Changeset 223a633 for doc


Ignore:
Timestamp:
Oct 15, 2020, 3:41:38 PM (11 months ago)
Author:
Thierry Delisle <tdelisle@…>
Branches:
arm-eh, jacob/cs343-translation, master, new-ast-unique-expr
Children:
b9537e6
Parents:
33c3ded (diff), 0b18db7 (diff)
Note: this is a merge changeset, the changes displayed below correspond to the merge itself.
Use the (diff) links above to see all the changes relative to each parent.
Message:

Merge branch 'master' of plg.uwaterloo.ca:software/cfa/cfa-cc

Location:
doc
Files:
13 added
12 edited
23 moved

Legend:

Unmodified
Added
Removed
  • doc/LaTeXmacros/common.tex

    r33c3ded r223a633  
    1111%% Created On       : Sat Apr  9 10:06:17 2016
    1212%% Last Modified By : Peter A. Buhr
    13 %% Last Modified On : Fri Sep  4 13:56:52 2020
    14 %% Update Count     : 383
     13%% Last Modified On : Mon Oct  5 09:34:46 2020
     14%% Update Count     : 464
    1515%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
    1616
     
    5555\newlength{\parindentlnth}
    5656\setlength{\parindentlnth}{\parindent}
    57 
    58 \newcommand{\LstBasicStyle}[1]{{\lst@basicstyle{#1}}}
    59 \newcommand{\LstKeywordStyle}[1]{{\lst@basicstyle{\lst@keywordstyle{#1}}}}
    60 \newcommand{\LstCommentStyle}[1]{{\lst@basicstyle{\lst@commentstyle{#1}}}}
    61 
    62 \newlength{\gcolumnposn}                                % temporary hack because lstlisting does not handle tabs correctly
    63 \newlength{\columnposn}
    64 \setlength{\gcolumnposn}{2.5in}
    65 \setlength{\columnposn}{\gcolumnposn}
    66 \newcommand{\C}[2][\@empty]{\ifx#1\@empty\else\global\setlength{\columnposn}{#1}\global\columnposn=\columnposn\fi\hfill\makebox[\textwidth-\columnposn][l]{\lst@basicstyle{\LstCommentStyle{#2}}}}
    67 \newcommand{\CRT}{\global\columnposn=\gcolumnposn}
    68 
    69 % allow escape sequence in lstinline
    70 %\usepackage{etoolbox}
    71 %\patchcmd{\lsthk@TextStyle}{\let\lst@DefEsc\@empty}{}{}{\errmessage{failed to patch}}
    7257
    7358\usepackage{pslatex}                                    % reduce size of san serif font
     
    244229\usepackage{listings}                                                                   % format program code
    245230\usepackage{lstlang}
    246 
    247 \newcommand{\CFADefaults}{%
     231\makeatletter
     232
     233\newcommand{\LstBasicStyle}[1]{{\lst@basicstyle{#1}}}
     234\newcommand{\LstKeywordStyle}[1]{{\lst@basicstyle{\lst@keywordstyle{#1}}}}
     235\newcommand{\LstCommentStyle}[1]{{\lst@basicstyle{\lst@commentstyle{#1}}}}
     236
     237\newlength{\gcolumnposn}                                % temporary hack because lstlisting does not handle tabs correctly
     238\newlength{\columnposn}
     239\setlength{\gcolumnposn}{2.75in}
     240\setlength{\columnposn}{\gcolumnposn}
     241\newcommand{\C}[2][\@empty]{\ifx#1\@empty\else\global\setlength{\columnposn}{#1}\global\columnposn=\columnposn\fi\hfill\makebox[\textwidth-\columnposn][l]{\lst@basicstyle{\LstCommentStyle{#2}}}}
     242\newcommand{\CRT}{\global\columnposn=\gcolumnposn}
     243
     244% allow escape sequence in lstinline
     245%\usepackage{etoolbox}
     246%\patchcmd{\lsthk@TextStyle}{\let\lst@DefEsc\@empty}{}{}{\errmessage{failed to patch}}
     247
     248% allow adding to lst literate
     249\def\addToLiterate#1{\protect\edef\lst@literate{\unexpanded\expandafter{\lst@literate}\unexpanded{#1}}}
     250\lst@Key{add to literate}{}{\addToLiterate{#1}}
     251\makeatother
     252
     253\newcommand{\CFAStyle}{%
    248254\lstset{
    249 language=CFA,
    250255columns=fullflexible,
    251256basicstyle=\linespread{0.9}\sf,                 % reduce line spacing and use sanserif font
     
    262267belowskip=3pt,
    263268% replace/adjust listing characters that look bad in sanserif
    264 literate={-}{\makebox[1ex][c]{\raisebox{0.4ex}{\rule{0.8ex}{0.1ex}}}}1 {^}{\raisebox{0.6ex}{$\scriptscriptstyle\land\,$}}1
     269literate={-}{\makebox[1ex][c]{\raisebox{0.4ex}{\rule{0.75ex}{0.1ex}}}}1 {^}{\raisebox{0.6ex}{$\scriptscriptstyle\land\,$}}1
    265270        {~}{\raisebox{0.3ex}{$\scriptstyle\sim\,$}}1 {`}{\ttfamily\upshape\hspace*{-0.1ex}`}1
    266271        {<-}{$\leftarrow$}2 {=>}{$\Rightarrow$}2 {->}{\makebox[1ex][c]{\raisebox{0.4ex}{\rule{0.8ex}{0.075ex}}}\kern-0.2ex\textgreater}2,
    267 moredelim=**[is][\color{red}]{?}{?},    % red highlighting ?...? (registered trademark symbol) emacs: C-q M-.
     272}% lstset
     273}% CFAStyle
     274
     275\ifdefined\CFALatin% extra Latin-1 escape characters
     276\lstnewenvironment{cfa}[1][]{
     277\lstset{
     278language=CFA,
     279moredelim=**[is][\color{red}]{®}{®},    % red highlighting ®...® (registered trademark symbol) emacs: C-q M-.
    268280moredelim=**[is][\color{blue}]{ß}{ß},   % blue highlighting ß...ß (sharp s symbol) emacs: C-q M-_
    269281moredelim=**[is][\color{OliveGreen}]{¢}{¢}, % green highlighting ¢...¢ (cent symbol) emacs: C-q M-"
    270282moredelim=[is][\lstset{keywords={}}]{¶}{¶}, % keyword escape ¶...¶ (pilcrow symbol) emacs: C-q M-^
     283% replace/adjust listing characters that look bad in sanserif
     284add to literate={`}{\ttfamily\upshape\hspace*{-0.1ex}`}1
    271285}% lstset
    272 }% CFADefaults
    273 \newcommand{\CFAStyle}{%
    274 \CFADefaults
     286\lstset{#1}
     287}{}
    275288% inline code ©...© (copyright symbol) emacs: C-q M-)
    276289\lstMakeShortInline©                                    % single-character for \lstinline
    277 }% CFAStyle
    278 
    279 \lstnewenvironment{cfa}[1][]
    280 {\CFADefaults\lstset{#1}}
    281 {}
     290\else% regular ASCI characters
     291\lstnewenvironment{cfa}[1][]{
     292\lstset{
     293language=CFA,
     294escapechar=\$,                                                  % LaTeX escape in CFA code
     295moredelim=**[is][\color{red}]{@}{@},    % red highlighting @...@
     296}% lstset
     297\lstset{#1}
     298}{}
     299% inline code @...@ (at symbol)
     300\lstMakeShortInline@                                    % single-character for \lstinline
     301\fi%
    282302
    283303% Local Variables: %
  • doc/LaTeXmacros/lstlang.sty

    r33c3ded r223a633  
    88%% Created On       : Sat May 13 16:34:42 2017
    99%% Last Modified By : Peter A. Buhr
    10 %% Last Modified On : Tue Jan  8 14:40:33 2019
    11 %% Update Count     : 21
     10%% Last Modified On : Wed Sep 23 22:40:04 2020
     11%% Update Count     : 24
    1212%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
    1313
     
    115115                auto, _Bool, catch, catchResume, choose, _Complex, __complex, __complex__, __const, __const__,
    116116                coroutine, disable, dtype, enable, exception, __extension__, fallthrough, fallthru, finally,
    117                 __float80, float80, __float128, float128, forall, ftype, _Generic, _Imaginary, __imag, __imag__,
     117                __float80, float80, __float128, float128, forall, ftype, generator, _Generic, _Imaginary, __imag, __imag__,
    118118                inline, __inline, __inline__, __int128, int128, __label__, monitor, mutex, _Noreturn, one_t, or,
    119                 otype, restrict, __restrict, __restrict__, __signed, __signed__, _Static_assert, thread,
     119                otype, restrict, __restrict, __restrict__, __signed, __signed__, _Static_assert, suspend, thread,
    120120                _Thread_local, throw, throwResume, timeout, trait, try, ttype, typeof, __typeof, __typeof__,
    121121                virtual, __volatile, __volatile__, waitfor, when, with, zero_t,
     
    125125
    126126% C++ programming language
    127 \lstdefinelanguage{C++}[ANSI]{C++}{}
     127\lstdefinelanguage{C++}[ANSI]{C++}{
     128        morekeywords={nullptr,}
     129}
    128130
    129131% uC++ programming language, based on ANSI C++
  • doc/bibliography/pl.bib

    r33c3ded r223a633  
    10051005    key         = {Cforall Benchmarks},
    10061006    author      = {{\textsf{C}{$\mathbf{\forall}$} Benchmarks}},
    1007     howpublished= {\href{https://plg.uwaterloo.ca/~cforall/doc/CforallConcurrentBenchmarks.tar}{https://\-plg.uwaterloo.ca/\-$\sim$cforall/\-doc/\-CforallConcurrentBenchmarks.tar}},
     1007    howpublished= {\href{https://github.com/cforall/ConcurrentBenchmarks_SPE20}{https://\-github.com/\-cforall/\-ConcurrentBenchmarks\_SPE20}},
    10081008}
    10091009
     
    19731973    title       = {Cooperating Sequential Processes},
    19741974    institution = {Technological University},
    1975     address     = {Eindhoven, Netherlands},
     1975    address     = {Eindhoven, Neth.},
    19761976    year        = 1965,
    19771977    note        = {Reprinted in \cite{Genuys68} pp. 43--112.}
  • doc/papers/concurrency/Paper.tex

    r33c3ded r223a633  
    224224{}
    225225\lstnewenvironment{C++}[1][]                            % use C++ style
    226 {\lstset{language=C++,moredelim=**[is][\protect\color{red}]{`}{`},#1}\lstset{#1}}
     226{\lstset{language=C++,moredelim=**[is][\protect\color{red}]{`}{`}}\lstset{#1}}
    227227{}
    228228\lstnewenvironment{uC++}[1][]
    229 {\lstset{language=uC++,moredelim=**[is][\protect\color{red}]{`}{`},#1}\lstset{#1}}
     229{\lstset{language=uC++,moredelim=**[is][\protect\color{red}]{`}{`}}\lstset{#1}}
    230230{}
    231231\lstnewenvironment{Go}[1][]
    232 {\lstset{language=Golang,moredelim=**[is][\protect\color{red}]{`}{`},#1}\lstset{#1}}
     232{\lstset{language=Golang,moredelim=**[is][\protect\color{red}]{`}{`}}\lstset{#1}}
    233233{}
    234234\lstnewenvironment{python}[1][]
    235 {\lstset{language=python,moredelim=**[is][\protect\color{red}]{`}{`},#1}\lstset{#1}}
     235{\lstset{language=python,moredelim=**[is][\protect\color{red}]{`}{`}}\lstset{#1}}
    236236{}
    237237\lstnewenvironment{java}[1][]
    238 {\lstset{language=java,moredelim=**[is][\protect\color{red}]{`}{`},#1}\lstset{#1}}
     238{\lstset{language=java,moredelim=**[is][\protect\color{red}]{`}{`}}\lstset{#1}}
    239239{}
    240240
     
    284284
    285285\begin{document}
    286 \linenumbers                            % comment out to turn off line numbering
     286%\linenumbers                           % comment out to turn off line numbering
    287287
    288288\maketitle
     
    450450\hline
    451451stateful                        & thread        & \multicolumn{1}{c|}{No} & \multicolumn{1}{c}{Yes} \\
    452 \hline   
    453 \hline   
     452\hline
     453\hline
    454454No                                      & No            & \textbf{1}\ \ \ @struct@                              & \textbf{2}\ \ \ @mutex@ @struct@              \\
    455 \hline   
     455\hline
    456456Yes (stackless)         & No            & \textbf{3}\ \ \ @generator@                   & \textbf{4}\ \ \ @mutex@ @generator@   \\
    457 \hline   
     457\hline
    458458Yes (stackful)          & No            & \textbf{5}\ \ \ @coroutine@                   & \textbf{6}\ \ \ @mutex@ @coroutine@   \\
    459 \hline   
     459\hline
    460460No                                      & Yes           & \textbf{7}\ \ \ {\color{red}rejected} & \textbf{8}\ \ \ {\color{red}rejected} \\
    461 \hline   
     461\hline
    462462Yes (stackless)         & Yes           & \textbf{9}\ \ \ {\color{red}rejected} & \textbf{10}\ \ \ {\color{red}rejected} \\
    463 \hline   
     463\hline
    464464Yes (stackful)          & Yes           & \textbf{11}\ \ \ @thread@                             & \textbf{12}\ \ @mutex@ @thread@               \\
    465465\end{tabular}
     
    28962896\label{s:RuntimeStructureCluster}
    28972897
    2898 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.
     2898A \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~\cite{Buhr90a}.
    28992899The term \newterm{virtual processor} is introduced as a synonym for kernel thread to disambiguate between user and kernel thread.
    29002900From the language perspective, a virtual processor is an actual processor (core).
     
