Building The Library¶
This document describes how to build Botan on Unix/POSIX and Windows systems. The POSIX oriented descriptions should apply to most common Unix systems (including OS X), along with POSIX-ish systems like BeOS, QNX, and Plan 9. Currently, systems other than Windows and POSIX (such as VMS, MacOS 9, OS/390, OS/400, …) are not supported by the build system, primarily due to lack of access. Please contact the maintainer if you would like to build Botan on such a system.
Botan’s build is controlled by configure.py, which is a Python script. Python 2.6 or later is required.
For the impatient, this works for most systems:
$ ./configure.py [--prefix=/some/directory] $ make $ make install
nmake, if you’re compiling on Windows with Visual C++. On
platforms that do not understand the ‘#!’ convention for beginning
script files, or that have Python installed in an unusual spot, you
might need to prefix the
configure.py command with
$ python ./configure.py [arguments]
Configuring the Build¶
The first step is to run
configure.py, which is a Python script
that creates various directories, config files, and a Makefile for
building everything. This script should run under a vanilla install of
Python 2.6, 2.7, or 3.x.
The script will attempt to guess what kind of system you are trying to
compile for (and will print messages telling you what it guessed).
You can override this process by passing the options
You can pass basically anything reasonable with
--cpu: the script
knows about a large number of different architectures, their
sub-models, and common aliases for them. You should only select the
64-bit version of a CPU (such as “sparc64” or “mips64”) if your
operating system knows how to handle 64-bit object code - a 32-bit
kernel on a 64-bit CPU will generally not like 64-bit code.
By default the script tries to figure out what will work on your system, and use that. It will print a display at the end showing which algorithms have and have not been enabled. For instance on one system we might see lines like:
INFO: Skipping (dependency failure): certstor_sqlite3 sessions_sqlite3 INFO: Skipping (incompatible CPU): aes_power8 INFO: Skipping (incompatible OS): darwin_secrandom getentropy win32_stats INFO: Skipping (incompatible compiler): aes_armv8 pmull sha1_armv8 sha2_32_armv8 INFO: Skipping (no enabled compression schemes): compression INFO: Skipping (requires external dependency): boost bzip2 lzma openssl sqlite3 tpm zlib
The ones that are skipped because they are require an external
dependency have to be explicitly asked for, because they rely on third
party libraries which your system might not have or that you might not
want the resulting binary to depend on. For instance to enable zlib
--with-zlib to your invocation of
All available modules can be listed with
You can control which algorithms and modules are built using the
Modules not listed on the command line will simply be loaded if needed
or if configured to load by default. If you use
only the most core modules will be included; you can then explicitly
enable things that you want to use with
--enable-modules. This is
useful for creating a minimal build targeting to a specific
application, especially in conjunction with the amalgamation option;
see The Amalgamation Build and Minimized Builds.
$ ./configure.py --minimized-build --enable-modules=rsa,eme_oaep,emsa_pssr
will set up a build that only includes RSA, OAEP, PSS along with any required dependencies. Note that a minimized build does not by default include any random number generator, which is needed for example to generate keys, nonces and IVs. See Random Number Generators on which random number generators are available.
Cross compiling refers to building software on one type of host (say Linux x86-64) but creating a binary for some other type (say MinGW x86-32). This is completely supported by the build system. To extend the example, we must tell configure.py to use the MinGW tools:
$ ./configure.py --os=mingw --cpu=x86_32 --cc-bin=i686-w64-mingw32-g++ --ar-command=i686-w64-mingw32-ar ... $ make ... $ file botan.exe botan.exe: PE32 executable (console) Intel 80386, for MS Windows
For whatever reason, some distributions of MinGW lack support for
threading or mutexes in the C++ standard library. You can work around
this by disabling thread support using
You can also specify the alternate tools by setting the CXX and AR environment variables (instead of the –cc-bin and –ar-command options), as is commonly done with autoconf builds.
