mirror/nasm
2
0
Fork 0
mirror of https://github.com/netwide-assembler/nasm.git synced 2024-11-21 03:14:19 +08:00
Code Issues Packages Projects Releases Wiki Activity
97fda4ce6e
nasm/configure.ac

239 lines
7.4 KiB
Plaintext
Raw Normal View History

Portability improvements and autoconf modernization Improve the portability of the code and modernize our use of autoconf. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-02-18 16:54:30 +08:00
dnl Process this file with autoconf 2.63 or later to produce
Check in the rest of the Cygwin support patch...
2002-05-19 10:26:38 +08:00
dnl a configure script.
Portability improvements and autoconf modernization Improve the portability of the code and modernize our use of autoconf. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-02-18 16:54:30 +08:00
AC_PREREQ(2.63)
Move config.h to a subdirectory, add MSVC-specific config file Instead of trying to do hacks in the Makefiles, define header files for specific compilers if they can't use autoconf. Currently defined for Microsoft Visual Studio, based on MSDN documentation. It is currently untested. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-10-05 08:01:59 +08:00
AC_INIT(config/config.h.in)
AC_CONFIG_HEADERS(config/config.h)
NASM 0.97
2002-05-01 04:53:16 +08:00
AC_PREFIX_PROGRAM(nasm)
configure.ac: smarter way to handle -O2 -> -O3 default change We don't want to tack on -O3 if the user has specified, say, CFLAGS='-g -O1' for debugging. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2017-03-02 04:57:46 +08:00
dnl Save initial CFLAGS, to see if -g -O2 came from configure or not
pa_init_cflags="$CFLAGS"
autoconf: fix handling of --enable/--disable options, WINE fix AC_ARG_ENABLE() doesn't really work the way you expect: one argument is called on *any* invocation. Create simple helper wrappers to get the effect we really want for boolean options. Define WINELOADER=/dev/null to prevent autoconf from inadvertently running Wine and think we are not cross-compiling even if we are. It is at the very best slow and buys us absolutely nothing. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2017-04-07 06:24:58 +08:00
dnl This prevents us from running Wine and thinking we are not
dnl cross-compiling when in fact we are; running Wine here is at
dnl the best very slow and doesn't buy us a single thing at all.
WINELOADER=/dev/null
export WINELOADER
configure.ac: smarter way to handle -O2 -> -O3 default change We don't want to tack on -O3 if the user has specified, say, CFLAGS='-g -O1' for debugging. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2017-03-02 04:57:46 +08:00
dnl Checks for programs and enable necessary CC extensions
Use autoconf to detect WORDS_LITTLEENDIAN Use autoconf to detect littleendian word order, and use WORDS_LITTLEENDIAN instead of X86_MEMORY when we don't require unaligned memory accesses to be permitted.
2008-06-09 11:53:29 +08:00
AC_USE_SYSTEM_EXTENSIONS
Platform compatibility improvements Windows: indicate that we support multibyte character sets in file names. Unix: support gigantic input and output files. Might be useful especially in the presence of the INCBIN directive. Signed-off-by: H. Peter Anvin <hpa@linux.kernel.org>
2016-02-19 03:57:53 +08:00
AC_SYS_LARGEFILE
NASM 0.97
2002-05-01 04:53:16 +08:00
AC_PROG_CC
Portability improvements and autoconf modernization Improve the portability of the code and modernize our use of autoconf. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-02-18 16:54:30 +08:00
AC_PROG_CC_STDC
NASM 0.97
2002-05-01 04:53:16 +08:00
AC_PROG_LN_S
AC_PROG_MAKE_SET
AC_PROG_INSTALL
autoconf: fix handling of --enable/--disable options, WINE fix AC_ARG_ENABLE() doesn't really work the way you expect: one argument is called on *any* invocation. Create simple helper wrappers to get the effect we really want for boolean options. Define WINELOADER=/dev/null to prevent autoconf from inadvertently running Wine and think we are not cross-compiling even if we are. It is at the very best slow and buys us absolutely nothing. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2017-04-07 06:24:58 +08:00
dnl If the user did not specify a CFLAGS default, change default -O2
dnl to either -O3 (normal) or -O0 (for debugging)
PA_ARG_DISABLED([optimization],
[compile without optimization (-O0) to help debugging],
[pa_optimize=-O0], [pa_optimize=-O3])
AS_IF([test x"$pa_init_cflags" = x],
[CFLAGS=`echo "$CFLAGS" | sed -e "s/-O2/$pa_optimize/"`])
configure.ac: default to -O3 Default to -O3 instead of -O2. The code gets somewhat bigger, but nasm is full of tiny functions called from all over the place. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2017-03-01 11:02:51 +08:00
configure: Add --enable-gdb option When one needs to run nasm under GDB Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2017-07-29 21:05:36 +08:00
dnl Compile and link with dwarf debug
PA_ARG_ENABLED([gdb],
[disable optimization and compile with extra debug information for GDB debugger],
[CFLAGS=`echo "$CFLAGS" | sed -e "s/\(-O2\|-O3\)/-O0/"`
PA_ADD_CLDFLAGS([-ggdb3])])
autoconf improvements: @SET_MAKE@, library extension Minor autoconf corner cases. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-03-10 06:11:40 +08:00
dnl Check for library extension
PA_LIBEXT
configure.in: Move AC_C_INLINE and friends to be checked before PA_ADD_CFLAGS Looks like -Werror=missing-declarations revealed problem in configure: the "inline" support has not been detected properly leading to problem in building procedure. Lets move AC_C_INLINE and etc to be tested before gcc flags. Reported-by: NAKAI Yuta <nak5124@live.jp> Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
2014-12-20 16:35:51 +08:00
dnl Checks for typedefs, structures, and compiler characteristics.
