Reorganize the source tree.

* Documentation is now in "docs". * Source code is now in "src". * Misc. fixes here and there so that everything still works. I think I got everything in this commit, but I only tested this on Linux (Make) and Windows (CMake), so I might've messed something up on other platforms...
2015-07-27 16:03:25 -04:00
parent 694d0fdade
commit 47e06c6d4e
45 changed files with 90 additions and 91 deletions
--- a/docs/Changelog
+++ b/docs/Changelog
@ -0,0 +1,520 @@
+Version 0.9.27:
+
+Licensing:
+
+- TinyCC partly relicensed to MIT license
+
+User interface:
+
+- define __STDC_HOSTED__ (Michael Matz, Urs Janssen)
+- added support for CPATH, C_INCLUDE_PATH and LD_LIBRARY_PATH (Andrew Aladjev
+and Urs Janssen)
+- added option -norunsrc to control argv[0] with tcc -run (James Lyon)
+- improve --with-libgcc configure help (grischka)
+- improve error message when memory is full (grischka)
+- improve wording about compiler switches in documentation (Thomas Preud'homme)
+- use GNU triplet prefix for cross compiler names (Thomas Preud'homme)
+- ignore unknown linker optimization and as-needed option (Austin English)
+
+Features:
+
+- added ABI tests with native compiler using libtcc (James Lyon)
+- added CMake build system with support for cross-compilation (James Lyon)
+- improved variable length array support (James Lyon)
+- add the possibility to use noname functions by ordinal (YX Hao)
+- add a install-strip target to install tcc (Thomas Preud'homme)
+- add runtime selection of float ABI on ARM (Thomas Preud'homme)
+- add shared lib support on x86-64 (Michael Matz)
+
+Platforms:
+- support Debian GNU/kfreeBSD 64bit userspace (Thomas Preud'homme)
+- fix GNU/Hurd interpreter path (Thomas Preud'homme)
+- fix configure script for FreeBSD host (Thomas Preud'homme)
+- make tcc -run work reliably on ARM by flushing caches (Thomas Preud'homme)
+- many x86-64 ABI fixes incl. XMM register passing (James Lyon)
+- improve compatibility with mingw's long double (James Lyon)
+- avoid .stabstr section name to be truncated on win32 (Roy)
+- add support for load/store of _Bool value (Thomas Preud'homme)
+- detect instruction with incorrect operands on x86-64 (Thomas Preud'homme)
+- improved relocations on ARM (Thomas Preud'homme)
+- add va_* macro implementation for ARM (Thomas Preud'homme)
+- define __ARM_EABI__, __ARMEL__ and __ARM_PCS_VFP (Thomas Preud'homme)
+- provide a runtime library for ARM (Thomas Preud'homme)
+- vastly improved support for ARM hard float calling convention
+(Thomas Preud'homme, Daniel Glöckner)
+- tcc can uses libtcc1 on ARM (Thomas Preud'homme)
+- use __fixdfdi for all float to integer conversion (grischka)
+- simplify startup code for unix platforms (grischka)
+- improve ELF generated on ARM (Thomas Preud'homme)
+- add support for thumb to ARM relocation (Thomas Preud'homme)
+- fix globbing to match MSVC on Windows (Thomas Preud'homme)
+- deprecate FPA and OABI support for ARM (Thomas Preud'homme)
+- warn about softfloat not being supported on ARM (Thomas Preud'homme)
+
+Bug fixes:
+- many code clean up (Urs Janssen, grischka)
+- fixes of other's patches (grischka, Ramsay Jones, Michael Matz)
+- fix documentation about __TINYC__ (Urs Janssen)
+- improve build of documentation (Urs Janssen)
+- improve build instructions (Jov)
+- switch from texi2html to makeinfo --html to build tcc-doc.html (James Lyon)
+- improve out of tree build (James Lyon)
+- improved passing and returning of struct (James Lyon)
+- fix CMake build on i386 and x86-64 (James Lyon)
+- fix i386 calling convention issue (James Lyon)
+- fix error in Windows build of tests (James Lyon)
+- fix x86-64 long double passing (James Lyon)
+- fix crash with undefined struct (grischka)
+- normalize slashes on win32 to always use backslashes (grischka)
+- use runtime function for float to int conversion on i386 (grischka)
+- improved documentation for include and lib lookup on win32 (grischka)
+- detect redefinition of function (Thomas Preud'homme)
+- detect the use of array of functions (Thomas Preud'homme)
+- detect use of enumerator with wrong enumeration (Thomas Preud'homme)
+- detect redefinition of enumerator or enumeration (Thomas Preud'homme)
+- set the user-defined library search paths first (Vittorio Giovara)
+- detect usage of incomplete types inside struct/union (Amine Najahi)
+- various macro bug fixes (Joseph Poirier)
+- avoid wrong trigger of assert on x86-64 platform (Thomas Preud'homme)
+- fix NaN comparison (Thomas Preud'homme)
+- use libtcc for static linking with runtime library (Thomas Preud'homme)
+- fix negation of 0.0 and -0.0 values (Thomas Preud'homme)
+- fix use of long long as if condition (Thomas Preud'homme)
+- disable bound check if libgcc is used (Thomas Preud'homme)
+- error out when casting to void (grischka)
+- remove circular dependency in Makefile (grischka)
+- stop preventing gcc to do strict aliasing (grischka)
+- fix Windows build of tcc (grischka)
+- build runtime library for arm cross compiler (Thomas Preud'homme)
