 10 0.0000 2 105 660 2325 4875 coroutine\001
    4351-6
    44 1 1 0 1 -1 -1 0 0 -1 0.000 1 0.0000 3450 4275 225 150 3450 4275 3675 4425
    45 1 1 0 1 -1 -1 0 0 -1 0.000 1 0.0000 4050 4275 225 150 4050 4275 4275 4425
     526 3450 1275 3750 1425
     531 3 0 1 -1 -1 0 0 20 0.000 1 0.0000 3525 1350 15 15 3525 1350 3540 1365
     541 3 0 1 -1 -1 0 0 20 0.000 1 0.0000 3600 1350 15 15 3600 1350 3615 1365
     551 3 0 1 -1 -1 0 0 20 0.000 1 0.0000 3675 1350 15 15 3675 1350 3690 1365
    4656-6
    47 6 6675 4125 7500 4425
    48 6 7200 4200 7500 4350
    49 1 3 0 1 -1 -1 0 0 20 0.000 1 0.0000 7275 4275 15 15 7275 4275 7290 4290
    50 1 3 0 1 -1 -1 0 0 20 0.000 1 0.0000 7350 4275 15 15 7350 4275 7365 4290
    51 1 3 0 1 -1 -1 0 0 20 0.000 1 0.0000 7425 4275 15 15 7425 4275 7440 4290
    52 -6
    53 1 1 0 1 -1 -1 0 0 -1 0.000 1 0.0000 6900 4275 225 150 6900 4275 7125 4425
    54 -6
    55 6 6675 3525 8025 3975
    56 2 1 0 1 -1 -1 0 0 -1 0.000 0 0 -1 1 0 2
    57         1 1 1.00 45.00 90.00
    58          6675 3750 6975 3750
    59 2 1 0 1 -1 -1 0 0 -1 0.000 0 0 -1 1 0 2
    60         1 1 1.00 45.00 90.00
    61          7125 3750 7350 3750
    62 2 2 0 1 -1 -1 0 0 -1 0.000 0 0 0 0 0 5
    63          7800 3975 7800 3525 7350 3525 7350 3975 7800 3975
    64 2 1 0 1 -1 -1 0 0 -1 0.000 0 0 -1 1 0 2
    65         1 1 1.00 45.00 90.00
    66          7800 3750 8025 3750
     576 5550 1275 5850 1425
     581 3 0 1 -1 -1 0 0 20 0.000 1 0.0000 5625 1350 15 15 5625 1350 5640 1365
     591 3 0 1 -1 -1 0 0 20 0.000 1 0.0000 5700 1350 15 15 5700 1350 5715 1365
     601 3 0 1 -1 -1 0 0 20 0.000 1 0.0000 5775 1350 15 15 5775 1350 5790 1365
    6761-6
    68621 3 0 1 -1 -1 0 0 -1 0.000 1 0.0000 5550 2625 150 150 5550 2625 5700 2625
    … …  
    73671 3 0 1 -1 -1 0 0 -1 0.000 1 0.0000 4425 2850 150 150 4425 2850 4575 2850
    74681 3 0 1 -1 -1 0 0 -1 0.000 1 0.0000 4650 2475 150 150 4650 2475 4800 2475
     691 3 0 1 -1 -1 0 0 -1 0.000 1 0.0000 3525 3600 150 150 3525 3600 3675 3600
    75701 3 0 1 -1 -1 0 0 -1 0.000 1 0.0000 3975 3600 150 150 3975 3600 4125 3600
    76711 3 0 1 0 0 0 0 0 0.000 1 0.0000 3525 3600 30 30 3525 3600 3555 3630
    … …  
    79741 3 0 1 -1 -1 0 0 -1 0.000 1 0.0000 3975 2850 150 150 3975 2850 4125 2850
    80751 3 0 1 -1 -1 0 0 -1 0.000 1 0.0000 7200 2775 150 150 7200 2775 7350 2775
    81 1 3 0 1 0 0 0 0 0 0.000 1 0.0000 2250 4830 30 30 2250 4830 2280 4860
    82 1 3 0 1 0 0 0 0 0 0.000 1 0.0000 7200 2775 30 30 7200 2775 7230 2805
    83 1 3 0 1 -1 -1 0 0 -1 0.000 1 0.0000 3525 3600 150 150 3525 3600 3675 3600
    84 1 3 0 1 -1 -1 0 0 -1 0.000 1 0.0000 3875 4800 100 100 3875 4800 3975 4800
    85 1 1 0 1 -1 -1 0 0 -1 0.000 1 0.0000 4650 4800 150 75 4650 4800 4800 4875
     761 1 0 1 -1 -1 0 0 -1 0.000 1 0.0000 4650 1350 225 150 4650 1350 4875 1500
     771 1 0 1 -1 -1 0 0 -1 0.000 1 0.0000 5250 1350 225 150 5250 1350 5475 1500
     781 1 0 1 -1 -1 0 0 -1 0.000 1 0.0000 4050 1350 225 150 4050 1350 4275 1500
    86792 2 0 1 -1 -1 0 0 -1 0.000 0 0 0 0 0 5
    8780         2400 4200 2400 3750 1950 3750 1950 4200 2400 4200
    … …  
    1471402 1 0 1 -1 -1 0 0 -1 0.000 0 0 -1 1 0 2
    148141        1 1 1.00 45.00 90.00
     142         6675 3975 6975 3975
     1432 1 0 1 -1 -1 0 0 -1 0.000 0 0 -1 1 0 2
     144        1 1 1.00 45.00 90.00
    149145         7050 2775 6825 2775
    1501462 1 0 1 -1 -1 0 0 -1 0.000 0 0 -1 0 0 2
    151          6825 2775 6825 3750
     147         6825 2775 6825 3975
     1482 1 0 1 -1 -1 0 0 -1 0.000 0 0 -1 1 0 2
     149        1 1 1.00 45.00 90.00
     150         7125 3975 7350 3975
     1512 2 0 1 -1 -1 0 0 -1 0.000 0 0 0 0 0 5
     152         7800 4200 7800 3750 7350 3750 7350 4200 7800 4200
     1532 1 0 1 -1 -1 0 0 -1 0.000 0 0 -1 1 0 2
     154        1 1 1.00 45.00 90.00
     155         7800 3975 8025 3975
    1521562 1 0 1 -1 -1 0 0 -1 0.000 0 0 -1 1 0 4
    153157        1 1 1.00 45.00 90.00
    154          7875 3750 7875 2325 7200 2325 7200 2550
    155 2 2 0 1 -1 -1 0 0 -1 0.000 0 0 0 0 0 5
    156          5850 4950 5850 4725 5625 4725 5625 4950 5850 4950
    157 2 2 1 1 -1 -1 0 0 -1 3.000 0 0 0 0 0 5
