morgan/systemd
1
0
Fork 0
mirror of https://github.com/morgan9e/systemd synced 2026-04-15 00:47:10 +09:00
Code Issues Packages Projects Releases Wiki Activity
Files
ac9391c55210e08981484fce07fe034ec0af21ab
systemd/meson.build

3214 lines
123 KiB
Meson
Raw Normal View History

license: LGPL-2.1+ -> LGPL-2.1-or-later
2020-11-09 13:23:58 +09:00
# SPDX-License-Identifier: LGPL-2.1-or-later
Add license headers and SPDX identifiers to meson.build files So far I avoided adding license headers to meson files, but they are pretty big and important and should carry license headers like everything else. I added my own copyright, even though other people modified those files too. But this is mostly symbolic, so I hope that's OK.
2017-11-18 18:32:01 +01:00
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
project('systemd', 'c',
meson: Start adding devel and rc suffixes to the project version Let's make sure that versions generated by meson-vcs-tag.sh always sort higher than official and stable releases. We achieve this by immediately updating the meson version in meson.build after a new release. To make sure this version always sorts lower than future rcs, we suffix it with "~devel" which will sort lower than "~rcX". The new release workflow is to update the version in meson.build for each rc and the official release and to also update the version number after a new release to the next development version. The full version is exposed as PROJECT_VERSION_FULL and used where it makes sense over PROJECT_VERSION. We also switch to reading the version from a meson.version file in the repo instead of hardcoding it in meson.build. This makes it easier to access both inside and outside of the project. The meson-vcs-tag.sh script is rewritten to query the version from meson.version instead of passing it in via the command line. This makes it easier to use outside of systemd since users don't have to query the version themselves first.
2024-02-14 11:04:36 +01:00
version : files('meson.version'),
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
license : 'LGPLv2+',
default_options: [
meson: bump C std to gnu17 GCC 8.4 (our baseline) defaults to gnu17 already.
2025-01-16 18:50:54 +01:00
'c_std=gnu17',
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-17 19:25:00 -04:00
'prefix=/usr',
'sysconfdir=/etc',
'localstatedir=/var',
meson: use warning_level=2 by default Let's bump up the warning level, and not add by -Wextra by hand. This is the approach recommended by meson. The idea is that all projects should be as similar as possible to make it easier for users to switch between projects.
2019-11-10 11:39:15 +01:00
'warning_level=2',
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
],
meson: bump required minimum version to 0.62.0 The commit 8442ac9c0264ac7beb5afd6c3bf922030a6edaf3 set install_tag option to install_emptydir() calls, but it requires meson-0.62.0. Hence, after the commit, we cannot build systemd with older meson anymore. As using install_tag is quite useful for building systemd package, let's bump the requirement of meson version to 0.62.0. Note, the current meson versions of major distributions are: CentOS 9: 0.63.3 CentOS 10: 1.4.1 Fedora 40: 1.4.1 Fedora 41: 1.5.1 Ubuntu 20.04 LTS (focal): 0.53.2 -- EOL on 2025-04 Ubuntu 22.04 LTS (jammy): 0.61.2 -- EOL on 2027-04 Ubuntu 24.04 LTS (noble): 1.3.2 Ubuntu 24.10 (oracular): 1.5.2 Debian 11 (bullseye): 0.56.2 (1.0.0 in backports) -- EOL on 2024-08 Debian 12 (bookworm): 1.0.1 (1.5.1 in backports) openSUSE Leap 15.6: 1.6.1 openSUSE Tumbleweed: 1.6.1 As the next version (v258) is not expected to be released before the end of 2025-04, it is OK to cut the support of Ubuntu 20.04 LTS and Debian 11. Also, our policy for support of distributions explicitly says only latest Ubuntu LTS and non-LTS releases are supported. Hence, we can also cut Ubuntu 22.04, even if it is not EOL. Follow-up for 8442ac9c0264ac7beb5afd6c3bf922030a6edaf3. Closes #35967.
2025-03-05 03:14:04 +09:00
meson_version : '>= 0.62.0',
)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
test: Use meson add_test_setup() instead of environment variables We add a default test setup that excludes the integration-tests suite so that the integration tests don't run by default. This allows us to get rid of $SYSTEMD_INTEGRATION_TESTS. Then, we add two extra setups: 'integration' and 'shell'. The 'integration' setup does not exclude the integration-tests suite, and so can be used to run the integration tests. The 'shell' setup does the same, but additionally sets $TEST_SHELL=1, allowing to get rid of $TEST_SHELL in the docs.
2025-04-17 13:29:16 +02:00
add_test_setup(
'default',
meson: Rework clang-tidy integration to be done via unit tests Instead of using run-clang-tidy.py with its own scheduling, let's just gather a list of source files ourselves and then use that to add a unit test for each source file that runs clang-tidy on the source file. We also add a bit of logic to run clang-tidy on most header files as well for extra coverage. This uncovered various header files that were not standalone so this commit also includes fixes to make sure the clang-tidy tests are all green. We can also use this in a later commit to run clang-include-cleaner on each source file in the same way.
2025-05-15 15:09:27 +02:00
exclude_suites : ['clang-tidy', 'integration-tests'],
test: Use meson add_test_setup() instead of environment variables We add a default test setup that excludes the integration-tests suite so that the integration tests don't run by default. This allows us to get rid of $SYSTEMD_INTEGRATION_TESTS. Then, we add two extra setups: 'integration' and 'shell'. The 'integration' setup does not exclude the integration-tests suite, and so can be used to run the integration tests. The 'shell' setup does the same, but additionally sets $TEST_SHELL=1, allowing to get rid of $TEST_SHELL in the docs.
2025-04-17 13:29:16 +02:00
is_default : true,
)
meson: Fix various versions Follow up for 8b3b01c4b7e0fde39b4be354990ee68f5e612c52 We switch to PROJECT_VERSION instead of PROJECT_VERSION_FULL where we report our version and which is likely being parsed to avoid breaking compat. If we didn't, the output would change from systemd 255 to systemd 255.1 which could break various tools.
2024-07-04 14:44:21 +02:00
project_major_version = meson.project_version().split('.')[0].split('~')[0]
if meson.project_version().contains('.')
project_minor_version = meson.project_version().split('.')[-1].split('~')[0]
else
project_minor_version = '0'
endif
meson: update version numbers for v258~rc1
2025-07-23 14:33:35 +01:00
libsystemd_version = '0.41.0'
libudev_version = '1.7.11'
meson: move library version defines to the top (#6939)
2017-09-28 19:24:16 +02:00
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
conf = configuration_data()
tree-wide: Update homepage to systemd.io
2022-05-19 13:58:15 +02:00
conf.set_quoted('PROJECT_URL', 'https://systemd.io/')
meson: Fix various versions Follow up for 8b3b01c4b7e0fde39b4be354990ee68f5e612c52 We switch to PROJECT_VERSION instead of PROJECT_VERSION_FULL where we report our version and which is likely being parsed to avoid breaking compat. If we didn't, the output would change from systemd 255 to systemd 255.1 which could break various tools.
2024-07-04 14:44:21 +02:00
conf.set('PROJECT_VERSION', project_major_version,
meson: explain GIT_VERSION and PROJECT_VERSION Fixes #11415.
2020-02-28 11:09:16 +01:00
description : 'Numerical project version (used where a simple number is expected)')
meson: introduce PROJECT_VERSION_STR and use it in udev
2024-11-12 13:53:13 +01:00
conf.set_quoted('PROJECT_VERSION_STR', project_major_version,
description: 'Stringified project version (used where a simple string is expected)')
meson: Start adding devel and rc suffixes to the project version Let's make sure that versions generated by meson-vcs-tag.sh always sort higher than official and stable releases. We achieve this by immediately updating the meson version in meson.build after a new release. To make sure this version always sorts lower than future rcs, we suffix it with "~devel" which will sort lower than "~rcX". The new release workflow is to update the version in meson.build for each rc and the official release and to also update the version number after a new release to the next development version. The full version is exposed as PROJECT_VERSION_FULL and used where it makes sense over PROJECT_VERSION. We also switch to reading the version from a meson.version file in the repo instead of hardcoding it in meson.build. This makes it easier to access both inside and outside of the project. The meson-vcs-tag.sh script is rewritten to query the version from meson.version instead of passing it in via the command line. This makes it easier to use outside of systemd since users don't have to query the version themselves first.
2024-02-14 11:04:36 +01:00
conf.set_quoted('PROJECT_VERSION_FULL', meson.project_version(), description : 'Full project version')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
basic/log: add a define for path relative to source root When using build/ directory inside of the source directory: __FILE__: ../src/test/test-log.c RELATIVE_SOURCE_PATH: .. PROJECT_FILE: src/test/test-log.c When using a build directory outside of the source directory: __FILE__: ../../../home/zbyszek/src/systemd-work/src/test/test-log.c RELATIVE_SOURCE_PATH: ../../../home/zbyszek/src/systemd-work PROJECT_FILE: src/test/test-log.c
2019-06-07 14:41:36 +02:00
relative_source_path = run_command('realpath',
meson: Drop project_source_root and project_build_root variables We can now rely on meson.project_source_root() and meson.project_build_root() which do the same thing.
2025-03-26 13:57:33 +01:00
'--relative-to=@0@'.format(meson.project_build_root()),
meson.project_source_root(),
meson: add check:true/false to all run_command() invocations meson-0.59.4-1.fc35.noarch says: WARNING: You should add the boolean check kwarg to the run_command call. It currently defaults to false, but it will default to true in future releases of meson. See also: https://github.com/mesonbuild/meson/issues/9300
2021-11-17 13:58:53 +01:00
check : true).stdout().strip()
basic/log: add a define for path relative to source root When using build/ directory inside of the source directory: __FILE__: ../src/test/test-log.c RELATIVE_SOURCE_PATH: .. PROJECT_FILE: src/test/test-log.c When using a build directory outside of the source directory: __FILE__: ../../../home/zbyszek/src/systemd-work/src/test/test-log.c RELATIVE_SOURCE_PATH: ../../../home/zbyszek/src/systemd-work PROJECT_FILE: src/test/test-log.c
2019-06-07 14:41:36 +02:00
conf.set_quoted('RELATIVE_SOURCE_PATH', relative_source_path)
meson: scope more git invocations with current_source_dir()
2019-02-27 11:19:07 -05:00
meson: simplify the BUILD_MODE conditional Using a enum is all nice and generic, but at this point it seems unlikely that we'll add further build modes. But having an enum means that we need to include the header file with the enumeration whenerever the conditional is used. I want to use the conditional in log.h, which makes it hard to avoid circular imports.
2021-04-14 13:17:22 +02:00
conf.set10('BUILD_MODE_DEVELOPER', get_option('mode') == 'developer',
description : 'tailor build to development or release builds')
meson: enable vmspawn by default in developer mode This should also implicitly enabled vmspawn in CI. It wasn't passing even the basic tests, which we didn't see, because it needs to be explicitly enabled.