+- fix installation of arm cross-compiler (Thomas Preud'homme)
+- add basic test for cross-compiler (Thomas Preud'homme)
+- fix failure when generating PE on x86-64 (Archidemon)
+- fix floating point unary minus and plus (Michael Matz)
+- add more tests for signed zero float (Michael Matz)
+- fix precision of double on x86-64 (Vincent Lefevre)
+- fix bound checking of argv with -run switch (Kirill Smelkov)
+- work around a wine cmd bug when building tcc on Windows (Austin English)
+- reenable some bound check tests (grischka)
+- boundtest.c lookup honors VPATH (grischka)
+- diff compared to CC in test[123]b? are now errors (grischka)
+- fix test3 on Windows (grischka)
+- prevent gcc from building (non functional) libtcc.a (grischka)
+- fix warning related to PE file generation on x86-64 (grischka)
+- stop mixing ordinary and implicit rule in Makefile (Iavael)
+- fix integer to double conversion on ARM (Thomas Preud'homme)
+- fix parameter passing of structure < 4 bytes on ARM (Thomas Preud'homme)
+- disable builtin_frame_address test on ARM due to gcc bug (Thomas Preud'homme)
+- fix initialization of struct on ARM (Thomas Preud'homme)
+- fix parameter passing of (unsigned) long long bitfield (Thomas Preud'homme)
+- improve float to integer tests (Thomas Preud'homme)
+- fix relocation of Thumb branch to ARM function (Thomas Preud'homme)
+- fix char wrong compatibility with [un]signed char (Thomas Preud'homme)
+- choose the code to compile based on target in libtcc1 (Thomas Preud'homme)
+- fix various clang warnings (Thomas Preud'homme)
+- don't hardcode tcc in Makefile for tests (Thomas Preud'homme)
+- fix relocation of __bound_init bound checking code (Thomas Preud'homme)
+- accept only one basic type for a given variable (Thomas Preud'homme)
+- fix error when using va_* with tcc using libgcc (Thomas Preud'homme)
+- support GOT32 and PLT32 reloc on the same symbol (Thomas Preud'homme)
+- fix memory leak due to symbol attributes (mingodad)
+- partially fix bound checking of argv and arge (Thomas Preud'homme)
+- fix possible dereference when getting name of symbol (grischka)
+- fix va_list type definition on x86-64 (Daniel Glöckner)
+- reduce number of scan-build false positive (mingodad)
+
+version 0.9.26:
+
+User interface:
+- -MD/-MF (automatically generate dependencies for make)
+- -pthread option (same as -D_REENTRANT -lpthread) (Henry Kroll III)
+- -m32/-m64 to re-exec cross compiler (Henry Kroll III)
+- -Wl, Mimic all GNU -option forms supported by ld (Kirill Smelkov)
+- new LIBTCCAPI tcc_set_options() (grischka)
+
+Platforms:
+- Many improvements for x86-64 target (Shinichiro Hamaji, Michael Matz, grischka)
+- x86-64 assembler (Frederic Feret)
+- Many improvements for ARM target (Daniel Glöckner, Thomas Preud'homme)
+- Support WinCE PE ARM (Timo VJ Lahde)
+- Support ARM hardfloat calling convention (Thomas Preud'homme)
+- Support SELinux (Security-Enhanced Linux) (Henry Kroll III)
+- Support Debian GNU/kFreeBSD kernels (Pierre Chifflier)
+- Support GNU/Hurd kernels (Thomas Preud'homme)
+- Support OSX (tcc -run only) (Milutin Jovanovic)
+- Support multiarch configuration (Thomas Preud'homme)
+- Support out-of-tree build (Akim Demaille)
+
+Features:
+- C99 variable length arrays (Thomas Preud'homme & Joe Soroka)
+- Asm labels for variables and functions (Thomas Preud'homme)
+- STT_GNU_IFUNC (Indirect functions as externals) (Thomas Preud'homme)
+- More tests (tests2) (Milutin Jovanovic)
+
+version 0.9.25:
+
+- first support for x86-64 target (Shinichiro Hamaji)
+- support µClibc
+- split tcc.c into tcc.h libtcc.c tccpp.c tccgen.c tcc.c
+- improved preprocess output with linenumbers and spaces preserved
+- tcc_relocate now copies code into user buffer
+- fix bitfields with non-int types and in unions
+- improve ARM cross-compiling (Daniel Glöckner)
+- link stabstr sections from multiple objects
+- better (still limited) support for multiple TCCStates
+
+version 0.9.24:
+
+- added verbosity levels -v, -vv, -vvv
+- Accept standard input as an inputstream (Hanzac Chen)
+- Support c89 compilers other than gcc (Hanzac Chen)
+- -soname linker option (Marc Andre Tanner)
+- Just warn about unknown directives, ignore quotes in #error/#warning
+- Define __STDC_VERSION__=199901L (477)
+- Switch to newer tccpe.c (includes support for resources)
+- Handle backslashes within #include/#error/#warning
+- Import changesets (part 4) 428,457,460,467: defines for openbsd etc.
+- Use _WIN32 for a windows hosted tcc and define it for the PE target,
+  otherwise define __unix / __linux (Detlef Riekenberg)
+- Import changesets (part 3) 409,410: ARM EABI by Daniel Glöckner
+- Some in-between fixes:
+  TCC -E no longer hangs with macro calls involving newlines.
+  (next_nomacro1 now advances the read-pointer with TOK_LINEFEED)
+  Global cast (int g_i = 1LL;) no longer crashes tcc.
+  (nocode_wanted is initially 1, and only 0 for gen_function)
+  On win32 now tcc.exe finds 'include' & 'lib' even if itself is in 'bin'.
