The QEMU build system architecture

This document aims to help developers understand the architecture of the QEMU build system. As with projects using GNU autotools, the QEMU build system has two stages; first the developer runs the “configure” script to determine the local build environment characteristics, then they run “make” to build the project. This is about where the similarities with GNU autotools end, so try to forget what you know about them.

The two general ways to perform a build are as follows:

• build artifacts outside of QEMU source tree entirely:

  cd ../
  mkdir build
  cd build
  ../qemu/configure
  make
  

• build artifacts in a subdir of QEMU source tree:

  mkdir build
  cd build
  ../configure
  make
  

Most of the actual build process uses Meson under the hood, therefore build artifacts cannot be placed in the source tree itself.

Stage 1: configure

The configure script has five tasks:

• detect the host architecture

• list the targets for which to build emulators; the list of targets also affects which firmware binaries and tests to build

• find the compilers (native and cross) used to build executables, firmware and tests. The results are written as either Makefile fragments (config-host.mak) or a Meson machine file (config-meson.cross)

• create a virtual environment in which all Python code runs during the build, and possibly install packages into it from PyPI

• invoke Meson in the virtual environment, to perform the actual configuration step for the emulator build

The configure script automatically recognizes command line options for which a same-named Meson option exists; dashes in the command line are replaced with underscores.

Almost all QEMU developers that need to modify the build system will only be concerned with Meson, and therefore can skip the rest of this section.

Modifying configure

configure is a shell script; it uses #!/bin/sh and therefore should be compatible with any POSIX shell. It is important to avoid using bash-isms to avoid breaking development platforms where bash is the primary host.

The configure script provides a variety of functions to help writing portable shell code and providing consistent behavior across architectures and operating systems:

error_exit $MESSAGE $MORE...

Print $MESSAGE to stderr, followed by $MORE… and then exit from the configure script with non-zero status.

has $COMMAND

Determine if $COMMAND exists in the current environment, either as a shell builtin, or executable binary, returning 0 on success. The replacement in Meson is find_program().

probe_target_compiler $TARGET

Detect a cross compiler and cross tools for the QEMU target $TARGET (e.g., $CPU-softmmu, $CPU-linux-user, $CPU-bsd-user). If a working compiler is present, return success and set variables $target_cc, $target_ar, etc. to non-empty values.

write_target_makefile

Write a Makefile fragment to stdout, exposing the result of the most probe_target_compiler call as the usual Make variables (CC, AR, LD, etc.).

Configure does not generally perform tests for compiler options beyond basic checks to detect the host platform and ensure the compiler is functioning. These are performed using a few more helper functions:

compile_object $CFLAGS

Attempt to compile a test program with the system C compiler using $CFLAGS. The test program must have been previously written to a file called $TMPC.

compile_prog $CFLAGS $LDFLAGS

Attempt to compile a test program with the system C compiler using $CFLAGS and link it with the system linker using $LDFLAGS. The test program must have been previously written to a file called $TMPC.

check_define $NAME

Determine if the macro $NAME is defined by the system C compiler.

do_compiler $CC $ARGS...

Attempt to run the C compiler $CC, passing it $ARGS… This function does not use flags passed via options such as --extra-cflags, and therefore can be used to check for cross compilers. However, most such checks are done at make time instead (see for example the cc-option macro in pc-bios/option-rom/Makefile).

write_c_skeleton

Write a minimal C program main() function to the temporary file indicated by $TMPC.

Python virtual environments and the build process

An important step in configure is to create a Python virtual environment (venv) during the configuration phase. The Python interpreter comes from the --python command line option, the $PYTHON variable from the environment, or the system PATH, in this order. The venv resides in the pyvenv directory in the build tree, and provides consistency in how the build process runs Python code.

At this stage, configure also queries the chosen Python interpreter about QEMU’s build dependencies. Note that the build process does not look for meson, sphinx-build or avocado binaries in the PATH; likewise, there are no options such as --meson or --sphinx-build. This avoids a potential mismatch, where Meson and Sphinx binaries on the PATH might operate in a different Python environment than the one chosen by the user during the build process. On the other hand, it introduces a potential source of confusion where the user installs a dependency but configure is not able to find it. When this happens, the dependency was installed in the site-packages directory of another interpreter, or with the wrong pip program.

If a package is available for the chosen interpreter, configure prepares a small script that invokes it from the venv itself[1]. If not, configure can also optionally install dependencies in the virtual environment with pip, either from wheels in python/wheels or by downloading the package with PyPI. Downloading can be disabled with --disable-download; and anyway, it only happens when a configure option (currently, only --enable-docs) is explicitly enabled but the dependencies are not present[2].