AC_C_CONST
Restore AC_C_INLINE Checkin 7729edf7224e39628bd342c3a3bb44c1753bdfb0 removed AC_C_INLINE, presumably inadvertently. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2015-11-05 03:29:09 +08:00
AC_C_INLINE
configure.in: Move AC_C_INLINE and friends to be checked before PA_ADD_CFLAGS Looks like -Werror=missing-declarations revealed problem in configure: the "inline" support has not been detected properly leading to problem in building procedure. Lets move AC_C_INLINE and etc to be tested before gcc flags. Reported-by: NAKAI Yuta <nak5124@live.jp> Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
2014-12-20 16:35:51 +08:00
AC_C_RESTRICT
AC_TYPE_SIZE_T
configure.in: if byte order is unknown or ill defined, don't define any We can always operate correctly if neither WORDS_BIGENDIAN nor WORDS_LITTLEENDIAN are defined, so if the word order is either indeterminable or universal (the compiler generates both bigendian and litteendian output from the same sources) then define neither. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2016-03-02 13:53:09 +08:00
AC_C_BIGENDIAN(AC_DEFINE(WORDS_BIGENDIAN),AC_DEFINE(WORDS_LITTLEENDIAN),,)
configure.in: Move AC_C_INLINE and friends to be checked before PA_ADD_CFLAGS Looks like -Werror=missing-declarations revealed problem in configure: the "inline" support has not been detected properly leading to problem in building procedure. Lets move AC_C_INLINE and etc to be tested before gcc flags. Reported-by: NAKAI Yuta <nak5124@live.jp> Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
2014-12-20 16:35:51 +08:00
AH_TEMPLATE(WORDS_BIGENDIAN,
[Define to 1 if your processor stores words with the most significant
byte first (like Motorola and SPARC, unlike Intel and VAX).])
AH_TEMPLATE(WORDS_LITTLEENDIAN,
[Define to 1 if your processor stores words with the least significant
byte first (like Intel and VAX, unlike Motorola and SPARC).])
Merge branch 'nasm-2.12.xx' Resolved Conflicts: configure.in output/codeview.c Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2016-03-02 13:51:40 +08:00
dnl Force gcc and gcc-compatible compilers treat signed integers
dnl as 2's complement
configure.in: add -fwrapv to gcc and compatible compilers Add the -fwrapv to gcc and compatible C compilers, to enforce deterministic 2's-complement behavior with regards to signed and unsigned integer types. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2016-02-27 16:15:13 +08:00
PA_ADD_CFLAGS([-fwrapv])
NASM 0.97
2002-05-01 04:53:16 +08:00
Better support for function attributes, autoconf cleanups Use autoconf to detect function attributes; some compilers like Sun CC do support some gcc-style attributes, but don't define __GNUC__. Also -U__STRICT_ANSI__ already in configure.ac so our tests match what we might eventually encounter. Add const_func and pure_func attributes. Decorate functions in nasmlib.h with const_func and pure_func. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-11-16 06:01:37 +08:00
dnl Some environments abuse __STRICT_ANSI__ to disable some
dnl function declarations
PA_ADD_CFLAGS([-U__STRICT_ANSI__])
configure.ac: disable common data to avoid problems with some linkers Disable common data to: a) discover problems with common data as quickly as possible (we should not use common data as some compilers may not handle it); b) work around a problem with the OSX linker causing it to not find zero_buffer even though it is defined in nasmlib/zerobuf.c. Reported-by: anonymous coward <nasm64developer@users.sf.net> Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-08-03 00:28:22 +08:00
dnl Don't put things in common if we can avoid it. We don't want to
dnl assume all compilers support common, and this will help find those
dnl problems. This also works around an OSX linker problem.
