|
|
@@ -56,7 +56,7 @@ writing. Since the compiler is under continuous development, some of the
|
|
|
things described here may be outdated. In case of doubt, consult the
|
|
|
\file{README} files distributed with the compiler.
|
|
|
The \file{README} files are, in case of conflict with this manual,
|
|
|
- authoritative.
|
|
|
+authoritative.
|
|
|
|
|
|
I hope, my poor english is quite understandable. Feel free to correct
|
|
|
spelling mistakes.
|
|
|
@@ -71,7 +71,7 @@ spelling mistakes.
|
|
|
|
|
|
The ultimate source for information about compiler internals is
|
|
|
the compiler source, though it isn't very well documented. If you
|
|
|
-need more infomration you should join the developers mailing
|
|
|
+need more information you should join the developers mailing
|
|
|
list or you can contact the developers.
|
|
|
|
|
|
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|
|
@@ -79,15 +79,46 @@ list or you can contact the developers.
|
|
|
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|
|
\chapter{Overview}
|
|
|
|
|
|
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|
|
+% History
|
|
|
+\section{History}
|
|
|
+
|
|
|
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|
|
+% The compiler passes
|
|
|
+\section{The compiler passes}
|
|
|
+
|
|
|
+It isn't easy to divide the compilation process of \fpc into passes
|
|
|
+how it is described by many thesis about compiler building,
|
|
|
+but I would say \fpc does the compilation in five passes:
|
|
|
+
|
|
|
+\begin{enumerate}
|
|
|
+\item Scanning and Parsing. The compiler reads the input file,
|
|
|
+does preprocessing (i. e.
|
|
|
+reading include files, expanding macros ...) (\ref{ch:scanner})
|
|
|
+and the parser (\ref{ch:parser}) creates a parse tree (\ref{ch:parse_tree}).
|
|
|
+While this pass the compiler builds also the symbol tables
|
|
|
+(\ref{ch:symbol_tables}).
|
|
|
+\item Semantic analysis. This pass checks if semantic of
|
|
|
+the code is correct, i.e. if the types of expressions matches
|
|
|
+to the operators (\ref{ch:semantical_analysis}). This pass determines
|
|
|
+also how many registers are needed to evalute an expression, this
|
|
|
+information is used by the code generator later.
|
|
|
+\item Code generation
|
|
|
+\item Optimizing of the assembler
|
|
|
+\item Assembler writing
|
|
|
+\end{enumerate}
|
|
|
+
|
|
|
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|
|
% The scanner
|
|
|
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|
|
%% \chapter{The scanner}
|
|
|
+\label{ch:scanner}
|
|
|
|
|
|
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|
|
% The symbol tables
|
|
|
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|
|
\chapter{The symbol tables}
|
|
|
+\label{ch:symbol_tables}
|
|
|
|
|
|
The symbol table is used to store information about all
|
|
|
symbols, declarations and definitions in a program.
|
|
|
@@ -130,37 +161,42 @@ To make it more clear let's have a look at the fields of tdef:
|
|
|
% The parse tree
|
|
|
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|
|
%% \chapter{The parse tree}
|
|
|
+\label{ch:parse_tree}
|
|
|
|
|
|
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|
|
% The parser
|
|
|
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|
|
%% \chapter{The parser}
|
|
|
+\label{ch:parser}
|
|
|
|
|
|
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|
|
% The semantical analysis
|
|
|
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|
|
%% \chapter{The semantical analysis}
|
|
|
+\label{ch:semantical_analysis}
|
|
|
|
|
|
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|
|
% The code generation
|
|
|
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|
|
%% \chapter{The code generation}
|
|
|
+\label{ch:code_generation}
|
|
|
|
|
|
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|
|
% The assembler writers
|
|
|
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|
|
\chapter{The assembler writers}
|
|
|
+\label{ch:assembler_writers}
|
|
|
|
|
|
\fpc doesn't generate machine language, it generates
|
|
|
assembler which must be assembled and linked.
|
|
|
|
|
|
The assembler output is configurable, \fpc can create
|
|
|
-assembler for the GNU AS, the NASM (Netwide assembler) and
|
|
|
+assembler for the \file{GNU AS}, the \file{NASM} (Netwide assembler) and
|
|
|
the assemblers of Borland and Microsoft. The default assembler
|
|
|
-is the GNU AS, because it is fast and and available on
|
|
|
-many platforms. Why don't we use the NASM? It is 2-4 times
|
|
|
-slower than the GNU AS and it is created for
|
|
|
-hand-written assembler, while the GNU AS is designed
|
|
|
+is the \file{GNU AS}, because it is fast and and available on
|
|
|
+many platforms. Why don't we use the \file{NASM}? It is 2-4 times
|
|
|
+slower than the \file{GNU AS} and it is created for
|
|
|
+hand-written assembler, while the \file{GNU AS} is designed
|
|
|
as back end for a compiler.
|
|
|
|
|
|
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|
|
@@ -366,7 +402,7 @@ procedure test;
|
|
|
\section{Future plans}
|
|
|
\label{se:future_plans}
|
|
|
|
|
|
-\Appendix
|
|
|
+\appendix
|
|
|
|
|
|
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|
|
% Coding style guide
|
florian