+  (new function w32_tcc_lib_path removes 'bin' if detected)
+  Added quick build batch file for mingw (win32/build-tcc.bat)
+  Last added case label optimization (455) produced wrong code. Reverted.
+
+- Import more changesets from Rob Landley's fork (part 2):
+  487: Handle long long constants in gen_opic() (Rob Landley)
+  484: Handle parentheses within __attribute__((...)) (Rob Landley)
+  480: Remove a goto in decl_initializer_alloc (Rob Landley)
+  475: Fix dereferences in inline assembly output (Joshua Phillips)
+  474: Cast ptrs to ints of different sizes correctly (Joshua Phillips)
+  473: Fix size of structs with empty array member (Joshua Phillips)
+  470: No warning for && and || with mixed pointers/integers (Rob Landley)
+  469: Fix symbol visibility problems in the linker (Vincent Pit)
+  468: Allow && and || involving pointer arguments (Rob Landley)
+  455: Optimize case labels with no code in between (Zdenek Pavlas)
+  450: Implement alloca for x86 (grischka)
+  415: Parse unicode escape sequences (Axel Liljencrantz)
+  407: Add a simple va_copy() in stdarg.h (Hasso Tepper)
+  400: Allow typedef names as symbols (Dave Dodge)
+
+- Import some changesets from Rob Landley's fork (part 1):
+  462: Use LGPL with bcheck.c and il-gen.c
+  458: Fix global compound literals (in unary: case '&':) (Andrew Johnson)
+  456: Use return code from tcc_output_file in main() (Michael Somos)
+  442: Fix indirections with function pointers (***fn)() (grischka)
+  441: Fix LL left shift in libtcc1.c:__shldi3 (grischka)
+  440: Pass structures and function ptrs through ?: (grischka)
+  439: Keep rvalue in bit assignment (bit2 = bit1 = x) (grischka)
+  438: Degrade nonportable pointer assignment to warning (grischka)
+  437: Call 'saveregs()' before jumping with logical and/or/not (grischka)
+  435: Put local static variables into global memory (grischka)
+  432/434: Cast double and ptr to bool (grischka)
+  420: Zero pad x87 tenbyte long doubles (Felix Nawothnig)
+  417: Make 'sizeof' unsigned (Rob Landley)
+  397: Fix save_reg for longlongs (Daniel Glöckner)
+  396: Fix "invalid relocation entry" problem on ubuntu - (Bernhard Fischer)
+
+- ignore AS_NEEDED ld command
+- mark executable sections as executable when running in memory
+- added support for win32 wchar_t (Filip Navara)
+- segment override prefix support (Filip Navara)
+- normalized slashes in paths (Filip Navara)
+- windows style fastcall (Filip Navara)
+- support for empty input register section in asm (Filip Navara)
+- anonymous union/struct support (Filip Navara)
+- fixed parsing of function parameters
+- workaround for function pointers in conditional expressions (Dave Dodge)
+- initial '-E' option support to use the C preprocessor alone
+- discard type qualifiers when comparing function parameters (Dave Dodge)
+- Bug fix: A long long value used as a test expression ignores the
+  upper 32 bits at runtime (Dave Dodge)
+- fixed multiple concatenation of PPNUM tokens (initial patch by Dave Dodge)
+- fixed multiple typedef specifiers handling
+- fixed sign extension in some type conversions (Dave Dodge)
+
+version 0.9.23:
+
+- initial PE executable format for windows version (grischka)
+- '#pragma pack' support (grischka)
+- '#include_next' support (Bernhard Fischer)
+- ignore '-pipe' option
+- added -f[no-]leading-underscore
+- preprocessor function macro parsing fix (grischka)
+
+version 0.9.22:
+
+- simple memory optimisations: kernel compilation is 30% faster
+- linker symbol definitions fixes
+- gcc 3.4 fixes
+- fixed value stack full error
+- 'packed' attribute support for variables and structure fields
+- ignore 'const' and 'volatile' in function prototypes
+- allow '_Bool' in bit fields
+
+version 0.9.21:
+
+- ARM target support (Daniel Glöckner)
+- added '-funsigned-char, '-fsigned-char' and
+  '-Wimplicit-function-declaration'
+- fixed assignment of const struct in struct
+- line comment fix (reported by Bertram Felgenhauer)
+- initial TMS320C67xx target support (TK)
+- win32 configure
+- regparm() attribute
+- many built-in assembler fixes
+- added '.org', '.fill' and '.previous' assembler directives
+- '-fno-common' option
+- '-Ttext' linker option
+- section alignment fixes
+- bit fields fixes
+- do not generate code for unused inline functions
+- '-oformat' linker option. 
+- added 'binary' output format.
+
+version 0.9.20:
+
+- added '-w' option
+- added '.gnu.linkonce' ELF sections support
+- fixed libc linking when running in memory (avoid 'stat' function
+  errors).
+- extended '-run' option to be able to give several arguments to a C
+  script.
+
+version 0.9.19:
+
+- "alacarte" linking (Dave Long)
+- simpler function call
+- more strict type checks
+- added 'const' and 'volatile' support and associated warnings
+- added -Werror, -Wunsupported, -Wwrite-strings, -Wall.
+- added __builtin_types_compatible_p() and __builtin_constant_p()
+- chars support in assembler (Dave Long)
+- .string, .globl, .section, .text, .data and .bss asm directive
+  support (Dave Long)
+- man page generated from tcc-doc.texi
+- fixed macro argument substitution
+- fixed zero argument macro parsing
+- changed license to LGPL
+- added -rdynamic option support
+
+version 0.9.18:
+
+- header fix (time.h)
+- fixed inline asm without operand case
+- fixed 'default:' or 'case x:' with '}' after (incorrect C construct accepted
+  by gcc)
+- added 'A' inline asm constraint.