2023-11-07 21:51:40 +01:00
meson: Convert more options to meson features The semantics for libidn2 and pwquality have changed slightly: We will pick a preferred one if both are enabled instead of making it an error.
2023-08-10 16:00:55 +02:00
feature = get_option('log-message-verification')
if feature.auto()
have = conf.get('BUILD_MODE_DEVELOPER') == 1
meson: turn on log-message-verification by default in developer builds I'm not _quite_ convinced that this a good idea… I'm at least keeping it separate to make it easy to revert ;)
2022-05-11 16:42:13 +02:00
else
meson: Convert more options to meson features The semantics for libidn2 and pwquality have changed slightly: We will pick a preferred one if both are enabled instead of making it an error.
2023-08-10 16:00:55 +02:00
have = feature.enabled()
meson: turn on log-message-verification by default in developer builds I'm not _quite_ convinced that this a good idea… I'm at least keeping it separate to make it easy to revert ;)
2022-05-11 16:42:13 +02:00
endif
meson: Convert more options to meson features The semantics for libidn2 and pwquality have changed slightly: We will pick a preferred one if both are enabled instead of making it an error.
2023-08-10 16:00:55 +02:00
conf.set10('LOG_MESSAGE_VERIFICATION', have)
meson: add "develop mode" config switch
2020-08-27 17:26:49 +02:00
meson: c++ is required for the fuzzer builds The configuration would fail with an error about cpp being an unknown language anyway, but it's nicer to fail early and explicitly. https://github.com/systemd/systemd/pull/10339#issuecomment-428279175
2018-10-10 11:56:45 +02:00
want_ossfuzz = get_option('oss-fuzz')
want_libfuzzer = get_option('llvm-fuzz')
meson.build: change operator combining bools from + to and upstream meson stopped allowing combining boolean with the plus operator, and now requires using the logical and operator reference: https://github.com/mesonbuild/meson/commit/43302d3296baff6aeaf8e03f5d701b0402e37a6c Fixes: #20632
2021-09-03 12:43:33 -04:00
if want_ossfuzz and want_libfuzzer
ci: turn off FuzzBuzz I don't think anyone uses it.
2020-03-30 06:42:19 +00:00
error('only one of oss-fuzz or llvm-fuzz can be specified')
meson: c++ is required for the fuzzer builds The configuration would fail with an error about cpp being an unknown language anyway, but it's nicer to fail early and explicitly. https://github.com/systemd/systemd/pull/10339#issuecomment-428279175
2018-10-10 11:56:45 +02:00
endif
tests: hook up fuzz targets to FuzzBuzz
2019-05-05 19:28:42 +00:00
ci: turn off FuzzBuzz I don't think anyone uses it.
2020-03-30 06:42:19 +00:00
fuzzer_build = want_ossfuzz or want_libfuzzer
meson: c++ is required for the fuzzer builds The configuration would fail with an error about cpp being an unknown language anyway, but it's nicer to fail early and explicitly. https://github.com/systemd/systemd/pull/10339#issuecomment-428279175
2018-10-10 11:56:45 +02:00
fuzzers: ignore size limits when compiled standalone This way we can still call fuzzers on old samples, but oss-fuzz will not waste its and our time finding overly large inputs.
2022-05-12 12:51:11 +02:00
# If we're building *not* for actual fuzzing, allow input samples of any size
# (for testing and for reproduction of issues discovered with previously-higher
# limits).
conf.set10('FUZZ_USE_SIZE_LIMIT', fuzzer_build)
meson: use 0|1 for SD_BOOT We converted to not using #ifdef for most of our defines because the syntax is nicer and we are protected against typos and can set -Werror=undef. Let's do the same for SD_BOOT. The define is nicely hidden in build.h for normal builds, and for EFI builds we were already setting SD_BOOT on the commandline.
2023-01-10 14:25:57 +01:00
# We'll set this to '1' for EFI builds in a different place.
conf.set10('SD_BOOT', false)
meson: move efi summary() section to src/boot/efi This way we can add the entries more naturally in the same place where they are defined.
2021-12-23 13:05:01 +01:00
# Create a title-less summary section early, so it ends up first in the output.
# More items are added later after they have been detected.
summary({'build mode' : get_option('mode')})
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
#####################################################################
meson: install the git hook (#6425) This was done autogen.sh previously and was dropped in 72cdb3e783174dcf9223a49f03e3b0e2ca95ddb8. Let's add it back. The meson configuration step is the only reasonable place. Note that this only works for the most standard git dirs, e.g. the hook will not be installed if git worktree is used or if $GIT_DIR is specified, etc. I think that's OK because most of the time meson will be run at least once in the original cloned dir.
2017-07-24 04:41:45 -04:00
# Try to install the git pre-commit hook
meson: Rename add-git-hook.sh to git-setup.sh and configure git in it Let's automatically apply the recommended git config with meson.
2024-03-14 10:29:55 +01:00
git_setup_sh = find_program('tools/git-setup.sh', required : false)
if git_setup_sh.found()
git_hook = run_command(git_setup_sh, check : false)
meson: use find_program() for add-git-hook.sh On azure systemd.systemd ci, the build would fail with: meson.build:53:0: ERROR: Program or command '/home/appuser/fuzzer/tools/add-git-hook.sh' not found or not executable We use find_program() for all helpers, so let's do it for this one too. This should solve the issue, whatever it exactly is.
2020-03-28 09:30:51 +01:00
if git_hook.returncode() == 0
message(git_hook.stdout().strip())
endif
meson: install the git hook (#6425) This was done autogen.sh previously and was dropped in 72cdb3e783174dcf9223a49f03e3b0e2ca95ddb8. Let's add it back. The meson configuration step is the only reasonable place. Note that this only works for the most standard git dirs, e.g. the hook will not be installed if git worktree is used or if $GIT_DIR is specified, etc. I think that's OK because most of the time meson will be run at least once in the original cloned dir.
2017-07-24 04:41:45 -04:00
endif
#####################################################################
meson: Use fs module Turns out that meson provides a fs module since 0.53.0, making it unnecessary to call out to test.
2021-11-13 17:01:16 +01:00
fs = import('fs')
meson: support both separate and merged sbin-bin directories Follow-up for ba7f4ae6178309dc937e10cf7dce0eca9dafb8de. By default, we detect if the real root has a separate /usr/sbin directory, but this can be overrides with -Dsplit-bin=true|false. The check assumes that /usr/sbin is split if it is not a symlink, so it'll return a false negative with some more complicated setups. But that's OK, in those cases this should be configured explicitly. This will copy the structure of the directories in the root file system to $DESTDIR. If a directory is a directory in $DESTDIR but a symlink in the root file system, this script will fail. This means that it's not possible to reuse a $DESTDIR from between ba7f4ae61 and this patch.
2018-03-01 10:28:29 +01:00
if get_option('split-bin') == 'auto'
meson: Use fs module Turns out that meson provides a fs module since 0.53.0, making it unnecessary to call out to test.
2021-11-13 17:01:16 +01:00
split_bin = not fs.is_symlink('/usr/sbin')
meson: support both separate and merged sbin-bin directories Follow-up for ba7f4ae6178309dc937e10cf7dce0eca9dafb8de. By default, we detect if the real root has a separate /usr/sbin directory, but this can be overrides with -Dsplit-bin=true|false. The check assumes that /usr/sbin is split if it is not a symlink, so it'll return a false negative with some more complicated setups. But that's OK, in those cases this should be configured explicitly. This will copy the structure of the directories in the root file system to $DESTDIR. If a directory is a directory in $DESTDIR but a symlink in the root file system, this script will fail. This means that it's not possible to reuse a $DESTDIR from between ba7f4ae61 and this patch.
2018-03-01 10:28:29 +01:00
else
split_bin = get_option('split-bin') == 'true'
endif
Remove /sbin from paths if split-bin is false (#8324) Follow-up for 157baa87e4.
2018-03-01 21:48:36 +01:00
conf.set10('HAVE_SPLIT_BIN', split_bin,
description : 'bin and sbin directories are separate')
meson: support both separate and merged sbin-bin directories Follow-up for ba7f4ae6178309dc937e10cf7dce0eca9dafb8de. By default, we detect if the real root has a separate /usr/sbin directory, but this can be overrides with -Dsplit-bin=true|false. The check assumes that /usr/sbin is split if it is not a symlink, so it'll return a false negative with some more complicated setups. But that's OK, in those cases this should be configured explicitly. This will copy the structure of the directories in the root file system to $DESTDIR. If a directory is a directory in $DESTDIR but a symlink in the root file system, this script will fail. This means that it's not possible to reuse a $DESTDIR from between ba7f4ae61 and this patch.
2018-03-01 10:28:29 +01:00
meson: add a new -Dstandalone-binaries=true option This adds an option to build standalone binaries that do not depend on the systemd-shared library. This option can be handy to build binaries that can be useful on a non-systemd system, binaries such as systemd-sysusers and systemd-tmpfiles have been previously requested, but installing them with all the required dependencies pulls in too much code that isn't really relevant for those use cases. The standalone use case is also relevant in containers, where minimizing the size of the container image is quite relevant. For now, only `systemd-sysusers` is also built as a standalone binary. The standalone binaries are installed as `/usr/bin/%{name}.standalone`, the packaging system is reponsible for renaming those into the correct names during the packaging step. RPM is able to do so with RemovePathPostfixes: The default behavior is to build shared binaries only, since this option is mainly intended for building distribution packages. Tested that a proper separate binary is built when using this option and that having it disabled (or using the default Meson configuration) does not produce a binary for this option.
2020-06-03 14:02:59 -07:00
have_standalone_binaries = get_option('standalone-binaries')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
sysvinit_path = get_option('sysvinit-path')
sysvrcnd_path = get_option('sysvrcnd-path')
meson: generate m4 preprocessor from config.h (#8914)
2018-05-07 18:17:35 +09:00
conf.set10('HAVE_SYSV_COMPAT', sysvinit_path != '' and sysvrcnd_path != '',
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 10:41:51 +02:00
description : 'SysV init scripts and rcN.d links are supported')
meson: disable rc-local-generator when an empty string is specified to rc-local= meson option In such case, the generator is meaningless. Let's kill it.
2025-04-02 22:09:12 +09:00
sysvrclocal_path = get_option('rc-local')
conf.set10('HAVE_SYSV_RC_LOCAL', sysvrclocal_path != '')
docs: move /var/log/README to a tmpfiles.d symlink This moves the /var/log/README content out of /var and into the docs location, replacing the previous file with a symlink created through a tmpfiles.d entry.