    29922992\end{cfa}
    29932993where CPU time in nanoseconds is from the appropriate language clock.
    2994 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.
     2994Each 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;
     2995each @N@ appears after the experiment name in the following tables.
    29952996The total time is divided by @N@ to obtain the average time for a benchmark.
    29962997Each benchmark experiment is run 13 times and the average appears in the table.
     2998For languages with a runtime JIT (Java, Node.js, Python), a single half-hour long experiment is run to check stability;
     2999all long-experiment results are statistically equivalent, \ie median/average/standard-deviation correlate with the short-experiment results, indicating the short experiments reached a steady state.
    29973000All omitted tests for other languages are functionally identical to the \CFA tests and available online~\cite{CforallConcurrentBenchmarks}.
    2998 % tar --exclude-ignore=exclude -cvhf benchmark.tar benchmark
    2999 % cp -p benchmark.tar /u/cforall/public_html/doc/concurrent_benchmark.tar
    30003001
    30013002\paragraph{Creation}
     
    30063007
    30073008\begin{multicols}{2}
    3008 \lstset{language=CFA,moredelim=**[is][\color{red}]{@}{@},deletedelim=**[is][]{`}{`}}
    3009 \begin{cfa}
    3010 @coroutine@ MyCoroutine {};
     3009\begin{cfa}[xleftmargin=0pt]
     3010`coroutine` MyCoroutine {};
    30113011void ?{}( MyCoroutine & this ) {
    30123012#ifdef EAGER
     
    30163016void main( MyCoroutine & ) {}
    30173017int main() {
    3018         BENCH( for ( N ) { @MyCoroutine c;@ } )
     3018        BENCH( for ( N ) { `MyCoroutine c;` } )
    30193019        sout | result;
    30203020}
     
    30303030
    30313031\begin{tabular}[t]{@{}r*{3}{D{.}{.}{5.2}}@{}}
    3032 \multicolumn{1}{@{}c}{} & \multicolumn{1}{c}{Median} & \multicolumn{1}{c}{Average} & \multicolumn{1}{c@{}}{Std Dev} \\
    3033 \CFA generator                  & 0.6           & 0.6           & 0.0           \\
    3034 \CFA coroutine lazy             & 13.4          & 13.1          & 0.5           \\
    3035 \CFA coroutine eager    & 144.7         & 143.9         & 1.5           \\
    3036 \CFA thread                             & 466.4         & 468.0         & 11.3          \\
    3037 \uC coroutine                   & 155.6         & 155.7         & 1.7           \\
    3038 \uC thread                              & 523.4         & 523.9         & 7.7           \\
    3039 Python generator                & 123.2         & 124.3         & 4.1           \\
    3040 Node.js generator               & 33.4          & 33.5          & 0.3           \\
    3041 Goroutine thread                & 751.0         & 750.5         & 3.1           \\
    3042 Rust tokio thread               & 1860.0        & 1881.1        & 37.6          \\
    3043 Rust thread                             & 53801.0       & 53896.8       & 274.9         \\
    3044 Java thread                             & 120274.0      & 120722.9      & 2356.7        \\
    3045 Pthreads thread                 & 31465.5       & 31419.5       & 140.4
     3032\multicolumn{1}{@{}r}{N\hspace*{10pt}} & \multicolumn{1}{c}{Median} & \multicolumn{1}{c}{Average} & \multicolumn{1}{c@{}}{Std Dev} \\
     3033\CFA generator (1B)                     & 0.6           & 0.6           & 0.0           \\
     3034\CFA coroutine lazy     (100M)  & 13.4          & 13.1          & 0.5           \\
     3035\CFA coroutine eager (10M)      & 144.7         & 143.9         & 1.5           \\
     3036\CFA thread (10M)                       & 466.4         & 468.0         & 11.3          \\
     3037\uC coroutine (10M)                     & 155.6         & 155.7         & 1.7           \\
     3038\uC thread (10M)                        & 523.4         & 523.9         & 7.7           \\
     3039Python generator (10M)          & 123.2         & 124.3         & 4.1           \\
     3040Node.js generator (10M)         & 33.4          & 33.5          & 0.3           \\
     3041Goroutine thread (10M)          & 751.0         & 750.5         & 3.1           \\
     3042Rust tokio thread (10M)         & 1860.0        & 1881.1        & 37.6          \\
     3043Rust thread     (250K)                  & 53801.0       & 53896.8       & 274.9         \\
     3044Java thread (250K)                      & 119256.0      & 119679.2      & 2244.0        \\
     3045% Java thread (1 000 000)               & 123100.0      & 123052.5      & 751.6         \\
     3046Pthreads thread (250K)          & 31465.5       & 31419.5       & 140.4
    30463047\end{tabular}
    30473048\end{multicols}
     
    30523053Internal scheduling is measured using a cycle of two threads signalling and waiting.
    30533054Figure~\ref{f:schedint} shows the code for \CFA, with results in Table~\ref{t:schedint}.
    3054 Note, the incremental cost of bulk acquire for \CFA, which is largely a fixed cost for small numbers of mutex objects.
    3055 Java scheduling is significantly greater because the benchmark explicitly creates multiple threads in order to prevent the JIT from making the program sequential, \ie removing all locking.
     3055Note, the \CFA incremental cost for bulk acquire is a fixed cost for small numbers of mutex objects.
     3056User-level threading has one kernel thread, eliminating contention between the threads (direct handoff of the kernel thread).
     3057Kernel-level threading has two kernel threads allowing some contention.
    30563058
    30573059\begin{multicols}{2}
    3058 \lstset{language=CFA,moredelim=**[is][\color{red}]{@}{@},deletedelim=**[is][]{`}{`}}
    3059 \begin{cfa}
     3060\setlength{\tabcolsep}{3pt}
     3061\begin{cfa}[xleftmargin=0pt]
    30603062volatile int go = 0;
    3061 @condition c;@
    3062 @monitor@ M {} m1/*, m2, m3, m4*/;
    3063 void call( M & @mutex p1/*, p2, p3, p4*/@ ) {
    3064         @signal( c );@
    3065 }
    3066 void wait( M & @mutex p1/*, p2, p3, p4*/@ ) {
     3063`condition c;`
     3064`monitor` M {} m1/*, m2, m3, m4*/;
     3065void call( M & `mutex p1/*, p2, p3, p4*/` ) {
     3066        `signal( c );`
     3067}
     3068void wait( M & `mutex p1/*, p2, p3, p4*/` ) {
    30673069        go = 1; // continue other thread
    3068         for ( N ) { @wait( c );@ } );
     3070        for ( N ) { `wait( c );` } );
    30693071}
    30703072thread T {};
     
    30913093
    30923094\begin{tabular}{@{}r*{3}{D{.}{.}{5.2}}@{}}
    3093 \multicolumn{1}{@{}c}{} & \multicolumn{1}{c}{Median} & \multicolumn{1}{c}{Average} & \multicolumn{1}{c@{}}{Std Dev} \\
    3094 \CFA @signal@, 1 monitor        & 364.4         & 364.2         & 4.4           \\
    3095 \CFA @signal@, 2 monitor        & 484.4         & 483.9         & 8.8           \\
    3096 \CFA @signal@, 4 monitor        & 709.1         & 707.7         & 15.0          \\
    3097 \uC @signal@ monitor            & 328.3         & 327.4         & 2.4           \\
    3098 Rust cond. variable                     & 7514.0        & 7437.4        & 397.2         \\
    3099 Java @notify@ monitor           & 9623.0        & 9654.6        & 236.2         \\
    3100 Pthreads cond. variable         & 5553.7        & 5576.1        & 345.6
     3095\multicolumn{1}{@{}r}{N\hspace*{10pt}} & \multicolumn{1}{c}{Median} & \multicolumn{1}{c}{Average} & \multicolumn{1}{c@{}}{Std Dev} \\
     3096\CFA @signal@, 1 monitor (10M)  & 364.4         & 364.2         & 4.4           \\
     3097\CFA @signal@, 2 monitor (10M)  & 484.4         & 483.9         & 8.8           \\
     3098\CFA @signal@, 4 monitor (10M)  & 709.1         & 707.7         & 15.0          \\
     3099\uC @signal@ monitor (10M)              & 328.3         & 327.4         & 2.4           \\
     3100Rust cond. variable     (1M)            & 7514.0        & 7437.4        & 397.2         \\
     3101Java @notify@ monitor (1M)              & 8717.0        & 8774.1        & 471.8         \\
     3102% Java @notify@ monitor (100 000 000)           & 8634.0        & 8683.5        & 330.5         \\
     3103Pthreads cond. variable (1M)    & 5553.7        & 5576.1        & 345.6
    31013104\end{tabular}
    31023105\end{multicols}
     
    31073110External scheduling is measured using a cycle of two threads calling and accepting the call using the @waitfor@ statement.
    31083111Figure~\ref{f:schedext} shows the code for \CFA with results in Table~\ref{t:schedext}.
    3109 Note, the incremental cost of bulk acquire for \CFA, which is largely a fixed cost for small numbers of mutex objects.
     3112Note, the \CFA incremental cost for bulk acquire is a fixed cost for small numbers of mutex objects.
    31103113
    31113114\begin{multicols}{2}
    3112 \lstset{language=CFA,moredelim=**[is][\color{red}]{@}{@},deletedelim=**[is][]{`}{`}}
     3115\setlength{\tabcolsep}{5pt}
    31133116\vspace*{-16pt}
    3114 \begin{cfa}
    3115 @monitor@ M {} m1/*, m2, m3, m4*/;
    3116 void call( M & @mutex p1/*, p2, p3, p4*/@ ) {}
    3117 void wait( M & @mutex p1/*, p2, p3, p4*/@ ) {
    3118         for ( N ) { @waitfor( call : p1/*, p2, p3, p4*/ );@ }
     3117\begin{cfa}[xleftmargin=0pt]
     3118`monitor` M {} m1/*, m2, m3, m4*/;
     3119void call( M & `mutex p1/*, p2, p3, p4*/` ) {}
     3120void wait( M & `mutex p1/*, p2, p3, p4*/` ) {
     3121        for ( N ) { `waitfor( call : p1/*, p2, p3, p4*/ );` }
    31193122}
    31203123thread T {};
     
    31333136\columnbreak
    31343137
    3135 \vspace*{-16pt}
     3138\vspace*{-18pt}
    31363139\captionof{table}{External-scheduling comparison (nanoseconds)}
    31373140\label{t:schedext}
    31383141\begin{tabular}{@{}r*{3}{D{.}{.}{3.2}}@{}}
    3139 \multicolumn{1}{@{}c}{} & \multicolumn{1}{c}{Median} &\multicolumn{1}{c}{Average} & \multicolumn{1}{c@{}}{Std Dev} \\
    3140 \CFA @waitfor@, 1 monitor       & 367.1 & 365.3 & 5.0   \\
    3141 \CFA @waitfor@, 2 monitor       & 463.0 & 464.6 & 7.1   \\
    3142 \CFA @waitfor@, 4 monitor       & 689.6 & 696.2 & 21.5  \\
    3143 \uC \lstinline[language=uC++]|_Accept| monitor  & 328.2 & 329.1 & 3.4   \\
    3144 Go \lstinline[language=Golang]|select| channel  & 365.0 & 365.5 & 1.2
     3142\multicolumn{1}{@{}r}{N\hspace*{10pt}} & \multicolumn{1}{c}{Median} &\multicolumn{1}{c}{Average} & \multicolumn{1}{c@{}}{Std Dev} \\
     3143\CFA @waitfor@, 1 monitor (10M) & 367.1 & 365.3 & 5.0   \\
     3144\CFA @waitfor@, 2 monitor (10M) & 463.0 & 464.6 & 7.1   \\
     3145\CFA @waitfor@, 4 monitor (10M) & 689.6 & 696.2 & 21.5  \\
     3146\uC \lstinline[language=uC++]|_Accept| monitor (10M)    & 328.2 & 329.1 & 3.4   \\
     3147Go \lstinline[language=Golang]|select| channel (10M)    & 365.0 & 365.5 & 1.2
    31453148\end{tabular}
    31463149\end{multicols}
     
    31553158
    31563159\begin{multicols}{2}
    3157 \lstset{language=CFA,moredelim=**[is][\color{red}]{@}{@},deletedelim=**[is][]{`}{`}}
    3158 \begin{cfa}
    3159 @monitor@ M {} m1/*, m2, m3, m4*/;
    3160 call( M & @mutex p1/*, p2, p3, p4*/@ ) {}
     3160\setlength{\tabcolsep}{3pt}
     3161\begin{cfa}[xleftmargin=0pt]
     3162`monitor` M {} m1/*, m2, m3, m4*/;
     3163call( M & `mutex p1/*, p2, p3, p4*/` ) {}
    31613164int main() {
    31623165        BENCH( for( N ) call( m1/*, m2, m3, m4*/ ); )
     
    31733176\label{t:mutex}
    31743177\begin{tabular}{@{}r*{3}{D{.}{.}{3.2}}@{}}
    3175 \multicolumn{1}{@{}c}{} & \multicolumn{1}{c}{Median} &\multicolumn{1}{c}{Average} & \multicolumn{1}{c@{}}{Std Dev} \\
    3176 test-and-test-set lock                  & 19.1  & 18.9  & 0.4   \\
    3177 \CFA @mutex@ function, 1 arg.   & 48.3  & 47.8  & 0.9   \\
    3178 \CFA @mutex@ function, 2 arg.   & 86.7  & 87.6  & 1.9   \\
    3179 \CFA @mutex@ function, 4 arg.   & 173.4 & 169.4 & 5.9   \\
    3180 \uC @monitor@ member rtn.               & 54.8  & 54.8  & 0.1   \\
    3181 Goroutine mutex lock                    & 34.0  & 34.0  & 0.0   \\
    3182 Rust mutex lock                                 & 33.0  & 33.2  & 0.8   \\
    3183 Java synchronized method                & 31.0  & 31.0  & 0.0   \\
    3184 Pthreads mutex Lock                             & 31.0  & 31.1  & 0.4
     3178\multicolumn{1}{@{}r}{N\hspace*{10pt}} & \multicolumn{1}{c}{Median} &\multicolumn{1}{c}{Average} & \multicolumn{1}{c@{}}{Std Dev} \\
     3179test-and-test-set lock (50M)            & 19.1  & 18.9  & 0.4   \\
     3180\CFA @mutex@ function, 1 arg. (50M)     & 48.3  & 47.8  & 0.9   \\
     3181\CFA @mutex@ function, 2 arg. (50M)     & 86.7  & 87.6  & 1.9   \\
     3182\CFA @mutex@ function, 4 arg. (50M)     & 173.4 & 169.4 & 5.9   \\
     3183\uC @monitor@ member rtn. (50M)         & 54.8  & 54.8  & 0.1   \\
     3184Goroutine mutex lock (50M)                      & 34.0  & 34.0  & 0.0   \\
     3185Rust mutex lock (50M)                           & 33.0  & 33.2  & 0.8   \\
     3186Java synchronized method (50M)          & 31.0  & 30.9  & 0.5   \\
     3187% Java synchronized method (10 000 000 000)             & 31.0 & 30.2 & 0.9 \\
     3188Pthreads mutex Lock (50M)                       & 31.0  & 31.1  & 0.4
    31853189\end{tabular}
    31863190\end{multicols}
     
    32013205% To: "Peter A. Buhr" <pabuhr@plg2.cs.uwaterloo.ca>
    32023206% Date: Fri, 24 Jan 2020 13:49:18 -0500
    3203 % 
     3207%
    32043208% I can also verify that the previous version, which just tied a bunch of promises together, *does not* go back to the
    32053209% event loop at all in the current version of Node. Presumably they're taking advantage of the fact that the ordering of
     
    32113215
    32123216\begin{multicols}{2}
    3213 \lstset{language=CFA,moredelim=**[is][\color{red}]{@}{@},deletedelim=**[is][]{`}{`}}
    3214 \begin{cfa}[aboveskip=0pt,belowskip=0pt]
    3215 @coroutine@ C {};
    3216 void main( C & ) { for () { @suspend;@ } }
     3217\begin{cfa}[xleftmargin=0pt]
     3218`coroutine` C {};
     3219void main( C & ) { for () { `suspend;` } }
    32173220int main() { // coroutine test
    32183221        C c;
    3219         BENCH( for ( N ) { @resume( c );@ } )
     3222        BENCH( for ( N ) { `resume( c );` } )
    32203223        sout | result;
    32213224}
    32223225int main() { // thread test
    3223         BENCH( for ( N ) { @yield();@ } )
     3226        BENCH( for ( N ) { `yield();` } )
    32243227        sout | result;
    32253228}
     