+
+version 0.9.17:
+
+- PLT generation fix
+- tcc doc fixes (Peter Lund)
+- struct parse fix (signaled by Pedro A. Aranda Gutierrez)
+- better _Bool lvalue support (signaled by Alex Measday)
+- function parameters must be converted to pointers (signaled by Neil Brown)
+- sanitized string and character constant parsing
+- fixed comment parse (signaled by Damian M Gryski)
+- fixed macro function bug (signaled by Philippe Ribet)
+- added configure (initial patch by Mitchell N Charity)
+- added '-run' and '-v' options (initial patch by vlindos)
+- added real date report in __DATE__ and __TIME__ macros
+
+version 0.9.16:
+
+- added assembler language support
+- added GCC inline asm() support
+- fixed multiple variable definitions : uninitialized variables are
+  created as COMMON symbols.
+- optimized macro processing
+- added GCC statement expressions support
+- added GCC local labels support
+- fixed array declaration in old style function parameters
+- support casts in static structure initializations
+- added various __xxx[__] keywords for GCC compatibility
+- ignore __extension__ GCC in an expression or in a type (still not perfect)
+- added '? :' GCC extension support
+
+version 0.9.15:
+
+- compilation fixes for glibc 2.2, gcc 2.95.3 and gcc 3.2.
+- FreeBSD compile fixes. Makefile patches still missing (Carl Drougge).
+- fixed file type guessing if '.' is in the path.
+- fixed tcc_compile_string()
+- add a dummy page in ELF files to fix RX/RW accesses (pageexec at
+  freemail dot hu).
+
+version 0.9.14:
+
+- added #warning. error message if invalid preprocessing directive.
+- added CType structure to ease typing (faster parse).
+- suppressed secondary hash tables (faster parse).
+- rewrote parser by optimizing common cases (faster parse).
+- fixed signed long long comparisons.
+- fixed 'int a(), b();' declaration case.
+- fixed structure init without '{}'.
+- correct alignment support in structures.
+- empty structures support.
+- gcc testsuite now supported.
+- output only warning if implicit integer/pointer conversions.
+- added static bitfield init.
+
+version 0.9.13:
+
+- correct preprocessing token pasting (## operator) in all cases (added
+  preprocessing number token).
+- fixed long long register spill.
+- fixed signed long long '>>'.
+- removed memory leaks.
+- better error handling : processing can continue on link errors. A
+  custom callback can be added to display error messages. Most
+  errors do not call exit() now.
+- ignore -O, -W, -m and -f options
+- added old style function declarations
+- added GCC __alignof__ support.
+- added GCC typeof support.
+- added GCC computed gotos support.
+- added stack backtrace in runtime error message. Improved runtime
+  error position display.
+
+version 0.9.12:
+
+- more fixes for || and && handling.
+- improved '? :' type handling.
+- fixed bound checking generation with structures
+- force '#endif' to be in same file as matching '#if'
+- #include file optimization with '#ifndef #endif' construct detection
+- macro handling optimization
+- added tcc_relocate() and tcc_get_symbol() in libtcc.
+
+version 0.9.11:
+
+- stdarg.h fix for double type (thanks to Philippe Ribet).
+- correct white space characters and added MSDOS newline support.
+- fixed invalid implicit function call type declaration.
+- special macros such as __LINE__ are defined if tested with defined().
+- fixed '!' operator with relocated address.
+- added symbol + offset relocation (fixes some static variable initializers)
+- '-l' option can be specified anywhere. '-c' option yields default
+  output name. added '-r' option for relocatable output.
+- fixed '\nnn' octal parsing.
+- fixed local extern variables declarations.
+
+version 0.9.10:
+
+- fixed lvalue type when saved in local stack.
+- fixed '#include' syntax when using macros.
+- fixed '#line' bug.
+- removed size limit on strings. Unified string constants handling
+  with variable declarations.
+- added correct support for '\xX' in wchar_t strings.
+- added support for bound checking in generated executables
+- fixed -I include order.
+- fixed incorrect function displayed in runtime error.
+
+version 0.9.9:
+
+- fixed preprocessor expression parsing for #if/#elif.
+- relocated debug info (.stab section).
+- relocated bounds info (.bounds section).
+- fixed cast to char of char constants ('\377' is -1 instead of 255)
+- fixed implicit cast for unary plus.
+- strings and '__func__' have now 'char[]' type instead of 'char *'
+  (fixes sizeof() return value).
+- added __start_xxx and __stop_xxx symbols in linker.
+- better DLL creation support (option -shared begins to work).
+- ELF sections and hash tables are resized dynamically.
+- executables and DLLs are stripped by default.
+
+version 0.9.8:
+
+- First version of full ELF linking support (generate objects, static
+  executable, dynamic executable, dynamic libraries). Dynamic library
+  support is not finished (need PIC support in compiler and some
+  patches in symbol exporting).
+- First version of ELF loader for object (.o) and archive (.a) files.
+- Support of simple GNU ld scripts (GROUP and FILE commands)
+- Separated runtime library and bound check code from TCC (smaller
+  compiler core).
+- fixed register reload in float compare.
+- fixed implicit char/short to int casting.
+- allow array type for address of ('&') operator.
+- fixed unused || or && result.
+- added GCC style variadic macro support.
+- optimized bound checking code for array access.
+- tcc includes are now in $(prefix)/lib/tcc/include.
+- more command line options - more consistent handling of multiple
+  input files.
+- added tcc man page (thanks to Cyril Bouthors).