    158          6975 4950 6750 4950 6750 4725 6975 4725 6975 4950
     158         7875 3975 7875 2325 7200 2325 7200 2550
    1591594 1 -1 0 0 0 10 0.0000 2 105 720 5550 4425 Processors\001
    1601604 1 -1 0 0 0 10 0.0000 2 120 1005 4200 3225 Blocked Tasks\001
    … …  
    1651654 1 -1 0 0 0 10 0.0000 2 105 990 2175 3525 Discrete-event\001
    1661664 1 -1 0 0 0 10 0.0000 2 135 795 2175 4350 preemption\001
    167 4 0 -1 0 0 0 10 0.0000 2 150 1290 2325 4875 genrator/coroutine\001
    168 4 0 -1 0 0 0 10 0.0000 2 120 270 4050 4875 task\001
    169 4 0 -1 0 0 0 10 0.0000 2 105 450 7050 4875 cluster\001
    170 4 0 -1 0 0 0 10 0.0000 2 105 660 5925 4875 processor\001
    171 4 0 -1 0 0 0 10 0.0000 2 105 555 4875 4875 monitor\001

  • doc/papers/concurrency/mail2

    r5452673 rc886f4b  
    2222Software: Practice and Experience Editorial Office
    2323
    24 
    25 
    26 Date: Tue, 12 Nov 2019 22:25:17 +0000
    27 From: Richard Jones <onbehalfof@manuscriptcentral.com>
    28 Reply-To: R.E.Jones@kent.ac.uk
    29 To: tdelisle@uwaterloo.ca, pabuhr@uwaterloo.ca
    30 Subject: Software: Practice and Experience - Decision on Manuscript ID
    31  SPE-19-0219
    32 
    33 12-Nov-2019
    34 
    35 Dear Dr Buhr,
    36 
    37 Many thanks for submitting SPE-19-0219 entitled "Advanced Control-flow and Concurrency in Cforall" to Software: Practice and Experience. The paper has now been reviewed and the comments of the referees are included at the bottom of this letter.
    38 
    39 The decision on this paper is that it requires substantial further work is required. The referees have a number of substantial concerns. All the reviewers found the submission very hard to read; two of the reviewers state that it needs very substantial restructuring. These concerns must be addressed before your submission can be considered further.
    40 
    41 A revised version of your manuscript that takes into account the comments of the referees will be reconsidered for publication.
    42 
    43 Please note that submitting a revision of your manuscript does not guarantee eventual acceptance, and that your revision will be subject to re-review by the referees before a decision is rendered.
    44 
    45 You have 90 days from the date of this email to submit your revision. If you are unable to complete the revision within this time, please contact me to request an extension.
    46 
    47 You can upload your revised manuscript and submit it through your Author Center. Log into https://mc.manuscriptcentral.com/spe  and enter your Author Center, where you will find your manuscript title listed under "Manuscripts with Decisions".
    48 
    49 When submitting your revised manuscript, you will be able to respond to the comments made by the referee(s) in the space provided.  You can use this space to document any changes you make to the original manuscript.
    50 
    51 If you feel that your paper could benefit from English language polishing, you may wish to consider having your paper professionally edited for English language by a service such as Wiley's at http://wileyeditingservices.com. Please note that while this service will greatly improve the readability of your paper, it does not guarantee acceptance of your paper by the journal.
    52  
    53 Once again, thank you for submitting your manuscript to Software: Practice and Experience and I look forward to receiving your revision.
    54 
    55 
    56 Sincerely,
    57 
    58 Prof. Richard Jones
    59 Software: Practice and Experience
    60 R.E.Jones@kent.ac.uk
    61 
    62 
    63 Referee(s)' Comments to Author:
    64 
    65 Reviewing: 1
    66 
    67 Comments to the Author
    68 This article presents the design and rationale behind the various
    69 threading and synchronization mechanisms of C-forall, a new low-level
    70 programming language.  This paper is very similar to a companion paper
    71 which I have also received: as the papers are similar, so will these
    72 reviews be --- in particular any general comments from the other
    73 review apply to this paper also.