2021-07-08 09:47:32 +00:00
conf.set10('CREATE_LOG_DIRS', get_option('create-log-dirs'))
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
allow removal of initrd services
2020-04-08 23:59:13 +02:00
if get_option('hibernate') and not get_option('initrd')
error('hibernate depends on initrd')
endif
main: bump fs.nr_open + fs.max-file to their largest possible values After discussions with kernel folks, a system with memcg really shouldn't need extra hard limits on file descriptors anymore, as they are properly accounted for by memcg anyway. Hence, let's bump these values to their maximums. This also adds a build time option to turn thiss off, to cover those users who do not want to use memcg.
2018-10-11 18:23:26 +02:00
conf.set10('BUMP_PROC_SYS_FS_FILE_MAX', get_option('bump-proc-sys-fs-file-max'))
conf.set10('BUMP_PROC_SYS_FS_NR_OPEN', get_option('bump-proc-sys-fs-nr-open'))
meson: let's bump RLIMIT_NOFILE hard limit to 512K Prompted by: https://lists.freedesktop.org/archives/systemd-devel/2018-October/041578.html
2018-11-15 09:50:42 +01:00
conf.set('HIGH_RLIMIT_NOFILE', 512*1024)
main: bump fs.nr_open + fs.max-file to their largest possible values After discussions with kernel folks, a system with memcg really shouldn't need extra hard limits on file descriptors anymore, as they are properly accounted for by memcg anyway. Hence, let's bump these values to their maximums. This also adds a build time option to turn thiss off, to cover those users who do not want to use memcg.
2018-10-11 18:23:26 +02:00
meson: use a/b instead of join_paths(a,b) It is nicer and shorter.
2021-07-27 19:32:35 +02:00
# Meson ignores the preceding arguments when joining paths if an absolute
# component is encountered, so this should canonicalize various paths when they
# are absolute or relative.
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
prefixdir = get_option('prefix')
if not prefixdir.startswith('/')
meson: using f-strings in meson Those were added before 0.61 which is our minimum version. Dots were dropped from the end of some messages. Most messages did not have the trailing dot. 0.63 added support for multi-line f-strings. We can't use those yet.
2025-10-03 15:26:20 +02:00
error(f'Prefix is not absolute: "@prefixdir@"')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
endif
meson: refuse when prefix is not a child of rootprefix This is most likely to happen when setting one but not the other. Note that we already warn when rootprefixdir != rootprefix_default, at the very end.
2021-02-24 18:02:36 +01:00
Drop split-usr and unmerged-usr support As previously announced, execute order 66: https://lists.freedesktop.org/archives/systemd-devel/2022-September/048352.html The meson options split-usr, rootlibdir and rootprefix become no-ops that print a warning if they are set to anything other than the default values. We can remove them in a future release.
2023-06-12 02:15:19 +01:00
prefixdir_noslash = '/' + prefixdir.strip('/')
meson: use a/b instead of join_paths(a,b) It is nicer and shorter.
2021-07-27 19:32:35 +02:00
bindir = prefixdir / get_option('bindir')
Drop split-usr and unmerged-usr support As previously announced, execute order 66: https://lists.freedesktop.org/archives/systemd-devel/2022-September/048352.html The meson options split-usr, rootlibdir and rootprefix become no-ops that print a warning if they are set to anything other than the default values. We can remove them in a future release.
2023-06-12 02:15:19 +01:00
sbindir = prefixdir / (split_bin ? 'sbin' : 'bin')
meson: use install_emptydir() and drop meson-make-symlink.sh The script is mostly equivalent to 'mkdir -p' and 'ln -sfr'. Let's replace it with install_emptydir() builtin function and inline meson call.
2023-08-08 10:38:47 +09:00
sbin_to_bin = split_bin ? '../bin/' : ''
meson: use a/b instead of join_paths(a,b) It is nicer and shorter.
2021-07-27 19:32:35 +02:00
libdir = prefixdir / get_option('libdir')
sysconfdir = prefixdir / get_option('sysconfdir')
includedir = prefixdir / get_option('includedir')
datadir = prefixdir / get_option('datadir')
localstatedir = '/' / get_option('localstatedir')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
Drop split-usr and unmerged-usr support As previously announced, execute order 66: https://lists.freedesktop.org/archives/systemd-devel/2022-September/048352.html The meson options split-usr, rootlibdir and rootprefix become no-ops that print a warning if they are set to anything other than the default values. We can remove them in a future release.
2023-06-12 02:15:19 +01:00
libexecdir = prefixdir / 'lib/systemd'
meson: use install_symlink() where applicable Now our baseline of meson is 0.62, hence install_symlink() can be used. Note, install_symlink() implies install_emptydir() for specified install_dir. Hence, this also drops several unnecessary install_emptydir() calls. Note, the function currently does not support 'relative' and 'force' flags, so several 'ln -frsT' inline calls cannot be replaced.
2025-03-05 06:27:08 +09:00
libexecdir_to_bin = '../../bin/'
Drop split-usr and unmerged-usr support As previously announced, execute order 66: https://lists.freedesktop.org/archives/systemd-devel/2022-September/048352.html The meson options split-usr, rootlibdir and rootprefix become no-ops that print a warning if they are set to anything other than the default values. We can remove them in a future release.
2023-06-12 02:15:19 +01:00
pkglibdir = libdir / 'systemd'
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
Add install-sysconfdir=no-samples option for (non-)installation of sample configs By default, systemd installs various sample configuration files containing commented-out defaults. Systems seeking to minimize the number of files in /etc may wish to install directories and configuration files that have semantic effects, but not install not commented-out sample configuration files. Turn install-sysconfdir into a multi-valued option, with a "no-samples" value to skip installing sample-only configuration files.
2020-12-30 15:11:30 -08:00
install_sysconfdir = get_option('install-sysconfdir') != 'false'
install_sysconfdir_samples = get_option('install-sysconfdir') == 'true'
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
# Dirs of external packages
meson: use a/b instead of join_paths(a,b) It is nicer and shorter.
2021-07-27 19:32:35 +02:00
pkgconfigdatadir = get_option('pkgconfigdatadir') != '' ? get_option('pkgconfigdatadir') : datadir / 'pkgconfig'
pkgconfiglibdir = get_option('pkgconfiglibdir') != '' ? get_option('pkgconfiglibdir') : libdir / 'pkgconfig'
polkitpolicydir = datadir / 'polkit-1/actions'
polkitrulesdir = datadir / 'polkit-1/rules.d'
polkitpkladir = localstatedir / 'lib/polkit-1/localauthority/10-vendor.d'
xinitrcdir = get_option('xinitrcdir') != '' ? get_option('xinitrcdir') : sysconfdir / 'X11/xinit/xinitrc.d'
meson: do not install RPM macros if rpmmacrosdir is set to no (#7430) Fixes #7427.
2017-11-23 22:20:22 +09:00
rpmmacrosdir = get_option('rpmmacrosdir')
if rpmmacrosdir != 'no'
meson: use a/b instead of join_paths(a,b) It is nicer and shorter.
2021-07-27 19:32:35 +02:00
rpmmacrosdir = prefixdir / rpmmacrosdir
meson: do not install RPM macros if rpmmacrosdir is set to no (#7430) Fixes #7427.
2017-11-23 22:20:22 +09:00
endif
Drop split-usr and unmerged-usr support As previously announced, execute order 66: https://lists.freedesktop.org/archives/systemd-devel/2022-September/048352.html The meson options split-usr, rootlibdir and rootprefix become no-ops that print a warning if they are set to anything other than the default values. We can remove them in a future release.
2023-06-12 02:15:19 +01:00
modprobedir = prefixdir / 'lib/modprobe.d'
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
# Our own paths
meson: use a/b instead of join_paths(a,b) It is nicer and shorter.
2021-07-27 19:32:35 +02:00
pkgdatadir = datadir / 'systemd'
environmentdir = prefixdir / 'lib/environment.d'
pkgsysconfdir = sysconfdir / 'systemd'
userunitdir = prefixdir / 'lib/systemd/user'
userpresetdir = prefixdir / 'lib/systemd/user-preset'
tmpfilesdir = prefixdir / 'lib/tmpfiles.d'
pkgconfig,rpm: expose vars for user-tmpfiles.d location Fixes https://bugzilla.redhat.com/show_bug.cgi?id=2098553.
2022-06-20 15:06:09 +02:00
usertmpfilesdir = prefixdir / 'share/user-tmpfiles.d'
meson: use a/b instead of join_paths(a,b) It is nicer and shorter.
2021-07-27 19:32:35 +02:00
sysusersdir = prefixdir / 'lib/sysusers.d'
sysctldir = prefixdir / 'lib/sysctl.d'
binfmtdir = prefixdir / 'lib/binfmt.d'
modulesloaddir = prefixdir / 'lib/modules-load.d'
Drop split-usr and unmerged-usr support As previously announced, execute order 66: https://lists.freedesktop.org/archives/systemd-devel/2022-September/048352.html The meson options split-usr, rootlibdir and rootprefix become no-ops that print a warning if they are set to anything other than the default values. We can remove them in a future release.
2023-06-12 02:15:19 +01:00
networkdir = prefixdir / 'lib/systemd/network'
systemgeneratordir = libexecdir / 'system-generators'
meson: use a/b instead of join_paths(a,b) It is nicer and shorter.
2021-07-27 19:32:35 +02:00
usergeneratordir = prefixdir / 'lib/systemd/user-generators'
systemenvgeneratordir = prefixdir / 'lib/systemd/system-environment-generators'
userenvgeneratordir = prefixdir / 'lib/systemd/user-environment-generators'
Drop split-usr and unmerged-usr support As previously announced, execute order 66: https://lists.freedesktop.org/archives/systemd-devel/2022-September/048352.html The meson options split-usr, rootlibdir and rootprefix become no-ops that print a warning if they are set to anything other than the default values. We can remove them in a future release.
2023-06-12 02:15:19 +01:00
systemshutdowndir = libexecdir / 'system-shutdown'
systemsleepdir = libexecdir / 'system-sleep'
systemunitdir = prefixdir / 'lib/systemd/system'
systempresetdir = prefixdir / 'lib/systemd/system-preset'
tree-wide: Add initrd presets Currently, when running "systemctl preset-all --root=xxx" in mkosi to enable/disable units for initrds, the system presets are used. The problem with this approach is that the system presets are written for the system, and that is not necessarily ideal for an initrd, but we still want to use the same packages in the initrd that we install in the system, so let's introduce a separate directory for initrd presets which is used to pick up preset files from when we detect that we're configuring an initrd (by looking for /etc/initrd-release). We also introduce a systemd preset file for the initrd, which is based on the system one, except with all the stuff unnecessary for the initrd removed.
2025-07-02 09:43:30 +02:00
initrdpresetdir = prefixdir / 'lib/systemd/initrd-preset'
Drop split-usr and unmerged-usr support As previously announced, execute order 66: https://lists.freedesktop.org/archives/systemd-devel/2022-September/048352.html The meson options split-usr, rootlibdir and rootprefix become no-ops that print a warning if they are set to anything other than the default values. We can remove them in a future release.