+- uClibc Makefile update
+- converted documentation to texinfo format.
+- added developper's guide in documentation.
+
+version 0.9.7:
+
+- added library API for easy dynamic compilation (see libtcc.h - first
+  draft).
+- fixed long long register spill bug.
+- fixed '? :' register spill bug.
+
+version 0.9.6:
+
+- added floating point constant propagation (fixes negative floating
+  point constants bug).
+
+version 0.9.5:
+
+ - uClibc patches (submitted by Alfonso Martone).
+ - error reporting fix
+ - added CONFIG_TCC_BCHECK to get smaller code if needed.
+
+version 0.9.4:
+
+ - windows port (currently cannot use -g, -b and dll functions).
+ - faster and simpler I/O handling.
+ - '-D' option works in all cases.
+ - preprocessor fixes (#elif and empty macro args)
+ - floating point fixes
+ - first code for CIL generation (does not work yet)
+
+version 0.9.3:
+
+ - better and smaller code generator.
+ - full ISOC99 64 bit 'long long' support.
+ - full 32 bit 'float', 64 bit 'double' and 96 bit 'long double' support.
+ - added '-U' option.
+ - added assembly sections support. 
+ - even faster startup time by mmaping sections instead of mallocing them.
+ - added GNUC __attribute__ keyword support (currently supports
+    'section' and 'aligned' attributes).
+ - added ELF file output (only usable for debugging now)
+ - added debug symbol generation (STAB format).
+ - added integrated runtime error analysis ('-g' option: print clear
+   run time error messages instead of "Segmentation fault").
+ - added first version of tiny memory and bound checker ('-b' option).
+
+version 0.9.2:
+
+ - even faster parsing.
+ - various syntax parsing fixes.
+ - fixed external relocation handling for variables or functions pointers.
+ - better function pointers type handling.
+ - can compile multiple files (-i option).
+ - ANSI C bit fields are supported.
+ - beginning of float/double/long double support.
+ - beginning of long long support.
+
+version 0.9.1:
+
+ - full ISOC99 initializers handling.
+ - compound literals.
+ - structures handle in assignments and as function param or return value.
+ - wide chars and strings.
+ - macro bug fix
+
+version 0.9:
+ - initial version.
--- a/docs/CodingStyle.md
+++ b/docs/CodingStyle.md
@ -0,0 +1,10 @@
+TinyCC Coding Style
+==========================
+Lines should be no more than 80 columns long.
+
+Indentation
+--------------------------------------
+Turn on a "fill tabs with spaces" option in your editor.
+
+Be also careful that some files are indented with 2 spaces (when they
+have large indentation) while most are indented with 4 spaces.
--- a/docs/TODO
+++ b/docs/TODO
@ -0,0 +1,92 @@
+TODO list:
+
+Bugs:
+
+- fix macro substitution with nested definitions (ShangHongzhang)
+- FPU st(0) is left unclean (kwisatz haderach). Incompatible with
+  optimized gcc/msc code
+
+- constructors
+- cast bug (Peter Wang)
+- define incomplete type if defined several times (Peter Wang).
+- test binutils/gcc compile
+- tci patch + argument.
+- see -lxxx bug (Michael Charity).
+- see transparent union pb in /urs/include/sys/socket.h
+- precise behaviour of typeof with arrays ? (__put_user macro)
+  but should suffice for most cases)
+- handle '? x, y : z' in unsized variable initialization (',' is
+  considered incorrectly as separator in preparser)
+- transform functions to function pointers in function parameters
+  (net/ipv4/ip_output.c)
+- fix function pointer type display
+- check lcc test suite -> fix bitfield binary operations
+- check section alignment in C
+- fix invalid cast in comparison 'if (v == (int8_t)v)'
+- finish varargs.h support (gcc 3.2 testsuite issue)
+- fix static functions declared inside block
+- fix multiple unions init
+- sizeof, alignof, typeof can still generate code in some cases.
+- Fix the remaining libtcc memory leaks.
+- make libtcc fully reentrant (except for the compilation stage itself).
+
+Bound checking:
+
+- '-b' bug.
+- fix bound exit on RedHat 7.3
+- setjmp is not supported properly in bound checking.
+- fix bound check code with '&' on local variables (currently done
+  only for local arrays).
+- bound checking and float/long long/struct copy code. bound
+  checking and symbol + offset optimization
+
+Missing features:
+
+- disable-asm and disable-bcheck options
+- __builtin_expect()
+- improve '-E' option.
+- atexit (Nigel Horne)
+- packed attribute
+- C99: add complex types (gcc 3.2 testsuite issue)
+- postfix compound literals (see 20010124-1.c)
+
+Optimizations:
+
+- suppress specific anonymous symbol handling
+- more parse optimizations (=even faster compilation)
+- memory alloc optimizations (=even faster compilation)
+- optimize VT_LOCAL + const
+- better local variables handling (needed for other targets)
+
+Not critical:
+
+- C99: fix multiple compound literals inits in blocks (ISOC99
+  normative example - only relevant when using gotos! -> must add
+  boolean variable to tell if compound literal was already
+  initialized).
+- add PowerPC or ARM code generator and improve codegen for RISC (need
+  to suppress VT_LOCAL and use a base register instead).
+- interactive mode / integrated debugger
+- fix preprocessor symbol redefinition
+- better constant opt (&&, ||, ?:)
+- add portable byte code generator and interpreter for other
+  unsupported architectures.
+- C++: variable declaration in for, minimal 'class' support.
+- win32: __intxx. use resolve for bchecked malloc et al.
+  check exception code (exception filter func).