    74 
    75 As far as I can tell, the article contains three main ideas: an
    76 asynchronous execution / threading model; a model for monitors to
    77 provide mutual exclusion; and an implementation.  The first two ideas
    78 are drawn together in Table 1: unfortunately this is on page 25 of 30
    79 pages of text. Implementation choices and descriptions are scattered
    80 throughout the paper - and the sectioning of the paper seems almost
    81 arbitrary.
    82 
    83 The article is about its contributions.  Simply adding feature X to
    84 language Y isn't by itself a contribution, (when feature X isn't
    85 already a contribution).  The contribution can be in the design: the
    86 motivation, the space of potential design options, the particular
    87 design chosen and the rationale for that choice, or the resulting
    88 performance.  For example: why support two kinds of generators as well
    89 as user-level threads?  Why support both low and high level
    90 synchronization constructs?  Similarly I would have found the article
    91 easier to follow if it was written top down, presenting the design
    92 principles, present the space of language features, justify chosen
    93 language features (and rationale) and those excluded, and then present
    94 implementation, and performance.
    95 
    96 Then the writing of the article is often hard to follow, to say the
    97 least. Two examples: section 3 "stateful functions" - I've some idea
    98 what that is (a function with Algol's "own" or C's "static" variables?
    99 but in fact the paper has a rather more specific idea than that. The
    100 top of page 3 throws a whole lot of defintions at the reader
    101 "generator" "coroutine" "stackful" "stackless" "symmetric"
    102 "asymmetric" without every stopping to define each one --- but then in
    103 footnote "C" takes the time to explain what C's "main" function is?  I
    104 cannot imagine a reader of this paper who doesn't know what "main" is
    105 in C; especially if they understand the other concepts already
    106 presented in the paper.  The start of section 3 then does the same
    107 thing: putting up a whole lot of definitions, making distinctions and
    108 comparisons, even talking about some runtime details, but the critical
    109 definition of a monitor doesn't appear until three pages later, at the
    110 start of section 5 on p15, lines 29-34 are a good, clear, description
    111 of what a monitor actually is.  That needs to come first, rather than
    112 being buried again after two sections of comparisons, discussions,
    113 implementations, and options that are ungrounded because they haven't
    114 told the reader what they are actually talking about.  First tell the
    115 reader what something is, then how they might use it (as programmers:
    116 what are the rules and restrictions) and only then start comparison
    117 with other things, other approaches, other languages, or
    118 implementations.
    119 
    120 The description of the implementation is similarly lost in the trees
    121 without ever really seeing the wood. Figure 19 is crucial here, but
    122 it's pretty much at the end of the paper, and comments about
    123 implementations are threaded throughout the paper without the context
    124 (fig 19) to understand what's going on.   The protocol for performance
    125 testing may just about suffice for C (although is N constantly ten
    126 million, or does it vary for each benchmark) but such evaluation isn't
    127 appropriate for garbage-collected or JITTed languages like Java or Go.
    128 
    129 other comments working through the paper - these are mostly low level
    130 and are certainly not comprehensive.
    131 
    132 p1 only a subset of C-forall extensions?
    133 
    134 p1 "has features often associated with object-oriented programming
    135 languages, such as constructors, destructors, virtuals and simple
    136 inheritance."   There's no need to quibble about this. Once a language
    137 has inheritance, it's hard to claim it's not object-oriented.
    138 
    139 
    140 p2 barging? signals-as-hints?
    141 
    142 p3 start your discussion of generations with a simple example of a
    143 C-forall generator.  Fig 1(b) might do: but put it inline instead of
    144 the python example - and explain the key rules and restrictions on the
    145 construct.  Then don't even start to compare with coroutines until
    146 you've presented, described and explained your coroutines...
    147 p3 I'd probably leave out the various "C" versions unless there are
    148 key points to make you can't make in C-forall. All the alternatives
    149 are just confusing.
    150 
    151 
    152 p4 but what's that "with" in Fig 1(B)
    153 
    154 p5 start with the high level features of C-forall generators...
    155 
    156 p5 why is the paper explaining networking protocols?
    157 
    158 p7 lines 1-9 (transforming generator to coroutine - why would I do any
    159 of this? Why would I want one instead of the other (do not use "stack"
    160 in your answer!)
    161 
    162 p10 last para "A coroutine must retain its last resumer to suspend
    163 back because the resumer is on a different stack. These reverse
    164 pointers allow suspend to cycle backwards, "  I've no idea what is
    165 going on here?  why should I care?  Shouldn't I just be using threads
    166 instead?  why not?
    167 
    168 p16 for the same reasons - what reasons?
    169 
    170 p17 if the multiple-monitor entry procedure really is novel, write a
    171 paper about that, and only about that.
    172 
    173 p23 "Loose Object Definitions" - no idea what that means.  in that
    174 section: you can't leave out JS-style dynamic properties.  Even in
    175 OOLs that (one way or another) allow separate definitions of methods
    176 (like Objective-C, Swift, Ruby, C#) at any time a runtime class has a
    177 fixed definition.  Quite why the detail about bit mask implementation
    178 is here anyway, I've no idea.
    179 
    180 p25 this cluster isn't a CLU cluster then?