    32343237\label{t:ctx-switch}
    32353238\begin{tabular}{@{}r*{3}{D{.}{.}{3.2}}@{}}
    3236 \multicolumn{1}{@{}c}{} & \multicolumn{1}{c}{Median} &\multicolumn{1}{c}{Average} & \multicolumn{1}{c@{}}{Std Dev} \\
    3237 C function                      & 1.8           & 1.8           & 0.0   \\
    3238 \CFA generator          & 1.8           & 2.0           & 0.3   \\
    3239 \CFA coroutine          & 32.5          & 32.9          & 0.8   \\
    3240 \CFA thread                     & 93.8          & 93.6          & 2.2   \\
    3241 \uC coroutine           & 50.3          & 50.3          & 0.2   \\
    3242 \uC thread                      & 97.3          & 97.4          & 1.0   \\
    3243 Python generator        & 40.9          & 41.3          & 1.5   \\
    3244 Node.js await           & 1852.2        & 1854.7        & 16.4  \\
    3245 Node.js generator       & 33.3          & 33.4          & 0.3   \\
    3246 Goroutine thread        & 143.0         & 143.3         & 1.1   \\
    3247 Rust async await        & 32.0          & 32.0          & 0.0   \\
    3248 Rust tokio thread       & 143.0         & 143.0         & 1.7   \\
    3249 Rust thread                     & 332.0         & 331.4         & 2.4   \\
    3250 Java thread                     & 405.0         & 415.0         & 17.6  \\
    3251 Pthreads thread         & 334.3         & 335.2         & 3.9
     3239\multicolumn{1}{@{}r}{N\hspace*{10pt}} & \multicolumn{1}{c}{Median} &\multicolumn{1}{c}{Average} & \multicolumn{1}{c@{}}{Std Dev} \\
     3240C function (10B)                        & 1.8           & 1.8           & 0.0   \\
     3241\CFA generator (5B)                     & 1.8           & 2.0           & 0.3   \\
     3242\CFA coroutine (100M)           & 32.5          & 32.9          & 0.8   \\
     3243\CFA thread (100M)                      & 93.8          & 93.6          & 2.2   \\
     3244\uC coroutine (100M)            & 50.3          & 50.3          & 0.2   \\
     3245\uC thread (100M)                       & 97.3          & 97.4          & 1.0   \\
     3246Python generator (100M)         & 40.9          & 41.3          & 1.5   \\
     3247Node.js await (5M)                      & 1852.2        & 1854.7        & 16.4  \\
     3248Node.js generator (100M)        & 33.3          & 33.4          & 0.3   \\
     3249Goroutine thread (100M)         & 143.0         & 143.3         & 1.1   \\
     3250Rust async await (100M)         & 32.0          & 32.0          & 0.0   \\
     3251Rust tokio thread (100M)        & 143.0         & 143.0         & 1.7   \\
     3252Rust thread (25M)                       & 332.0         & 331.4         & 2.4   \\
     3253Java thread (100M)                      & 405.0         & 415.0         & 17.6  \\
     3254% Java thread (  100 000 000)                   & 413.0 & 414.2 & 6.2 \\
     3255% Java thread (5 000 000 000)                   & 415.0 & 415.2 & 6.1 \\
     3256Pthreads thread (25M)           & 334.3         & 335.2         & 3.9
    32523257\end{tabular}
    32533258\end{multicols}
     
    32583263Languages using 1:1 threading based on pthreads can at best meet or exceed, due to language overhead, the pthread results.
    32593264Note, pthreads has a fast zero-contention mutex lock checked in user space.
    3260 Languages with M:N threading have better performance than 1:1 because there is no operating-system interactions.
     3265Languages with M:N threading have better performance than 1:1 because there is no operating-system interactions (context-switching or locking).
     3266As well, for locking experiments, M:N threading has less contention if only one kernel thread is used.
    32613267Languages with stackful coroutines have higher cost than stackless coroutines because of stack allocation and context switching;
    32623268however, stackful \uC and \CFA coroutines have approximately the same performance as stackless Python and Node.js generators.
    32633269The \CFA stackless generator is approximately 25 times faster for suspend/resume and 200 times faster for creation than stackless Python and Node.js generators.
     3270The Node.js context-switch is costly when asynchronous await must enter the event engine because a promise is not fulfilled.
     3271Finally, the benchmark results correlate across programming languages with and without JIT, indicating the JIT has completed any runtime optimizations.
    32643272
    32653273
     
    33193327
    33203328The 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.
    3321 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.
     3329This research is funded by the NSERC/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.
    33223330
    33233331{%
  • doc/papers/concurrency/annex/local.bib

    r33c3ded r223a633  
    5959@manual{Cpp-Transactions,
    6060        keywords        = {C++, Transactional Memory},
    61         title           = {Technical Specification for C++ Extensions for Transactional Memory},
     61        title           = {Tech. Spec. for C++ Extensions for Transactional Memory},
    6262        organization= {International Standard ISO/IEC TS 19841:2015 },
    6363        publisher   = {American National Standards Institute},
  • doc/papers/concurrency/mail2

    r33c3ded r223a633  
    959959Software: Practice and Experience Editorial Office
    960960
     961
     962
     963Date: Wed, 2 Sep 2020 20:55:34 +0000
     964From: Richard Jones <onbehalfof@manuscriptcentral.com>
     965Reply-To: R.E.Jones@kent.ac.uk
     966To: tdelisle@uwaterloo.ca, pabuhr@uwaterloo.ca
     967Subject: Software: Practice and Experience - Decision on Manuscript ID
     968 SPE-19-0219.R2
     969
     97002-Sep-2020
     971
     972Dear Dr Buhr,
     973
     974Many thanks for submitting SPE-19-0219.R2 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. I apologise for the length of time it has taken to get these.
     975
     976Both reviewers consider this paper to be close to acceptance. However, before I can accept this paper, I would like you address the comments of Reviewer 2, particularly with regard to the description of the adaptation Java harness to deal with warmup. I would expect to see a convincing argument that the computation has reached a steady state. I would also like you to provide the values for N for each benchmark run. This should be very straightforward for you to do. There are a couple of papers on steady state that you may wish to consult (though I am certainly not pushing my own work).
     977
     9781) Barrett, Edd; Bolz-Tereick, Carl Friedrich; Killick, Rebecca; Mount, Sarah and Tratt, Laurence. Virtual Machine Warmup Blows Hot and Cold. OOPSLA 2017. https://doi.org/10.1145/3133876
     979Virtual Machines (VMs) with Just-In-Time (JIT) compilers are traditionally thought to execute programs in two phases: the initial warmup phase determines which parts of a program would most benefit from dynamic compilation, before JIT compiling those parts into machine code; subsequently the program is said to be at a steady state of peak performance. Measurement methodologies almost always discard data collected during the warmup phase such that reported measurements focus entirely on peak performance. We introduce a fully automated statistical approach, based on changepoint analysis, which allows us to determine if a program has reached a steady state and, if so, whether that represents peak performance or not. Using this, we show that even when run in the most controlled of circumstances, small, deterministic, widely studied microbenchmarks often fail to reach a steady state of peak performance on a variety of common VMs. Repeating our experiment on 3 different machines, we found that at most 43.5% of pairs consistently reach a steady state of peak performance.
     980
     9812) Kalibera, Tomas and Jones, Richard. Rigorous Benchmarking in Reasonable Time. ISMM  2013. https://doi.org/10.1145/2555670.2464160
     982Experimental evaluation is key to systems research. Because modern systems are complex and non-deterministic, good experimental methodology demands that researchers account for uncertainty. To obtain valid results, they are expected to run many iterations of benchmarks, invoke virtual machines (VMs) several times, or even rebuild VM or benchmark binaries more than once. All this repetition costs time to complete experiments. Currently, many evaluations give up on sufficient repetition or rigorous statistical methods, or even run benchmarks only in training sizes. The results reported often lack proper variation estimates and, when a small difference between two systems is reported, some are simply unreliable.In contrast, we provide a statistically rigorous methodology for repetition and summarising results that makes efficient use of experimentation time. Time efficiency comes from two key observations. First, a given benchmark on a given platform is typically prone to much less non-determinism than the common worst-case of published corner-case studies. Second, repetition is most needed where most uncertainty arises (whether between builds, between executions or between iterations). We capture experimentation cost with a novel mathematical model, which we use to identify the number of repetitions at each level of an experiment necessary and sufficient to obtain a given level of precision.We present our methodology as a cookbook that guides researchers on the number of repetitions they should run to obtain reliable results. We also show how to present results with an effect size confidence interval. As an example, we show how to use our methodology to conduct throughput experiments with the DaCapo and SPEC CPU benchmarks on three recent platforms.
     983
     984You have 42 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 a short extension.
     985
     986You 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".
     987
     988When 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.
     989
     990If you would like help with English language editing, or other article preparation support, Wiley Editing Services offers expert help with English Language Editing, as well as translation, manuscript formatting, and figure formatting at www.wileyauthors.com/eeo/preparation. You can also check out our resources for Preparing Your Article for general guidance about writing and preparing your manuscript at www.wileyauthors.com/eeo/prepresources.
     991 
     992Once again, thank you for submitting your manuscript to Software: Practice and Experience. I look forward to receiving your revision.
     993
     994Sincerely,
     995Richard
     996
     997Prof. Richard Jones
     998Editor, Software: Practice and Experience
     999R.E.Jones@kent.ac.uk
     1000
     1001Referee(s)' Comments to Author:
     1002
     1003Reviewing: 1
     1004
     1005Comments to the Author
     1006Overall, I felt that this draft was an improvement on previous drafts and I don't have further changes to request.
     1007
     1008I appreciated the new language to clarify the relationship of external and internal scheduling, for example, as well as the new measurements of Rust tokio. Also, while I still believe that the choice between thread/generator/coroutine and so forth could be made crisper and clearer, the current draft of Section 2 did seem adequate to me in terms of specifying the considerations that users would have to take into account to make the choice.
     1009
     1010
     1011Reviewing: 2
     1012
     1013Comments to the Author
     1014First: let me apologise for the delay on this review. I'll blame the global pandemic combined with my institution's senior management's counterproductive decisions for taking up most of my time and all of my energy.
     1015
     1016At this point, reading the responses, I think we've been around the course enough times that further iteration is unlikely to really improve the paper any further, so I'm happy to recommend acceptance.    My main comments are that there were some good points in the responses to *all* the reviews and I strongly encourage the authors to incorporate those discursive responses into the final paper so they may benefit readers as well as reviewers.   I agree with the recommendations of reviewer #2 that the paper could usefully be split in to two, which I think I made to a previous revision, but I'm happy to leave that decision to the Editor.
     1017
     1018Finally, the paper needs to describe how the Java harness was adapted to deal with warmup; why the computation has warmed up and reached a steady state - similarly for js and Python. The tables should also give the "N" chosen for each benchmark run.
     1019 
     1020minor points
     1021* don't start sentences with "However"
     1022* most downloaded isn't an "Award"
     1023
     1024
     1025
     1026Date: Thu, 1 Oct 2020 05:34:29 +0000
     1027From: Richard Jones <onbehalfof@manuscriptcentral.com>
     1028Reply-To: R.E.Jones@kent.ac.uk
     1029To: pabuhr@uwaterloo.ca
     1030Subject: Revision reminder - SPE-19-0219.R2
     1031
     103201-Oct-2020
     1033
     1034Dear Dr Buhr
     1035
     1036SPE-19-0219.R2
     1037
     1038This is a reminder that your opportunity to revise and re-submit your manuscript will expire 14 days from now. If you require more time please contact me directly and I may grant an extension to this deadline, otherwise the option to submit a revision online, will not be available.
     1039
     1040If your article is of potential interest to the general public, (which means it must be timely, groundbreaking, interesting and impact on everyday society) then please e-mail ejp@wiley.co.uk explaining the public interest side of the research. Wiley will then investigate the potential for undertaking a global press campaign on the article.
     1041
     1042I look forward to receiving your revision.
     1043
     1044Sincerely,
     1045
     1046Prof. Richard Jones
     1047Editor, Software: Practice and Experience
     1048
     1049https://mc.manuscriptcentral.com/spe
     1050
     1051
     1052
     1053Date: Tue, 6 Oct 2020 15:29:41 +0000
     1054From: Mayank Roy Chowdhury <onbehalfof@manuscriptcentral.com>
     1055Reply-To: speoffice@wiley.com
     1056To: tdelisle@uwaterloo.ca, pabuhr@uwaterloo.ca
     1057Subject: SPE-19-0219.R3 successfully submitted
     1058
     105906-Oct-2020
     1060
     1061Dear Dr Buhr,
     1062
     1063Your manuscript entitled "Advanced Control-flow and Concurrency in Cforall" has been successfully submitted online and is presently being given full consideration for publication in Software: Practice and Experience.
     1064
     1065Your manuscript number is SPE-19-0219.R3.  Please mention this number in all future correspondence regarding this submission.
     1066
     1067You can view the status of your manuscript at any time by checking your Author Center after logging into https://mc.manuscriptcentral.com/spe.  If you have difficulty using this site, please click the 'Get Help Now' link at the top right corner of the site.
     1068
     1069
     1070Thank you for submitting your manuscript to Software: Practice and Experience.
     1071
     1072Sincerely,
     1073
     1074Software: Practice and Experience Editorial Office
     1075
  • doc/refrat/refrat.tex

    r33c3ded r223a633  
    1111%% Created On       : Wed Apr  6 14:52:25 2016
    1212%% Last Modified By : Peter A. Buhr
    13 %% Last Modified On : Wed Jan 31 17:30:23 2018
    14 %% Update Count     : 108
     13%% Last Modified On : Mon Oct  5 09:02:53 2020
     14%% Update Count     : 110
    1515%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
    1616
     
    3030\usepackage{upquote}                                                                    % switch curled `'" to straight
    3131\usepackage{calc}
    32 \usepackage{xspace}
    3332\usepackage{varioref}                                                                   % extended references
    34 \usepackage{listings}                                                                   % format program code
    3533\usepackage[flushmargin]{footmisc}                                              % support label/reference in footnote
    3634\usepackage{latexsym}                                   % \Box glyph
    3735\usepackage{mathptmx}                                   % better math font with "times"
    3836\usepackage[usenames]{color}
    39 \input{common}                                          % common CFA document macros
    40 \usepackage[dvips,plainpages=false,pdfpagelabels,pdfpagemode=UseNone,colorlinks=true,pagebackref=true,linkcolor=blue,citecolor=blue,urlcolor=blue,pagebackref=true,breaklinks=true]{hyperref}
    41 \usepackage{breakurl}
    42 \renewcommand{\UrlFont}{\small\sf}
    43 
    44 \usepackage[pagewise]{lineno}
    45 \renewcommand{\linenumberfont}{\scriptsize\sffamily}
    46 \usepackage[firstpage]{draftwatermark}
    47 \SetWatermarkLightness{0.9}
    48 
    49 % Default underscore is too low and wide. Cannot use lstlisting "literate" as replacing underscore
    50 % removes it as a variable-name character so keywords in variables are highlighted. MUST APPEAR
    51 % AFTER HYPERREF.
    52 \renewcommand{\textunderscore}{\leavevmode\makebox[1.2ex][c]{\rule{1ex}{0.075ex}}}
    53 
    54 \setlength{\topmargin}{-0.45in}                                                 % move running title into header
    55 \setlength{\headsep}{0.25in}
    56 
    57 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
    58 
    59 \CFAStyle                                                                                               % use default CFA format-style
    60 \lstnewenvironment{C++}[1][]                            % use C++ style
    61 {\lstset{language=C++,moredelim=**[is][\protect\color{red}]{®}{®}#1}}
    62 {}
    63 
     37\newcommand{\CFALatin}{}
    6438% inline code ©...© (copyright symbol) emacs: C-q M-)
    6539% red highlighting ®...® (registered trademark symbol) emacs: C-q M-.
     