2023-06-12 02:15:19 +01:00
udevlibexecdir = prefixdir / 'lib/udev'
meson: use a/b instead of join_paths(a,b) It is nicer and shorter.
2021-07-27 19:32:35 +02:00
udevrulesdir = udevlibexecdir / 'rules.d'
udevhwdbdir = udevlibexecdir / 'hwdb.d'
catalogdir = prefixdir / 'lib/systemd/catalog'
kernel-install: replace 00-entry-directory with K_I_LAYOUT in k-i 341890de866f2ee34919a47ce3fc6c8cd3c1924c made "bootctl install" create ESP\MID, in preparation of cf73f650890b56a59bfb713c4c82b4e29daa7316 that followed it and created 00-entry-directory.install to make ESP\MID\KVER if ESP\MID existed ‒ this meant that "bootctl install" followed by "kernel-install $(uname -r) /boot/vml*$(uname -r) /boot/ini*$(uname -r)" actually installed the kernel correctly. Later, 31e57550b552e113bd3d44355b237c41e42beb58 reverted the first commit, meaning, that now running those two commands first installs sd-boot, but then does nothing. Everything appears to work right, nothing errors out, but no changes are actually done. To the untrained eye (all of them), even running with -v appears to work: all the hooks are run, as is depmod, but, again, nothing happens. This is horrible. Nothing in either manpage suggests what to do (nor should it, really), but the user is left with a bootloader that appears fully funxional, since nothing suggests a failure in the output, but with an unbootable machine, /no way to boot it/, even if they drop to an EFI shell, since the boot bundle isn't present on the ESP, and no real recourse even if they boot into a recovery system, apart from installing like GRUB or whatever. 00- is purely instrumentation for 90-, and separating one from the other has led to downstream dissatisfaxion (indeed, the last mentioned commit cited cited exactly that as the reversion reason), while creating $ENTRY_DIR_ABS is only required for bootloaders using the BLS, and shouldn't itself toggle anything. To that end, introduce an /{e,l}/k/install.conf file that allows overriding the detected layout, and detect it as "bls" if $BOOT_ROOT/$MACHINE_ID ($ENTRY_DIR_ABS/..) exists, otherwise "other" ‒ if a user wishes to select a different bootloader, like GRUB, they (or, indeed, the postinst script) can specify layout=grub. This disables 90- and $ENTRY_DIR_ABS manipulation.
2021-03-16 16:47:34 +01:00
kerneldir = prefixdir / 'lib/kernel'
kernelinstalldir = kerneldir / 'install.d'
meson: use a/b instead of join_paths(a,b) It is nicer and shorter.
2021-07-27 19:32:35 +02:00
factorydir = datadir / 'factory'
bootlibdir = prefixdir / 'lib/systemd/boot/efi'
testsdir = prefixdir / 'lib/systemd/tests'
test: install unit tests in a dedicated subdirectory below '$testsdir' /usr/lib/systemd/tests may contain more than the unit tests. For example on SUSE we also install the integration tests there. Putting the unit tests in a dedicated directory named 'unit-tests' makes the layout cleaner. Note that `run-unit-tests.py` has not been moved so we don't need to adjust (Fedora) packaging and users also don't need to descend into the subdirectory.
2023-01-26 16:07:17 +01:00
unittestsdir = testsdir / 'unit-tests'
meson: define testdata_dir globally
2023-03-17 14:14:16 +01:00
testdata_dir = testsdir / 'testdata'
meson: use a/b instead of join_paths(a,b) It is nicer and shorter.
2021-07-27 19:32:35 +02:00
systemdstatedir = localstatedir / 'lib/systemd'
catalogstatedir = systemdstatedir / 'catalog'
randomseeddir = localstatedir / 'lib/systemd'
Drop split-usr and unmerged-usr support As previously announced, execute order 66: https://lists.freedesktop.org/archives/systemd-devel/2022-September/048352.html The meson options split-usr, rootlibdir and rootprefix become no-ops that print a warning if they are set to anything other than the default values. We can remove them in a future release.
2023-06-12 02:15:19 +01:00
profiledir = libexecdir / 'portable' / 'profile'
repart: make it easy to generate DDIs This adds --make-ddi=confext, --make-ddi=sysext, --make-ddi=portable, to make it really easiy to generate DDIs of the specified class. It it's ultimately just a fancy wrapper around some defaults and in particular --definitions=. This makes it very easy to generate a confext: $ systemd-repart -C --private-key=privkey.pem --certificate=cert.crt -s mytree/ mytree.confext.raw
2023-09-28 18:16:59 +02:00
repartdefinitionsdir = libexecdir / 'repart/definitions'
Drop split-usr and unmerged-usr support As previously announced, execute order 66: https://lists.freedesktop.org/archives/systemd-devel/2022-September/048352.html The meson options split-usr, rootlibdir and rootprefix become no-ops that print a warning if they are set to anything other than the default values. We can remove them in a future release.
2023-06-12 02:15:19 +01:00
ntpservicelistdir = prefixdir / 'lib/systemd/ntp-units.d'
meson: Create credstore directories Let's make the creds directories a bit more discoverable and make it easier for users to use them. This also allows us to fix the mode to 0700 for /etc instead of the usual 0755 which is what probably would happen if users had to create this directory themselves.
2023-05-24 15:32:17 +02:00
credstoredir = prefixdir / 'lib/credstore'
pcrlock: add pre-defined pcrlock files These cover well-known measurements done by the UEFI firmware or systemd.
2023-08-16 11:40:01 +02:00
pcrlockdir = prefixdir / 'lib/pcrlock.d'
mime: register confext/sysext images in shared-mime-info This make them recognized by file managers and stuff. Maybe one day we should properly register mime types in the "vnd." namespace with IANA, but I am too lazy to deal with the bureaucracy for that, hence let's stick with the x. namespace for now. This defines confext/sysext DDIs as subtype of: https://www.iana.org/assignments/media-types/application/vnd.efi.img Which is what everyone appears to use for raw disk images, in particular if they contain a GPT partition table.
2023-11-20 11:51:47 +01:00
mimepackagesdir = prefixdir / 'share/mime/packages'
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: add build option for install path of main config files This allows distros to install configuration file templates in /usr/lib/systemd for example. Currently we install "empty" config files in /etc/systemd/. They serve two purposes: - The file contains commented-out values that show the default settings. - It is easier to edit the right file if it is already there, the user doesn't have to type in the path correctly, and the basic file structure is already in place so it's easier to edit. Things that have happened since this approach was put in place: - We started supporting drop-ins for config files, and drop-ins are the recommended way to create local configuration overrides. - We have systemd-analyze cat-config which takes care of iterating over all possible locations (/etc, /run, /usr, /usr/local) and figuring out the right file. - Because of the first two points, systemd-analyze cat-config is much better, because it takes care of finding all the drop-ins and figuring out the precedence. Looking at files manually is still possible of course, but not very convenient. The disadvantages of the current approach with "empty" files in /etc: - We clutter up /etc so it's harder to see what the local configuration actually is. - If a user edits the file, package updates will not override the file (e.g. systemd.rpm uses %config(noreplace). This means that the "documented defaults" will become stale over time, if the user ever edits the main config file. Thus, I think that it's reasonable to: - Install the main config file to /usr/lib so that it serves as reference for syntax and option names and default values and is properly updated on package upgrades. - Recommend to users to always use drop-ins for configuration and systemd-analyze cat-config to view the documentation. This setting makes this change opt-in. Fixes #18420. [zjs: add more text to the description]
2023-08-21 12:37:00 +02:00
configfiledir = get_option('configfiledir')
if configfiledir == ''
shell: define three system credentials we can propagate into shell prompts and welcome messages
2024-09-06 16:30:54 +02:00
configfiledir = sysconfdir
meson: add build option for install path of main config files This allows distros to install configuration file templates in /usr/lib/systemd for example. Currently we install "empty" config files in /etc/systemd/. They serve two purposes: - The file contains commented-out values that show the default settings. - It is easier to edit the right file if it is already there, the user doesn't have to type in the path correctly, and the basic file structure is already in place so it's easier to edit. Things that have happened since this approach was put in place: - We started supporting drop-ins for config files, and drop-ins are the recommended way to create local configuration overrides. - We have systemd-analyze cat-config which takes care of iterating over all possible locations (/etc, /run, /usr, /usr/local) and figuring out the right file. - Because of the first two points, systemd-analyze cat-config is much better, because it takes care of finding all the drop-ins and figuring out the precedence. Looking at files manually is still possible of course, but not very convenient. The disadvantages of the current approach with "empty" files in /etc: - We clutter up /etc so it's harder to see what the local configuration actually is. - If a user edits the file, package updates will not override the file (e.g. systemd.rpm uses %config(noreplace). This means that the "documented defaults" will become stale over time, if the user ever edits the main config file. Thus, I think that it's reasonable to: - Install the main config file to /usr/lib so that it serves as reference for syntax and option names and default values and is properly updated on package upgrades. - Recommend to users to always use drop-ins for configuration and systemd-analyze cat-config to view the documentation. This setting makes this change opt-in. Fixes #18420. [zjs: add more text to the description]
2023-08-21 12:37:00 +02:00
endif
pkgconfigfiledir = configfiledir / 'systemd'
meson.build: make docdir configurable (#8068) SUSE is using a different docdir, so the path should be configurable
2018-02-01 22:46:15 +01:00
docdir = get_option('docdir')
if docdir == ''
meson: use a/b instead of join_paths(a,b) It is nicer and shorter.
2021-07-27 19:32:35 +02:00
docdir = datadir / 'doc/systemd'
meson.build: make docdir configurable (#8068) SUSE is using a different docdir, so the path should be configurable
2018-02-01 22:46:15 +01:00
endif
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
pamlibdir = get_option('pamlibdir')
if pamlibdir == ''
Drop split-usr and unmerged-usr support As previously announced, execute order 66: https://lists.freedesktop.org/archives/systemd-devel/2022-September/048352.html The meson options split-usr, rootlibdir and rootprefix become no-ops that print a warning if they are set to anything other than the default values. We can remove them in a future release.
2023-06-12 02:15:19 +01:00
pamlibdir = libdir / 'security'
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
endif
pamconfdir = get_option('pamconfdir')
if pamconfdir == ''
meson: use a/b instead of join_paths(a,b) It is nicer and shorter.
2021-07-27 19:32:35 +02:00
pamconfdir = prefixdir / 'lib/pam.d'
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
endif
ssh-proxy: add ssh ProxyCommand tool that can connect to AF_UNIX + AF_VSOCK sockets This adds a tiny binary that is hooked into SSH client config via ProxyCommand and which simply connects to an AF_UNIX or AF_VSOCK socket of choice. The syntax is as simple as this: ssh unix/some/path # (this connects to AF_UNIX socket /some/path) or: ssh vsock/4711 I used "/" as separator of the protocol ID and the value since ":" is already taken by SSH itself when doing sftp. And "@" is already taken for separating the user name.