[1]

The scripts are created based on the package’s metadata, specifically the console_script entry points. This is the same mechanism that pip uses when installing a package. Currently, in all cases it would be possible to use python -m instead of an entry point script, which makes this approach a bit overkill. On the other hand, creating the scripts is future proof and it makes the contents of the pyvenv/bin directory more informative. Portability is also not an issue, because the Python Packaging Authority provides a package distlib.scripts to perform this task.

[2]

pip might also be used when running make check-avocado if downloading is enabled, to ensure that Avocado is available.

The required versions of the packages are stored in a configuration file pythondeps.toml. The format is custom to QEMU, but it is documented at the top of the file itself and it should be easy to understand. The requirements should make it possible to use the version that is packaged that is provided by supported distros.

When dependencies are downloaded, instead, configure uses a “known good” version that is also listed in pythondeps.toml. In this scenario, pythondeps.toml behaves like the “lock file” used by cargo, poetry or other dependency management systems.

Bundled Python packages

Python packages that are mandatory dependencies to build QEMU, but are not available in all supported distros, are bundled with the QEMU sources. The only one is currently Meson (outdated in Ubuntu 22.04 and openSUSE Leap).

In order to include a new or updated wheel, modify and rerun the python/scripts/vendor.py script. The script embeds the sha256 hash of package sources and checks it. The pypi.org web site provides an easy way to retrieve the sha256 hash of the sources.

Stage 2: Meson

The Meson build system describes the build and install process for:

1. executables, which include:

   • Tools - qemu-img, qemu-nbd, qemu-ga (guest agent), etc

   • System emulators - qemu-system-$ARCH

   • Userspace emulators - qemu-$ARCH

   • Unit tests

2. documentation

3. ROMs, whether provided as binary blobs in the QEMU distributions or cross compiled under the direction of the configure script

4. other data files, such as icons or desktop files

All executables are built by default, except for some contrib/ binaries that are known to fail to build on some platforms (for example 32-bit or big-endian platforms). Tests are also built by default, though that might change in the future.

The source code is highly modularized, split across many files to facilitate building of all of these components with as little duplicated compilation as possible. Using the Meson “sourceset” functionality, meson.build files group the source files in rules that are enabled according to the available system libraries and to various configuration symbols. Sourcesets belong to one of four groups:

Subsystem sourcesets:

Various subsystems that are common to both tools and emulators have their own sourceset, for example block_ss for the block device subsystem, chardev_ss for the character device subsystem, etc. These sourcesets are then turned into static libraries as follows:

libchardev = static_library('chardev', chardev_ss.sources(),
                            build_by_default: false)

chardev = declare_dependency(objects: libchardev.extract_all_objects(recursive: false),
                             dependencies: chardev_ss.dependencies())

Target-independent emulator sourcesets:

Various general purpose helper code is compiled only once and the .o files are linked into all output binaries that need it. This includes error handling infrastructure, standard data structures, platform portability wrapper functions, etc.

Target-independent code lives in the common_ss, system_ss and user_ss sourcesets. common_ss is linked into all emulators, system_ss only in system emulators, user_ss only in user-mode emulators.

Target-dependent emulator sourcesets:

In the target-dependent set lives CPU emulation, some device emulation and much glue code. This sometimes also has to be compiled multiple times, once for each target being built. Target-dependent files are included in the specific_ss sourceset.

Each emulator also includes sources for files in the hw/ and target/ subdirectories. The subdirectory used for each emulator comes from the target’s definition of TARGET_BASE_ARCH or (if missing) TARGET_ARCH, as found in default-configs/targets/*.mak.

Each subdirectory in hw/ adds one sourceset to the hw_arch dictionary, for example:

arm_ss = ss.source_set()
arm_ss.add(files('boot.c'), fdt)
...
hw_arch += {'arm': arm_ss}

The sourceset is only used for system emulators.

Each subdirectory in target/ instead should add one sourceset to each of the target_arch and target_system_arch, which are used respectively for all emulators and for system emulators only. For example:

arm_ss = ss.source_set()
arm_system_ss = ss.source_set()
...
target_arch += {'arm': arm_ss}
target_system_arch += {'arm': arm_system_ss}

Module sourcesets:

There are two dictionaries for modules: modules is used for target-independent modules and target_modules is used for target-dependent modules. When modules are disabled the module source sets are added to system_ss and the target_modules source sets are added to specific_ss.