    181 
    182 * conclusion should conclude the paper, not the related.
    183 
    184 
    185 Reviewing: 2
    186 
    187 Comments to the Author
    188 This paper describes the concurrency features of an extension of C (whose name I will write as "C\/" here, for convenience), including much design-level discussion of the coroutine- and monitor-based features and some microbenchmarks exploring the current implementation's performance. The key message of the latter is that the system's concurrency abstractions are much lighter-weight than the threading found in mainstream C or Java implementations.
    189 
    190 There is much description of the system and its details, but nothing about (non-artificial) uses of it. Although the microbenchmark data is encouraging, arguably not enough practical experience with the system has been reported here to say much about either its usability advantages or its performance.
    191 
    192 As such, the main contribution of the paper seem to be to document the existence of the described system and to provide a detailed design rationale and (partial) tutorial. I believe that could be of interest to some readers, so an acceptable manuscript is lurking in here somewhere.
    193 
    194 Unfortunately, at present the writing style is somewhere between unclear and infuriating. It omits to define terms; it uses needlessly many terms for what are apparently (but not clearly) the same things; it interrupts itself rather than deliver the natural consequent of whatever it has just said; and so on. Section 5 is particularly bad in these regards -- see my detailed comments below. Fairly major additional efforts will be needed to turn the present text into a digestible design-and-tutorial document. I suspect that a shorter paper could do this job better than the present manuscript, which is overwrought in parts.
    195 
    196 p2: lines 4--9 are a little sloppy. It is not the languages but their popular implementations which "adopt" the 1:1 kernel threading model.
    197 
    198 line 10: "medium work" -- "medium-sized work"?
    199 
    200 line 18: "is all sequential to the compiler" -- not true in modern compilers, and in 2004 H-J Boehm wrote a tech report describing exactly why ("Threads cannot be implemented as a library", HP Labs).
    201 
    202 line 20: "knows the optimization boundaries" -- I found this vague. What's an example?
    203 
    204 line 31: this paragraph has made a lot of claims. Perhaps forward-reference to the parts of the paper that discuss each one.
    205 
    206 line 33: "so the reader can judge if" -- this reads rather passive-aggressively. Perhaps better: "... to support our argument that..."
    207 
    208 line 41: "a dynamic partitioning mechanism" -- I couldn't tell what this meant
    209 
    210 p3. Presenting concept of a "stateful function" as a new language feature seems odd. In C, functions often have local state thanks to static local variables (or globals, indeed). Of course, that has several limitations. Can you perhaps present your contributions by enumerating these limitations? See also my suggestion below about a possible framing centred on a strawman.
    211 
    212 line 2: "an old idea that is new again" -- this is too oblique
    213 
    214 lines 2--15: I found this to be a word/concept soup. Stacks, closures, generators, stackless stackful, coroutine, symmetric, asymmetric, resume/suspend versus resume/resume... there needs to be a more gradual and structured way to introduce all this, and ideally one that minimises redundancy. Maybe present it as a series of "definitions" each with its own heading, e.g. "A closure is stackless if its local state has statically known fixed size"; "A generator simply means a stackless closure." And so on. Perhaps also strongly introduce the word "activate" as a direct contrast with resume and suspend. These are just a flavour of the sort of changes that might make this paragraph into something readable.
    215 
    216 Continuing the thought: I found it confusing that by these definitinos, a stackful closure is not a stack, even though logically the stack *is* a kind of closure (it is a representation of the current thread's continuation).
    217 
    218 lines 24--27: without explaining what the boost functor types mean, I don't think the point here comes across.
    219 
    220 line 34: "semantically coupled" -- I wasn't surew hat this meant
    221 
    222 p4: the point of Figure 1 (C) was not immediately clear. It seem to be showing how one might "compile down" Figure 1 (B). Or is that Figure 1 (A)?
    223 
    224 It's right that the incidental language features of the system are not front-and-centre, but I'd appreciate some brief glossing of non-C languages features as they appear. Examples are the square bracket notation, the pipe notation and the constructor syntax. These explanations could go in the caption of the figure which first uses them, perhaps. Overall I found the figure captions to be terse, and a missed opportunity to explain clearly what was going on.
    225 
    226 p5 line 23: "This restriction is removed..." -- give us some up-front summary of your contributions and the elements of the language design that will be talked about, so that this isn't an aside. This will reduce the "twisty passages" feeling that characterises much of the paper.
    227 
    228 line 40: "a killer asymmetric generator" -- this is stylistically odd, and the sentence about failures doesn't convincigly argue that C\/ will help with them. Have you any experience writing device drivers using C\/? Or any argument that the kinds of failures can be traced to the "stack-ripping" style that one is forced to use without coroutines? Also, a typo on line 41: "device drives". And saying "Windows/Linux" is sloppy... what does the cited paper actually say?
    229 
    230 p6 lines 13--23: this paragraph is difficult to understand. It seems to be talking about a control-flow pattern roughly equivalent to tail recursion. What is the high-level point, other than that this is possible?
    231 
    232 line 34: "which they call coroutines" -- a better way to make this point is presumably that the C++20 proposal only provides a specialised kind of coroutine, namely generators, despite its use of the more general word.
    233 
    234 line 47: "... due to dynamic stack allocation, execution..." -- this sentence doesn't scan. I suggest adding "and for" in the relevant places where currently there are only commas.