    6943% keyword escape ¶...¶ (pilcrow symbol) emacs: C-q M-^
    7044% math escape $...$ (dollar symbol)
     45\input{common}                                          % common CFA document macros
     46\usepackage[dvips,plainpages=false,pdfpagelabels,pdfpagemode=UseNone,colorlinks=true,pagebackref=true,linkcolor=blue,citecolor=blue,urlcolor=blue,pagebackref=true,breaklinks=true]{hyperref}
     47\usepackage{breakurl}
     48\renewcommand{\UrlFont}{\small\sf}
     49
     50\usepackage[pagewise]{lineno}
     51\renewcommand{\linenumberfont}{\scriptsize\sffamily}
     52\usepackage[firstpage]{draftwatermark}
     53\SetWatermarkLightness{0.9}
     54
     55% Default underscore is too low and wide. Cannot use lstlisting "literate" as replacing underscore
     56% removes it as a variable-name character so keywords in variables are highlighted. MUST APPEAR
     57% AFTER HYPERREF.
     58\renewcommand{\textunderscore}{\leavevmode\makebox[1.2ex][c]{\rule{1ex}{0.075ex}}}
     59
     60\setlength{\topmargin}{-0.45in}                                                 % move running title into header
     61\setlength{\headsep}{0.25in}
    7162
    7263%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
    7364
     65\CFAStyle                                                                                               % use default CFA format-style
     66\lstnewenvironment{C++}[1][]                            % use C++ style
     67{\lstset{language=C++,moredelim=**[is][\protect\color{red}]{®}{®},#1}}
     68{}
     69
     70%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
     71
    7472% Names used in the document.
    75 \newcommand{\Version}{\input{../../version}}
     73\newcommand{\Version}{\input{build/version}}
    7674\newcommand{\Textbf}[2][red]{{\color{#1}{\textbf{#2}}}}
    7775\newcommand{\Emph}[2][red]{{\color{#1}\textbf{\emph{#2}}}}
  • doc/theses/andrew_beach_MMath/thesis.tex

    r33c3ded r223a633  
    3434\usepackage[toc,abbreviations]{glossaries-extra}
    3535
    36 % Main glossary entries -- definitions of relevant terminology
    37 \newglossaryentry{computer}
    38 {
    39 name=computer,
    40 description={A programmable machine that receives input data,
    41                stores and manipulates the data, and provides
    42                formatted output}
    43 }
    44 
    45 % Nomenclature glossary entries -- New definitions, or unusual terminology
    46 \newglossary*{nomenclature}{Nomenclature}
    47 \newglossaryentry{dingledorf}
    48 {
    49 type=nomenclature,
    50 name=dingledorf,
    51 description={A person of supposed average intelligence who makes incredibly
    52                brainless misjudgments}
    53 }
    54 
    55 % List of Abbreviations (abbreviations are from the glossaries-extra package)
    56 \newabbreviation{aaaaz}{AAAAZ}{American Association of Amature Astronomers
    57                and Zoologists}
    58 
    59 % List of Symbols
    60 \newglossary*{symbols}{List of Symbols}
    61 \newglossaryentry{rvec}
    62 {
    63 name={$\mathbf{v}$},
    64 sort={label},
    65 type=symbols,
    66 description={Random vector: a location in n-dimensional Cartesian space, where
    67                each dimensional component is determined by a random process}
    68 }
     36% Define all the glossaries.
     37\input{glossaries}
    6938
    7039% Generate the glossaries defined above.
  • doc/theses/fangren_yu_COOP_S20/Makefile

    r33c3ded r223a633  
    4646# File Dependencies #
    4747
    48 
    4948${DOCUMENT} : ${BASE}.ps
    5049        ps2pdf $<
  • doc/theses/fangren_yu_COOP_S20/Report.tex

    r33c3ded r223a633  
    1 \documentclass[twoside,12pt]{article}
     1\documentclass[twoside,11pt]{article}
    22
    33%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
     
    1111\usepackage[labelformat=simple,aboveskip=0pt,farskip=0pt]{subfig}
    1212\renewcommand{\thesubfigure}{\alph{subfigure})}
     13\usepackage[flushmargin]{footmisc}                                              % support label/reference in footnote
    1314\usepackage{latexsym}                                   % \Box glyph
    1415\usepackage{mathptmx}                                   % better math font with "times"
     16\usepackage[toc]{appendix}                                                              % article does not have appendix
    1517\usepackage[usenames]{color}
    1618\input{common}                                          % common CFA document macros
    1719\usepackage[dvips,plainpages=false,pdfpagelabels,pdfpagemode=UseNone,colorlinks=true,pagebackref=true,linkcolor=blue,citecolor=blue,urlcolor=blue,pagebackref=true,breaklinks=true]{hyperref}
    1820\usepackage{breakurl}
     21\urlstyle{sf}
     22
     23% reduce spacing
     24\setlist[itemize]{topsep=5pt,parsep=0pt}% global
     25\setlist[enumerate]{topsep=5pt,parsep=0pt}% global
    1926
    2027\usepackage[pagewise]{lineno}
     
    2633\renewcommand{\textunderscore}{\leavevmode\makebox[1.2ex][c]{\rule{1ex}{0.075ex}}}
    2734\newcommand{\NOTE}{\textbf{NOTE}}
     35\newcommand{\TODO}[1]{{\color{Purple}#1}}
    2836
    2937\setlength{\topmargin}{-0.45in}                                                 % move running title into header
     
    3240%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
    3341
    34 \CFADefaults
     42\CFAStyle                                                                                               % CFA code-style for all languages
    3543\lstset{
    36 language=C++,                                                                                   % make C++ the default language
    37 escapechar=\$,                                                                                  % LaTeX escape in CFA code
    38 moredelim=**[is][\color{red}]{`}{`},
     44language=C++,moredelim=**[is][\color{red}]{@}{@}                % make C++ the default language
    3945}% lstset
    40 \lstMakeShortInline@%
    4146\lstnewenvironment{C++}[1][]                            % use C++ style
    42 {\lstset{language=C++,moredelim=**[is][\protect\color{red}]{`}{`},#1}}
    43 {}
     47{\lstset{language=C++,moredelim=**[is][\color{red}]{@}{@}}\lstset{#1}}{}
    4448
    4549%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
     
    8488\section{Overview}
    8589
    86 cfa-cc is the reference compiler for the \CFA programming language, which is a non-
    87 object-oriented extension to C.
    88 \CFA attempts to introduce productive modern programming language features to C
    89 while maintaining as much backward-compatibility as possible, so that most existing C
    90 programs can seamlessly work with \CFA.
    91 
    92 Since the \CFA project was dated back to the early 2000s, and only restarted in the past
    93 few years, there is a significant amount of legacy code in the current compiler codebase,
    94 with little proper documentation available. This becomes a difficulty while developing new
    95 features based on the previous implementations, and especially while diagnosing
    96 problems.
    97 
    98 Currently, the \CFA team is also facing another problem: bad compiler performance. For
    99 the development of a new programming language, writing a standard library is an
    100 important part. The incompetence of the compiler causes building the library files to take
    101 tens of minutes, making iterative development and testing almost impossible. There is
    102 ongoing effort to rewrite the core data structure of the compiler to overcome the
    103 performance issue, but many bugs may appear during the work, and lack of documentation
    104 makes debugging extremely difficult.
    105 
    106 This developer's reference will be continuously improved and eventually cover the
    107 compiler codebase. For now, the focus is mainly on the parts being rewritten, and also the
    108 performance bottleneck, namely the resolution algorithm. It is aimed to provide new
    109 developers to the project enough guidance and clarify the purposes and behavior of certain
    110 functions which are not mentioned in the previous \CFA research papers.
     90@cfa-cc@ is the reference compiler for the \CFA programming language, which is a non-object-oriented extension to C.
     91\CFA attempts to introduce productive modern programming language features to C while maintaining as much backward-compatibility as possible, so that most existing C programs can seamlessly work with \CFA.
     92
     93Since the \CFA project dates back to the early 2000s, and only restarted in the past few years, there is a significant amount of legacy code in the current compiler codebase with little documentation.
     94The lack of documentation makes it difficult to develop new features from the current implementation and diagnose problems.
     95
     96Currently, the \CFA team is also facing poor compiler performance.
     97For the development of a new programming language, writing standard libraries is an important component.
     98The slow compiler causes building of the library files to take tens of minutes, making iterative development and testing almost impossible.
     99There is an ongoing effort to rewrite the core data-structure of the compiler to overcome the performance issue, but many bugs have appeared during this work, and lack of documentation is hampering debugging.
     100
     101This developer's reference manual begins the documentation and should be continuously im\-proved until it eventually covers the entire compiler codebase.
     102For now, the focus is mainly on the parts being rewritten, and also the primary performance bottleneck, namely the resolution algorithm.
     103Its aimed is to provide new project developers with guidance in understanding the codebase, and clarify the purpose and behaviour of certain functions that are not mentioned in the previous \CFA research papers~\cite{Bilson03,Ditchfield92,Moss19}.
    111104
    112105
    113106\section{Compiler Framework}
    114107
     108\CFA source code is first transformed into an abstract syntax tree (AST) by the parser before analyzed by the compiler.
     109
     110
    115111\subsection{AST Representation}
    116112
    117 Source code input is first transformed into abstract syntax tree (AST) representation by the
    118 parser before analyzed by the compiler.
    119 
    120 There are 4 major categories of AST nodes used by the compiler, along with some derived
    121 structures.
    122 
    123 \subsubsection{Declaration nodes}
     113
     114There are 4 major categories of AST nodes used by the compiler, along with some derived structures.
     115
     116\subsubsection{Declaration Nodes}
    124117
    125118A declaration node represents either of:
    126119\begin{itemize}
    127120\item
    128 Type declaration: struct, union, typedef or type parameter (see Appendix A.3)
    129 \item
    130 Variable declaration
    131 \item
    132 Function declaration
     121type declaration: @struct@, @union@, @typedef@ or type parameter (see \VRef[Appendix]{s:KindsTypeParameters})
     122\item
     123variable declaration
     124\item
     125function declaration
    133126\end{itemize}
    134127Declarations are introduced by standard C declarations, with the usual scoping rules.
    135 In addition, declarations can also be introduced by the forall clause (which is the origin
    136 of \CFA's name):
     128In addition, declarations can also be qualified by the \lstinline[language=CFA]@forall@ clause (which is the origin of \CFA's name):
    137129\begin{cfa}
    138 forall (<$\emph{TypeParameterList}$> | <$\emph{AssertionList}$>)
     130forall ( <$\emph{TypeParameterList}$> | <$\emph{AssertionList}$> )
    139131        $\emph{declaration}$
    140132\end{cfa}
    141 Type parameters in \CFA are similar to \CC template type parameters. The \CFA
    142 declaration
     133Type parameters in \CFA are similar to \CC template type parameters.
     134The \CFA declaration
    143135\begin{cfa}
    144136forall (dtype T) ...
    145137\end{cfa}
    146 behaves similarly as the \CC template declaration
     138behaves similarly to the \CC template declaration
    147139\begin{C++}
    148140template <typename T> ...
    149141\end{C++}
    150142
    151 Assertions are a distinctive feature of \CFA: contrary to the \CC template where
    152 arbitrary functions and operators can be used in a template definition, in a \CFA
    153 parametric function, operations on parameterized types must be declared in assertions.
    154 
     143Assertions are a distinctive feature of \CFA, similar to \emph{interfaces} in D and Go, and \emph{traits} in Rust.
     144Contrary to the \CC template where arbitrary functions and operators can be used in a template definition, in a \CFA parametric function, operations on parameterized types must be declared in assertions.
    155145Consider the following \CC template:
    156146\begin{C++}
    157 template <typename T> int foo(T t) {
    158         return bar(t) + baz(t);
     147@template@ forall<typename T> T foo( T t ) {
     148        return t + t * t;
    159149}
    160150\end{C++}
    161 Unless bar and baz are also parametric functions taking any argument type, they must be
    162 declared in the assertions, or otherwise the code will not compile:
     151where there are no explicit requirements on the type @T@.
     152Therefore, the \CC compiler must deduce what operators are required during textual (macro) expansion of the template at each usage.
     153As a result, templates cannot be compiled.
     154\CFA assertions specify restrictions on type parameters:
    163155\begin{cfa}
    164 forall (dtype T | { int bar(T); int baz(t); }) int foo (T t) {
    165         return bar(t) + baz(t);
     156forall( dtype T | @{ T ?+?( T, T ); T ?*?( T, T ) }@ ) int foo ( T t ) {
     157        return t + t * t;
    166158}
    167159\end{cfa}
    168 Assertions are written using the usual function declaration syntax. The scope of type
    169 parameters and assertions is the following declaration.
    170 
    171 \subsubsection{Type nodes}
    172 
    173 A type node represents the type of an object or expression.
    174 Named types reference the corresponding type declarations. The type of a function is its
    175 function pointer type (same as standard C).
    176 With the addition of type parameters, named types may contain a list of parameter values
    177 (actual parameter types).
    178 
    179 \subsubsection{Statement nodes}
    180 
    181 Statement nodes represent the statements in the program, including basic expression
    182 statements, control flows and blocks.
     160Assertions are written using the usual \CFA function declaration syntax.
     161Only types with operators ``@+@'' and ``@*@'' work with this function, and the function prototype is sufficient to allow separate compilation.
     162
     163Type parameters and assertions are used in the following compiler data-structures.
     164
     165
     166\subsubsection{Type Nodes}
     167
     168Type nodes represent the type of an object or expression.
     169Named types reference the corresponding type declarations.
     170The type of a function is its function pointer type (same as standard C).
     171With the addition of type parameters, named types may contain a list of parameter values (actual parameter types).
     172
     173
     174\subsubsection{Statement Nodes}
     175
     176Statement nodes represent the executable statements in the program, including basic expression statements, control flows and blocks.
    183177Local declarations (within a block statement) are represented as declaration statements.
    184178
    185 \subsubsection{Expression nodes}
    186 
    187 Some expressions are represented differently in the compiler before and after resolution
    188 stage:
     179
     180\subsubsection{Expression Nodes}
     181
     182Some expressions are represented differently before and after the resolution stage:
    189183\begin{itemize}
    190184\item
    191 Name expressions: NameExpr pre-resolution, VariableExpr post-resolution
    192 \item
    193 Member expressions: UntypedMemberExpr pre-resolution, MemberExpr post-resolution
    194 \item
    195 Function call expressions (including overloadable operators): UntypedExpr pre-resolution, ApplicationExpr post-resolution
     185Name expressions: @NameExpr@ pre-resolution, @VariableExpr@ post-resolution
     186\item
     187Member expressions: @UntypedMemberExpr@ pre-resolution, @MemberExpr@ post-resolution
     188\item
     189\begin{sloppypar}
     190Function call expressions (including overloadable operators): @UntypedExpr@ pre-resolution, @ApplicationExpr@ post-resolution
     191\end{sloppypar}
    196192\end{itemize}
    197 The pre-resolution representations contain only the symbols. Post-resolution results link
    198 them to the actual variable and function declarations.
     193The pre-resolution representation contains only the symbols.
     194Post-resolution links them to the actual variable and function declarations.
    199195
    200196
    201197\subsection{Compilation Passes}
    202198
    203 Compilation steps are implemented as passes, which follows a general structural recursion
    204 pattern on the syntax tree.
    205 
    206 The basic work flow of compilation passes follows preorder and postorder traversal on
    207 tree data structure, implemented with visitor pattern, and can be loosely described with
    208 the following pseudocode:
    209 \begin{C++}
    210 Pass::visit (node_t node) {
    211         previsit(node);
    212         if (visit_children)
     199Compilation steps are implemented as passes, which follows a general structural recursion pattern on the syntax tree.
     200
     201The basic workflow of compilation passes follows preorder and postorder traversal on the AST data-structure, implemented with visitor pattern, and can be loosely described with the following pseudocode:
     202\begin{C++}
     203Pass::visit( node_t node ) {
     204        previsit( node );
     205        if ( visit_children )
    213206                for each child of node:
    214                         child.accept(this);
    215         postvisit(node);
     207                        child.accept( this );
     208        postvisit( node );
    216209}
    217210\end{C++}
    218 Operations in previsit() happen in preorder (top to bottom) and operations in
    219 postvisit() happen in postorder (bottom to top). The precise order of recursive
    220 operations on child nodes can be found in @Common/PassVisitor.impl.h@ (old) and
    221 @AST/Pass.impl.hpp@ (new).
    222 Implementations of compilation passes need to follow certain conventions:
     211Operations in @previsit@ happen in preorder (top to bottom) and operations in @postvisit@ happen in postorder (bottom to top).
     212The precise order of recursive operations on child nodes can be found in @Common/PassVisitor.impl.h@ (old) and @AST/Pass.impl.hpp@ (new).
     213
     214Implementations of compilation passes follow certain conventions:
    223215\begin{itemize}
    224216\item
    225 Passes \textbf{should not} directly override the visit method (Non-virtual Interface
    226 principle); if a pass desires different recursion behavior, it should set
    227 @visit_children@ to false and perform recursive calls manually within previsit or
    228 postvisit procedures. To enable this option, inherit from @WithShortCircuiting@ mixin.
    229 \item
    230 previsit may mutate the node but \textbf{must not} change the node type or return null.
    231 \item
    232 postvisit may mutate the node, reconstruct it to a different node type, or delete it by
    233 returning null.
     217Passes \textbf{should not} directly override the visit method (Non-virtual Interface principle);
     218if a pass desires different recursion behaviour, it should set @visit_children@ to false and perform recursive calls manually within previsit or postvisit procedures.
     219To enable this option, inherit from the @WithShortCircuiting@ mixin.
     220\item
     221previsit may mutate the node but \textbf{must not} change the node type or return @nullptr@.
     222\item
     223postvisit may mutate the node, reconstruct it to a different node type, or delete it by returning @nullptr@.
    234224\item
    235225If the previsit or postvisit method is not defined for a node type, the step is skipped.
    236 If the return type is declared as void, the original node is returned by default. These
    237 behaviors are controlled by template specialization rules; see
    238 @Common/PassVisitor.proto.h@ (old) and @AST/Pass.proto.hpp@ (new) for details.
     226If the return type is declared as @void@, the original node is returned by default.
     227These behaviours are controlled by template specialization rules;
     228see @Common/PassVisitor.proto.h@ (old) and @AST/@ @Pass.proto.hpp@ (new) for details.
    239229\end{itemize}
    240230Other useful mixin classes for compilation passes include:
    241231\begin{itemize}
    242232\item
    243 WithGuards allows saving values of variables and restore automatically upon exiting
    244 the current node.
    245 \item
    246 WithVisitorRef creates a wrapped entity of current pass (the actual argument
    247 passed to recursive calls internally) for explicit recursion, usually used together
    248 with WithShortCircuiting.
    249 \item
    250 WithSymbolTable gives a managed symbol table with built-in scoping rule handling
    251 (\eg on entering and exiting a block statement)
     233@WithGuards@ allows saving and restoring variable values automatically upon entering/exiting the current node.
     234\item
     235@WithVisitorRef@ creates a wrapped entity for the current pass (the actual argument passed to recursive calls internally) for explicit recursion, usually used together with @WithShortCircuiting@.
     236\item
     237@WithSymbolTable@ gives a managed symbol table with built-in scoping-rule handling (\eg on entering and exiting a block statement)
    252238\end{itemize}
    253 \NOTE: If a pass extends the functionality of another existing pass, due to \CC overloading
    254 resolution rules, it \textbf{must} explicitly introduce the inherited previsit and postvisit procedures
    255 to its own scope, or otherwise they will not be picked up by template resolution:
     239\NOTE: If a pass extends the functionality of another existing pass, due to \CC overloading resolution rules, it \textbf{must} explicitly introduce the inherited previsit and postvisit procedures to its own scope, or otherwise they are not picked up by template resolution:
    256240\begin{C++}
    257241class Pass2: public Pass1 {
    258         using Pass1::previsit;
    259         using Pass1::postvisit;
     242        @using Pass1::previsit;@
     243        @using Pass1::postvisit;@
    260244        // new procedures
    261245}
     