PA_ADD_CFLAGS([-fno-common])
doc: for ps to pdf conversion, use "acrodist", "ps2pdf" or "pstopdf" For PostScript to PDF conversion, use whichever of "acrodist", "ps2pdf", or "pstopdf" which we can find on the system. I haven't tried either acrodist or pstopdf myself, only going my the documentation, but prefer acrodist since it is claimed to produce smaller output files than ps2pdf.
2008-06-20 02:39:23 +08:00
dnl Look for programs...
Make the errors for missing man page tools comprehensible A lot of confusion was had due to missing asciidoc... Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2013-05-18 03:02:49 +08:00
AC_CHECK_PROGS(NROFF, nroff, false)
Revert "build: Include pregenerated nasm manpages" This reverts commit 99427bdb6c85c812665f6d7b36ac520a631c5b23. We don't include generated files in the repository, instead we should pregenerate them for the tarball generation, just as we do for the Perl-generated files. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2013-10-24 20:22:19 +08:00
AC_CHECK_PROGS(ASCIIDOC, asciidoc, false)
AC_CHECK_PROGS(XMLTO, xmlto, false)
NASM 0.98p6
2002-05-01 04:57:38 +08:00
Revert "build: Include pregenerated nasm manpages" This reverts commit 99427bdb6c85c812665f6d7b36ac520a631c5b23. We don't include generated files in the repository, instead we should pregenerate them for the tarball generation, just as we do for the Perl-generated files. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2013-10-24 20:22:19 +08:00
dnl Check for progs needed for manpage generation
autoconf: fix handling of --enable/--disable options, WINE fix AC_ARG_ENABLE() doesn't really work the way you expect: one argument is called on *any* invocation. Create simple helper wrappers to get the effect we really want for boolean options. Define WINELOADER=/dev/null to prevent autoconf from inadvertently running Wine and think we are not cross-compiling even if we are. It is at the very best slow and buys us absolutely nothing. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2017-04-07 06:24:58 +08:00
AS_IF([test $ASCIIDOC = false],
[AC_MSG_WARN([No asciidoc package found])]
)
AS_IF([test $XMLTO = false],
[AC_MSG_WARN([No xmlto package found])]
)
Revert "build: Include pregenerated nasm manpages" This reverts commit 99427bdb6c85c812665f6d7b36ac520a631c5b23. We don't include generated files in the repository, instead we should pregenerate them for the tarball generation, just as we do for the Perl-generated files. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2013-10-24 20:22:19 +08:00
Makefiles: use actual archive managers There is no reason to not use an archive manager to build our executables. If there really are systems which don't have any kind of archive manager, we can simply link all the objects. This also drops any use of configure to detect library objects. Instead just use HAVE_* and let the archive manager delete them. A lot of additional functions could be declared library functions and reorganized. ***FIX*** Mkfiles/*.mak have not yet been updated. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2016-03-08 18:06:39 +08:00
dnl Check for host compiler tools
AC_CHECK_TOOL(AR, ar)
AC_CHECK_TOOL(RANLIB, ranlib, :)
AC_CHECK_TOOL(STRIP, strip)
NASM 0.97
2002-05-01 04:53:16 +08:00
dnl Checks for header files.
AC_HEADER_STDC
lib: split library into stdlib and nasmlib; header handling fixes Split lib/ into nasmlib/ (for nasm-specific functions) and stdlib/ (for replacements for C library functions which may be missing.) Rename the ersatz inttypes.h to nasmint.h so we can use a simple test in compiler.h instead of dealing with include path magic. Remove tests in configure.in for ancient missing functions (which will break the build anyway.) Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2016-03-09 04:14:55 +08:00
AC_CHECK_HEADERS(inttypes.h)
Portability fixes Concentrate compiler dependencies to compiler.h; make sure compiler.h is included first in every .c file (since some prototypes may depend on the presence of feature request macros.) Actually use the conditional inclusion of various functions (totally broken in previous releases.)