    235 
    236 p8 / Figure 5 (B) -- the GNU C extension of unary "&&" needs to be explained. The whole figure needs a better explanation, in fact.
    237 
    238 p9, lines 1--10: I wasn't sure this stepping-through really added much value. What are the truly important points to note about this code?
    239 
    240 p10: similarly, lines 3--27 again are somewhere between tedious and confusing. I'm sure the motivation and details of "starter semantics" can both be stated much more pithily.
    241 
    242 line 32: "a self-resume does not overwrite the last resumer" -- is this a hack or a defensible principled decision?
    243 
    244 p11: "a common source of errors" -- among beginners or among production code? Presumably the former.
    245 
    246 line 23: "with builtin and library" -- not sure what this means
    247 
    248 lines 31--36: these can be much briefer. The only important point here seems to be that coroutines cannot be copied.
    249 
    250 p12: line 1: what is a "task"? Does it matter?
    251 
    252 line 7: calling it "heap stack" seems to be a recipe for confusion. "Stack-and-heap" might be better, and contrast with "stack-and-VLS" perhaps. When "VLS" is glossed, suggest actually expanding its initials: say "length" not "size".
    253 
    254 line 21: are you saying "cooperative threading" is the same as "non-preemptive scheduling", or that one is a special case (kind) of the other? Both are defensible, but be clear.
    255 
    256 line 27: "mutual exclusion and synchronization" -- the former is a kind of the latter, so I suggest "and other forms of synchronization".
    257 
    258 line 30: "can either be a stackless or stackful" -- stray "a", but also, this seems to be switching from generic/background terminology to C\/-specific terminology.
    259 
    260 An expositional idea occurs: start the paper with a strawman naive/limited realisation of coroutines -- say, Simon Tatham's popular "Coroutines in C" web page -- and identify point by point what the limitations are and how C\/ overcomes them. Currently the presentation is often flat (lacking motivating contrasts) and backwards (stating solutions before problems). The foregoing approach might fix both of these.
    261 
    262 page 13: line 23: it seems a distraction to mention the Python feature here.
    263 
    264 p14 line 5: it seems odd to describe these as "stateless" just because they lack shared mutable state. It means the code itself is even more stateful. Maybe the "stack ripping" argument could usefully be given here.
    265 
    266 line 16: "too restrictive" -- would be good to have a reference to justify this, or at least give a sense of what the state-of-the-art performance in transactional memory systems is (both software and hardware)
    267 
    268 line 22: "simulate monitors" -- what about just *implementing* monitors? isn't that what these systems do? or is the point more about refining them somehow into something more specialised?
    269 
    270 p15: sections 4.1 and 4.2 seem adrift and misplaced. Split them into basic parts (which go earlier) and more advanced parts (e.g. barging, which can be explained later).
    271 
    272 line 31: "acquire/release" -- misses an opportunity to contrast the monitor's "enter/exit" abstraction with the less structured acquire/release of locks.
    273 
    274 p16 line 12: the "implicit" versus "explicit" point is unclear. Is it perhaps about the contract between an opt-in *discipline* and a language-enforced *guarantee*?
    275 
    276 line 28: no need to spend ages dithering about which one is default and which one is the explicit qualifier. Tell us what you decided, briefly justify it, and move on.
    277 
    278 p17: Figure 11: since the main point seems to be to highlight bulk acquire, include a comment which identifies the line where this is happening.
    279 
    280 line 2: "impossible to statically..." -- or dynamically. Doing it dynamically would be perfectly acceptable (locking is a dynamic operation after all)
    281 
    282 "guarantees acquisition order is consistent" -- assuming it's done in a single bulk acquire.
    283 
    284 p18: section 5.3: the text here is a mess. The explanations of "internal" versus "external" scheduling are unclear, and "signals as hints" is not explained. "... can cause thread starvation" -- means including a while loop, or not doing so? "There are three signalling mechanisms.." but the text does not follow that by telling us what they are. My own scribbled attempt at unpicking the internal/external thing: "threads already in the monitor, albeit waiting, have priority over those trying to enter".
    285 
    286 p19: line 3: "empty condition" -- explain that condition variables don't store anything. So being "empty" means that the queue of waiting threads (threads waiting to be signalled that the condition has become true) is empty.
    287 
    288 line 6: "... can be transformed into external scheduling..." -- OK, but give some motivation.
    289 
    290 p20: line 6: "mechnaism"
    291 
    292 lines 16--20: this is dense and can probably only be made clear with an example
    293 
    294 p21 line 21: clarify that nested monitor deadlock was describe earlier (in 5.2). (Is the repetition necessary?)
    295 
    296 line 27: "locks, and by extension monitors" -- this is true but the "by extension" argument is faulty. It is perfectly possible to use locks as a primitive and build a compositional mechanism out of them, e.g. transactions.
    297 
    298 p22 line 2: should say "restructured"
    299 
    300 line 33: "Implementing a fast subset check..." -- make clear that the following section explains how to do this. Restructuring the sections themselves could do this, or noting in the text.
    301 
    302 p23: line 3: "dynamic member adding, eg, JavaScript" -- needs to say "as permitted in JavaScript", and "dynamically adding members" is stylistically better
    303 
    304 p23: line 18: "urgent stack" -- back-reference to where this was explained before
    305 
    306 p24 line 7: I did not understand what was more "direct" about "direct communication". Also, what is a "passive monitor" -- just a monitor, given that monitors are passive by design?