    263247
    264248
    265 \subsection{Data Structure Change WIP (new-ast)}
    266 
    267 It has been observed that excessive copying of syntax tree structures accounts for a
    268 majority of computation cost and significantly slows down the compiler. In the previous
    269 implementation of the syntax tree, every internal node has a unique parent; therefore all
    270 copies are required to duplicate everything down to the bottom. A new, experimental
    271 re-implementation of the syntax tree (source under directory AST/ hereby referred to as
    272 ``new-ast'') attempts to overcome this issue with a functional approach that allows sharing
    273 of common sub-structures and only makes copies when necessary.
    274 
    275 The core of new-ast is a customized implementation of smart pointers, similar to
    276 @std::shared_ptr@ and @std::weak_ptr@ in \CC standard library. Reference counting is
    277 used to detect sharing and allows optimization. For a purely functional (a.k.a. immutable)
    278 data structure, all mutations are modelled by shallow copies along the path of mutation.
     249\subsection{Data Structure Change (new-ast)}
     250
     251It has been observed that excessive copying of syntax tree structures accounts for a majority of computation cost and significantly slows down the compiler.
     252In the previous implementation of the syntax tree, every internal node has a unique parent;
     253therefore all copies are required to duplicate the entire subtree.
     254A new, experimental re-implementation of the syntax tree (source under directory @AST/@ hereby referred to as ``new-ast'') attempts to overcome this issue with a functional approach that allows sharing of common sub-structures and only makes copies when necessary.
     255
     256The core of new-ast is a customized implementation of smart pointers, similar to @std::shared_ptr@ and @std::weak_ptr@ in the \CC standard library.
     257Reference counting is used to detect sharing and allowing certain optimizations.
     258For a purely functional (immutable) data-structure, all mutations are modelled by shallow copies along the path of mutation.
    279259With reference counting optimization, unique nodes are allowed to be mutated in place.
    280 This however, may potentially introduce some complications and bugs; a few issues are
    281 discussed near the end of this section.
    282 
    283 \subsubsection{Source: AST/Node.hpp}
    284 
    285 class @ast::Node@ is the base class of all new-ast node classes, which implements
    286 reference counting mechanism. Two different counters are recorded: ``strong'' reference
    287 count for number of nodes semantically owning it; ``weak'' reference count for number of
    288 nodes holding a mere reference and only need to observe changes.
    289 class @ast::ptr_base@ is the smart pointer implementation and also takes care of
    290 resource management.
    291 
    292 Direct access through the smart pointer is read-only. A mutable access should be obtained
    293 by calling shallowCopy or mutate as below.
    294 
    295 Currently, the weak pointers are only used to reference declaration nodes from a named
    296 type, or a variable expression. Since declaration nodes are intended to denote unique
    297 entities in the program, weak pointers always point to unique (unshared) nodes. This may
    298 change in the future, and weak references to shared nodes may introduce some problems;
     260This however, may potentially introduce some complications and bugs;
     261a few issues are discussed near the end of this section.
     262
     263
     264\subsubsection{Source: \lstinline{AST/Node.hpp}}
     265
     266Class @ast::Node@ is the base class of all new-ast node classes, which implements reference counting mechanism.
     267Two different counters are recorded: ``strong'' reference count for number of nodes semantically owning it;
     268``weak'' reference count for number of nodes holding a mere reference and only need to observe changes.
     269Class @ast::ptr_base@ is the smart pointer implementation and also takes care of resource management.
     270
     271Direct access through the smart pointer is read-only.
     272A mutable access should be obtained by calling @shallowCopy@ or mutate as below.
     273
     274Currently, the weak pointers are only used to reference declaration nodes from a named type, or a variable expression.
     275Since declaration nodes are intended to denote unique entities in the program, weak pointers always point to unique (unshared) nodes.
     276This property may change in the future, and weak references to shared nodes may introduce some problems;
    299277see mutate function below.
    300278
    301 All node classes should always use smart pointers in the structure and should not use raw
    302 pointers.
    303 
     279All node classes should always use smart pointers in structure definitions versus raw pointers.
     280Function
    304281\begin{C++}
    305282void ast::Node::increment(ref_type ref)
    306283\end{C++}
    307 Increments this node's strong or weak reference count.
     284increments this node's strong or weak reference count.
     285Function
    308286\begin{C++}
    309287void ast::Node::decrement(ref_type ref, bool do_delete = true)
    310288\end{C++}
    311 Decrements this node's strong or weak reference count. If strong reference count reaches
    312 zero, the node is deleted by default.
    313 \NOTE: Setting @do_delete@ to false may result in a detached node. Subsequent code should
    314 manually delete the node or assign it to a strong pointer to prevent memory leak.
     289decrements this node's strong or weak reference count.
     290If strong reference count reaches zero, the node is deleted.
     291\NOTE: Setting @do_delete@ to false may result in a detached node.
     292Subsequent code should manually delete the node or assign it to a strong pointer to prevent memory leak.
     293
    315294Reference counting functions are internally called by @ast::ptr_base@.
     295Function
    316296\begin{C++}
    317297template<typename node_t>
    318298node_t * shallowCopy(const node_t * node)
    319299\end{C++}
    320 Returns a mutable, shallow copy of node: all child pointers are pointing to the same child
    321 nodes.
     300returns a mutable, shallow copy of node: all child pointers are pointing to the same child nodes.
     301Function
    322302\begin{C++}
    323303template<typename node_t>
    324304node_t * mutate(const node_t * node)
    325305\end{C++}
    326 If node is unique (strong reference count is 1), returns a mutable pointer to the same node.
    327 Otherwise, returns shallowCopy(node).
    328 It is an error to mutate a shared node that is weak-referenced. Currently this does not
    329 happen. The problem may appear once weak pointers to shared nodes (\eg expression
    330 nodes) are used; special care will be needed.
    331 
    332 \NOTE: This naive uniqueness check may not be sufficient in some cases. A discussion of the
    333 issue is presented at the end of this section.
     306returns a mutable pointer to the same node, if the node is unique (strong reference count is 1);
     307otherwise, it returns @shallowCopy(node)@.
     308It is an error to mutate a shared node that is weak-referenced.
     309Currently this does not happen.
     310A problem may appear once weak pointers to shared nodes (\eg expression nodes) are used;
     311special care is needed.
     312
     313\NOTE: This naive uniqueness check may not be sufficient in some cases.
     314A discussion of the issue is presented at the end of this section.
     315Functions
    334316\begin{C++}
    335317template<typename node_t, typename parent_t, typename field_t, typename assn_t>
    336 const node_t * mutate_field(const node_t * node, field_t parent_t::*field, assn_t && val)
     318const node_t * mutate_field(const node_t * node, field_t parent_t::* field, assn_t && val)
    337319\end{C++}
    338320\begin{C++}
     
    342324                field_t && val)
    343325\end{C++}
    344 Helpers for mutating a field on a node using pointer to member (creates shallow copy
    345 when necessary).
    346 
    347 \subsubsection{Issue: Undetected sharing}
    348 
    349 The @mutate@ behavior described above has a problem: deeper shared nodes may be
     326are helpers for mutating a field on a node using pointer to a member function (creates shallow copy when necessary).
     327
     328
     329\subsubsection{Issue: Undetected Sharing}
     330
     331The @mutate@ behaviour described above has a problem: deeper shared nodes may be
    350332mistakenly considered as unique. \VRef[Figure]{f:DeepNodeSharing} shows how the problem could arise:
    351333\begin{figure}
     
    355337\label{f:DeepNodeSharing}
    356338\end{figure}
    357 Suppose that we are working on the tree rooted at P1, which
    358 is logically the chain P1-A-B and P2 is irrelevant, and then
    359 mutate(B) is called. The algorithm considers B as unique since
    360 it is only directly owned by A. However, the other tree P2-A-B
    361 indirectly shares the node B and is therefore wrongly mutated.
    362 
    363 To partly address this problem, if the mutation is called higher up the tree, a chain
    364 mutation helper can be used:
    365 
    366 \subsubsection{Source: AST/Chain.hpp}
    367 
     339Given the tree rooted at P1, which is logically the chain P1-A-B, and P2 is irrelevant, assume @mutate(B)@ is called.
     340The algorithm considers B as unique since it is only directly owned by A.
     341However, the other tree P2-A-B indirectly shares the node B and is therefore wrongly mutated.
     342
     343To partly address this problem, if the mutation is called higher up the tree, a chain mutation helper can be used.
     344
     345\subsubsection{Source: \lstinline{AST/Chain.hpp}}
     346
     347Function
    368348\begin{C++}
    369349template<typename node_t, Node::ref_type ref_t>
    370350auto chain_mutate(ptr_base<node_t, ref_t> & base)
    371351\end{C++}
    372 This function returns a chain mutator handle which takes pointer-to-member to go down
    373 the tree while creating shallow copies as necessary; see @struct _chain_mutator@ in the
    374 source code for details.
    375 
    376 For example, in the above diagram, if mutation of B is wanted while at P1, the call using
    377 @chain_mutate@ looks like the following:
     352returns a chain mutator handle that takes pointer-to-member to go down the tree, while creating shallow copies as necessary;
     353see @struct _chain_mutator@ in the source code for details.
     354
     355For example, in the above diagram, if mutation of B is wanted while at P1, the call using @chain_mutate@ looks like the following:
    378356\begin{C++}
    379357chain_mutate(P1.a)(&A.b) = new_value_of_b;
    380358\end{C++}
    381 Note that if some node in chain mutate is shared (therefore shallow copied), it implies that
    382 every node further down will also be copied, thus correctly executing the functional
    383 mutation algorithm. This example code creates copies of both A and B and performs
    384 mutation on the new nodes, so that the other tree P2-A-B is untouched.
    385 However, if a pass traverses down to node B and performs mutation, for example, in
    386 @postvisit(B)@, information on sharing higher up is lost. Since the new-ast structure is only in
    387 experimental use with the resolver algorithm, which mostly rebuilds the tree bottom-up,
    388 this issue does not actually happen. It should be addressed in the future when other
    389 compilation passes are migrated to new-ast and many of them contain procedural
    390 mutations, where it might cause accidental mutations to other logically independent trees
    391 (\eg common sub-expression) and become a bug.
    392 
    393 
    394 \vspace*{20pt} % FIX ME, spacing problem with this heading ???
     359\NOTE: if some node in chain mutate is shared (therefore shallow copied), it implies that every node further down is also copied, thus correctly executing the functional mutation algorithm.
     360This example code creates copies of both A and B and performs mutation on the new nodes, so that the other tree P2-A-B is untouched.
     361However, if a pass traverses down to node B and performs mutation, for example, in @postvisit(B)@, information on sharing higher up is lost.
     362Since the new-ast structure is only in experimental use with the resolver algorithm, which mostly rebuilds the tree bottom-up, this issue does not actually happen.
     363It should be addressed in the future when other compilation passes are migrated to new-ast and many of them contain procedural mutations, where it might cause accidental mutations to other logically independent trees (\eg common sub-expression) and become a bug.
     364
     365
    395366\section{Compiler Algorithm Documentation}
    396367
    397 This documentation currently covers most of the resolver, data structures used in variable
    398 and expression resolution, and a few directly related passes. Later passes involving code
    399 generation is not included yet; documentation for those will be done afterwards.
     368This compiler algorithm documentation covers most of the resolver, data structures used in variable and expression resolution, and a few directly related passes.
     369Later passes involving code generation are not included yet;
     370documentation for those will be done latter.
     371
    400372
    401373\subsection{Symbol Table}
    402374
    403 \NOTE: For historical reasons, the symbol table data structure was called ``indexer'' in the
    404 old implementation. Hereby we will be using the name SymbolTable everywhere.
    405 The symbol table stores a mapping from names to declarations and implements a similar
    406 name space separation rule, and the same scoping rules in standard C.\footnote{ISO/IEC 9899:1999, Sections 6.2.1 and 6.2.3} The difference in
    407 name space rule is that typedef aliases are no longer considered ordinary identifiers.
    408 In addition to C tag types (struct, union, enum), \CFA introduces another tag type, trait,
    409 which is a named collection of assertions.
    410 
    411 \subsubsection{Source: AST/SymbolTable.hpp}
    412 
    413 \subsubsection{Source: SymTab/Indexer.h}
    414 
     375\NOTE: For historical reasons, the symbol-table data-structure is called @indexer@ in the old implementation.
     376Hereby, the name is changed to @SymbolTable@.
     377The symbol table stores a mapping from names to declarations, implements a similar name-space separation rule, and provides the same scoping rules as standard C.\footnote{ISO/IEC 9899:1999, Sections 6.2.1 and 6.2.3.}
     378The difference in name-space rule is that @typedef@ aliases are no longer considered ordinary identifiers.
     379In addition to C tag-types (@struct@, @union@, @enum@), \CFA introduces another tag type, @trait@, which is a named collection of assertions.
     380
     381
     382\subsubsection{Source: \lstinline{AST/SymbolTable.hpp}}
     383
     384Function
    415385\begin{C++}
    416386SymbolTable::addId(const DeclWithType * decl)
    417387\end{C++}
    418 Since \CFA allows overloading of variables and functions, ordinary identifier names need
    419 to be mangled. The mangling scheme is closely based on the Itanium \CC ABI,\footnote{\url{https://itanium-cxx-abi.github.io/cxx-abi/abi.html}, Section 5.1} while
    420 making adaptations to \CFA specific features, mainly assertions and overloaded variables
    421 by type. Naming conflicts are handled by mangled names; lookup by name returns a list of
    422 declarations with the same literal identifier name.
    423 
     388provides name mangling of identifiers, since \CFA allows overloading of variables and functions.
     389The mangling scheme is closely based on the Itanium \CC ABI,\footnote{\url{https://itanium-cxx-abi.github.io/cxx-abi/abi.html}, Section 5.1} while making adaptations to \CFA specific features, mainly assertions and overloaded variables by type.
     390
     391Naming conflicts are handled by mangled names;
     392lookup by name returns a list of declarations with the same identifier name.
     393Functions
    424394\begin{C++}
    425395SymbolTable::addStruct(const StructDecl * decl)
     