2007-10-03 12:53:51 +08:00
AC_CHECK_HEADERS(strings.h)
Portability improvements and autoconf modernization Improve the portability of the code and modernize our use of autoconf. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-02-18 16:54:30 +08:00
AC_HEADER_STDBOOL
If we have new features introduced by C11, use them Instead of using hacks or compiler-specific features, if we have standard features as defined in ISO C11, use them. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-12-20 18:29:58 +08:00
AC_CHECK_HEADERS(stdnoreturn.h)
nasmlib: allow writing of sparse files On systems that support it, allow the writing of sparse files. This can be useful for some file formats (like binary, or ELF if the alignments are very large) that can contain large amounts of zeroes. This is not inherently portable code, so condition it on certain known systems. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-02-19 06:24:40 +08:00
AC_CHECK_HEADERS(io.h)
file: improve the file access interface A number of fairly common operations are invoked way too many times, especially when using incbin. Drastically reduce the number of system calls that need to be executed, and use memory mapping to reduce unnecessary double buffering. We could improve this further by leaving files open once used; however, that might run into file count problems on some systems. Still unclear is why we seem to invoke nasm_file_size() twice per pass for incbin. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-09-22 05:03:18 +08:00
AC_CHECK_HEADERS(fcntl.h)
Portability improvements to nasm_realpath() Try harder to nasm_realpath() to be as portable as possible. Move it to a separate file since it has gotten complex enough that it is cleaner that way. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2016-01-27 04:01:34 +08:00
AC_CHECK_HEADERS(unistd.h)
AC_CHECK_HEADERS(sys/param.h)
file: improve the file access interface A number of fairly common operations are invoked way too many times, especially when using incbin. Drastically reduce the number of system calls that need to be executed, and use memory mapping to reduce unnecessary double buffering. We could improve this further by leaving files open once used; however, that might run into file count problems on some systems. Still unclear is why we seem to invoke nasm_file_size() twice per pass for incbin. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-09-22 05:03:18 +08:00
AC_CHECK_HEADERS(sys/mman.h)
compiler.h: include <sys/types.h> Apparently, at least on some OS/2 compilers, <sys/types.h> needs to be included for off_t to be defined. This seems like a generically good idea to include this header whenever it is available. Remove reference to <types.h>. This was supposedly for MSVC, but there is no actual evidence that it is useful beyond <sys/types.h>. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-10-20 04:13:38 +08:00
AC_CHECK_HEADERS(sys/types.h)
MSVC compatibility: clean up the handling of _stati64/_fstati64 On Windows we need to use _stati64/_fstati64 in order to handle large file sizes, but the handling was broken in the canned MSVC++ configuration. Clean it up and fix it. This addresses BR 3392398. Reported-by: Nikolai Saoukh <nms@otdel-1.org> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2017-04-06 12:19:03 +08:00
AC_CHECK_HEADERS(sys/stat.h)
Portability improvements to nasm_realpath() Try harder to nasm_realpath() to be as portable as possible. Move it to a separate file since it has gotten complex enough that it is cleaner that way. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2016-01-27 04:01:34 +08:00
NASM 0.97
2002-05-01 04:53:16 +08:00
dnl Checks for library functions.
Portability fixes Concentrate compiler dependencies to compiler.h; make sure compiler.h is included first in every .c file (since some prototypes may depend on the presence of feature request macros.) Actually use the conditional inclusion of various functions (totally broken in previous releases.)
2007-10-03 12:53:51 +08:00
AC_CHECK_FUNCS(strcasecmp stricmp)
AC_CHECK_FUNCS(strncasecmp strnicmp)
Windows portability fixes Windows needs strsep() and defines a P_WAIT constant that we need to undefine. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-02-19 06:40:37 +08:00
AC_CHECK_FUNCS(strsep)
nasmlib: break up nasmlib.c into logical components nasmlib.c had basically turned into a garbage bin of various functions with very little in common. Break it up into logical components for isolation and manageability. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2016-05-25 20:42:47 +08:00
AC_CHECK_FUNCS(strnlen)
Portability fixes Concentrate compiler dependencies to compiler.h; make sure compiler.h is included first in every .c file (since some prototypes may depend on the presence of feature request macros.) Actually use the conditional inclusion of various functions (totally broken in previous releases.)
2007-10-03 12:53:51 +08:00
Remove obsolete types; add <inttypes.h> where needed; header fixes - Remove obsolete types like "uint32"; use "uint32_t" consistently. - Make sure we include <inttypes.h> where needed. - Header file guards should be FOO_H or SUBDIR_FOO_H; _FOO_H infringes on the C implementation's namespace and should only be used when writing libc! - Change a few "int8_t" back to "char" where appropriate. There are a lot more places where that should be done, though. - Clean up the check for getuid/getgid in rdoff/rdlar.h.
