8 1.2. Platform-specific notes
9 1.2.1. Darwin (Mac OS X)
15 1.3. Adding support for new platforms
17 3. xzgrep and other scripts
21 4.1. "No C99 compiler was found."
22 4.1. "No POSIX conforming shell (sh) was found."
23 4.2. configure works but build fails at crc32_x86.S
29 If you aren't familiar with building packages that use GNU Autotools,
30 see the file INSTALL.generic for generic instructions before reading
33 If you are going to build a package for distribution, see also the
34 file PACKAGERS. It contains information that should help making the
35 binary packages as good as possible, but the information isn't very
36 interesting to those making local builds for private use or for use
37 in special situations like embedded systems.
40 1. Supported platforms
41 ----------------------
43 XZ Utils are developed on GNU/Linux, but they should work on many
44 POSIX-like operating systems like *BSDs and Solaris, and even on
45 a few non-POSIX operating systems.
50 A C99 compiler is required to compile XZ Utils. If you use GCC, you
51 need at least version 3.x.x. GCC version 2.xx.x doesn't support some
52 C99 features used in XZ Utils source code, thus GCC 2 won't compile
55 XZ Utils takes advantage of some GNU C extensions when building
56 with GCC. Because these extensions are used only when building
57 with GCC, it should be possible to use any C99 compiler.
60 1.2. Platform-specific notes
62 1.2.1. Darwin (Mac OS X)
64 You may need --disable-assembler if building universal binaries on
65 Darwin. This is because different files are built when assembler is
66 enabled, and there's no way to make it work with universal build.
67 If you want to keep the assembler code, consider building one
68 architecture at a time, and then combining the results to create
69 universal binaries (see lipo(1)).
74 If you try to use the native C compiler on Tru64 (passing CC=cc to
75 configure), it is possible that the configure script will complain
76 that no C99 compiler was found even when the native compiler supports
77 C99. You can safely override the test for C99 compiler by passing
78 ac_cv_prog_cc_c99= as the argument to the configure script.
83 Building XZ Utils on Windows is supported under MinGW and Cygwin.
84 If the Autotools based build gives you trouble with MinGW, you may
85 want try the alternative method found from the "windows" directory.
87 MSVC doesn't support C99, thus it is not possible to use MSVC to
88 compile XZ Utils. However, it is possible to use liblzma.dll from
89 MSVC once liblzma.dll has been built with MinGW. The required
90 import library for MSVC can be created from liblzma.def using the
91 "lib" command shipped in MSVC:
93 lib /def:liblzma.def /out:liblzma.lib /machine:ix86
95 On x86-64, the /machine argument has to naturally be changed:
97 lib /def:liblzma.def /out:liblzma.lib /machine:x64
102 There is an experimental Makefile in the "dos" directory to build
103 XZ Utils on DOS using DJGPP. Support for long file names (LFN) is
106 GNU Autotools based build hasn't been tried on DOS.
111 You will need to pass --disable-assembler to configure when building
114 To omit large number of harmless warnings about visibility support,
115 pass gl_cv_cc_visibility=no as an argument to the configure script.
116 This isn't mandatory since it should have no effect on the resulting
122 XZ Utils can be built for OpenVMS, but the build system files are
123 currently not included in the XZ Utils source package. The required
124 OpenVMS-specific files are maintained by Jouk Jansen and can be
127 http://nchrem.tnw.tudelft.nl/openvms/software2.html#xzutils
130 1.3. Adding support for new platforms
132 If you have written patches to make XZ Utils to work on previously
133 unsupported platform, please send the patches to me! I will consider
134 including them to the official version. It's nice to minimize the
135 need of third-party patching.
137 One exception: Don't request or send patches to change the whole
138 source package to C89. I find C99 substantially nicer to write and
139 maintain. However, the public library headers must be in C89 to
140 avoid frustrating those who maintain programs, which are strictly
147 In most cases, the defaults are what you want. Most of the options
148 below are useful only when building a size-optimized version of
149 liblzma or command line tools.