Both dictionaries are nested. One dictionary is created per subdirectory, and these per-subdirectory dictionaries are added to the toplevel dictionaries. For example:

hw_display_modules = {}
qxl_ss = ss.source_set()
...
hw_display_modules += { 'qxl': qxl_ss }
modules += { 'hw-display': hw_display_modules }

Utility sourcesets:

All binaries link with a static library libqemuutil.a. This library is built from several sourcesets; most of them however host generated code, and the only two of general interest are util_ss and stub_ss.

The separation between these two is purely for documentation purposes. util_ss contains generic utility files. Even though this code is only linked in some binaries, sometimes it requires hooks only in some of these and depend on other functions that are not fully implemented by all QEMU binaries. stub_ss links dummy stubs that will only be linked into the binary if the real implementation is not present. In a way, the stubs can be thought of as a portable implementation of the weak symbols concept.

The following files concur in the definition of which files are linked into each emulator:

default-configs/devices/*.mak

The files under default-configs/devices/ control the boards and devices that are built into each QEMU system emulation targets. They merely contain a list of config variable definitions such as:

include arm-softmmu.mak
CONFIG_XLNX_ZYNQMP_ARM=y
CONFIG_XLNX_VERSAL=y

*/Kconfig

These files are processed together with default-configs/devices/*.mak and describe the dependencies between various features, subsystems and device models. They are described in QEMU and Kconfig

default-configs/targets/*.mak

These files mostly define symbols that appear in the *-config-target.h file for each emulator[3]. However, the TARGET_ARCH and TARGET_BASE_ARCH will also be used to select the hw/ and target/ subdirectories that are compiled into each target.

[3]

This header is included by qemu/osdep.h when compiling files from the target-specific sourcesets.

These files rarely need changing unless you are adding a completely new target, or enabling new devices or hardware for a particular system/userspace emulation target

Adding checks

Compiler checks can be as simple as the following:

config_host_data.set('HAVE_BTRFS_H', cc.has_header('linux/btrfs.h'))

A more complex task such as adding a new dependency usually comprises the following tasks:

• Add a Meson build option to meson_options.txt.

• Add code to perform the actual feature check.

• Add code to include the feature status in config-host.h

• Add code to print out the feature status in the configure summary upon completion.

Taking the probe for SDL2_Image as an example, we have the following in meson_options.txt:

option('sdl_image', type : 'feature', value : 'auto',
       description: 'SDL Image support for icons')

Unless the option was given a non-auto value (on the configure command line), the detection code must be performed only if the dependency will be used:

sdl_image = not_found
if not get_option('sdl_image').auto() or have_system
  sdl_image = dependency('SDL2_image', required: get_option('sdl_image'),
                         method: 'pkg-config')
endif

This avoids warnings on static builds of user-mode emulators, for example. Most of the libraries used by system-mode emulators are not available for static linking.

The other supporting code is generally simple:

# Create config-host.h (if applicable)
config_host_data.set('CONFIG_SDL_IMAGE', sdl_image.found())

# Summary
summary_info += {'SDL image support': sdl_image.found()}

For the configure script to parse the new option, the scripts/meson-buildoptions.sh file must be up-to-date; make update-buildoptions (or just make) will take care of updating it.

Support scripts

Meson has a special convention for invoking Python scripts: if their first line is #! /usr/bin/env python3 and the file is not executable, find_program() arranges to invoke the script under the same Python interpreter that was used to invoke Meson. This is the most common and preferred way to invoke support scripts from Meson build files, because it automatically uses the value of configure’s –python= option.

In case the script is not written in Python, use a #! /usr/bin/env ... line and make the script executable.

Scripts written in Python, where it is desirable to make the script executable (for example for test scripts that developers may want to invoke from the command line, such as tests/qapi-schema/test-qapi.py), should be invoked through the python variable in meson.build. For example:

test('QAPI schema regression tests', python,
     args: files('test-qapi.py'),
     env: test_env, suite: ['qapi-schema', 'qapi-frontend'])

This is needed to obey the –python= option passed to the configure script, which may point to something other than the first python3 binary on the path.

By the time Meson runs, Python dependencies are available in the virtual environment and should be invoked through the scripts that configure places under pyvenv. One way to do so is as follows, using Meson’s find_program function:

sphinx_build = find_program(
     fs.parent(python.full_path()) / 'sphinx-build',
     required: get_option('docs'))

Stage 3: Make

The next step in building QEMU is to invoke make. GNU Make is required to build QEMU, and may be installed as gmake on some hosts.