    307 
    308 line 14 / section 5.9: this table was useful and it (or something like it) could be used much earlier on to set the structure of the rest of the paper. The explanation at present is too brief, e.g. I did not really understand the point about cases 7 and 8.
    309 
    310 p25 line 2: instead of casually dropping in a terse explanation for the newly intrdouced term "virtual processor", introduce it properly. Presumably the point is to give a less ambiguous meaning to "thread" by reserving it only for C\/'s green threads.
    311 
    312 Table 1: what does "No / Yes" mean?
    313 
    314 p26 line 15: "transforms user threads into fibres" -- a reference is needed to explain what "fibres" means... guessing it's in the sense of Adya et al.
    315 
    316 line 20: "Microsoft runtime" -- means Windows?
    317 
    318 lines 21--26: don't say "interrupt" to mean "signal", especially not without clear introduction. You can use "POSIX signal" to disambiguate from condition variables' "signal".
    319 
    320 p27 line 3: "frequency is usually long" -- that's a "time period" or "interval", not a frequency
    321 
    322 line 5: the lengthy quotation is not really necessary; just paraphrase the first sentence and move on.
    323 
    324 line 20: "to verify the implementation" -- I don't think that means what is intended
    325 
    326 Tables in section 7 -- too many significant figures. How many overall runs are described? What is N in each case?
    327 
    328 p29 line 2: "to eliminate this cost" -- arguably confusing since nowadays on commodity CPUs most of the benefits of inlining are not to do with call overheads, but from later optimizations enabled as a consequence of the inlining
    329 
    330 line 41: "a hierarchy" -- are they a hierarchy? If so, this could be explained earlier. Also, to say these make up "an integrated set... of control-flow features" verges on the tautologous.
    331 
    332 p30 line 15: "a common case being web servers and XaaS" -- that's two cases
    333 
    334 
    335 Reviewing: 3
    336 
    337 Comments to the Author
    338 # Cforall review
    339 
    340 Overall, I quite enjoyed reading the paper. Cforall has some very interesting ideas. I did have some suggestions that I think would be helpful before final publication. I also left notes on various parts of the paper that I find confusing when reading, in hopes that it may be useful to you.
    341 
    342 ## Summary
    343 
    344 * Expand on the motivations for including both generator and coroutines, vs trying to build one atop the other
    345 * Expand on the motivations for having Why both symmetric and asymettric coroutines?
    346 * Comparison to async-await model adopted by other languages
    347     * C#, JS
    348     * Rust and its async/await model
    349 * Consider performance comparisons against node.js and Rust frameworks
    350 * Discuss performance of monitors vs finer-grained memory models and atomic operations found in other languages
    351 * Why both internal/external scheduling for synchronization?
    352 
    353 ## Generator/coroutines
    354 
    355 In general, this section was clear, but I thought it would be useful to provide a somewhat deeper look into why Cforall opted for the particular combination of features that it offers. I see three main differences from other languages:
    356 
    357 * Generators are not exposed as a "function" that returns a generator object, but rather as a kind of struct, with communication happening via mutable state instead of "return values". That is, the generator must be manually resumed and (if I understood) it is expected to store values that can then later be read (perhaps via methods), instead of having a `yield <Expr>` statement that yields up a value explicitly.
    358 * Both "symmetric" and "asymmetric" generators are supported, instead of only asymmetric.
    359 * Coroutines (multi-frame generators) are an explicit mechanism.
    360 
    361 In most other languages, coroutines are rather built by layering single-frame generators atop one another (e.g., using a mechanism like async-await), and symmetric coroutines are basically not supported. I'd like to see a bit more justification for Cforall including all the above mechanisms -- it seemed like symmetric coroutines were a useful building block for some of the user-space threading and custom scheduler mechanisms that were briefly mentioned later in the paper.
    362 
    363 In the discussion of coroutines, I would have expected a bit more of a comparison to the async-await mechanism offered in other languages. Certainly the semantics of async-await in JavaScript implies significantly more overhead (because each async fn is a distinct heap object). [Rust's approach avoids this overhead][zc], however, and might be worthy of a comparison (see the Performance section).
    364 
    365 ## Locks and threading
    366 
    367 ### Comparison to atomics overlooks performance
    368 
    369 There are several sections in the paper that compare against atomics -- for example, on page 15, the paper shows a simple monitor that encapsulates an integer and compares that to C++ atomics. Later, the paper compares the simplicity of monitors against the `volatile` quantifier from Java. The conclusion in section 8 also revisits this point.
    370 
    371 While I agree that monitors are simpler, they are obviously also significantly different from a performance perspective -- the paper doesn't seem to address this at all. It's plausible that (e.g.) the `Aint` monitor type described in the paper can be compiled and mapped to the specialized instructions offered by hardware, but I didn't see any mention of how this would be done. There is also no mention of the more nuanced memory ordering relations offered by C++11 and how one might achieve similar performance characteristics in Cforall (perhaps the answer is that one simply doesn't need to; I think that's defensible, but worth stating explicitly).
    372 
    373 ### Justification for external scheduling feels lacking
    374 
    375 Cforall includes both internal and external scheduling; I found the explanation for the external scheduling mechanism to be lacking in justification. Why include both mechanisms when most languages seem to make do with only internal scheduling? It would be useful to show some scenarios where external scheduling is truly more powerful.