151 --enable-encoders=LIST
153 Specify a comma-separated LIST of filter encoders to
154 build. See "./configure --help" for exact list of
155 available filter encoders. The default is to build all
158 If LIST is empty or --disable-encoders is used, no filter
159 encoders will be built and also the code shared between
160 encoders will be omitted.
162 Disabling encoders will remove some symbols from the
163 liblzma ABI, so this option should be used only when it
164 is known to not cause problems.
166 --enable-decoders=LIST
168 This is like --enable-encoders but for decoders. The
169 default is to build all supported decoders.
171 --enable-match-finders=LIST
172 liblzma includes two categories of match finders:
173 hash chains and binary trees. Hash chains (hc3 and hc4)
174 are quite fast but they don't provide the best compression
175 ratio. Binary trees (bt2, bt3 and bt4) give excellent
176 compression ratio, but they are slower and need more
177 memory than hash chains.
179 You need to enable at least one match finder to build the
180 LZMA1 or LZMA2 filter encoders. Usually hash chains are
181 used only in the fast mode, while binary trees are used to
182 when the best compression ratio is wanted.
184 The default is to build all the match finders if LZMA1
185 or LZMA2 filter encoders are being built.
188 liblzma support multiple integrity checks. CRC32 is
189 mandatory, and cannot be omitted. See "./configure --help"
190 for exact list of available integrity check types.
192 liblzma and the command line tools can decompress files
193 which use unsupported integrity check type, but naturally
194 the file integrity cannot be verified in that case.
196 Disabling integrity checks may remove some symbols from
197 the liblzma ABI, so this option should be used only when
198 it is known to not cause problems.
201 liblzma includes some assembler optimizations. Currently
202 there is only assembler code for CRC32 and CRC64 for
205 All the assembler code in liblzma is position-independent
206 code, which is suitable for use in shared libraries and
207 position-independent executables. So far only i386
208 instructions are used, but the code is optimized for i686
209 class CPUs. If you are compiling liblzma exclusively for
210 pre-i686 systems, you may want to disable the assembler
213 --enable-unaligned-access
214 Allow liblzma to use unaligned memory access for 16-bit
215 and 32-bit loads and stores. This should be enabled only
216 when the hardware supports this, i.e. when unaligned
217 access is fast. Some operating system kernels emulate
218 unaligned access, which is extremely slow. This option
219 shouldn't be used on systems that rely on such emulation.
221 Unaligned access is enabled by default on x86, x86-64,
222 and big endian PowerPC.
225 Reduce the size of liblzma by selecting smaller but
226 semantically equivalent version of some functions, and
227 omit precomputed lookup tables. This option tends to
228 make liblzma slightly slower.
230 Note that while omitting the precomputed tables makes
231 liblzma smaller on disk, the tables are still needed at
232 run time, and need to be computed at startup. This also
233 means that the RAM holding the tables won't be shared
234 between applications linked against shared liblzma.
236 This option doesn't modify CFLAGS to tell the compiler
237 to optimize for size. You need to add -Os or equivalent
238 flag(s) to CFLAGS manually.
240 --enable-assume-ram=SIZE
241 On the most common operating systems, XZ Utils is able to
242 detect the amount of physical memory on the system. This
243 information is used to set the default memory usage limit.
245 On some systems, there is no code to detect the amount of
246 RAM though. Using --enable-assume-ram one can set how much
247 memory to assume on these systems. SIZE is given as MiB.
248 The default is 32 MiB, which is probably too low for most
249 systems, but it is enough to allow decompressing .xz files
250 created with the default settings.
252 Feel free to send patches to add support for detecting
253 the amount of RAM on the operating system you use. See
254 src/common/tuklib_physmem.c for details.
257 Disable threading support. This makes some things
258 thread-unsafe, meaning that if multithreaded application
259 calls liblzma functions from more than one thread,
260 something bad may happen.