2007-04-13 00:54:50 +08:00
AC_CHECK_FUNCS(getuid)
AC_CHECK_FUNCS(getgid)
Portability improvements to nasm_realpath() Try harder to nasm_realpath() to be as portable as possible. Move it to a separate file since it has gotten complex enough that it is cleaner that way. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2016-01-27 04:01:34 +08:00
AC_CHECK_FUNCS(realpath)
AC_CHECK_FUNCS(canonicalize_file_name)
AC_CHECK_FUNCS(_fullpath)
AC_CHECK_FUNCS(pathconf)
nasmlib: allow writing of sparse files On systems that support it, allow the writing of sparse files. This can be useful for some file formats (like binary, or ELF if the alignments are very large) that can contain large amounts of zeroes. This is not inherently portable code, so condition it on certain known systems. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-02-19 06:24:40 +08:00
AC_FUNC_FSEEKO
configure: smarter way to handle fseeko() searching Better handling of the replacement for fseeko() in its absence; also look for the Windows _fseeki64() function. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2016-02-22 13:20:45 +08:00
AC_CHECK_FUNCS([_fseeki64])
nasmlib: allow writing of sparse files On systems that support it, allow the writing of sparse files. This can be useful for some file formats (like binary, or ELF if the alignments are very large) that can contain large amounts of zeroes. This is not inherently portable code, so condition it on certain known systems. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-02-19 06:24:40 +08:00
AC_CHECK_FUNCS([ftruncate _chsize _chsize_s])
Move config.h to a subdirectory, add MSVC-specific config file Instead of trying to do hacks in the Makefiles, define header files for specific compilers if they can't use autoconf. Currently defined for Microsoft Visual Studio, based on MSDN documentation. It is currently untested. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-10-05 08:01:59 +08:00
AC_CHECK_FUNCS([fileno _fileno])
nasmlib: allow writing of sparse files On systems that support it, allow the writing of sparse files. This can be useful for some file formats (like binary, or ELF if the alignments are very large) that can contain large amounts of zeroes. This is not inherently portable code, so condition it on certain known systems. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-02-19 06:24:40 +08:00
Move config.h to a subdirectory, add MSVC-specific config file Instead of trying to do hacks in the Makefiles, define header files for specific compilers if they can't use autoconf. Currently defined for Microsoft Visual Studio, based on MSDN documentation. It is currently untested. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-10-05 08:01:59 +08:00
AC_CHECK_FUNCS(_filelengthi64)
configure.ac: use AC_FUNC_MMAP instead of AC_CHECK_FUNCS(mmap) There are a couple of extra checks in autoconf for mmap, supposedly. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2017-02-24 12:34:38 +08:00
AC_FUNC_MMAP
file: improve the file access interface A number of fairly common operations are invoked way too many times, especially when using incbin. Drastically reduce the number of system calls that need to be executed, and use memory mapping to reduce unnecessary double buffering. We could improve this further by leaving files open once used; however, that might run into file count problems on some systems. Still unclear is why we seem to invoke nasm_file_size() twice per pass for incbin. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-09-22 05:03:18 +08:00
AC_CHECK_FUNCS(getpagesize)
AC_CHECK_FUNCS(sysconf)
Move config.h to a subdirectory, add MSVC-specific config file Instead of trying to do hacks in the Makefiles, define header files for specific compilers if they can't use autoconf. Currently defined for Microsoft Visual Studio, based on MSDN documentation. It is currently untested. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-10-05 08:01:59 +08:00
AC_CHECK_FUNCS([access _access faccessat])
file: improve the file access interface A number of fairly common operations are invoked way too many times, especially when using incbin. Drastically reduce the number of system calls that need to be executed, and use memory mapping to reduce unnecessary double buffering. We could improve this further by leaving files open once used; however, that might run into file count problems on some systems. Still unclear is why we seem to invoke nasm_file_size() twice per pass for incbin. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-09-22 05:03:18 +08:00
Various fixes to the ilog2 functions Fix several bugs in the ilog2 functions. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-11-16 06:23:54 +08:00
PA_HAVE_FUNC(__builtin_clz, (0U))
PA_HAVE_FUNC(__builtin_clzl, (0UL))
PA_HAVE_FUNC(__builtin_clzll, (0ULL))
Better support for function attributes, autoconf cleanups Use autoconf to detect function attributes; some compilers like Sun CC do support some gcc-style attributes, but don't define __GNUC__. Also -U__STRICT_ANSI__ already in configure.ac so our tests match what we might eventually encounter. Add const_func and pure_func attributes. Decorate functions in nasmlib.h with const_func and pure_func. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-11-16 06:01:37 +08:00
PA_HAVE_FUNC(__builtin_expect, (1,1))
Portability improvements and autoconf modernization Improve the portability of the code and modernize our use of autoconf. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-02-18 16:54:30 +08:00
dnl Functions for which we have replacements available in lib/
Makefiles: use actual archive managers There is no reason to not use an archive manager to build our executables. If there really are systems which don't have any kind of archive manager, we can simply link all the objects. This also drops any use of configure to detect library objects. Instead just use HAVE_* and let the archive manager delete them. A lot of additional functions could be declared library functions and reorganized. ***FIX*** Mkfiles/*.mak have not yet been updated. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2016-03-08 18:06:39 +08:00
AC_CHECK_FUNCS([vsnprintf _vsnprintf])
AC_CHECK_FUNCS([snprintf _snprintf])
AC_CHECK_FUNCS([strlcpy])
Portability improvements and autoconf modernization Improve the portability of the code and modernize our use of autoconf. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-02-18 16:54:30 +08:00
Windows: clean up the handling of stat on Windows [f]stat on Windows is messy: we need to use _stati64 for maximum compatibility, but because there is a bunch of stuff wrapped in macros, autoconf sometimes gets the wrong answers. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2017-04-07 06:30:56 +08:00
dnl These types are POSIX-specific, and Windows does it differently...