2024-01-04 23:31:51 +01:00
sshconfdir = get_option('sshconfdir')
if sshconfdir == ''
sshconfdir = sysconfdir / 'ssh/ssh_config.d'
endif
meson: don't put a symlink pointing to '20-systemd-ssh-proxy.conf' in /etc in all cases On distros like SUSE where ssh config dropins in /usr are supported, there's no need for a symlink in /etc/ssh/ssh_config.d/ that points to the dropin installed somewhere in /usr (that is not reachable by ssh).
2024-05-24 14:11:54 +02:00
conf.set10('LINK_SSH_PROXY_DROPIN', sshconfdir != 'no' and not sshconfdir.startswith('/usr/'))
ssh-proxy: add ssh ProxyCommand tool that can connect to AF_UNIX + AF_VSOCK sockets This adds a tiny binary that is hooked into SSH client config via ProxyCommand and which simply connects to an AF_UNIX or AF_VSOCK socket of choice. The syntax is as simple as this: ssh unix/some/path # (this connects to AF_UNIX socket /some/path) or: ssh vsock/4711 I used "/" as separator of the protocol ID and the value since ":" is already taken by SSH itself when doing sftp. And "@" is already taken for separating the user name.
2024-01-04 23:31:51 +01:00
userdbctl: enable ssh-authorized-keys logic by default sshd now supports config file drop-ins, hence let's install one to hook up "userdb ssh-authorized-keys", so that things just work. We put the drop-in relatively early, so that other drop-ins generally will override this. Ideally sshd would support such drop-ins in /usr/ rather than /etc/, but let's take what we can get. It's not that sshd's upstream was particularly open to weird ideas from Linux people.
2023-11-16 18:47:52 +01:00
sshdconfdir = get_option('sshdconfdir')
if sshdconfdir == ''
sshdconfdir = sysconfdir / 'ssh/sshd_config.d'
endif
meson: don't put a symlink pointing to '20-systemd-userdb.conf' in /etc in all cases It's only needed on distros where sshd doesn't support drop-ins in /usr, which is not the case on SUSE.
2024-05-24 15:38:13 +02:00
conf.set10('LINK_SSHD_USERDB_DROPIN', sshdconfdir != 'no' and not sshdconfdir.startswith('/usr/'))
userdbctl: enable ssh-authorized-keys logic by default sshd now supports config file drop-ins, hence let's install one to hook up "userdb ssh-authorized-keys", so that things just work. We put the drop-in relatively early, so that other drop-ins generally will override this. Ideally sshd would support such drop-ins in /usr/ rather than /etc/, but let's take what we can get. It's not that sshd's upstream was particularly open to weird ideas from Linux people.
2023-11-16 18:47:52 +01:00
ssh-generator: create privsep dir via tmpfiles.d/ if we are told to To make it easy to have a workable ssh-generator on various distros, let's optionally generate the ssh privsep dir via tmpfiles.d/ drop-in. This enables the concept with a path of /run/sshd/ as default. This is the path Debian/Ubuntu uses, and means that we just work on those distros. Debian/Ubuntu is the only distro (apparently?) that puts the privsep dir under /run/, hence always needs the dir to be created manually. Other distros don't need it that much, because they place the dir in /usr/ (fedora, best choice!) or /var/ (others, not ideal, because still mutable). Also adds a longer explanation about this in NEWS, in the hope that distro maintaines read that and maybe start cleaning this up. Alternative to: #31543
2024-03-01 11:25:52 +01:00
sshdprivsepdir = get_option('sshdprivsepdir')
conf.set10('CREATE_SSHDPRIVSEPDIR', sshdprivsepdir != 'no' and not sshdprivsepdir.startswith('/usr/'))
conf.set('SSHDPRIVSEPDIR', sshdprivsepdir, description : 'SSH privilege separation directory')
Add support for systemd-tpm2 libcryptsetup plugin. Add support for systemd-tpm2 based LUKS2 device activation via libcryptsetup plugin. This make the feature (tpm2 sealed LUKS2 keyslot passphrase) usable from both systemd utilities and cryptsetup cli. The feature is configured via -Dlibcryptsetup-plugins combo with default value set to 'auto'. It get's enabled automatically when cryptsetup 2.4.0 or later is installed in build system.
2021-03-16 20:13:28 +01:00
libcryptsetup_plugins_dir = get_option('libcryptsetup-plugins-dir')
if libcryptsetup_plugins_dir == ''
Drop split-usr and unmerged-usr support As previously announced, execute order 66: https://lists.freedesktop.org/archives/systemd-devel/2022-September/048352.html The meson options split-usr, rootlibdir and rootprefix become no-ops that print a warning if they are set to anything other than the default values. We can remove them in a future release.
2023-06-12 02:15:19 +01:00
libcryptsetup_plugins_dir = libdir / 'cryptsetup'
Add support for systemd-tpm2 libcryptsetup plugin. Add support for systemd-tpm2 based LUKS2 device activation via libcryptsetup plugin. This make the feature (tpm2 sealed LUKS2 keyslot passphrase) usable from both systemd utilities and cryptsetup cli. The feature is configured via -Dlibcryptsetup-plugins combo with default value set to 'auto'. It get's enabled automatically when cryptsetup 2.4.0 or later is installed in build system.
2021-03-16 20:13:28 +01:00
endif
shell: define three system credentials we can propagate into shell prompts and welcome messages
2024-09-06 16:30:54 +02:00
shellprofiledir = get_option('shellprofiledir')
if shellprofiledir == ''
shellprofiledir = sysconfdir / 'profile.d'
endif
conf.set10('LINK_SHELL_EXTRA_DROPIN', shellprofiledir != 'no' and not shellprofiledir.startswith('/usr/'))
profile: generate shell + command OSC events
2025-01-30 15:09:21 +01:00
conf.set10('LINK_OSC_CONTEXT_DROPIN', shellprofiledir != 'no' and not shellprofiledir.startswith('/usr/'))
shell: define three system credentials we can propagate into shell prompts and welcome messages
2024-09-06 16:30:54 +02:00
conf.set('SHELLPROFILEDIR', shellprofiledir, description : 'shell profile directory')
meson: add -Dmemory-accounting-default=true|false This makes it easy to set the default for distributions and users which want to default to off because they primarily use older kernels.
2018-02-15 11:43:08 +01:00
memory_accounting_default = get_option('memory-accounting-default')
Add config and kernel commandline option to use short identifiers No functional change, just docs and configuration and parsing. v2: - change ShortIdentifiers=yes|no to StatusUnitFormat=name|description.
2019-06-06 19:22:20 +02:00
status_unit_format_default = get_option('status-unit-format-default')
meson: use status-unit-format-default=name in developer mode I think developers are particularly unlikely to find the descriptions useful, and would benefit from being able to copy&paste unit names. Let's make this choice automatically.
2022-06-09 20:33:29 +02:00
if status_unit_format_default == 'auto'
status_unit_format_default = conf.get('BUILD_MODE_DEVELOPER') == 1 ? 'name' : 'description'
endif
meson: add -Dmemory-accounting-default=true|false This makes it easy to set the default for distributions and users which want to default to off because they primarily use older kernels.
2018-02-15 11:43:08 +01:00
Drop split-usr and unmerged-usr support As previously announced, execute order 66: https://lists.freedesktop.org/archives/systemd-devel/2022-September/048352.html The meson options split-usr, rootlibdir and rootprefix become no-ops that print a warning if they are set to anything other than the default values. We can remove them in a future release.
2023-06-12 02:15:19 +01:00
conf.set_quoted('BINDIR', bindir)
meson: sort conf paths alphabetically The order was a complete mess. Let's make it a bit more tidy.
2021-05-16 19:34:52 +02:00
conf.set_quoted('BINFMT_DIR', binfmtdir)
conf.set_quoted('BOOTLIBDIR', bootlibdir)
meson: use a/b instead of join_paths(a,b) It is nicer and shorter.
2021-07-27 19:32:35 +02:00
conf.set_quoted('CATALOG_DATABASE', catalogstatedir / 'database')
meson: sort conf paths alphabetically The order was a complete mess. Let's make it a bit more tidy.
2021-05-16 19:34:52 +02:00
conf.set_quoted('CERTIFICATE_ROOT', get_option('certificate-root'))
docs: move /var/log/README to a tmpfiles.d symlink This moves the /var/log/README content out of /var and into the docs location, replacing the previous file with a symlink created through a tmpfiles.d entry.
2021-07-08 09:47:32 +00:00
conf.set_quoted('DOC_DIR', docdir)
meson: use a/b instead of join_paths(a,b) It is nicer and shorter.
2021-07-27 19:32:35 +02:00
conf.set_quoted('DOCUMENT_ROOT', pkgdatadir / 'gatewayd')
meson: sort conf paths alphabetically The order was a complete mess. Let's make it a bit more tidy.
2021-05-16 19:34:52 +02:00
conf.set_quoted('ENVIRONMENT_DIR', environmentdir)
conf.set_quoted('INCLUDE_DIR', includedir)
conf.set_quoted('LIBDIR', libdir)
Drop split-usr and unmerged-usr support As previously announced, execute order 66: https://lists.freedesktop.org/archives/systemd-devel/2022-September/048352.html The meson options split-usr, rootlibdir and rootprefix become no-ops that print a warning if they are set to anything other than the default values. We can remove them in a future release.
2023-06-12 02:15:19 +01:00
conf.set_quoted('LIBEXECDIR', libexecdir)
rpm/macros: add %_kernel_install_dir This makes it easier for people packaging kernel-install plugins to get the path right. E.g. https://src.fedoraproject.org/rpms/python-virt-firmware/pull-request/3 fixes an issue where %{_libdir}/kernel/install.d was used, which gives incorrect results on 64-bit architectures. %_kernel_install_dir will make this even easier.
2024-02-29 21:38:03 +01:00
conf.set_quoted('KERNEL_INSTALL_DIR', kernelinstalldir)
meson: sort conf paths alphabetically The order was a complete mess. Let's make it a bit more tidy.
2021-05-16 19:34:52 +02:00
conf.set_quoted('MODPROBE_DIR', modprobedir)
conf.set_quoted('MODULESLOAD_DIR', modulesloaddir)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
conf.set_quoted('PKGSYSCONFDIR', pkgsysconfdir)
meson: use a/b instead of join_paths(a,b) It is nicer and shorter.