262 Use this option if threading support causes you trouble,
263 or if you know that you will use liblzma only from
264 single-threaded applications and want to avoid dependency
267 --enable-dynamic=TYPE
268 Specify how command line tools should be linked against
269 liblzma. Possible TYPES:
271 yes All command line tools are linked against
272 shared liblzma (if shared liblzma was built).
273 This is equivalent to --enable-dynamic (i.e.
276 mixed Some tools are linked against static liblzma
277 and some against shared liblzma. This is the
278 default and recommended way.
280 no All command line tools are linked against
281 static liblzma (if static liblzma was built).
282 This is equivalent to --disable-dynamic.
284 This option is mostly useful for packagers, if distro
285 policy requires linking against shared libaries. See the
286 file PACKAGERS for more information about pros and cons
290 This enables the assert() macro and possibly some other
291 run-time consistency checks. It makes the code slower, so
292 you normally don't want to have this enabled.
295 If building with GCC, make all compiler warnings an error,
296 that abort the compilation. This may help catching bugs,
297 and should work on most systems. This has no effect on the
301 3. xzgrep and other scripts
302 ---------------------------
306 POSIX shell (sh) and bunch of other standard POSIX tools are required
307 to run the scripts. The configure script tries to find a POSIX
308 compliant sh, but if it fails, you can force the shell by passing
309 gl_cv_posix_shell=/path/to/posix-sh as an argument to the configure
312 Some of the scripts require also mktemp. The original mktemp can be
313 found from <http://www.mktemp.org/>. On GNU, most will use the mktemp
314 program from GNU coreutils instead of the original implementation.
315 Both mktemp versions are fine for XZ Utils (and practically for
316 everything else too).
321 The scripts assume that the required tools (standard POSIX utilities,
322 mktemp, and xz) are in PATH; the scripts don't set the PATH themselves.
323 Some people like this while some think this is a bug. Those in the
324 latter group can easily patch the scripts before running the configure
325 script by taking advantage of a placeholder line in the scripts.
327 For example, to make the scripts prefix /usr/bin:/bin to PATH:
329 perl -pi -e 's|^#SET_PATH.*$|PATH=/usr/bin:/bin:\$PATH|' \
336 4.1. "No C99 compiler was found."
338 You need a C99 compiler to build XZ Utils. If the configure script
339 cannot find a C99 compiler and you think you have such a compiler
340 installed, set the compiler command by passing CC=/path/to/c99 as
341 an argument to the configure script.
343 If you get this error even when you think your compiler supports C99,
344 you can override the test by passing ac_cv_prog_cc_c99= as an argument
345 to the configure script. The test for C99 compiler is not perfect (and
346 it is not as easy to make it perfect as it sounds), so sometimes this
347 may be needed. You will get a compile error if your compiler doesn't
351 4.1. "No POSIX conforming shell (sh) was found."
353 xzgrep and other scripts need a shell that (roughly) conforms
354 to POSIX. The configure script tries to find such a shell. If
355 it fails, you can force the shell to be used by passing
356 gl_cv_posix_shell=/path/to/posix-sh as an argument to the configure
360 4.2. configure works but build fails at crc32_x86.S
362 The easy fix is to pass --disable-assembler to the configure script.
364 The configure script determines if assembler code can be used by
365 looking at the configure triplet; there is currently no check if
366 the assembler code can actually actually be built. The x86 assembler
367 code should work on x86 GNU/Linux, *BSDs, Solaris, Darwin, MinGW,
368 Cygwin, and DJGPP. On other x86 systems, there may be problems and
369 the assembler code may need to be disabled with the configure option.
371 If you get this error when building for x86-64, you have specified or
372 the configure script has misguessed your architecture. Pass the
373 correct configure triplet using the --build=CPU-COMPANY-SYSTEM option
374 (see INSTALL.generic).