AC_CHECK_TYPES([struct _stati64])
AC_CHECK_TYPES([struct stat])
AC_CHECK_FUNCS([stat _stati64])
AC_CHECK_FUNCS([fstat _fstati64])
Better handling of platforms which hide "extended" functionality Some platforms apparently feel -std=c99, which defines __STRICT_ANSI__, should also hide a bunch of function prototypes. This rather sucks. At least try to deal with it. MinGW and DJGPP both have this problem, in particular.
2007-10-27 09:49:29 +08:00
dnl Check for functions that might not be declared in the headers for
dnl various idiotic reasons (mostly because of library authors
dnl abusing the meaning of __STRICT_ANSI__)
AC_CHECK_DECLS(strcasecmp)
AC_CHECK_DECLS(stricmp)
AC_CHECK_DECLS(strncasecmp)
AC_CHECK_DECLS(strnicmp)
AC_CHECK_DECLS(strsep)
Add strlcpy() function Add strlcpy() function and implementation, and use configure to detect if strlcpy() is natively available on the system. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2009-08-11 06:56:52 +08:00
AC_CHECK_DECLS(strlcpy)
nasmlib: break up nasmlib.c into logical components nasmlib.c had basically turned into a garbage bin of various functions with very little in common. Break it up into logical components for isolation and manageability. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2016-05-25 20:42:47 +08:00
AC_CHECK_DECLS(strnlen)
Better handling of platforms which hide "extended" functionality Some platforms apparently feel -std=c99, which defines __STRICT_ANSI__, should also hide a bunch of function prototypes. This rather sucks. At least try to deal with it. MinGW and DJGPP both have this problem, in particular.
2007-10-27 09:49:29 +08:00
file: improve the file access interface A number of fairly common operations are invoked way too many times, especially when using incbin. Drastically reduce the number of system calls that need to be executed, and use memory mapping to reduce unnecessary double buffering. We could improve this further by leaving files open once used; however, that might run into file count problems on some systems. Still unclear is why we seem to invoke nasm_file_size() twice per pass for incbin. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-09-22 05:03:18 +08:00
dnl Check for missing types
AC_TYPE_UINTPTR_T
doc: add the option to generate an uncompressed PDF An uncompressed PDF is about twice as big, but if one is using an external compression program (e.g. .pdf.xz) it compresses far better. Use it for the RPM specfile. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2017-04-11 07:10:28 +08:00
dnl Documentation: should we generate an uncompressed PDF? It is
dnl about twice as big, but it can be externally compressed (e.g. with xz)
dnl and becomes significantly smaller than the original.
PA_ARG_DISABLED([pdf-compression],
[generate an uncompressed documentation PDF],
[PDFOPT='-nocompress'])
AC_SUBST([PDFOPT])
Better support for function attributes, autoconf cleanups Use autoconf to detect function attributes; some compilers like Sun CC do support some gcc-style attributes, but don't define __GNUC__. Also -U__STRICT_ANSI__ already in configure.ac so our tests match what we might eventually encounter. Add const_func and pure_func attributes. Decorate functions in nasmlib.h with const_func and pure_func. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-11-16 06:01:37 +08:00
dnl
dnl Check for supported gcc attributes; some compilers (e.g. Sun CC)
dnl support these, but don't define __GNUC__ as they don't support
dnl some other features of gcc.