2021-07-27 19:32:35 +02:00
conf.set_quoted('POLKIT_AGENT_BINARY_PATH', bindir / 'pkttyagent')
login: add polkit example rules for allowing root to ignore inhibitors The semantics of strong inhibitors require that POLKIT_ALWAYS_QUERY always be set when checking if we can allow blocking inhibitors to be ignored on shutdown, reboot, etc. With the default polkit rules and policy, users may experience a situation where users in the sudo group are authorized to run: systemctl reboot --check-inhibitors=no but the root user is not authorized. Instead, the following error is given: Call to Reboot failed: Interactive authentication required. While this is correct according to the semantics of strong inhibitors, it is confusing. To help the situation, provide example polkit rules that allow root to perform these actions. Finally, when root receives SD_BUS_ERROR_INTERACTIVE_AUTHORIZATION_REQUIRED when calling e.g. systemctl reboot, print a message explaining that this is due to the current polkit policy, and point to the new example rule. Related: https://github.com/systemd/systemd/issues/36786
2025-03-21 15:14:20 -04:00
conf.set_quoted('POLKIT_RULES_DIR', polkitrulesdir)
meson: sort conf paths alphabetically The order was a complete mess. Let's make it a bit more tidy.
2021-05-16 19:34:52 +02:00
conf.set_quoted('PREFIX', prefixdir)
Drop split-usr and unmerged-usr support As previously announced, execute order 66: https://lists.freedesktop.org/archives/systemd-devel/2022-September/048352.html The meson options split-usr, rootlibdir and rootprefix become no-ops that print a warning if they are set to anything other than the default values. We can remove them in a future release.
2023-06-12 02:15:19 +01:00
conf.set_quoted('PREFIX_NOSLASH', prefixdir_noslash)
meson: use a/b instead of join_paths(a,b) It is nicer and shorter.
2021-07-27 19:32:35 +02:00
conf.set_quoted('RANDOM_SEED', randomseeddir / 'random-seed')
meson: sort conf paths alphabetically The order was a complete mess. Let's make it a bit more tidy.
2021-05-16 19:34:52 +02:00
conf.set_quoted('RANDOM_SEED_DIR', randomseeddir)
ssh-proxy: add ssh ProxyCommand tool that can connect to AF_UNIX + AF_VSOCK sockets This adds a tiny binary that is hooked into SSH client config via ProxyCommand and which simply connects to an AF_UNIX or AF_VSOCK socket of choice. The syntax is as simple as this: ssh unix/some/path # (this connects to AF_UNIX socket /some/path) or: ssh vsock/4711 I used "/" as separator of the protocol ID and the value since ":" is already taken by SSH itself when doing sftp. And "@" is already taken for separating the user name.
2024-01-04 23:31:51 +01:00
conf.set_quoted('SSHCONFDIR', sshconfdir)
userdbctl: enable ssh-authorized-keys logic by default sshd now supports config file drop-ins, hence let's install one to hook up "userdb ssh-authorized-keys", so that things just work. We put the drop-in relatively early, so that other drop-ins generally will override this. Ideally sshd would support such drop-ins in /usr/ rather than /etc/, but let's take what we can get. It's not that sshd's upstream was particularly open to weird ideas from Linux people.
2023-11-16 18:47:52 +01:00
conf.set_quoted('SSHDCONFDIR', sshdconfdir)
shell: define three system credentials we can propagate into shell prompts and welcome messages
2024-09-06 16:30:54 +02:00
conf.set_quoted('SHELLPROFILEDIR', shellprofiledir)
meson: sort conf paths alphabetically The order was a complete mess. Let's make it a bit more tidy.
2021-05-16 19:34:52 +02:00
conf.set_quoted('SYSCONF_DIR', sysconfdir)
conf.set_quoted('SYSCTL_DIR', sysctldir)
Drop split-usr and unmerged-usr support As previously announced, execute order 66: https://lists.freedesktop.org/archives/systemd-devel/2022-September/048352.html The meson options split-usr, rootlibdir and rootprefix become no-ops that print a warning if they are set to anything other than the default values. We can remove them in a future release.
2023-06-12 02:15:19 +01:00
conf.set_quoted('SYSTEMCTL_BINARY_PATH', bindir / 'systemctl')
conf.set_quoted('SYSTEMD_BINARY_PATH', libexecdir / 'systemd')
core: add systemd-executor binary Currently we spawn services by forking a child process, doing a bunch of work, and then exec'ing the service executable. There are some advantages to this approach: - quick: we immediately have access to all the enourmous amount of state simply by virtue of sharing the memory with the parent - easy to refactor and add features - part of the same binary, will never be out of sync There are however significant drawbacks: - doing work after fork and before exec is against glibc's supported case for several APIs we call - copy-on-write trap: anytime any memory is touched in either parent or child, a copy of that page will be triggered - memory footprint of the child process will be memory footprint of PID1, but using the cgroup memory limits of the unit The last issue is especially problematic on resource constrained systems where hard memory caps are enforced and swap is not allowed. As soon as PID1 is under load, with no page out due to no swap, and a service with a low MemoryMax= tries to start, hilarity ensues. Add a new systemd-executor binary, that is able to receive all the required state via memfd, deserialize it, prepare the appropriate data structures and call exec_child. Use posix_spawn which uses CLONE_VM + CLONE_VFORK, to ensure there is no copy-on-write (same address space will be used, and parent process will be frozen, until exec). The sd-executor binary is pinned by FD on startup, so that we can guarantee there will be no incompatibilities during upgrades.
2023-06-01 19:51:42 +01:00
conf.set_quoted('SYSTEMD_EXECUTOR_BINARY_PATH', libexecdir / 'systemd-executor')
meson: sort conf paths alphabetically The order was a complete mess. Let's make it a bit more tidy.
2021-05-16 19:34:52 +02:00
conf.set_quoted('SYSTEMD_CATALOG_DIR', catalogdir)
meson: move systemd-cryptsetup to /usr/bin This was requested, though I think an issue was never filed. If people are supposed to invoke it, even for testing, then it's reasonable to make it "public".
2023-09-22 18:47:05 +02:00
conf.set_quoted('SYSTEMD_CRYPTSETUP_PATH', bindir / 'systemd-cryptsetup')
Drop split-usr and unmerged-usr support As previously announced, execute order 66: https://lists.freedesktop.org/archives/systemd-devel/2022-September/048352.html The meson options split-usr, rootlibdir and rootprefix become no-ops that print a warning if they are set to anything other than the default values. We can remove them in a future release.
2023-06-12 02:15:19 +01:00
conf.set_quoted('SYSTEMD_EXPORT_PATH', libexecdir / 'systemd-export')
conf.set_quoted('SYSTEMD_FSCK_PATH', libexecdir / 'systemd-fsck')
conf.set_quoted('SYSTEMD_GROWFS_PATH', libexecdir / 'systemd-growfs')
conf.set_quoted('SYSTEMD_HOMEWORK_PATH', libexecdir / 'systemd-homework')
conf.set_quoted('SYSTEMD_IMPORT_FS_PATH', libexecdir / 'systemd-import-fs')
conf.set_quoted('SYSTEMD_IMPORT_PATH', libexecdir / 'systemd-import')
conf.set_quoted('SYSTEMD_INTEGRITYSETUP_PATH', libexecdir / 'systemd-integritysetup')
meson: use a/b instead of join_paths(a,b) It is nicer and shorter.
2021-07-27 19:32:35 +02:00
conf.set_quoted('SYSTEMD_KBD_MODEL_MAP', pkgdatadir / 'kbd-model-map')
conf.set_quoted('SYSTEMD_LANGUAGE_FALLBACK_MAP', pkgdatadir / 'language-fallback-map')
Drop split-usr and unmerged-usr support As previously announced, execute order 66: https://lists.freedesktop.org/archives/systemd-devel/2022-September/048352.html The meson options split-usr, rootlibdir and rootprefix become no-ops that print a warning if they are set to anything other than the default values. We can remove them in a future release.
2023-06-12 02:15:19 +01:00
conf.set_quoted('SYSTEMD_MAKEFS_PATH', libexecdir / 'systemd-makefs')
conf.set_quoted('SYSTEMD_PULL_PATH', libexecdir / 'systemd-pull')
conf.set_quoted('SYSTEMD_SHUTDOWN_BINARY_PATH', libexecdir / 'systemd-shutdown')
sysupdate: Implement systemd-sysupdated dbus service Co-authored-by: Tom Coldrick <thomas.coldrick@codethink.co.uk> Co-authored-by: Abderrahim Kitouni <abderrahim.kitouni@codethink.co.uk>
2023-06-29 22:58:41 -04:00
conf.set_quoted('SYSTEMD_SYSUPDATE_PATH', libexecdir / 'systemd-sysupdate')
meson: define testdata_dir globally
2023-03-17 14:14:16 +01:00
conf.set_quoted('SYSTEMD_TEST_DATA', testdata_dir)
Drop split-usr and unmerged-usr support As previously announced, execute order 66: https://lists.freedesktop.org/archives/systemd-devel/2022-September/048352.html The meson options split-usr, rootlibdir and rootprefix become no-ops that print a warning if they are set to anything other than the default values. We can remove them in a future release.
2023-06-12 02:15:19 +01:00
conf.set_quoted('SYSTEMD_TTY_ASK_PASSWORD_AGENT_BINARY_PATH', bindir / 'systemd-tty-ask-password-agent')
conf.set_quoted('SYSTEMD_UPDATE_HELPER_PATH', libexecdir / 'systemd-update-helper')
conf.set_quoted('SYSTEMD_USERWORK_PATH', libexecdir / 'systemd-userwork')
mountfsd: add new systemd-mountfsd component
2023-03-09 12:27:29 +01:00
conf.set_quoted('SYSTEMD_MOUNTWORK_PATH', libexecdir / 'systemd-mountwork')
nsresourced: add new daemon for granting clients user namespaces and assigning resources to them This adds a small, socket-activated Varlink daemon that can delegate UID ranges for user namespaces to clients asking for it. The primary call is AllocateUserRange() where the user passes in an uninitialized userns fd, which is then set up. There are other calls that allow assigning a mount fd to a userns allocated that way, to set up permissions for a cgroup subtree, and to allocate a veth for such a user namespace. Since the UID assignments are supposed to be transitive, i.e. not permanent, care is taken to ensure that users cannot create inodes owned by these UIDs, so that persistancy cannot be acquired. This is implemented via a BPF-LSM module that ensures that any member of a userns allocated that way cannot create files unless the mount it operates on is owned by the userns itself, or is explicitly allowelisted. BPF LSM program with contributions from Alexei Starovoitov.
2023-04-20 19:07:33 +02:00
conf.set_quoted('SYSTEMD_NSRESOURCEWORK_PATH', libexecdir / 'systemd-nsresourcework')
Drop split-usr and unmerged-usr support As previously announced, execute order 66: https://lists.freedesktop.org/archives/systemd-devel/2022-September/048352.html The meson options split-usr, rootlibdir and rootprefix become no-ops that print a warning if they are set to anything other than the default values. We can remove them in a future release.
2023-06-12 02:15:19 +01:00
conf.set_quoted('SYSTEMD_VERITYSETUP_PATH', libexecdir / 'systemd-veritysetup')
meson: use a/b instead of join_paths(a,b) It is nicer and shorter.