The basic build procedure on Unix and Unix-like systems is:
$ ./configure.py [--enable-modules=<list>] [--cc=CC] $ make $ make check
If the tests look OK, install:
$ make install
On Unix systems the script will default to using GCC; use
you want something else. For instance use
--cc=icc for Intel C++
--cc=clang for Clang.
make install target has a default directory in which it will
install Botan (typically
/usr/local). You can override this by
--prefix argument to
configure.py, like so:
$ ./configure.py --prefix=/opt <other arguments>
On some systems shared libraries might not be immediately visible to
the runtime linker. For example, on Linux you may have to edit
/etc/ld.so.conf and run
ldconfig (as root) in order for new
shared libraries to be picked up by the linker. An alternative is to
LD_LIBRARY_PATH shell variable to include the directory
that the Botan libraries were installed into.
A build on macOS works much like that on any other Unix-like system.
To build a universal binary for macOS, you need to set some additional build flags. Do this with the configure.py flag –cc-abi-flags:
--cc-abi-flags="-force_cpusubtype_ALL -mmacosx-version-min=10.4 -arch i386 -arch ppc"
The earliest versions of Windows supported are Windows 7 and Windows 2008 R2
You need to have a copy of Python installed, and have both Python and your chosen compiler in your path. Open a command shell (or the SDK shell), and run:
$ python configure.py --cc=msvc --os=windows $ nmake $ nmake check $ nmake install
Botan supports the nmake replacement Jom which enables you to run multiple build jobs in parallel.
For MinGW, use:
$ python configure.py --cc=gcc --os=mingw $ make
By default the install target will be
C:\botan; you can modify
this with the
When building your applications, all you have to do is tell the
compiler to look for both include files and library files in
C:\botan, and it will find both. Or you can move them to a
place where they will be in the default compiler search paths (consult
your documentation and/or local expert for details).
For iOS using XCode¶
For iOS, you typically build for 3 architectures: armv7 (32 bit, older iOS devices), armv8-a (64 bit, recent iOS devices) and x86_64 for the iPhone simulator. You can build for these 3 architectures and then create a universal binary containing code for all of these architectures, so you can link to Botan for the simulator as well as for an iOS device.
To cross compile for armv7, configure and make with:
$ ./configure.py --os=ios --prefix="iphone-32" --cpu=armv7 --cc=clang \ --cc-abi-flags="-arch armv7" $ xcrun --sdk iphoneos make install
To cross compile for armv8-a, configure and make with:
$ ./configure.py --os=ios --prefix="iphone-64" --cpu=armv8-a --cc=clang \ --cc-abi-flags="-arch arm64" $ xcrun --sdk iphoneos make install
To compile for the iPhone Simulator, configure and make with:
$ ./configure.py --os=ios --prefix="iphone-simulator" --cpu=x86_64 --cc=clang \ --cc-abi-flags="-arch x86_64" $ xcrun --sdk iphonesimulator make install
Now create the universal binary and confirm the library is compiled for all three architectures:
$ xcrun --sdk iphoneos lipo -create -output libbotan-2.a \ iphone-32/lib/libbotan-2.a \ iphone-64/lib/libbotan-2.a \ iphone-simulator/lib/libbotan-2.a $ xcrun --sdk iphoneos lipo -info libbotan-2.a Architectures in the fat file: libbotan-2.a are: armv7 x86_64 armv64
The resulting static library can be linked to your app in Xcode.
Modern versions of Android NDK use Clang and support C++11. Simply configure using the appropriate NDK compiler:
$ export CXX=/opt/android-ndk/toolchains/llvm/prebuilt/linux-x86_64/bin/aarch64-linux-android28-clang++ $ ./configure.py --os=android --cc=clang --cpu=arm64
To build android version, there is the possibility to use the docker way:
sudo ANDROID_SDK_VER=21 ANDROID_ARCH=arm64 src/scripts/docker-android.sh
This will produce the docker-builds/android folder containing each architecture compiled.