dnl
configure: make sure we flag invalid attributes Use -Werror=attributes so we can actually test for non-supported attributes without spewing warnings. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2017-02-24 11:22:27 +08:00
PA_ADD_CFLAGS([-Werror=attributes])
Better support for function attributes, autoconf cleanups Use autoconf to detect function attributes; some compilers like Sun CC do support some gcc-style attributes, but don't define __GNUC__. Also -U__STRICT_ANSI__ already in configure.ac so our tests match what we might eventually encounter. Add const_func and pure_func attributes. Decorate functions in nasmlib.h with const_func and pure_func. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-11-16 06:01:37 +08:00
PA_FUNC_ATTRIBUTE(noreturn)
PA_FUNC_ATTRIBUTE(returns_nonnull)
PA_FUNC_ATTRIBUTE(malloc)
PA_FUNC_ATTRIBUTE(alloc_size, (1))
PA_FUNC_ATTRIBUTE(format, [(printf,1,2)], int, [const char *, ...], ["%d",1])
PA_FUNC_ATTRIBUTE(const)
PA_FUNC_ATTRIBUTE(pure)
If we have new features introduced by C11, use them Instead of using hacks or compiler-specific features, if we have standard features as defined in ISO C11, use them. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-12-20 18:29:58 +08:00
PA_FUNC_ATTRIBUTE_ERROR
NASM 0.97
2002-05-01 04:53:16 +08:00
autoconf: Add --with-ccache option Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
2010-11-11 16:18:44 +08:00
dnl
configure.ac: make possible to compile with -ffunction/data-sections The -ffunction-sections and -fdata-sections, if supported, can enable the linker to do some amount of dead code elimination. Basically a very basic form of LTO. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2017-03-01 11:34:36 +08:00
dnl support function sections
autoconf: Add --with-ccache option Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
2010-11-11 16:18:44 +08:00
dnl
autoconf: fix handling of --enable/--disable options, WINE fix AC_ARG_ENABLE() doesn't really work the way you expect: one argument is called on *any* invocation. Create simple helper wrappers to get the effect we really want for boolean options. Define WINELOADER=/dev/null to prevent autoconf from inadvertently running Wine and think we are not cross-compiling even if we are. It is at the very best slow and buys us absolutely nothing. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2017-04-07 06:24:58 +08:00
PA_ARG_ENABLED([sections],
[compile with function/data section support],
configure.ac: fix --enable-sections The --enable-sections option isn't too useful without -Wl,--gc-sections. It's unclear if gcc will provide that option by default. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2017-04-24 12:42:31 +08:00
[PA_ADD_CLDFLAGS([-ffunction-sections])
PA_ADD_CLDFLAGS([-fdata-sections])
PA_ADD_CLDFLAGS([-Wl,--gc-sections])],
autoconf: fix handling of --enable/--disable options, WINE fix AC_ARG_ENABLE() doesn't really work the way you expect: one argument is called on *any* invocation. Create simple helper wrappers to get the effect we really want for boolean options. Define WINELOADER=/dev/null to prevent autoconf from inadvertently running Wine and think we are not cross-compiling even if we are. It is at the very best slow and buys us absolutely nothing. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2017-04-07 06:24:58 +08:00
[])
Makefile: make it possible to compile with gcc link-time optimization Put in the necessary machinery to compile with gcc link-time optimization. This means compiling and linking with -flto, as well as using gcc-ar and gcc-ranlib instead of the normal tools. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2017-03-01 10:48:01 +08:00
dnl
dnl support LTO
dnl
autoconf: fix handling of --enable/--disable options, WINE fix AC_ARG_ENABLE() doesn't really work the way you expect: one argument is called on *any* invocation. Create simple helper wrappers to get the effect we really want for boolean options. Define WINELOADER=/dev/null to prevent autoconf from inadvertently running Wine and think we are not cross-compiling even if we are. It is at the very best slow and buys us absolutely nothing. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2017-04-07 06:24:58 +08:00
PA_ARG_ENABLED([lto],
[compile with gcc-style link time optimization],
[PA_ADD_CLDFLAGS([-flto])
dnl Note: we use _PROG rather than _TOOL since we are prepending the full
dnl CC name which ought to already contain the host triplet if needed
ccbase=`echo "$CC" | awk '{ print $1; }'`
AC_CHECK_PROGS(CC_AR, [${ccbase}-ar], [$ac_cv_prog_AR])
AR="$CC_AR"
AC_CHECK_PROGS(CC_RANLIB, [${ccbase}-ranlib], [$ac_cv_prog_RANLIB])
RANLIB="$CC_RANLIB"], [])
Makefile: make it possible to compile with gcc link-time optimization Put in the necessary machinery to compile with gcc link-time optimization. This means compiling and linking with -flto, as well as using gcc-ar and gcc-ranlib instead of the normal tools. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2017-03-01 10:48:01 +08:00
Merge branch 'nasm-2.12.xx' Resolved Conflicts: configure.in output/codeview.c Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2016-03-02 13:51:40 +08:00
dnl If we have gcc, add appropriate code cleanliness options
Portability improvements and autoconf modernization Improve the portability of the code and modernize our use of autoconf. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2016-02-18 16:54:30 +08:00
PA_ADD_CFLAGS([-W])
PA_ADD_CFLAGS([-Wall])
configure.in: Add -Werror flags last (BR 2292332) configure.in: Add the -Werror flags as the last thing. Otherwise, with development versions of autoconf, all AC_CHECK_FUNCS calls fails with: conftest.c:49: error: function declaration isn't a prototype conftest.c:49: warning: conflicting types for built-in function 'strcspn' Signed-off-by: Ozkan Sezer <sezero@users.sourceforge.net> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2016-02-13 15:49:49 +08:00
PA_ADD_CFLAGS([-pedantic])
configure: correctly discover -W options for clang clang doesn't error out on unknown -W options unless -Werror=unknown-warning-option is specified first, so do that so the configure script can do its job. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2016-05-17 12:35:48 +08:00
dnl LLVM doesn't error out on invalid -W options unless this option is
dnl specified first. Enable this so this script can actually discover
dnl which -W options are possible for this compiler.