    428398SymbolTable::addTrait(const TraitDecl * decl)
    429399\end{C++}
    430 Adds a tag type declaration to the symbol table.
     400add a tag-type declaration to the symbol table.
     401Function
    431402\begin{C++}
    432403SymbolTable::addType(const NamedTypeDecl * decl)
    433404\end{C++}
    434 Adds a typedef alias to the symbol table.
    435 
    436 \textbf{C Incompatibility Note}: Since Cforall allows using struct, union and enum type names
    437 without the keywords, typedef names and tag type names cannot be disambiguated by
    438 syntax rules. Currently the compiler puts them together and disallows collision. The
    439 following program is valid C but not valid Cforall:
     405adds a @typedef@ alias to the symbol table.
     406
     407\textbf{C Incompatibility Note}: Since \CFA allows using @struct@, @union@ and @enum@ type-names without a prefix keyword, as in \CC, @typedef@ names and tag-type names cannot be disambiguated by syntax rules.
     408Currently the compiler puts them together and disallows collision.
     409The following program is valid C but invalid \CFA (and \CC):
    440410\begin{C++}
    441411struct A {};
     412typedef int A; // gcc: ok, cfa: Cannot redefine typedef A
     413struct A sa; // C disambiguates via struct prefix
     414A ia;
     415\end{C++}
     416In practices, such usage is extremely rare, and hence, this change (as in \CC) has minimal impact on existing C programs.
     417The declaration
     418\begin{C++}
     419struct A {};
     420typedef struct A A; // A is an alias for struct A
     421A a;
     422struct A b;
     423\end{C++}
     424is not an error because the alias name is identical to the original.
     425Finally, the following program is allowed in \CFA:
     426\begin{C++}
    442427typedef int A;
    443 // gcc: ok, cfa: Cannot redefine typedef A
    444 \end{C++}
    445 In actual practices however, such usage is extremely rare, and typedef struct A A; is
    446 not considered an error, but silently discarded. Therefore, we expect this change to have
    447 minimal impact on existing C programs.
    448 Meanwhile, the following program is allowed in Cforall:
    449 \begin{C++}
    450 typedef int A;
    451 void A();
     428void A(); // name mangled
    452429// gcc: A redeclared as different kind of symbol, cfa: ok
    453430\end{C++}
     431because the function name is mangled.
     432
    454433
    455434\subsection{Type Environment and Unification}
    456435
    457 The core of parametric type resolution algorithm.
    458 Type Environment organizes type parameters in \textbf{equivalent classes} and maps them to
    459 actual types. Unification is the algorithm that takes two (possibly parametric) types and
    460 parameter mappings and attempts to produce a common type by matching the type
    461 environments.
     436The following core ideas underlie the parametric type-resolution algorithm.
     437A type environment organizes type parameters into \textbf{equivalent classes} and maps them to actual types.
     438Unification is the algorithm that takes two (possibly parametric) types and parameter mappings, and attempts to produce a common type by matching information in the type environments.
    462439
    463440The unification algorithm is recursive in nature and runs in two different modes internally:
    464441\begin{itemize}
    465442\item
    466 \textbf{Exact} unification mode requires equivalent parameters to match perfectly;
    467 \item
    468 \textbf{Inexact} unification mode allows equivalent parameters to be converted to a
    469 common type.
     443Exact unification mode requires equivalent parameters to match perfectly.
     444\item
     445Inexact unification mode allows equivalent parameters to be converted to a common type.
    470446\end{itemize}
    471 For a pair of matching parameters (actually, their equivalent classes), if either side is open
    472 (not bound to a concrete type yet), they are simply combined.
    473 
    474 Within inexact mode, types are allowed to differ on their cv-qualifiers; additionally, if a
    475 type never appear either in parameter list or as the base type of a pointer, it may also be
    476 widened (i.e. safely converted). As Cforall currently does not implement subclassing similar
    477 to object-oriented languages, widening conversions are on primitive types only, for
    478 example the conversion from int to long.
    479 
    480 The need for two unification modes come from the fact that parametric types are
    481 considered compatible only if all parameters are exactly the same (not just compatible).
    482 Pointer types also behaves similarly; in fact, they may be viewed as a primitive kind of
    483 parametric types. @int*@ and @long*@ are different types, just like @vector(int)@ and
    484 @vector(long)@ are, for the parametric type @vector(T)@.
    485 
    486 The resolver should use the following ``@public@'' functions:\footnote{
    487 Actual code also tracks assertions on type parameters; those extra arguments are omitted here for
    488 conciseness.}
    489 
    490 
    491 \subsubsection{Source: ResolvExpr/Unify.cc}
    492 
    493 \begin{C++}
    494 bool unify(const Type *type1, const Type *type2, TypeEnvironment &env,
    495 OpenVarSet &openVars, const SymbolTable &symtab, Type *&commonType)
    496 \end{C++}
    497 Attempts to unify @type1@ and @type2@ with current type environment.
    498 
    499 If operation succeeds, @env@ is modified by combining the equivalence classes of matching
    500 parameters in @type1@ and @type2@, and their common type is written to commonType.
    501 
    502 If operation fails, returns false.
    503 \begin{C++}
    504 bool typesCompatible(const Type * type1, const Type * type2, const
    505 SymbolTable &symtab, const TypeEnvironment &env)
    506 bool typesCompatibleIgnoreQualifiers(const Type * type1, const Type *
    507 type2, const SymbolTable &symtab, const TypeEnvironment &env)
    508 \end{C++}
    509 
    510 Determines if type1 and type2 can possibly be the same type. The second version ignores
    511 the outermost cv-qualifiers if present.\footnote{
    512 In const \lstinline@int * const@, only the second \lstinline@const@ is ignored.}
    513 
    514 The call has no side effect.
    515 
    516 \NOTE: No attempts are made to widen the types (exact unification is used), although the
    517 function names may suggest otherwise. E.g. @typesCompatible(int, long)@ returns false.
     447For a pair of matching parameters (actually, their equivalent classes), if either side is open (not bound to a concrete type yet), they are combined.
     448
     449Within the inexact mode, types are allowed to differ on their cv-qualifiers (\eg @const@, @volatile@, \etc);
     450additionally, if a type never appear either in a parameter list or as the base type of a pointer, it may also be widened (\ie safely converted).
     451As \CFA currently does not implement subclassing as in object-oriented languages, widening conversions are only on the primitive types, \eg conversion from @int@ to @long int@.
     452
     453The need for two unification modes comes from the fact that parametric types are considered compatible only if all parameters are exactly the same (not just compatible).
     454Pointer types also behaves similarly;
     455in fact, they may be viewed as a primitive kind of parametric types.
     456@int *@ and @long *@ are different types, just like @vector(int)@ and @vector(long)@ are, for the parametric type @*(T)@ / @vector(T)@, respectively.
     457
     458The resolver uses the following @public@ functions:\footnote{
     459Actual code also tracks assertions on type parameters; those extra arguments are omitted here for conciseness.}
     460
     461
     462\subsubsection{Source: \lstinline{ResolvExpr/Unify.cc}}
     463
     464Function
     465\begin{C++}
     466bool unify(const Type * type1, const Type * type2, TypeEnvironment & env,
     467        OpenVarSet & openVars, const SymbolTable & symtab, Type *& commonType)
     468\end{C++}
     469returns a boolean indicating if the unification succeeds or fails after attempting to unify @type1@ and @type2@ within current type environment.
     470If the unify succeeds, @env@ is modified by combining the equivalence classes of matching parameters in @type1@ and @type2@, and their common type is written to @commonType@.
     471If the unify fails, nothing changes.
     472Functions
     473\begin{C++}
     474bool typesCompatible(const Type * type1, const Type * type2, const SymbolTable & symtab,
     475        const TypeEnvironment & env)
     476bool typesCompatibleIgnoreQualifiers(const Type * type1, const Type * type2,
     477        const SymbolTable & symtab, const TypeEnvironment & env)
     478\end{C++}
     479return a boolean indicating if types @type1@ and @type2@ can possibly be the same type.
     480The second version ignores the outermost cv-qualifiers if present.\footnote{
     481In \lstinline@const int * const@, only the second \lstinline@const@ is ignored.}
     482These function have no side effects.
     483
     484\NOTE: No attempt is made to widen the types (exact unification is used), although the function names may suggest otherwise, \eg @typesCompatible(int, long)@ returns false.
    518485
    519486
    520487\subsection{Expression Resolution}
    521488
    522 The design of the current version of expression resolver is outlined in the Ph.D. Thesis from
    523 Aaron Moss~\cite{Moss19}.
    524 
     489The design of the current version of expression resolver is outlined in the Ph.D.\ thesis by Aaron Moss~\cite{Moss19}.
    525490A summary of the resolver algorithm for each expression type is presented below.
    526491
    527 All overloadable operators are modelled as function calls. For a function call,
    528 interpretations of the function and arguments are found recursively. Then the following
    529 steps produce a filtered list of valid interpretations:
     492All overloadable operators are modelled as function calls.
     493For a function call, interpretations of the function and arguments are found recursively.
     494Then the following steps produce a filtered list of valid interpretations:
    530495\begin{enumerate}
    531496\item
    532 From all possible combinations of interpretations of the function and arguments,
    533 those where argument types may be converted to function parameter types are
    534 considered valid.
     497From all possible combinations of interpretations of the function and arguments, those where argument types may be converted to function parameter types are considered valid.
    535498\item
    536499Valid interpretations with the minimum sum of argument costs are kept.
    537500\item
    538 Argument costs are then discarded; the actual cost for the function call expression is
    539 the sum of conversion costs from the argument types to parameter types.
    540 \item
    541 For each return type, the interpretations with satisfiable assertions are then sorted
    542 by actual cost computed in step 3. If for a given type, the minimum cost
    543 interpretations are not unique, it is said that for that return type the interpretation
    544 is ambiguous. If the minimum cost interpretation is unique but contains an
    545 ambiguous argument, it is also considered ambiguous.
     501\label{p:argcost}
     502Argument costs are then discarded; the actual cost for the function call expression is the sum of conversion costs from the argument types to parameter types.
     503\item
     504\label{p:returntype}
     505For each return type, the interpretations with satisfiable assertions are then sorted by actual cost computed in step~\ref{p:argcost}.
     506If for a given type, the minimum cost interpretations are not unique, that return type is ambiguous.
     507If the minimum cost interpretation is unique but contains an ambiguous argument, it is also ambiguous.
    546508\end{enumerate}
    547 Therefore, for each return type, the resolver produces either of:
     509Therefore, for each return type, the resolver produces:
    548510\begin{itemize}
    549511\item
    550 No alternatives
    551 \item
    552 A single valid alternative
    553 \item
    554 An ambiguous alternative
     512no alternatives
     513\item
     514a single valid alternative
     515\item
     516an ambiguous alternative
    555517\end{itemize}
    556 Note that an ambiguous alternative may be discarded at the parent expressions because a
    557 different return type matches better for the parent expressions.
    558 
    559 The non-overloadable expressions in Cforall are: cast expressions, address-of (unary @&@)
    560 expressions, short-circuiting logical expressions (@&&@, @||@) and ternary conditional
    561 expression (@?:@).
    562 
    563 For a cast expression, the convertible argument types are kept. Then the result is selected
    564 by lowest argument cost, and further by lowest conversion cost to target type. If the lowest
    565 cost is still not unique, or an ambiguous argument interpretation is selected, the cast
    566 expression is ambiguous. In an expression statement, the top level expression is implicitly
    567 cast to void.
     518\NOTE: an ambiguous alternative may be discarded at the parent expressions because a different return type matches better for the parent expressions.
     519
     520The \emph{non}-overloadable expressions in \CFA are: cast expressions, address-of (unary @&@) expressions, short-circuiting logical expressions (@&&@, @||@) and ternary conditional expression (@?:@).
     521
     522For a cast expression, the convertible argument types are kept.
     523Then the result is selected by lowest argument cost, and further by lowest conversion cost to target type.
     524If the lowest cost is still not unique or an ambiguous argument interpretation is selected, the cast expression is ambiguous.
     525In an expression statement, the top level expression is implicitly cast to @void@.
    568526
    569527For an address-of expression, only lvalue results are kept and the minimum cost is selected.
    570528
    571 For logical expressions @&&@ and @||@, arguments are implicitly cast to bool, and follow the rule
    572 of cast expression as above.
    573 
    574 For the ternary conditional expression, the condition is implicitly cast to bool, and the
    575 branch expressions must have compatible types. Each pair of compatible branch
    576 expression types produce a possible interpretation, and the cost is defined as the sum of
    577 expression costs plus the sum of conversion costs to the common type.
    578 
    579 TODO: Write a specification for expression costs.
     529For logical expressions @&&@ and @||@, arguments are implicitly cast to @bool@, and follow the rules fr cast expression above.
     530
     531For the ternary conditional expression, the condition is implicitly cast to @bool@, and the branch expressions must have compatible types.
     532Each pair of compatible branch expression types produce a possible interpretation, and the cost is defined as the sum of the expression costs plus the sum of conversion costs to the common type.
     533
     534
     535\subsection{Conversion and Application Cost}
     536
     537There were some unclear parts in the previous documentation in the cost system, as described in the Moss thesis~\cite{Moss19}, section 4.1.2.
     538Some clarification are presented in this section.
     539
     540\begin{enumerate}
     541\item
     542Conversion to a type denoted by parameter may incur additional cost if the match is not exact.
     543For example, if a function is declared to accept @(T, T)@ and receives @(int, long)@, @T@ is deducted @long@ and an additional widening conversion cost is added for @int@ to @T@.
     544
     545\item
     546The specialization level of a function is the sum of the least depth of an appearance of a type parameter (counting pointers, references and parameterized types), plus the number of assertions.
     547A higher specialization level is favoured if argument conversion costs are equal.
     548
     549\item
     550Coercion of pointer types is only allowed in explicit cast expressions;
     551the only allowed implicit pointer casts are adding qualifiers to the base type and cast to @void*@, and these counts as safe conversions.
     552Note that implicit cast from @void *@ to other pointer types is no longer valid, as opposed to standard C.
     553\end{enumerate}
    580554
    581555
    582556\subsection{Assertion Satisfaction}
    583557
    584 The resolver tries to satisfy assertions on expressions only when it is needed: either while
    585 selecting from multiple alternatives of a same result type for a function call (step 4 of
    586 resolving function calls), or upon reaching the top level of an expression statement.
    587 
    588 Unsatisfiable alternatives are discarded. Satisfiable alternatives receive \textbf{implicit
    589 parameters}: in Cforall, parametric functions are designed such that they can be compiled
    590 separately, as opposed to \CC templates which are only compiled at instantiation. Given a
    591 parametric function definition:
     558The resolver tries to satisfy assertions on expressions only when it is needed: either while selecting from multiple alternatives of a same result type for a function call (step \ref{p:returntype} of resolving function calls) or upon reaching the top level of an expression statement.
     559
     560Unsatisfiable alternatives are discarded.
     561Satisfiable alternatives receive \textbf{implicit parameters}: in \CFA, parametric functions may be separately compiled, as opposed to \CC templates which are only compiled at instantiation.
     562Given the parametric function-definition:
    592563\begin{C++}
    593564forall (otype T | {void foo(T);})
    594565void bar (T t) { foo(t); }
    595566\end{C++}
    596 The function bar does not know which @foo@ to call when compiled without knowing the call
    597 site, so it requests a function pointer to be passed as an extra argument. At the call site,
    598 implicit parameters are automatically inserted by the compiler.
    599 
    600 \textbf{TODO}: Explain how recursive assertion satisfaction and polymorphic recursion work.
    601 
     567the function @bar@ does not know which @foo@ to call when compiled without knowing the call site, so it requests a function pointer to be passed as an extra argument.
     568At the call site, implicit parameters are automatically inserted by the compiler.
     569
     570Implementation of implicit parameters is discussed in \VRef[Appendix]{s:ImplementationParametricFunctions}.
    602571
    603572\section{Tests}
     