To build for WebAssembly using Emscripten, try:
CXX=em++ ./configure.py --cc=clang --cpu=llvm --os=emscripten make
This will produce bitcode files
along with a static archive
libbotan-2.a which can linked with
other modules. To convert the tests into a WASM file which can be
executed on a browser, use:
em++ -s ALLOW_MEMORY_GROWTH=1 -s DISABLE_EXCEPTION_CATCHING=0 -s WASM=1 \ --preload-file src/tests/data botan-test.bc -o botan-test.html
Supporting Older Distros¶
Some “stable” distributions, notably RHEL/CentOS, ship very obsolete versions of binutils, which do not support more recent CPU instructions. As a result when building you may receive errors like:
Error: no such instruction: `sha256rnds2 %xmm0,%xmm4,%xmm3'
Depending on how old your binutils is, you may need to disable BMI2,
AVX2, SHA-NI, and/or RDSEED. These can be disabled by passing the
Botan usually links in several different system libraries (such as
libz), depending on which modules are configured at
compile time. In many environments, particularly ones using static
libraries, an application has to link against the same libraries as
Botan for the linking step to succeed. But how does it figure out what
libraries it is linked against?
The answer is to ask the
botan command line tool using
botan version: Print the Botan version number.
botan config prefix: If no argument, print the prefix where Botan is
installed (such as
botan config cflags: Print options that should be passed to the
compiler whenever a C++ file is compiled. Typically this is used for
setting include paths.
botan config libs: Print options for which libraries to link to
(this will include a reference to the botan library itself).
Makefile can run
botan config and get the options
necessary for getting your application to compile and link, regardless
of whatever crazy libraries Botan might be linked against.
No special help exists for building applications on Windows. However, given that typically Windows software is distributed as binaries, this is less of a problem - only the developer needs to worry about it. As long as they can remember where they installed Botan, they just have to set the appropriate flags in their Makefile/project file.
Many developers wish to configure a minimized build which contains only the
specific features their application will use. In general this is straighforward:
--enable-modules= to enable the specific modules
you wish to use. Any such configurations should build and pass the tests; if you
encounter a case where it doesn’t please file an issue.
The only trick is knowing which features you want to enable. The most common
difficulty comes with entropy sources. By default, none are enabled, which means
if you attempt to use
AutoSeeded_RNG, it will fail. The easiest resolution
is to also enable
system_rng which can act as either an entropy source or
used directly as the RNG.
If you are building for x86, ARM, or POWER, it can be beneficial to enable
hardware support for the relevant instruction sets with modules such as
clmul for x86, or
sha2_32_armv8 on ARMv8. SIMD optimizations such as
chacha_avx2 also can
provide substantial performance improvements.
In a future release, hardware specific modules will be enabled by default if the underlying “base” module is enabled.
If you are building a TLS application, you may (or may not) want to include
tls_cbc which enables support for CBC ciphersuites. If
disabled, then it will not be possible to negotiate TLS v1.0/v1.1. In general
this should be considered a feature; only enable this if you need backward
compatability with obsolete clients or servers.
For TLS another useful feature which is not enabled by default is the
ChaCha20Poly1305 ciphersuites. To enable these, add
Configure Script Options¶
Set the target CPU architecture. If not used, the arch of the current system is detected (using Python’s platform module) and used.
Set the target operating system.
Set the desired build compiler
Set the minimal version of the target compiler. Use –cc-min-version=0.0 to support all compiler versions. Default is auto detection.
Set path to compiler binary
If not provided, the value of the
CXX environment variable is used if set.
Set ABI flags, which for the purposes of this option mean options which should be passed to both the compiler and linker.
Override all compiler flags. This is equivalent to setting
in the environment.
Set extra compiler flags, which are appended to the default set. This is useful if you want to set just one or two additional options but leave the normal logic for selecting flags alone.
Set flags to pass to the linker. This is equivalent to setting
Set the path to the tool to use to create static archives (
This is normally only used for cross-compilation.
If not provided, the value of the
AR environment variable is used if set.
Specify the options to pass to
If not provided, the value of the
AR_OPTIONS environment variable is used if set.
Specify the MSVC runtime to use (MT, MD, MTd, or MDd). If not specified, picks either MD or MDd depending on if debug mode is set.