PA_ADD_CFLAGS([-Werror=unknown-warning-option])
configure: enable additional code cleanliness warnings We are supposed to handle compiling on a "C90 plus long long" compiler, so make gcc (our most common development platform compiler) complain when we don't. However, suppress the complaints about the Microsoft definitions of the <inttypes.h> strings. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2016-03-02 14:20:10 +08:00
dnl Suppress format warning on Windows targets due to their <inttypes.h>
PA_ADD_CFLAGS([-Wpedantic-ms-format],[-Wno-pedantic-ms-format])
PA_ADD_CFLAGS([-Wc90-c99-compat])
PA_ADD_CFLAGS([-Wlong-long],[-Wno-long-long])
configure.in: disable -Wwrite-strings for now At the moment, NASM is not clean with -Wwrite-strings due to passing string constants to functions that under some conditions modify their arguments. This is problematic if nothing else for guaranteeing correctness, but will take some work to remedy. In the meantime, disable -Wwrite-strings. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2016-03-07 13:44:14 +08:00
dnl PA_ADD_CFLAGS([-Wwrite-strings])
autoconf: fix handling of --enable/--disable options, WINE fix AC_ARG_ENABLE() doesn't really work the way you expect: one argument is called on *any* invocation. Create simple helper wrappers to get the effect we really want for boolean options. Define WINELOADER=/dev/null to prevent autoconf from inadvertently running Wine and think we are not cross-compiling even if we are. It is at the very best slow and buys us absolutely nothing. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2017-04-07 06:24:58 +08:00
PA_ARG_ENABLED([werror],
[compile with -Werror to error out on any warning],
[PA_ADD_CFLAGS([-Werror])],
[PA_ADD_CFLAGS([-Werror=implicit])
PA_ADD_CFLAGS([-Werror=missing-braces])
PA_ADD_CFLAGS([-Werror=return-type])
PA_ADD_CFLAGS([-Werror=trigraphs])
PA_ADD_CFLAGS([-Werror=pointer-arith])
PA_ADD_CFLAGS([-Werror=strict-prototypes])
PA_ADD_CFLAGS([-Werror=missing-prototypes])
PA_ADD_CFLAGS([-Werror=missing-declarations])
PA_ADD_CFLAGS([-Werror=comment])
PA_ADD_CFLAGS([-Werror=vla])]
)
configure.in: Add -Werror flags last (BR 2292332) configure.in: Add the -Werror flags as the last thing. Otherwise, with development versions of autoconf, all AC_CHECK_FUNCS calls fails with: conftest.c:49: error: function declaration isn't a prototype conftest.c:49: warning: conflicting types for built-in function 'strcspn' Signed-off-by: Ozkan Sezer <sezero@users.sourceforge.net> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2016-02-13 15:49:49 +08:00
configure.ac: make possible to compile with -ffunction/data-sections The -ffunction-sections and -fdata-sections, if supported, can enable the linker to do some amount of dead code elimination. Basically a very basic form of LTO. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2017-03-01 11:34:36 +08:00
dnl
dnl support ccache
dnl
autoconf: fix handling of --enable/--disable options, WINE fix AC_ARG_ENABLE() doesn't really work the way you expect: one argument is called on *any* invocation. Create simple helper wrappers to get the effect we really want for boolean options. Define WINELOADER=/dev/null to prevent autoconf from inadvertently running Wine and think we are not cross-compiling even if we are. It is at the very best slow and buys us absolutely nothing. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2017-04-07 06:24:58 +08:00
PA_ARG_ENABLED([ccache], [compile with ccache], [CC="ccache $CC"], [])
configure.ac: make possible to compile with -ffunction/data-sections The -ffunction-sections and -fdata-sections, if supported, can enable the linker to do some amount of dead code elimination. Basically a very basic form of LTO. Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2017-03-01 11:34:36 +08:00
configure.ac, doc/Makefile.in: fix out-of-directory build Missed some necessary updates. Signed-off-by: Ed Beroset <beroset@ieee.org> Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2017-04-15 05:08:15 +08:00
AC_OUTPUT_COMMANDS([mkdir -p config nasmlib nsis output stdlib x86 asm disasm rdoff macros common])
rdoff: move rdoff build into main Makefile to make it build on Windows NSIS packages rdoff, and we really want to be able to build the full NSIS package on an MSVC platform. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2017-04-07 08:24:03 +08:00
AC_OUTPUT(Makefile doc/Makefile)
Reference in New Issue Copy Permalink