    605574\subsection{Test Suites}
    606575
    607 Automatic test suites are located under the @tests/@ directory. A test case consists of an
    608 input CFA source file (name ending with @.cfa@), and an expected output file located
    609 in @.expect/@ directory relative to the source file, with the same file name ending with @.txt@.
    610 So a test named @tuple/tupleCast@ has the following files, for example:
     576Automatic test suites are located under the @tests/@ directory.
     577A test case consists of an input CFA source file (suffix @.cfa@), and an expected output file located in the @tests/.expect/@ directory, with the same file name ending with suffix @.txt@.
     578For example, the test named @tests/tuple/tupleCast.cfa@ has the following files, for example:
    611579\begin{C++}
    612580tests/
    613 ..     tuple/
    614 ......     .expect/
    615 ..........       tupleCast.txt
    616 ......     tupleCast.cfa
    617 \end{C++}
    618 If compilation fails, the error output is compared to the expect file. If compilation succeeds,
    619 the built program is run and its output compared to the expect file.
    620 To run the tests, execute the test script @test.py@ under the @tests/@ directory, with a list of
    621 test names to be run, or @--all@ to run all tests. The test script reports test cases
    622 fail/success, compilation time and program run time.
     581        tuple/
     582                .expect/
     583                        tupleCast.txt
     584                tupleCast.cfa
     585\end{C++}
     586If compilation fails, the error output is compared to the expect file.
     587If the compilation succeeds but does not generate an executable, the compilation output is compared to the expect file.
     588If the compilation succeeds and generates an executable, the executable is run and its output is compared to the expect file.
     589To run the tests, execute the test script @test.py@ under the @tests/@ directory, with a list of test names to be run, or @--all@ (or @make all-tests@) to run all tests.
     590The test script reports test cases fail/success, compilation time and program run time.
     591To see all the options available for @test.py@ using the @--help@ option.
    623592
    624593
    625594\subsection{Performance Reports}
    626595
    627 To turn on performance reports, pass @-S@ flag to the compiler.
    628 
    629 3 kinds of performance reports are available:
     596To turn on performance reports, pass the @-XCFA -S@ flag to the compiler.
     597Three kinds of performance reports are available:
    630598\begin{enumerate}
    631599\item
     
    639607@Common/Stats/Counter.h@.
    640608\end{enumerate}
    641 It is suggested to run performance tests with optimized build (@g++@ flag @-O3@)
    642 
     609It is suggested to run performance tests with optimization (@g++@ flag @-O3@).
     610
     611
     612\appendix
     613\section{Appendix}
     614
     615\subsection{Kinds of Type Parameters}
     616\label{s:KindsTypeParameters}
     617
     618A type parameter in a @forall@ clause has 3 kinds:
     619\begin{enumerate}[listparindent=0pt]
     620\item
     621@dtype@: any data type (built-in or user defined) that is not a concrete type.
     622
     623A non-concrete type is an incomplete type such as an opaque type or pointer/reference with an implicit (pointer) size and implicitly generated reference and dereference operations.
     624\item
     625@otype@: any data type (built-in or user defined) that is concrete type.
     626
     627A concrete type is a complete type, \ie types that can be used to create a variable, which also implicitly asserts the existence of default and copy constructors, assignment, and destructor\footnote{\CFA implements the same automatic resource management (RAII) semantics as \CC.}.
     628% \item
     629% @ftype@: any function type.
     630%
     631% @ftype@ provides two purposes:
     632% \begin{itemize}
     633% \item
     634% Differentiate function pointer from data pointer because (in theory) some systems have different sizes for these pointers.
     635% \item
     636% Disallow a function pointer to match an overloaded data pointer, since variables and functions can have the same names.
     637% \end{itemize}
     638
     639\item
     640@ttype@: tuple (variadic) type.
     641
     642Restricted to the type for the last parameter in a function, it provides a type-safe way to implement variadic functions.
     643Note however, that it has certain restrictions, as described in the implementation section below.
     644\end{enumerate}
     645
     646
     647\subsection{GNU C Nested Functions}
     648
     649\CFA is designed to be mostly compatible with GNU C, an extension to ISO C99 and C11 standards. The \CFA compiler also implements some language features by GCC extensions, most notably nested functions.
     650
     651In ISO C, function definitions are not allowed to be nested. GCC allows nested functions with full lexical scoping. The following example is taken from GCC documentation\footnote{\url{https://gcc.gnu.org/onlinedocs/gcc/Nested-Functions.html}}:
     652\begin{C++}
     653void bar( int * array, int offset, int size ) {
     654        int access( int * array, int index ) { return array[index + offset]; }
     655        int i;
     656        /* ... */
     657        for ( i = 0; i < size; i++ )
     658                /* ... */ access (array, i) /* ... */
     659}
     660\end{C++}
     661GCC nested functions behave identically to \CC lambda functions with default by-reference capture (stack-allocated, lifetime ends upon exiting the declared block), while also possible to be passed as arguments with standard function pointer types.
     662
     663
     664\subsection{Implementation of Parametric Functions}
     665\label{s:ImplementationParametricFunctions}
     666
     667\CFA implements parametric functions using the implicit parameter approach: required assertions are passed to the callee by function pointers;
     668size of a parametric type must also be known if referenced directly (\ie not as a pointer).
     669
     670The implementation is similar to the one from Scala\footnote{\url{https://www.scala-lang.org/files/archive/spec/2.13/07-implicits.html}}, with some notable differences in resolution:
     671\begin{enumerate}
     672\item
     673All types, variables, and functions are candidates of implicit parameters
     674\item
     675The parameter (assertion) name must match the actual declarations.
     676\end{enumerate}
     677
     678For example, the \CFA function declaration
     679\begin{cfa}
     680forall( otype T | { int foo( T, int ); } )
     681int bar(T);
     682\end{cfa}
     683after implicit parameter expansion, has the actual signature\footnote{\textbf{otype} also requires the type to have constructor and destructor, which are the first two function pointers preceding the one for \textbf{foo}.}
     684\begin{C++}
     685int bar( T, size_t, void (*)(T&), void (*)(T&), int (*)(T, int) );
     686\end{C++}
     687The implicit parameter approach has an apparent issue: when the satisfying declaration is also parametric, it may require its own implicit parameters too.
     688That also causes the supplied implicit parameter to have a different \textbf{actual} type than the \textbf{nominal} type, so it cannot be passed directly.
     689Therefore, a wrapper with matching actual type must be created, and it is here where GCC nested functions are used internally by the compiler.
     690
     691Consider the following program:
     692\begin{cfa}
     693int assertion(int);
     694
     695forall( otype T | { int assertion(T); } )
     696void foo(T);
     697
     698forall(otype T | { void foo(T); } )
     699void bar(T t) {
     700        foo(t);
     701}
     702\end{cfa}
     703The \CFA compiler translates the program to non-parametric form\footnote{In the final code output, \lstinline@T@ needs to be replaced by an opaque type, and arguments must be accessed by a frame pointer offset table, due to the unknown sizes. The presented code here is simplified for better understanding.}
     704\begin{C++}
     705// ctor, dtor and size arguments are omitted
     706void foo(T, int (*)(T));
     707
     708void bar(T t, void (*foo)(T)) {
     709        foo(t);
     710}
     711\end{C++}
     712However, when @bar(1)@ is called, @foo@ cannot be directly provided as an argument:
     713\begin{C++}
     714bar(1, foo); // WRONG: foo has different actual type
     715\end{C++}
     716and an additional step is required:
     717\begin{C++}
     718{
     719        void _foo_wrapper(int t) {
     720                foo( t, assertion );
     721        }
     722        bar( 1, _foo_wrapper );
     723}
     724\end{C++}
     725Nested assertions and implicit parameter creation may continue indefinitely.
     726This issue is a limitation of implicit parameter implementation.
     727In particular, polymorphic variadic recursion must be structural (\ie the number of arguments decreases in any possible recursive calls), otherwise code generation gets into an infinite loop.
     728The \CFA compiler sets a limit on assertion depth and reports an error if assertion resolution does not terminate within the limit (as for \lstinline[language=C++]@templates@ in \CC).
    643729
    644730\bibliographystyle{plain}
  • doc/user/Makefile

    r33c3ded r223a633  
    5555
    5656${DOCUMENT} : ${BASE}.ps
    57         ps2pdf $<
     57        ps2pdf -dPDFSETTINGS=/prepress $<
    5858
    5959${BASE}.ps : ${BASE}.dvi
  • doc/user/user.tex

    r33c3ded r223a633  
    1111%% Created On       : Wed Apr  6 14:53:29 2016
    1212%% Last Modified By : Peter A. Buhr
    13 %% Last Modified On : Fri Mar  6 13:34:52 2020
    14 %% Update Count     : 3924
     13%% Last Modified On : Mon Oct  5 08:57:29 2020
     14%% Update Count     : 3998
    1515%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
    1616
     
    3030\usepackage{upquote}                                                                    % switch curled `'" to straight
    3131\usepackage{calc}
    32 \usepackage{xspace}
    3332\usepackage{varioref}                                                                   % extended references
    34 \usepackage{listings}                                                                   % format program code
     33\usepackage[labelformat=simple,aboveskip=0pt,farskip=0pt]{subfig}
     34\renewcommand{\thesubfigure}{\alph{subfigure})}
    3535\usepackage[flushmargin]{footmisc}                                              % support label/reference in footnote
    3636\usepackage{latexsym}                                   % \Box glyph
    3737\usepackage{mathptmx}                                   % better math font with "times"
    3838\usepackage[usenames]{color}
    39 \input{common}                                          % common CFA document macros
    40 \usepackage[dvips,plainpages=false,pdfpagelabels,pdfpagemode=UseNone,colorlinks=true,pagebackref=true,linkcolor=blue,citecolor=blue,urlcolor=blue,pagebackref=true,breaklinks=true]{hyperref}
    41 \usepackage{breakurl}
    42 
    43 \usepackage[pagewise]{lineno}
    44 \renewcommand{\linenumberfont}{\scriptsize\sffamily}
    45 \usepackage[firstpage]{draftwatermark}
    46 \SetWatermarkLightness{0.9}
    47 
    48 % Default underscore is too low and wide. Cannot use lstlisting "literate" as replacing underscore
    49 % removes it as a variable-name character so keywords in variables are highlighted. MUST APPEAR
    50 % AFTER HYPERREF.
    51 \renewcommand{\textunderscore}{\leavevmode\makebox[1.2ex][c]{\rule{1ex}{0.075ex}}}
    52 
    53 \setlength{\topmargin}{-0.45in}                                                 % move running title into header
    54 \setlength{\headsep}{0.25in}
    55 
    56 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
    57 
    58 \CFAStyle                                                                                               % use default CFA format-style
    59 \lstnewenvironment{C++}[1][]                            % use C++ style
    60 {\lstset{language=C++,moredelim=**[is][\protect\color{red}]{®}{®},#1}}
    61 {}
    62 
     39\newcommand{\CFALatin}{}
    6340% inline code ©...© (copyright symbol) emacs: C-q M-)
    6441% red highlighting ®...® (registered trademark symbol) emacs: C-q M-.
     
    6845% keyword escape ¶...¶ (pilcrow symbol) emacs: C-q M-^
    6946% math escape $...$ (dollar symbol)
     47\input{common}                                          % common CFA document macros
     48\usepackage[dvips,plainpages=false,pdfpagelabels,pdfpagemode=UseNone,colorlinks=true,pagebackref=true,linkcolor=blue,citecolor=blue,urlcolor=blue,pagebackref=true,breaklinks=true]{hyperref}
     49\usepackage{breakurl}
     50
     51\renewcommand\footnoterule{\kern -3pt\rule{0.3\linewidth}{0.15pt}\kern 2pt}
     52
     53\usepackage[pagewise]{lineno}
     54\renewcommand{\linenumberfont}{\scriptsize\sffamily}
     55\usepackage[firstpage]{draftwatermark}
     56\SetWatermarkLightness{0.9}
     57
     58% Default underscore is too low and wide. Cannot use lstlisting "literate" as replacing underscore
     59% removes it as a variable-name character so keywords in variables are highlighted. MUST APPEAR
     60% AFTER HYPERREF.
     61\renewcommand{\textunderscore}{\leavevmode\makebox[1.2ex][c]{\rule{1ex}{0.075ex}}}
     62
     63\setlength{\topmargin}{-0.45in}                                                 % move running title into header
     64\setlength{\headsep}{0.25in}
     65
     66%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
     67
     68\CFAStyle                                                                                               % use default CFA format-style
     69\lstnewenvironment{C++}[1][]                            % use C++ style
     70{\lstset{language=C++,moredelim=**[is][\protect\color{red}]{®}{®},#1}}
     71{}
     72
     73\newsavebox{\myboxA}
     74\newsavebox{\myboxB}
    7075
    7176%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
     
    7984\newcommand{\G}[1]{{\Textbf[OliveGreen]{#1}}}
    8085\newcommand{\KWC}{K-W C\xspace}
    81 
    82 \newsavebox{\LstBox}
    8386
    8487%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
     
    253256
    254257The signature feature of \CFA is \emph{\Index{overload}able} \Index{parametric-polymorphic} functions~\cite{forceone:impl,Cormack90,Duggan96} with functions generalized using a ©forall© clause (giving the language its name):
    255 \begin{lstlisting}
     258\begin{cfa}
    256259®forall( otype T )® T identity( T val ) { return val; }
    257260int forty_two = identity( 42 ); §\C{// T is bound to int, forty\_two == 42}§
    258 \end{lstlisting}
     261\end{cfa}
    259262% extending the C type system with parametric polymorphism and overloading, as opposed to the \Index*[C++]{\CC{}} approach of object-oriented extensions.
    260263\CFA{}\hspace{1pt}'s polymorphism was originally formalized by \Index*{Glen Ditchfield}\index{Ditchfield, Glen}~\cite{Ditchfield92}, and first implemented by \Index*{Richard Bilson}\index{Bilson, Richard}~\cite{Bilson03}.
     
    275278\begin{comment}
    276279A simple example is leveraging the existing type-unsafe (©void *©) C ©bsearch© to binary search a sorted floating array:
    277 \begin{lstlisting}
     280\begin{cfa}
    278281void * bsearch( const void * key, const void * base, size_t dim, size_t size,
    279282                                int (* compar)( const void *, const void * ));
     
    284287double key = 5.0, vals[10] = { /* 10 sorted floating values */ };
    285288double * val = (double *)bsearch( &key, vals, 10, sizeof(vals[0]), comp ); §\C{// search sorted array}§
    286 \end{lstlisting}
     289\end{cfa}
    287290which can be augmented simply with a polymorphic, type-safe, \CFA-overloaded wrappers:
    288 \begin{lstlisting}
     291\begin{cfa}
    289292forall( otype T | { int ?<?( T, T ); } ) T * bsearch( T key, const T * arr, size_t size ) {
    290293        int comp( const void * t1, const void * t2 ) { /* as above with double changed to T */ }
     
    297300double * val = bsearch( 5.0, vals, 10 ); §\C{// selection based on return type}§
    298301int posn = bsearch( 5.0, vals, 10 );
    299 \end{lstlisting}
     302\end{cfa}
    300303The nested function ©comp© provides the hidden interface from typed \CFA to untyped (©void *©) C, plus the cast of the result.
    301304Providing a hidden ©comp© function in \CC is awkward as lambdas do not use C calling-conventions and template declarations cannot appear at block scope.
     