    376 
    377 I would have liked to see some more discussion of external scheduling and how it  interacts with software engineering best practices. It seems somewhat similar to AOP in certain regards. It seems to add a bit of "extra semantics" to monitor methods, in that any method may now also become a kind of synchronization point. The "open-ended" nature of this feels like it could easily lead to subtle bugs, particularly when code refactoring occurs (which may e.g. split an existing method into two). This seems particularly true if external scheduling can occur across compilation units -- the paper suggested that this is true, but I wasn't entirely clear.
    378 
    379 I would have also appreciated a few more details on how external scheduling is implemented. It seems to me that there must be some sort of "hooks" on mutex methods so that they can detect whether some other function is waiting on them and awaken those blocked threads. I'm not sure how such hooks are inserted, particularly across compilation units. The material in Section 5.6 didn't quite clarify the matter for me. For example, it left me somewhat confused about whether the `f` and `g` functions declared were meant to be local to a translation unit, or shared with other unit.
    380 
    381 ### Presentation of monitors is somewhat confusing
    382 
    383 I found myself confused fairly often in the section on monitors. I'm just going to leave some notes here on places that I got confused in how that it could be useful to you as feedback on writing that might want to be clarified.
    384 
    385 To start, I did not realize that the `mutex_opt` notation was a keyword, I thought it was a type annotation. I think this could be called out more explicitly.
    386 
    387 Later, in section 5.2, the paper discusses `nomutex` annotations, which initially threw me, as they had not been introduced (now I realize that this paragraph is there to justify why there is no such keyword). The paragraph might be rearranged to make that clearer, perhaps by leading with the choice that Cforall made.
    388 
    389 On page 17, the paper states that "acquiring multiple monitors is safe from deadlock", but this could be stated a bit more precisely: acquiring multiple monitors in a bulk-acquire is safe from deadlock (deadlock can still result from nested acquires).
    390 
    391 On page 18, the paper states that wait states do not have to be enclosed in loops, as there is no concern of barging. This seems true but there are also other reasons to use loops (e.g., if there are multiple reasons to notify on the same condition). Thus the statement initially surprised me, as barging is only one of many reasons that I typically employ loops around waits.
    392 
    393 I did not understand the diagram in Figure 12 for some time. Initially, I thought that it was generic to all monitors, and I could not understand the state space. It was only later that I realized it was specific to your example. Updating the caption from "Monitor scheduling to "Monitor scheduling in the example from Fig 13" might have helped me quite a bit.
    394 
    395 I spent quite some time reading the boy/girl dating example (\*) and I admit I found it somewhat confusing. For example, I couldn't tell whether there were supposed to be many "girl" threads executing at once, or if there was only supposed to be one girl and one boy thread executing in a loop. Are the girl/boy threads supposed to invoke the girl/boy methods or vice versa? Surely there is some easier way to set this up? I believe that when reading the paper I convinced myself of how it was supposed to be working, but I'm writing this review some days later, and I find myself confused all over again and not able to easily figure it out.
    396 
    397 (\*) as an aside, I would consider modifying the example to some other form of matching, like customers and support personnel.
    398 
    399 ## Related work
    400 
    401 The paper offered a number of comparisons to Go, C#, Scala, and so forth, but seems to have overlooked another recent language, Rust. In many ways, Rust seems to be closest in philosophy to Cforall, so it seems like an odd omission. I already mentioned above that Rust is in the process of shipping [async-await syntax][aa], which is definitely an alternative to the generator/coroutine approach in Cforall (though one with clear pros/cons).
    402 
    403 ## Performance
    404 
    405 In the performance section in particular, you might consider comparing against some of the Rust web servers and threading systems. For example, actix is top of the [single query TechEmpower Framework benchmarks], and tokio is near the top of the [plainthreading benchmarks][pt] (hyper, the top, is more of an HTTP framework, though it is also written in Rust). It would seem worth trying to compare their "context switching" costs as well -- I believe both actix and tokio have a notion of threads that could be readily compared.
    406 
    407 Another addition that might be worth considering is to compare against node.js promises, although I think the comparison to process creation is not as clean.
    408 
    409 That said, I think that the performance comparison is not a big focus of the paper, so it may not be necessary to add anything to it.
    410 
    411 ## Authorship of this review
    412 
    413 I'm going to sign this review. This review was authored by Nicholas D. Matsakis. In the intrerest of full disclosure, I'm heavily involved in the Rust project, although I dont' think that influenced this review in particular. Feel free to reach out to me for clarifying questions.
    414 
    415 ## Links
    416 
    417 [aa]: https://blog.rust-lang.org/2019/09/30/Async-await-hits-beta.html
    418 [zc]: https://aturon.github.io/blog/2016/08/11/futures/
    419 [sq]: https://www.techempower.com/benchmarks/#section=data-r18&hw=ph&test=db
    420 [pt]: https://www.techempower.com/benchmarks/#section=data-r18&hw=ph&test=plaintext
    421 
    422 
    423 
    424 Subject: Re: manuscript SPE-19-0219
    425 To: "Peter A. Buhr" <pabuhr@uwaterloo.ca>
    426 From: Richard Jones <R.E.Jones@kent.ac.uk>
    427 Date: Tue, 12 Nov 2019 22:43:55 +0000
    428 
    429 Dear Dr Buhr
    430 
    431 Your should have received a decision letter on this today. I am sorry that this
    432 has taken so long. Unfortunately SP&E receives a lot of submissions and getting
    433 reviewers is a perennial problem.