The parameter should be either “little” or “big”. If not used then if the target architecture has a default, that is used. Otherwise left unspecified, which causes less optimal codepaths to be used but will work on either little or big endian.
Specify an OS feature to enable. See
doc/os.rst for more information.
Specify an OS feature to disable.
Disable use of SSE2 intrinsics
Disable use of SSSE3 intrinsics
Disable use of SSE4.1 intrinsics
Disable use of SSE4.2 intrinsics
Disable use of AVX2 intrinsics
Disable use of BMI2 intrinsics
Disable use of RDRAND intrinsics
Disable use of RDSEED intrinsics
Disable use of AES-NI intrinsics
Disable use of SHA-NI intrinsics
Disable use of AltiVec intrinsics
Disable use of NEON intrinsics
Disable use of ARMv8 Crypto intrinsics
Disable use of POWER Crypto intrinsics
Include debug symbols.
Enable some default set of sanitizer checks. What exactly is enabled depends on the compiler.
Enable specific sanitizers. See
src/build-data/cc for more information.
Disable stack smashing protections. not recommended
Add coverage info and disable optimizations
Add coverage info, but leave optimizations alone
Disable building static library
Optimize for code size.
Disable all optimizations for debugging.
Enable debug info and disable optimizations
Use amalgamation to build
Set a path to a file containing one or more trusted CA certificates in PEM format. If not given, some default locations are checked.
Setup the build in a specified directory instead of
Search for includes in this directory. Provide this parameter multiple times to define multiple additional include directories.
Add DIR to the link path. Provide this parameter multiple times to define multiple additional library link directories.
Set a compile-time pre-processor definition (i.e. add a -D… to the compiler invocations). Provide this parameter multiple times to add multiple compile-time definitions. Both KEY=VALUE and KEY (without specific value) are supported.
Use specified dir for system root while cross-compiling
Enable use of OpenMP
Include the contents of FILE into the generated build.h
Set distribution specific version information
A build configuration used by library developers, which enables extra warnings and turns most warnings into errors.
When this option is used, all relevant warnings available in the most recent release of GCC/Clang are enabled, so it may fail to build if your compiler is not sufficiently recent. In addition there may be non-default configurations or unusual platforms which cause warnings which are converted to errors. Patches addressing such warnings are welcome, but otherwise no support is available when using this option.
Turns most warnings into errors.
Skip installing Python module.
Where to install botan2.py. By default this is chosen to be the
version of Python that is running
Use valgrind API to perform additional checks. Not needed by end users.
Disable essential checks for testing. UNSAFE FOR PRODUCTION
Select which interface the fuzzer uses. Options are “afl”, “libfuzzer”, “klee”, or “test”. The “test” mode builds fuzzers that read one input from stdin and then exit.
Specify an additional library that fuzzer binaries must link with.
Build only the specific targets and tools
Provide an alternative name for a boost library. Depending on the platform and
boost’s build configuration these library names differ significantly (see Boost docs).
The provided library name must be suitable as identifier in a linker parameter,
e.g on unix:
boost_system or windows:
Skip building/installing documentation
Use Sphinx to generate the handbook
Use Sphinx to generate PDF doc
Use rst2man to generate a man page for the CLI
Use Doxygen to generate API reference
--module-policy=POL enables modules required by and
disables modules prohibited by a text policy in
Additional modules can be enabled if not prohibited by the policy.
Currently available policies include
$ ./configure.py --module-policy=bsi --enable-modules=tls,xts
Enable some specific modules
Disable some specific modules
Start with the bare minimum. This is mostly useful in conjuction with
--enable-modules to get a build that has just the features a
particular application requires.
Use Boost.Asio for networking support. This primarily affects the command line utils.
Enable bzip2 compression
Enable lzma compression
Enable using zlib compression
Enable using OpenSSL for certain operations
Enable using CommonCrypto for certain operations
Enable using sqlite3 for data storage
Enable support for TPM
A string to append to all program binaries.
A string to append to all library names.
Set the install prefix.
Set the documentation installation dir.
Set the binary installation dir.
Set the library installation dir.
Set the man page installation dir.
Set the include file installation dir.