+- handle void (__attribute__() *ptr)()
+
+Fixed (probably):
+
+- bug with defines:
+    #define spin_lock(lock) do { } while (0)
+    #define wq_spin_lock spin_lock
+    #define TEST() wq_spin_lock(a)
+- typedefs can be structure fields
+- see bugfixes.diff + improvement.diff from Daniel Glockner
+- long long constant evaluation
+- add alloca()
+- gcc '-E' option.
+- #include_next support for /usr/include/limits ?
+- function pointers/lvalues in ? : (linux kernel net/core/dev.c)
+- win32: add __stdcall, check GetModuleHandle for dlls.
--- a/docs/config.texi.in
+++ b/docs/config.texi.in
@ -0,0 +1 @@
+@set VERSION ${TCC_VERSION}
--- a/docs/tcc-doc.texi
+++ b/docs/tcc-doc.texi
--- a/docs/texi2pod.pl
+++ b/docs/texi2pod.pl
@ -0,0 +1,427 @@
+#! /usr/bin/perl -w
+
+#   Copyright (C) 1999, 2000, 2001 Free Software Foundation, Inc.
+
+# This file is part of GNU CC.
+
+# GNU CC is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2, or (at your option)
+# any later version.
+
+# GNU CC is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+
+# You should have received a copy of the GNU General Public License
+# along with GNU CC; see the file COPYING.  If not, write to
+# the Free Software Foundation, 59 Temple Place - Suite 330,
+# Boston MA 02111-1307, USA.
+
+# This does trivial (and I mean _trivial_) conversion of Texinfo
+# markup to Perl POD format.  It's intended to be used to extract
+# something suitable for a manpage from a Texinfo document.
+
+$output = 0;
+$skipping = 0;
+%sects = ();
+$section = "";
+@icstack = ();
+@endwstack = ();
+@skstack = ();
+@instack = ();
+$shift = "";
+%defs = ();
+$fnno = 1;
+$inf = "";
+$ibase = "";
+
+while ($_ = shift) {
+    if (/^-D(.*)$/) {
+	if ($1 ne "") {
+	    $flag = $1;
+	} else {
+	    $flag = shift;
+	}
+	$value = "";
+	($flag, $value) = ($flag =~ /^([^=]+)(?:=(.+))?/);
+	die "no flag specified for -D\n"
+	    unless $flag ne "";
+	die "flags may only contain letters, digits, hyphens, dashes and underscores\n"
+	    unless $flag =~ /^[a-zA-Z0-9_-]+$/;
+	$defs{$flag} = $value;
+    } elsif (/^-/) {
+	usage();
+    } else {
+	$in = $_, next unless defined $in;
+	$out = $_, next unless defined $out;
+	usage();
+    }
+}
+
+if (defined $in) {
+    $inf = gensym();
+    open($inf, "<$in") or die "opening \"$in\": $!\n";
+    $ibase = $1 if $in =~ m|^(.+)/[^/]+$|;
+} else {
+    $inf = \*STDIN;
+}
+
+if (defined $out) {
+    open(STDOUT, ">$out") or die "opening \"$out\": $!\n";
+}
+
+while(defined $inf) {
+while(<$inf>) {
+    # Certain commands are discarded without further processing.
+    /^\@(?:
+	 [a-z]+index		# @*index: useful only in complete manual
+	 |need			# @need: useful only in printed manual
+	 |(?:end\s+)?group	# @group .. @end group: ditto
+	 |page			# @page: ditto
+	 |node			# @node: useful only in .info file
+	 |(?:end\s+)?ifnottex   # @ifnottex .. @end ifnottex: use contents
+	)\b/x and next;
+
+    chomp;
+
+    # Look for filename and title markers.
+    /^\@setfilename\s+([^.]+)/ and $fn = $1, next;
+    /^\@settitle\s+([^.]+)/ and $tl = postprocess($1), next;
+
+    # Identify a man title but keep only the one we are interested in.
+    /^\@c\s+man\s+title\s+([A-Za-z0-9-]+)\s+(.+)/ and do {
+	if (exists $defs{$1}) {
+	    $fn = $1;
+	    $tl = postprocess($2);
+	}
+	next;
+    };
+
+    # Look for blocks surrounded by @c man begin SECTION ... @c man end.
+    # This really oughta be @ifman ... @end ifman and the like, but such
+    # would require rev'ing all other Texinfo translators.
+    /^\@c\s+man\s+begin\s+([A-Z]+)\s+([A-Za-z0-9-]+)/ and do {
+	$output = 1 if exists $defs{$2};
+        $sect = $1;
+	next;
+    };
+    /^\@c\s+man\s+begin\s+([A-Z]+)/ and $sect = $1, $output = 1, next;
+    /^\@c\s+man\s+end/ and do {
+	$sects{$sect} = "" unless exists $sects{$sect};
+	$sects{$sect} .= postprocess($section);
+	$section = "";
+	$output = 0;
+	next;
+    };
+
+    # handle variables
+    /^\@set\s+([a-zA-Z0-9_-]+)\s*(.*)$/ and do {
+	$defs{$1} = $2;
+	next;
+    };
+    /^\@clear\s+([a-zA-Z0-9_-]+)/ and do {
+	delete $defs{$1};
+	next;
+    };
+
+    next unless $output;
+
+    # Discard comments.  (Can't do it above, because then we'd never see
+    # @c man lines.)
+    /^\@c\b/ and next;
+
+    # End-block handler goes up here because it needs to operate even
+    # if we are skipping.
+    /^\@end\s+([a-z]+)/ and do {
+	# Ignore @end foo, where foo is not an operation which may
+	# cause us to skip, if we are presently skipping.