    434 
    435 Regards
    436 Richard
    437 
    438 Peter A. Buhr wrote on 11/11/2019 13:10:
    439 >     26-Jun-2019
    440 >     Your manuscript entitled "Advanced Control-flow and Concurrency in Cforall"
    441 >     has been received by Software: Practice and Experience. It will be given
    442 >     full consideration for publication in the journal.
    443 >
    444 > Hi, it has been over 4 months since submission of our manuscript SPE-19-0219
    445 > with no response.
    446 >
    447 > Currently, I am refereeing a paper for IEEE that already cites our prior SP&E
    448 > paper and the Master's thesis forming the bases of the SP&E paper under
    449 > review. Hence our work is apropos and we want to get it disseminates as soon as
    450 > possible.
    451 >
    452 > [3] A. Moss, R. Schluntz, and P. A. Buhr, "Cforall: Adding modern programming
    453 >      language features to C," Software - Practice and Experience, vol. 48,
    454 >      no. 12, pp. 2111-2146, 2018.
    455 >
    456 > [4] T. Delisle, "Concurrency in C for all," Master's thesis, University of
    457 >      Waterloo, 2018.  [Online].  Available:
    458 >      https://uwspace.uwaterloo.ca/bitstream/handle/10012/12888
    459 
    460 
    461 
    462 Date: Mon, 13 Jan 2020 05:33:15 +0000
    463 From: Richard Jones <onbehalfof@manuscriptcentral.com>
    464 Reply-To: R.E.Jones@kent.ac.uk
    465 To: pabuhr@uwaterloo.ca
    466 Subject: Revision reminder - SPE-19-0219
    467 
    468 13-Jan-2020
    469 Dear Dr Buhr
    470 SPE-19-0219
    471 
    472 This is a reminder that your opportunity to revise and re-submit your
    473 manuscript will expire 28 days from now. If you require more time please
    474 contact me directly and I may grant an extension to this deadline, otherwise
    475 the option to submit a revision online, will not be available.
    476 
    477 I look forward to receiving your revision.
    478 
    479 Sincerely,
    480 
    481 Prof. Richard Jones
    482 Editor, Software: Practice and Experience
    483 https://mc.manuscriptcentral.com/spe
    484 
    485 
    486 
    487 Date: Wed, 5 Feb 2020 04:22:18 +0000
    488 From: Aaron Thomas <onbehalfof@manuscriptcentral.com>
    489 Reply-To: speoffice@wiley.com
    490 To: tdelisle@uwaterloo.ca, pabuhr@uwaterloo.ca
    491 Subject: SPE-19-0219.R1 successfully submitted
    492 
    493 04-Feb-2020
    494 
    495 Dear Dr Buhr,
    496 
    497 Your manuscript entitled "Advanced Control-flow and Concurrency in Cforall" has
    498 been successfully submitted online and is presently being given full
    499 consideration for publication in Software: Practice and Experience.
    500 
    501 Your manuscript number is SPE-19-0219.R1.  Please mention this number in all
    502 future correspondence regarding this submission.
    503 
    504 You can view the status of your manuscript at any time by checking your Author
    505 Center after logging into https://mc.manuscriptcentral.com/spe.  If you have
    506 difficulty using this site, please click the 'Get Help Now' link at the top
    507 right corner of the site.
    508 
    509 Thank you for submitting your manuscript to Software: Practice and Experience.
    510 
    511 Sincerely,
    512 Software: Practice and Experience Editorial Office
    513 

  • src/Parser/parser.yy

    r5452673 rc886f4b  
    1010// Created On       : Sat Sep  1 20:22:55 2001
    1111// Last Modified By : Peter A. Buhr
    12 // Last Modified On : Wed Feb 26 14:27:39 2020
    13 // Update Count     : 4472
     12// Last Modified On : Fri Feb 21 14:47:29 2020
     13// Update Count     : 4468
    1414//
    1515
    … …  
    20772077aggregate_control:                                                                              // CFA
    20782078        GENERATOR
    2079                 { SemanticError( yylloc, "generator is currently unimplemented." ); $$ = AggregateDecl::NoAggregate; }
    2080         | MONITOR GENERATOR
    2081                 { SemanticError( yylloc, "monitor generator is currently unimplemented." ); $$ = AggregateDecl::NoAggregate; }
     2079                { yyy = true; $$ = AggregateDecl::Coroutine; }
    20822080        | COROUTINE
    20832081                { yyy = true; $$ = AggregateDecl::Coroutine; }
    20842082        | MONITOR
    20852083                { yyy = true; $$ = AggregateDecl::Monitor; }
    2086         | MONITOR COROUTINE
    2087                 { SemanticError( yylloc, "monitor coroutine is currently unimplemented." ); $$ = AggregateDecl::NoAggregate; }
    20882084        | THREAD
    20892085                { yyy = true; $$ = AggregateDecl::Thread; }
    2090         | MONITOR THREAD
    2091                 { SemanticError( yylloc, "monitor thread is currently unimplemented." ); $$ = AggregateDecl::NoAggregate; }
    20922086        ;
    20932087
Note: See TracChangeset for help on using the changeset viewer.