2021-07-27 19:32:35 +02:00
conf.set_quoted('SYSTEM_CONFIG_UNIT_DIR', pkgsysconfdir / 'system')
meson: sort conf paths alphabetically The order was a complete mess. Let's make it a bit more tidy.
2021-05-16 19:34:52 +02:00
conf.set_quoted('SYSTEM_DATA_UNIT_DIR', systemunitdir)
Rename _PATH variables to _DIR when they refer to a directory We were very inconsistent in this, but in general _PATH signifies a search path (separated with :), and _DIR signifies a single directory.
2020-03-24 16:28:26 +01:00
conf.set_quoted('SYSTEM_ENV_GENERATOR_DIR', systemenvgeneratordir)
meson: sort conf paths alphabetically The order was a complete mess. Let's make it a bit more tidy.
2021-05-16 19:34:52 +02:00
conf.set_quoted('SYSTEM_GENERATOR_DIR', systemgeneratordir)
conf.set_quoted('SYSTEM_PRESET_DIR', systempresetdir)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
conf.set_quoted('SYSTEM_SHUTDOWN_PATH', systemshutdowndir)
conf.set_quoted('SYSTEM_SLEEP_PATH', systemsleepdir)
meson: sort conf paths alphabetically The order was a complete mess. Let's make it a bit more tidy.
2021-05-16 19:34:52 +02:00
conf.set_quoted('SYSTEM_SYSVINIT_PATH', sysvinit_path)
conf.set_quoted('SYSTEM_SYSVRCND_PATH', sysvrcnd_path)
meson: rename RC_LOCAL_PATH -> SYSTEM_SYSVRCLOCAL_PATH No functional change, but just for emphasizing that this is for SysV compatibility.
2025-04-02 22:10:29 +09:00
conf.set_quoted('SYSTEM_SYSVRCLOCAL_PATH', sysvrclocal_path)
meson: use jinja2 for rpm templates The naming of variables is very inconsistent. I tried to use more modern style naming (UNDERSCORED_TITLE_CASE), but I didn't change existing names too much. Only SYSTEM_DATA_UNIT_PATH is renamed to SYSTEM_DATA_UNIT_DIR to match SYSTEM_CONFIG_UNIT_DIR.
2021-05-16 16:44:47 +02:00
conf.set_quoted('SYSUSERS_DIR', sysusersdir)
conf.set_quoted('TMPFILES_DIR', tmpfilesdir)
pkgconfig,rpm: expose vars for user-tmpfiles.d location Fixes https://bugzilla.redhat.com/show_bug.cgi?id=2098553.
2022-06-20 15:06:09 +02:00
conf.set_quoted('USER_TMPFILES_DIR', usertmpfilesdir)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
conf.set_quoted('UDEVLIBEXECDIR', udevlibexecdir)
meson: sort conf paths alphabetically The order was a complete mess. Let's make it a bit more tidy.
2021-05-16 19:34:52 +02:00
conf.set_quoted('UDEV_HWDB_DIR', udevhwdbdir)
conf.set_quoted('UDEV_RULES_DIR', udevrulesdir)
meson: use a/b instead of join_paths(a,b) It is nicer and shorter.
2021-07-27 19:32:35 +02:00
conf.set_quoted('USER_CONFIG_UNIT_DIR', pkgsysconfdir / 'user')
meson: sort conf paths alphabetically The order was a complete mess. Let's make it a bit more tidy.
2021-05-16 19:34:52 +02:00
conf.set_quoted('USER_DATA_UNIT_DIR', userunitdir)
conf.set_quoted('USER_ENV_GENERATOR_DIR', userenvgeneratordir)
conf.set_quoted('USER_GENERATOR_DIR', usergeneratordir)
import: rename keyring extension from .gpg to .pgp The format is not specific to gpg, it's defined by the pgp protocols. Lintian even warns about it: systemd-container: openpgp-file-has-implementation-specific-extension [usr/lib/systemd/import-pubring.gpg] Rename it accordingly, as new pgp implementations are slowly becoming more popular. Keep using /etc/systemd/import-pubring.gpg for backward compatibility.
2025-06-22 17:43:35 +01:00
conf.set_quoted('USER_KEYRING_PATH', pkgsysconfdir / 'import-pubring.pgp')
conf.set_quoted('USER_KEYRING_PATH_LEGACY', pkgsysconfdir / 'import-pubring.gpg')
meson: sort conf paths alphabetically The order was a complete mess. Let's make it a bit more tidy.
2021-05-16 19:34:52 +02:00
conf.set_quoted('USER_PRESET_DIR', userpresetdir)
import: rename keyring extension from .gpg to .pgp The format is not specific to gpg, it's defined by the pgp protocols. Lintian even warns about it: systemd-container: openpgp-file-has-implementation-specific-extension [usr/lib/systemd/import-pubring.gpg] Rename it accordingly, as new pgp implementations are slowly becoming more popular. Keep using /etc/systemd/import-pubring.gpg for backward compatibility.
2025-06-22 17:43:35 +01:00
conf.set_quoted('VENDOR_KEYRING_PATH', libexecdir / 'import-pubring.pgp')
meson: sort conf paths alphabetically The order was a complete mess. Let's make it a bit more tidy.
2021-05-16 19:34:52 +02:00
conf.set('ANSI_OK_COLOR', 'ANSI_' + get_option('ok-color').underscorify().to_upper())
Add meson option to disable urlify. Useful for systems that don't use a version of less with hyperlink support.
2021-07-11 04:39:33 -06:00
conf.set10('ENABLE_URLIFY', get_option('urlify'))
meson: sort conf paths alphabetically The order was a complete mess. Let's make it a bit more tidy.
2021-05-16 19:34:52 +02:00
conf.set10('ENABLE_FEXECVE', get_option('fexecve'))
meson: simplify definition of MEMORY_ACCOUNTING_DEFAULT Let's just use the simplest form, it doesn't really matter how the define looks after preprocessing.
2018-10-17 14:36:09 +02:00
conf.set10('MEMORY_ACCOUNTING_DEFAULT', memory_accounting_default)
Add config and kernel commandline option to use short identifiers No functional change, just docs and configuration and parsing. v2: - change ShortIdentifiers=yes|no to StatusUnitFormat=name|description.
2019-06-06 19:22:20 +02:00
conf.set('STATUS_UNIT_FORMAT_DEFAULT', 'STATUS_UNIT_FORMAT_' + status_unit_format_default.to_upper())
meson: use jinja2 for various files in src/core/
2021-05-16 17:52:42 +02:00
conf.set_quoted('STATUS_UNIT_FORMAT_DEFAULT_STR', status_unit_format_default)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
core: split system/user job timeouts and make them configurable Config options are -Ddefault-timeout-sec= and -Ddefault-user-timeout-sec=. Existing -Dupdate-helper-user-timeout= is renamed to -Dupdate-helper-user-timeout-sec= for consistency. All three options take an integer value in seconds. The renaming and type-change of the option is a small compat break, but it's just at compile time and result in a clear error message. I also doubt that anyone was actually using the option. This commit separates the user manager timeouts, but keeps them unchanged at 90 s. The timeout for the user manager is set to 4/3*user-timeout, which means that it is still 120 s. Fedora wants to experiment with lower timeouts, but doing this via a patch would be annoying and more work than necessary. Let's make this easy to configure.
2021-01-26 20:23:12 +01:00
conf.set('DEFAULT_TIMEOUT_SEC', get_option('default-timeout-sec'))
conf.set('DEFAULT_USER_TIMEOUT_SEC', get_option('default-user-timeout-sec'))
conf.set('UPDATE_HELPER_USER_TIMEOUT_SEC', get_option('update-helper-user-timeout-sec'))
meson: fix printing of first-boot-full-preset The meson summary logic checks for ENABLE_* and HAVE_*, but we used a define with no prefix. Let's make it ENABLE_… for consistency with other config options. Obviously this also fixes the summary output.
2023-11-08 08:31:13 +01:00
conf.set10('ENABLE_FIRST_BOOT_FULL_PRESET', get_option('first-boot-full-preset'))
manager: optionally, do a full preset on first boot A compile time option is added to select behaviour: by default UNIT_FILE_PRESET_ENABLE_ONLY is still used, but the intent is to change to UNIT_FILE_PRESET_FULL at some point in the future. Distros that want to opt-in can use the config option to change the behaviour. (The option is just a boolean: it would be possible to make it multi-valued, and allow full, enable-only, disable-only, none. But so far nobody has asked for this, and it's better not to complicate things needlessly.) With the configuration option flipped, instead of only doing enablements, perform a full preset on first boot. The reason is that although `/etc/machine-id` might be missing, there may be other files provisioned in `/etc` (in fact, this use case is mentioned in `log_execution_mode`). Some of those possible files include enablement symlinks even if presets dictate it should be disabled. Such a seemingly contradictory situation occurs in {RHEL,Fedora} CoreOS, where we ship `/etc` as if `preset-all` were called. However, we want to allow users to disable default-enabled services via Ignition, which does this by creating preset dropins before switchroot. (For why we do `preset-all` at compose time, see: https://github.com/coreos/fedora-coreos-config/pull/77). For example, the composed FCOS image has a `enable zincati.service` preset and an enablement for that in `/etc`, while at boot time when we switch root, there may be a `disable zincati.service` preset with higher precedence. In that case, we want systemd to disable the service. This is essentially a revert of 304b3079a203. It seems like systemd *used* to do this, but it was changed to try to make the container workflow a bit faster. Resolves: https://github.com/coreos/fedora-coreos-tracker/issues/392 Co-authored-by: Zbigniew Jędrzejewski-Szmek <zbyszek@in.waw.pl>
2020-03-23 12:25:19 -04:00
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
#####################################################################
cc = meson.get_compiler('c')
meson: Introduce userspace dep This will help in a later commit to separate userspace from EFI builds.
2023-02-26 11:07:24 +01:00
userspace_c_args = []
userspace_c_ld_args = []
meson: fix coverage build This moves coverage.h to src/coverage/, and specifies path to coverage.h with files() directive, to make it can be included even when located outside of the include directories. Otherwise, libc-wrapper cannot be built when -Db_coverage=true option is enabled. Follow-up for 543a48b653ff25cc562ce4dcffec377a875f604e.
2025-07-12 02:14:00 +09:00
userspace_sources = []
meson: use files() not find_program() for helper scripts We went back and forth between 'prog.sh', files('prog.sh'), and find_program('prog.sh'). We want to use files() or find_program() so that we get a good error message if the file is missing. Behaviour of meson changed over time, and in the past not all forms could be used in all places. For example 0f4c4f3824891aea6a76b2e861f38a6045b2a246 added find_program() in many places to avoid repeated messages. But it seems that all recent meson versions work fine with files(). find_program prints silly messages: Program tools/make-man-index.py found: YES (/home/zbyszek/src/systemd/tools/make-man-index.py) Program tools/meson-render-jinja2.py found: YES (/home/zbyszek/src/systemd/tools/meson-render-jinja2.py) ... We know that those files will be found, they are part of the git checkout. With files() this is gone and the meson output is easier to read.