+	my $ended = $1;
+	next if $skipping && $ended !~ /^(?:ifset|ifclear|ignore|menu|iftex)$/;
+
+	die "\@end $ended without \@$ended at line $.\n" unless defined $endw;
+	die "\@$endw ended by \@end $ended at line $.\n" unless $ended eq $endw;
+
+	$endw = pop @endwstack;
+
+	if ($ended =~ /^(?:ifset|ifclear|ignore|menu|iftex)$/) {
+	    $skipping = pop @skstack;
+	    next;
+	} elsif ($ended =~ /^(?:example|smallexample|display)$/) {
+	    $shift = "";
+	    $_ = "";	# need a paragraph break
+	} elsif ($ended =~ /^(?:itemize|enumerate|[fv]?table)$/) {
+	    $_ = "\n=back\n";
+	    $ic = pop @icstack;
+	} else {
+	    die "unknown command \@end $ended at line $.\n";
+	}
+    };
+
+    # We must handle commands which can cause skipping even while we
+    # are skipping, otherwise we will not process nested conditionals
+    # correctly.
+    /^\@ifset\s+([a-zA-Z0-9_-]+)/ and do {
+	push @endwstack, $endw;
+	push @skstack, $skipping;
+	$endw = "ifset";
+	$skipping = 1 unless exists $defs{$1};
+	next;
+    };
+
+    /^\@ifclear\s+([a-zA-Z0-9_-]+)/ and do {
+	push @endwstack, $endw;
+	push @skstack, $skipping;
+	$endw = "ifclear";
+	$skipping = 1 if exists $defs{$1};
+	next;
+    };
+
+    /^\@(ignore|menu|iftex)\b/ and do {
+	push @endwstack, $endw;
+	push @skstack, $skipping;
+	$endw = $1;
+	$skipping = 1;
+	next;
+    };
+
+    next if $skipping;
+
+    # Character entities.  First the ones that can be replaced by raw text
+    # or discarded outright:
+    s/\@copyright\{\}/(c)/g;
+    s/\@dots\{\}/.../g;
+    s/\@enddots\{\}/..../g;
+    s/\@([.!? ])/$1/g;
+    s/\@[:-]//g;
+    s/\@bullet(?:\{\})?/*/g;
+    s/\@TeX\{\}/TeX/g;
+    s/\@pounds\{\}/\#/g;
+    s/\@minus(?:\{\})?/-/g;
+    s/\\,/,/g;
+
+    # Now the ones that have to be replaced by special escapes
+    # (which will be turned back into text by unmunge())
+    s/&/&amp;/g;
+    s/\@\{/&lbrace;/g;
+    s/\@\}/&rbrace;/g;
+    s/\@\@/&at;/g;
+
+    # Inside a verbatim block, handle @var specially.
+    if ($shift ne "") {
+	s/\@var\{([^\}]*)\}/<$1>/g;
+    }
+
+    # POD doesn't interpret E<> inside a verbatim block.
+    if ($shift eq "") {
+	s/</&lt;/g;
+	s/>/&gt;/g;
+    } else {
+	s/</&LT;/g;
+	s/>/&GT;/g;
+    }
+
+    # Single line command handlers.
+
+    /^\@include\s+(.+)$/ and do {
+	push @instack, $inf;
+	$inf = gensym();
+
+	# Try cwd and $ibase.
+	open($inf, "<" . $1)
+	    or open($inf, "<" . $ibase . "/" . $1)
+		or die "cannot open $1 or $ibase/$1: $!\n";
+	next;
+    };
+
+    /^\@(?:section|unnumbered|unnumberedsec|center)\s+(.+)$/
+	and $_ = "\n=head2 $1\n";
+    /^\@subsection\s+(.+)$/
+	and $_ = "\n=head3 $1\n";
+
+    # Block command handlers:
+    /^\@itemize\s+(\@[a-z]+|\*|-)/ and do {
+	push @endwstack, $endw;
+	push @icstack, $ic;
+	$ic = $1;
+	$_ = "\n=over 4\n";
+	$endw = "itemize";
+    };
+
+    /^\@enumerate(?:\s+([a-zA-Z0-9]+))?/ and do {
+	push @endwstack, $endw;
+	push @icstack, $ic;
+	if (defined $1) {
+	    $ic = $1 . ".";
+	} else {
+	    $ic = "1.";
+	}
+	$_ = "\n=over 4\n";
+	$endw = "enumerate";
+    };
+
+    /^\@([fv]?table)\s+(\@[a-z]+)/ and do {
+	push @endwstack, $endw;
+	push @icstack, $ic;
+	$endw = $1;
+	$ic = $2;
+	$ic =~ s/\@(?:samp|strong|key|gcctabopt|option|env)/B/;
+	$ic =~ s/\@(?:code|kbd)/C/;
+	$ic =~ s/\@(?:dfn|var|emph|cite|i)/I/;
+	$ic =~ s/\@(?:file)/F/;
+	$_ = "\n=over 4\n";
+    };
+
+    /^\@((?:small)?example|display)/ and do {
+	push @endwstack, $endw;
+	$endw = $1;
+	$shift = "\t";
+	$_ = "";	# need a paragraph break
+    };
+
+    /^\@itemx?\s*(.+)?$/ and do {
+	if (defined $1) {
+	    # Entity escapes prevent munging by the <> processing below.
+	    $_ = "\n=item $ic\&LT;$1\&GT;\n";
+	} else {
+	    $_ = "\n=item $ic\n";
+	    $ic =~ y/A-Ya-y/B-Zb-z/;
+	    $ic =~ s/(\d+)/$1 + 1/eg;
+	}
+    };
+
+    $section .= $shift.$_."\n";
+}
+# End of current file.