The output of Meson is a build.ninja file, which is used with the Ninja build tool. However, QEMU’s build comprises other components than just the emulators (namely firmware and the tests in tests/tcg) which need different cross compilers. The QEMU Makefile wraps both Ninja and the smaller build systems for firmware and tests; it also takes care of running configure again when the script changes. Apart from invoking these sub-Makefiles, the resulting build is largely non-recursive.

Tests, whether defined in meson.build or not, are also ran by the Makefile with the traditional make check phony target, while benchmarks are run with make bench. Meson test suites such as unit can be ran with make check-unit, and make check-tcg builds and runs “non-Meson” tests for all targets.

If desired, it is also possible to use ninja and meson test, respectively to build emulators and run tests defined in meson.build. The main difference is that make needs the -jN flag in order to enable parallel builds or tests.

Useful make targets

help

Print a help message for the most common build targets.

print-VAR

Print the value of the variable VAR. Useful for debugging the build system.

Important files for the build system

Statically defined files

The following key files are statically defined in the source tree, with the rules needed to build QEMU. Their behaviour is influenced by a number of dynamically created files listed later.

Makefile

The main entry point used when invoking make to build all the components of QEMU. The default ‘all’ target will naturally result in the build of every component.

*/meson.build

The meson.build file in the root directory is the main entry point for the Meson build system, and it coordinates the configuration and build of all executables. Build rules for various subdirectories are included in other meson.build files spread throughout the QEMU source tree.

python/scripts/mkvenv.py

A wrapper for the Python venv and distlib.scripts packages. It handles creating the virtual environment, creating scripts in pyvenv/bin, and calling pip to install dependencies.

tests/Makefile.include

Rules for external test harnesses. These include the TCG tests and the Avocado-based integration tests.

tests/docker/Makefile.include

Rules for Docker tests. Like tests/Makefile.include, this file is included directly by the top level Makefile, anything defined in this file will influence the entire build system.

tests/vm/Makefile.include

Rules for VM-based tests. Like tests/Makefile.include, this file is included directly by the top level Makefile, anything defined in this file will influence the entire build system.

Dynamically created files

The following files are generated at run-time in order to control the behaviour of the Makefiles. This avoids the need for QEMU makefiles to go through any pre-processing as seen with autotools, where configure generates Makefile from Makefile.in.

Built by configure:

config-host.mak

When configure has determined the characteristics of the build host it will write the paths to various tools to this file, for use in Makefile and to a smaller extent meson.build.

config-host.mak is also used as a dependency checking mechanism. If make sees that the modification timestamp on configure is newer than that on config-host.mak, then configure will be re-run.

config-meson.cross

A Meson “cross file” (or native file) used to communicate the paths to the toolchain and other configuration options.

config.status

A small shell script that will invoke configure again with the same environment variables that were set during the first run. It’s used to rerun configure after changes to the source code, but it can also be inspected manually to check the contents of the environment.

Makefile.prereqs

A set of Makefile dependencies that order the build and execution of firmware and tests after the container images and emulators that they need.

pc-bios/*/config.mak, tests/tcg/config-host.mak, tests/tcg/*/config-target.mak

Configuration variables used to build the firmware and TCG tests, including paths to cross compilation toolchains.

pyvenv

A Python virtual environment that is used for all Python code running during the build. Using a virtual environment ensures that even code that is run via sphinx-build, meson etc. uses the same interpreter and packages.

Built by Meson:

config-host.h

Used by C code to determine the properties of the build environment and the set of enabled features for the entire build.

${TARGET-NAME}-config-devices.mak

TARGET-NAME is the name of a system emulator. The file is generated by Meson using files under configs/devices as input.

${TARGET-NAME}-config-target.mak

TARGET-NAME is the name of a system or usermode emulator. The file is generated by Meson using files under configs/targets as input.

$TARGET_NAME-config-target.h, $TARGET_NAME-config-devices.h

Used by C code to determine the properties and enabled features for each target. enabled. They are generated from the contents of the corresponding *.mak files using Meson’s configure_file() function; each target can include them using the CONFIG_TARGET and CONFIG_DEVICES macro respectively.

build.ninja

The build rules.

Built by Makefile:

Makefile.ninja

A Makefile include that bridges to ninja for the actual build. The Makefile is mostly a list of targets that Meson included in build.ninja.

Makefile.mtest

The Makefile definitions that let “make check” run tests defined in meson.build. The rules are produced from Meson’s JSON description of tests (obtained with “meson introspect –tests”) through the script scripts/mtest2make.py.