    305308\CFA has replacement libraries condensing hundreds of existing C functions into tens of \CFA overloaded functions, all without rewriting the actual computations.
    306309For example, it is possible to write a type-safe \CFA wrapper ©malloc© based on the C ©malloc©:
    307 \begin{lstlisting}
     310\begin{cfa}
    308311forall( dtype T | sized(T) ) T * malloc( void ) { return (T *)malloc( sizeof(T) ); }
    309312int * ip = malloc(); §\C{// select type and size from left-hand side}§
    310313double * dp = malloc();
    311314struct S {...} * sp = malloc();
    312 \end{lstlisting}
     315\end{cfa}
    313316where the return type supplies the type/size of the allocation, which is impossible in most type systems.
    314317\end{comment}
     
    943946the same level as a ©case© clause; the target label may be case ©default©, but only associated
    944947with the current ©switch©/©choose© statement.
    945 
    946 
    947 \subsection{Loop Control}
    948 
    949 The ©for©/©while©/©do-while© loop-control allows empty or simplified ranges (see Figure~\ref{f:LoopControlExamples}).
    950 \begin{itemize}
    951 \item
    952 The loop index is polymorphic in the type of the comparison value N (when the start value is implicit) or the start value M.
    953 \item
    954 An empty conditional implies comparison value of ©1© (true).
    955 \item
    956 A comparison N is implicit up-to exclusive range [0,N©®)®©.
    957 \item
    958 A comparison ©=© N is implicit up-to inclusive range [0,N©®]®©.
    959 \item
    960 The up-to range M ©~©\index{~@©~©} N means exclusive range [M,N©®)®©.
    961 \item
    962 The up-to range M ©~=©\index{~=@©~=©} N means inclusive range [M,N©®]®©.
    963 \item
    964 The down-to range M ©-~©\index{-~@©-~©} N means exclusive range [N,M©®)®©.
    965 \item
    966 The down-to range M ©-~=©\index{-~=@©-~=©} N means inclusive range [N,M©®]®©.
    967 \item
    968 ©0© is the implicit start value;
    969 \item
    970 ©1© is the implicit increment value.
    971 \item
    972 The up-to range uses operator ©+=© for increment;
    973 \item
    974 The down-to range uses operator ©-=© for decrement.
    975 \item
    976 ©@© means put nothing in this field.
    977 \item
    978 ©:© means start another index.
    979 \end{itemize}
    980948
    981949\begin{figure}
     
    10861054
    10871055
     1056\subsection{Loop Control}
     1057
     1058The ©for©/©while©/©do-while© loop-control allows empty or simplified ranges (see Figure~\ref{f:LoopControlExamples}).
     1059\begin{itemize}
     1060\item
     1061The loop index is polymorphic in the type of the comparison value N (when the start value is implicit) or the start value M.
     1062\item
     1063An empty conditional implies comparison value of ©1© (true).
     1064\item
     1065A comparison N is implicit up-to exclusive range [0,N©®)®©.
     1066\item
     1067A comparison ©=© N is implicit up-to inclusive range [0,N©®]®©.
     1068\item
     1069The up-to range M ©~©\index{~@©~©} N means exclusive range [M,N©®)®©.
     1070\item
     1071The up-to range M ©~=©\index{~=@©~=©} N means inclusive range [M,N©®]®©.
     1072\item
     1073The down-to range M ©-~©\index{-~@©-~©} N means exclusive range [N,M©®)®©.
     1074\item
     1075The down-to range M ©-~=©\index{-~=@©-~=©} N means inclusive range [N,M©®]®©.
     1076\item
     1077©0© is the implicit start value;
     1078\item
     1079©1© is the implicit increment value.
     1080\item
     1081The up-to range uses operator ©+=© for increment;
     1082\item
     1083The down-to range uses operator ©-=© for decrement.
     1084\item
     1085©@© means put nothing in this field.
     1086\item
     1087©:© means start another index.
     1088\end{itemize}
     1089
     1090
    10881091%\subsection{\texorpdfstring{Labelled \protect\lstinline@continue@ / \protect\lstinline@break@}{Labelled continue / break}}
    10891092\subsection{\texorpdfstring{Labelled \LstKeywordStyle{continue} / \LstKeywordStyle{break} Statement}{Labelled continue / break Statement}}
     
    10951098for ©break©, the target label can also be associated with a ©switch©, ©if© or compound (©{}©) statement.
    10961099\VRef[Figure]{f:MultiLevelExit} shows ©continue© and ©break© indicating the specific control structure, and the corresponding C program using only ©goto© and labels.
    1097 The innermost loop has 7 exit points, which cause continuation or termination of one or more of the 7 \Index{nested control-structure}s.
     1100The innermost loop has 8 exit points, which cause continuation or termination of one or more of the 7 \Index{nested control-structure}s.
    10981101
    10991102\begin{figure}
    1100 \begin{tabular}{@{\hspace{\parindentlnth}}l@{\hspace{\parindentlnth}}l@{\hspace{\parindentlnth}}l@{}}
    1101 \multicolumn{1}{@{\hspace{\parindentlnth}}c@{\hspace{\parindentlnth}}}{\textbf{\CFA}}   & \multicolumn{1}{@{\hspace{\parindentlnth}}c}{\textbf{C}}      \\
    1102 \begin{cfa}
    1103 ®LC:® {
    1104         ... §declarations§ ...
    1105         ®LS:® switch ( ... ) {
    1106           case 3:
    1107                 ®LIF:® if ( ... ) {
    1108                         ®LF:® for ( ... ) {
    1109                                 ®LW:® while ( ... ) {
    1110                                         ... break ®LC®; ...
    1111                                         ... break ®LS®; ...
    1112                                         ... break ®LIF®; ...
    1113                                         ... continue ®LF;® ...
    1114                                         ... break ®LF®; ...
    1115                                         ... continue ®LW®; ...
    1116                                         ... break ®LW®; ...
    1117                                 } // while
    1118                         } // for
    1119                 } else {
    1120                         ... break ®LIF®; ...
    1121                 } // if
    1122         } // switch
     1103\centering
     1104\begin{lrbox}{\myboxA}
     1105\begin{cfa}[tabsize=3]
     1106®Compound:® {
     1107        ®Try:® try {
     1108                ®For:® for ( ... ) {
     1109                        ®While:® while ( ... ) {
     1110                                ®Do:® do {
     1111                                        ®If:® if ( ... ) {
     1112                                                ®Switch:® switch ( ... ) {
     1113                                                        case 3:
     1114                                                                ®break Compound®;
     1115                                                                ®break Try®;
     1116                                                                ®break For®;      /* or */  ®continue For®;
     1117                                                                ®break While®;  /* or */  ®continue While®;
     1118                                                                ®break Do®;      /* or */  ®continue Do®;
     1119                                                                ®break If®;
     1120                                                                ®break Switch®;
     1121                                                        } // switch
     1122                                                } else {
     1123                                                        ... ®break If®; ...     // terminate if
     1124                                                } // if
     1125                                } while ( ... ); // do
     1126                        } // while
     1127                } // for
     1128        } ®finally® { // always executed
     1129        } // try
    11231130} // compound
    11241131\end{cfa}
    1125 &
    1126 \begin{cfa}
     1132\end{lrbox}
     1133
     1134\begin{lrbox}{\myboxB}
     1135\begin{cfa}[tabsize=3]
    11271136{
    1128         ... §declarations§ ...
    1129         switch ( ... ) {
    1130           case 3:
    1131                 if ( ... ) {
    1132                         for ( ... ) {
    1133                                 while ( ... ) {
    1134                                         ... goto ®LC®; ...
    1135                                         ... goto ®LS®; ...
    1136                                         ... goto ®LIF®; ...
    1137                                         ... goto ®LFC®; ...
    1138                                         ... goto ®LFB®; ...
    1139                                         ... goto ®LWC®; ...
    1140                                         ... goto ®LWB®; ...
    1141                                   ®LWC®: ; } ®LWB:® ;
    1142                           ®LFC:® ; } ®LFB:® ;
    1143                 } else {
    1144                         ... goto ®LIF®; ...
    1145                 } ®L3:® ;
    1146         } ®LS:® ;
    1147 } ®LC:® ;
    1148 \end{cfa}
    1149 &
    1150 \begin{cfa}
    1151 
    1152 
    1153 
    1154 
    1155 
    1156 
    1157 
    1158 // terminate compound
    1159 // terminate switch
    1160 // terminate if
    1161 // continue loop
    1162 // terminate loop
    1163 // continue loop
    1164 // terminate loop
    1165 
    1166 
    1167 
    1168 // terminate if
    1169 
    1170 
    1171 
    1172 \end{cfa}
    1173 \end{tabular}
     1137
     1138                ®ForC:® for ( ... ) {
     1139                        ®WhileC:® while ( ... ) {
     1140                                ®DoC:® do {
     1141                                        if ( ... ) {
     1142                                                switch ( ... ) {
     1143                                                        case 3:
     1144                                                                ®goto Compound®;
     1145                                                                ®goto Try®;
     1146                                                                ®goto ForB®;      /* or */  ®goto ForC®;
     1147                                                                ®goto WhileB®;  /* or */  ®goto WhileC®;
     1148                                                                ®goto DoB®;      /* or */  ®goto DoC®;
     1149                                                                ®goto If®;
     1150                                                                ®goto Switch®;
     1151                                                        } ®Switch:® ;
     1152                                                } else {
     1153                                                        ... ®goto If®; ...      // terminate if
     1154                                                } ®If:®;
     1155                                } while ( ... ); ®DoB:® ;
     1156                        } ®WhileB:® ;
     1157                } ®ForB:® ;
     1158
     1159
     1160} ®Compound:® ;
     1161\end{cfa}
     1162\end{lrbox}
     1163
     1164\subfloat[\CFA]{\label{f:CFibonacci}\usebox\myboxA}
     1165\hspace{2pt}
     1166\vrule
     1167\hspace{2pt}
     1168\subfloat[C]{\label{f:CFAFibonacciGen}\usebox\myboxB}
    11741169\caption{Multi-level Exit}
    11751170\label{f:MultiLevelExit}
     
    14261421try {
    14271422        f(...);
    1428 } catch( E e ; §boolean-predicate§ ) {          §\C[8cm]{// termination handler}§
     1423} catch( E e ; §boolean-predicate§ ) {          §\C{// termination handler}§
    14291424        // recover and continue
    1430 } catchResume( E e ; §boolean-predicate§ ) { §\C{// resumption handler}\CRT§
     1425} catchResume( E e ; §boolean-predicate§ ) { §\C{// resumption handler}§
    14311426        // repair and return
    14321427} finally {
     
    34913486For implicit formatted input, the common case is reading a sequence of values separated by whitespace, where the type of an input constant must match with the type of the input variable.
    34923487\begin{cquote}
    3493 \begin{lrbox}{\LstBox}
     3488\begin{lrbox}{\myboxA}
    34943489\begin{cfa}[aboveskip=0pt,belowskip=0pt]
    34953490int x;   double y   char z;
     
    34973492\end{lrbox}
    34983493\begin{tabular}{@{}l@{\hspace{3em}}l@{\hspace{3em}}l@{}}
    3499 \multicolumn{1}{@{}l@{}}{\usebox\LstBox} \\
     3494\multicolumn{1}{@{}l@{}}{\usebox\myboxA} \\
    35003495\multicolumn{1}{c@{\hspace{2em}}}{\textbf{\CFA}}        & \multicolumn{1}{c@{\hspace{2em}}}{\textbf{\CC}}       & \multicolumn{1}{c}{\textbf{Python}}   \\
    35013496\begin{cfa}[aboveskip=0pt,belowskip=0pt]
     
    66726667For example, an initial alignment and fill capability are preserved during a resize copy so the copy has the same alignment and extended storage is filled.
    66736668Without sticky properties it is dangerous to use ©realloc©, resulting in an idiom of manually performing the reallocation to maintain correctness.
     6669\begin{cfa}
     6670
     6671\end{cfa}
    66746672
    66756673\CFA memory management extends allocation to support constructors for initialization of allocated storage, \eg in
     
    67216719
    67226720        // §\CFA§ safe general allocation, fill, resize, alignment, array
    6723         T * alloc( void );§\indexc{alloc}§
    6724         T * alloc( size_t dim );
    6725         T * alloc( T ptr[], size_t dim );
    6726         T * alloc_set( char fill );§\indexc{alloc_set}§
    6727         T * alloc_set( T fill );
    6728         T * alloc_set( size_t dim, char fill );
    6729         T * alloc_set( size_t dim, T fill );
    6730         T * alloc_set( size_t dim, const T fill[] );
    6731         T * alloc_set( T ptr[], size_t dim, char fill );
    6732 
    6733         T * alloc_align( size_t align );
    6734         T * alloc_align( size_t align, size_t dim );
    6735         T * alloc_align( T ptr[], size_t align ); // aligned realloc array
    6736         T * alloc_align( T ptr[], size_t align, size_t dim ); // aligned realloc array
    6737         T * alloc_align_set( size_t align, char fill );
    6738         T * alloc_align_set( size_t align, T fill );
    6739         T * alloc_align_set( size_t align, size_t dim, char fill );
    6740         T * alloc_align_set( size_t align, size_t dim, T fill );
    6741         T * alloc_align_set( size_t align, size_t dim, const T fill[] );
    6742         T * alloc_align_set( T ptr[], size_t align, size_t dim, char fill );
     6721        T * alloc( void );§\indexc{alloc}§                                      §\C[3.5in]{// variable, T size}§
     6722        T * alloc( size_t dim );                                                        §\C{// array[dim], T size elements}§
     6723        T * alloc( T ptr[], size_t dim );                                       §\C{// realloc array[dim], T size elements}§
     6724
     6725        T * alloc_set( char fill );§\indexc{alloc_set}§         §\C{// variable, T size, fill bytes with value}§
     6726        T * alloc_set( T fill );                                                        §\C{// variable, T size, fill with value}§
     6727        T * alloc_set( size_t dim, char fill );                         §\C{// array[dim], T size elements, fill bytes with value}§
     6728        T * alloc_set( size_t dim, T fill );                            §\C{// array[dim], T size elements, fill elements with value}§
     6729        T * alloc_set( size_t dim, const T fill[] );            §\C{// array[dim], T size elements, fill elements with array}§
     6730        T * alloc_set( T ptr[], size_t dim, char fill );        §\C{// realloc array[dim], T size elements, fill bytes with value}§
     6731
     6732        T * alloc_align( size_t align );                                        §\C{// aligned variable, T size}§
     6733        T * alloc_align( size_t align, size_t dim );            §\C{// aligned array[dim], T size elements}§
     6734        T * alloc_align( T ptr[], size_t align );                       §\C{// realloc new aligned array}§
     6735        T * alloc_align( T ptr[], size_t align, size_t dim ); §\C{// realloc new aligned array[dim]}§
     6736
     6737        T * alloc_align_set( size_t align, char fill );         §\C{// aligned variable, T size, fill bytes with value}§
     6738        T * alloc_align_set( size_t align, T fill );            §\C{// aligned variable, T size, fill with value}§
     6739        T * alloc_align_set( size_t align, size_t dim, char fill ); §\C{// aligned array[dim], T size elements, fill bytes with value}§
     6740        T * alloc_align_set( size_t align, size_t dim, T fill ); §\C{// aligned array[dim], T size elements, fill elements with value}§
     6741        T * alloc_align_set( size_t align, size_t dim, const T fill[] ); §\C{// aligned array[dim], T size elements, fill elements with array}§
     6742        T * alloc_align_set( T ptr[], size_t align, size_t dim, char fill ); §\C{// realloc new aligned array[dim], fill new bytes with value}§
    67436743
    67446744        // §\CFA§ safe initialization/copy, i.e., implicit size specification
Note: See TracChangeset for help on using the changeset viewer.