+close($inf);
+$inf = pop @instack;
+}
+
+die "No filename or title\n" unless defined $fn && defined $tl;
+
+$sects{NAME} = "$fn \- $tl\n";
+$sects{FOOTNOTES} .= "=back\n" if exists $sects{FOOTNOTES};
+
+for $sect (qw(NAME SYNOPSIS DESCRIPTION OPTIONS ENVIRONMENT FILES
+	      BUGS NOTES FOOTNOTES SEEALSO AUTHOR COPYRIGHT)) {
+    if(exists $sects{$sect}) {
+	$head = $sect;
+	$head =~ s/SEEALSO/SEE ALSO/;
+	print "=head1 $head\n\n";
+	print scalar unmunge ($sects{$sect});
+	print "\n";
+    }
+}
+
+sub usage
+{
+    die "usage: $0 [-D toggle...] [infile [outfile]]\n";
+}
+
+sub postprocess
+{
+    local $_ = $_[0];
+
+    # @value{foo} is replaced by whatever 'foo' is defined as.
+    while (m/(\@value\{([a-zA-Z0-9_-]+)\})/g) {
+	if (! exists $defs{$2}) {
+	    print STDERR "Option $2 not defined\n";
+	    s/\Q$1\E//;
+	} else {
+	    $value = $defs{$2};
+	    s/\Q$1\E/$value/;
+	}
+    }
+
+    # Formatting commands.
+    # Temporary escape for @r.
+    s/\@r\{([^\}]*)\}/R<$1>/g;
+    s/\@(?:dfn|var|emph|cite|i)\{([^\}]*)\}/I<$1>/g;
+    s/\@(?:code|kbd)\{([^\}]*)\}/C<$1>/g;
+    s/\@(?:gccoptlist|samp|strong|key|option|env|command|b)\{([^\}]*)\}/B<$1>/g;
+    s/\@sc\{([^\}]*)\}/\U$1/g;
+    s/\@file\{([^\}]*)\}/F<$1>/g;
+    s/\@w\{([^\}]*)\}/S<$1>/g;
+    s/\@(?:dmn|math)\{([^\}]*)\}/$1/g;
+
+    # Cross references are thrown away, as are @noindent and @refill.
+    # (@noindent is impossible in .pod, and @refill is unnecessary.)
+    # @* is also impossible in .pod; we discard it and any newline that
+    # follows it.  Similarly, our macro @gol must be discarded.
+
+    s/\(?\@xref\{(?:[^\}]*)\}(?:[^.<]|(?:<[^<>]*>))*\.\)?//g;
+    s/\s+\(\@pxref\{(?:[^\}]*)\}\)//g;
+    s/;\s+\@pxref\{(?:[^\}]*)\}//g;
+    s/\@noindent\s*//g;
+    s/\@refill//g;
+    s/\@gol//g;
+    s/\@\*\s*\n?//g;
+
+    # @uref can take one, two, or three arguments, with different
+    # semantics each time.  @url and @email are just like @uref with
+    # one argument, for our purposes.
+    s/\@(?:uref|url|email)\{([^\},]*)\}/&lt;B<$1>&gt;/g;
+    s/\@uref\{([^\},]*),([^\},]*)\}/$2 (C<$1>)/g;
+    s/\@uref\{([^\},]*),([^\},]*),([^\},]*)\}/$3/g;
+
+    # Turn B<blah I<blah> blah> into B<blah> I<blah> B<blah> to
+    # match Texinfo semantics of @emph inside @samp.  Also handle @r
+    # inside bold.
+    s/&LT;/</g;
+    s/&GT;/>/g;
+    1 while s/B<((?:[^<>]|I<[^<>]*>)*)R<([^>]*)>/B<$1>${2}B</g;
+    1 while (s/B<([^<>]*)I<([^>]+)>/B<$1>I<$2>B</g);
+    1 while (s/I<([^<>]*)B<([^>]+)>/I<$1>B<$2>I</g);
+    s/[BI]<>//g;
+    s/([BI])<(\s+)([^>]+)>/$2$1<$3>/g;
+    s/([BI])<([^>]+?)(\s+)>/$1<$2>$3/g;
+
+    # Extract footnotes.  This has to be done after all other
+    # processing because otherwise the regexp will choke on formatting
+    # inside @footnote.
+    while (/\@footnote/g) {
+	s/\@footnote\{([^\}]+)\}/[$fnno]/;
+	add_footnote($1, $fnno);
+	$fnno++;
+    }
+
+    return $_;
+}
+
+sub unmunge
+{
+    # Replace escaped symbols with their equivalents.
+    local $_ = $_[0];
+
+    s/&lt;/E<lt>/g;
+    s/&gt;/E<gt>/g;
+    s/&lbrace;/\{/g;
+    s/&rbrace;/\}/g;
+    s/&at;/\@/g;
+    s/&amp;/&/g;
+    return $_;
+}
+
+sub add_footnote
+{
+    unless (exists $sects{FOOTNOTES}) {
+	$sects{FOOTNOTES} = "\n=over 4\n\n";
+    }
+
+    $sects{FOOTNOTES} .= "=item $fnno.\n\n"; $fnno++;
+    $sects{FOOTNOTES} .= $_[0];
+    $sects{FOOTNOTES} .= "\n\n";
+}
+
+# stolen from Symbol.pm
+{
+    my $genseq = 0;
+    sub gensym
+    {
+	my $name = "GEN" . $genseq++;
+	my $ref = \*{$name};
+	delete $::{$name};
+	return $ref;
+    }
+}