2025-07-25 12:18:19 +02:00
meson_build_sh = files('tools/meson-build.sh')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: disable "slow tests" too when tests are generally disabled We would have a strange situation where after setting -Dslow-tests=true -Dtests=false we'd get mostly the slow tests (plus some other ones which I'll fix in subsequent commit). Let's simplify things by making -Dtests=false just disable those tests too.
2018-09-12 11:02:58 +02:00
want_tests = get_option('tests')
test: Only set environment variable if integration tests are enabled. If we set it to '0' if integration tests are not enabled then we can't enable them from the command line since environment from meson takes priority over environment variables from the command line. We also rename the related variables to avoid conflicts with the existing integration_tests variable.
2024-05-02 20:21:29 +02:00
want_slow_tests = want_tests != 'false' and get_option('slow-tests')
want_fuzz_tests = want_tests != 'false' and get_option('fuzz-tests')
meson: do not attempt to install tests when they are disabled If -Dtests=false but -Dinstall-tests=true the build will fail, as some tests will be pulled in the build but not their prerequisites. It doesn't make sense to ask for tests to be installed if they are disabled. FAILED: test-acd cc -o test-acd test-acd.p/src_libsystemd-network_test-acd.c.o -flto -Wl,--as-needed -Wl,--no-undefined -pie -fstack-protector -Wl,-z,relro -specs=/usr/share/debhelper/dh_package_notes/debian-package-notes.specs -g -O2 -ffile-prefix-map=/tmp/s=. -fstack-protector-strong -fstack-clash-protection -Wformat -Werror=format-security -fcf-protection -ffat-lto-objects -Wdate-time -D_FORTIFY_SOURCE=2 '-Wl,-rpath,$ORIGIN/src/shared:XXXXXXXXXXXXXXX' -Wl,-rpath-link,/tmp/s/obj-x86_64-linux-gnu/src/shared -Wl,--start-group src/shared/libsystemd-shared-255.so src/libsystemd-network/libsystemd-network.a -Wl,--end-group -Wl,--fatal-warnings -Wl,-z,now -Wl,-z,relro -Wl,--warn-common -Wl,--gc-sections -Wl,--fatal-warnings -Wl,-z,now -Wl,-z,relro -Wl,--warn-common -Wl,--gc-sections /usr/bin/ld: /tmp/cc0oYwFZ.ltrans0.ltrans.o: in function `main': ./obj-x86_64-linux-gnu/./obj-x86_64-linux-gnu/<artificial>:85:(.text.startup+0x33): undefined reference to `test_setup_logging' collect2: error: ld returned 1 exit status
2024-02-13 13:43:49 +00:00
install_tests = want_tests != 'false' and get_option('install-tests')
meson: disable "slow tests" too when tests are generally disabled We would have a strange situation where after setting -Dslow-tests=true -Dtests=false we'd get mostly the slow tests (plus some other ones which I'll fix in subsequent commit). Let's simplify things by making -Dtests=false just disable those tests too.
2018-09-12 11:02:58 +02:00
meson: bump required version to 0.54.0
2022-05-16 03:21:16 +09:00
if add_languages('cpp', native : false, required : fuzzer_build)
meson: always allow compilation of tests binaries
2018-09-12 16:52:08 +02:00
# Used only for tests
meson: apply our -Wno-* options also in c++ calls We compile some c++ code for tests. We would simply use the default options for those. When the previous commit raised the default warning level, we started getting warnings from c++ code. Let's add the most important options to the c++ command, so that we get a compilation without any warnings again. I don't think it makes sense to add *all* the options that we add for c to the c++ flags, because testing them takes quite a while, and the c++ compilations are for small amounts of code, mostly to check that the headers have compatible syntax.
2019-11-10 12:16:41 +01:00
cxx = meson.get_compiler('cpp')
cxx_cmd = ' '.join(cxx.cmd_array())
meson: we cannot call get_compiler('cpp') if we didn't "add" it Follow-up for f6d783ac3d145f3f3ffc2023d4ad90f76cf7fe34.
2018-10-09 18:48:09 +02:00
else
meson: rename cpp_cmd to cxx_cmd cpp is a really bad alias for c++ because it's also the name of the preprocessor. Let's rename the variable.
2018-10-10 11:50:57 +02:00
cxx_cmd = ''
tests,meson: add test-bus-vtable, compiled as C and C++ This test is mostly a compilation test that checks that various defines in sd-bus-vtable.h are valid C++. The code is executed, but the results are not checked (apart from sd-bus functions not returning an error). test-bus-objects contains pretty extensive tests for this functionality. The C++ version is only added to meson, since it's simpler there. Because of the .cc extension, meson will compile the executable with c++. This test is necessary to properly check the macros in sd-bus-vtable.h. Just running the headers through g++ is not enough, because the macros are not exercised. Follow-up for #5941.
2017-05-13 13:23:28 -04:00
endif
fuzz: allow building fuzzers outside of oss-fuzz Add a new -Dllvm-fuzz=true option that can be used to build against libFuzzer and update the oss-fuzz script to work outside of the oss-fuzz build environment.
2018-01-16 10:25:43 -05:00
if want_libfuzzer
meson: Stop doing nested build when fuzzers are enabled Currently, when fuzzers are enabled, we run meson from within meson to build the fuzzer executables with sanitizers. The idea is that we can build the fuzzers with different kinds of sanitizers independently from the main build. The issue with this setup is that we don't actually make use of it. We only build the fuzzers with one set of sanitizers (address,undefined) so we're adding a bunch of extra complexity without any benefit as we can just setup the top level meson build with these sanitizers and get the same result. The other issue with this setup is that we don't pass on all the options passed to the top level meson build to the nested meson build. The only things we pass on are extra compiler arguments and the value of the auto_features option, but none of the individual feature options if overridden are passed on, which can lead to very hard to debug issues as an option enabled in the top level build is not enabled in the nested build. Since we're not getting anything useful out of this setup, let's simplify and get rid of the nested meson build. Instead, sanitizers should be enabled for the top level meson.build. This currently didn't work as we were overriding the sanitizers passed to the meson build with the fuzzer sanitizer, so we fix that as well by making sure we combine the fuzzer sanitizer with the ones passed in by the user. We also drop support for looking up libFuzzer as a separate library as it has been shipped builtin in clang since clang 6.0, so we can assume that -fsanitize=fuzzer is available. To make sure we still run the fuzzing tests, we enable the fuzz-tests option by default now to make sure they still always run (without instrumentation unless one of llvm-fuzz or oss-fuzz is enabled).
2025-05-11 09:42:28 +02:00
if cc.has_argument('-fsanitize=fuzzer-no-link')
meson: Introduce userspace dep This will help in a later commit to separate userspace from EFI builds.
2023-02-26 11:07:24 +01:00
userspace_c_args += '-fsanitize=fuzzer-no-link'
fuzzers: use -fsanitizer=fuzzer if clang supports it Closes https://github.com/systemd/systemd/issues/10645
2019-05-09 00:03:41 +02:00
else
meson: Stop doing nested build when fuzzers are enabled Currently, when fuzzers are enabled, we run meson from within meson to build the fuzzer executables with sanitizers. The idea is that we can build the fuzzers with different kinds of sanitizers independently from the main build. The issue with this setup is that we don't actually make use of it. We only build the fuzzers with one set of sanitizers (address,undefined) so we're adding a bunch of extra complexity without any benefit as we can just setup the top level meson build with these sanitizers and get the same result. The other issue with this setup is that we don't pass on all the options passed to the top level meson build to the nested meson build. The only things we pass on are extra compiler arguments and the value of the auto_features option, but none of the individual feature options if overridden are passed on, which can lead to very hard to debug issues as an option enabled in the top level build is not enabled in the nested build. Since we're not getting anything useful out of this setup, let's simplify and get rid of the nested meson build. Instead, sanitizers should be enabled for the top level meson.build. This currently didn't work as we were overriding the sanitizers passed to the meson build with the fuzzer sanitizer, so we fix that as well by making sure we combine the fuzzer sanitizer with the ones passed in by the user. We also drop support for looking up libFuzzer as a separate library as it has been shipped builtin in clang since clang 6.0, so we can assume that -fsanitize=fuzzer is available. To make sure we still run the fuzzing tests, we enable the fuzz-tests option by default now to make sure they still always run (without instrumentation unless one of llvm-fuzz or oss-fuzz is enabled).
2025-05-11 09:42:28 +02:00
error('Looks like -fsanitize=fuzzer-no-link is not supported')
fuzzers: use -fsanitizer=fuzzer if clang supports it Closes https://github.com/systemd/systemd/issues/10645
2019-05-09 00:03:41 +02:00
endif
meson: c++ is required for the fuzzer builds The configuration would fail with an error about cpp being an unknown language anyway, but it's nicer to fail early and explicitly. https://github.com/systemd/systemd/pull/10339#issuecomment-428279175
2018-10-10 11:56:45 +02:00
elif want_ossfuzz
fuzz: add initial fuzzing infrastructure The fuzzers will be used by oss-fuzz to automatically and continuously fuzz systemd. This commit includes the build tooling necessary to build fuzz targets, and a fuzzer for the DNS packet parser.
2018-01-13 19:51:07 -05:00
fuzzing_engine = meson.get_compiler('cpp').find_library('FuzzingEngine')
endif
meson: apply our -Wno-* options also in c++ calls We compile some c++ code for tests. We would simply use the default options for those. When the previous commit raised the default warning level, we started getting warnings from c++ code. Let's add the most important options to the c++ command, so that we get a compilation without any warnings again. I don't think it makes sense to add *all* the options that we add for c to the c++ flags, because testing them takes quite a while, and the c++ compilations are for small amounts of code, mostly to check that the headers have compatible syntax.
2019-11-10 12:16:41 +01:00
# Those generate many false positives, and we do not want to change the code to
# avoid them.
basic_disabled_warnings = [
'-Wno-missing-field-initializers',
macro: Add DISABLE_WARNING_STRINGOP_OVERREAD While at it, also add -Wunknown-warning-option to basic_disabled_warnings to fix compilation with clang
2024-10-11 16:44:54 +03:00
'-Wno-unknown-warning-option',
meson: sort compiler flags
2021-06-02 15:53:29 +09:00
'-Wno-unused-parameter',
meson: disable -Wnonnull-compare This gets enabled by default in gcc-14 and complains everywhere where we use assert() on an expression that is always true (i.e. using `int x[static 2]` in function declaration, etc.): [153/2414] Compiling C object src/basic/libbasic.a.p/fs-util.c.o In file included from ../src/basic/macro.h:13, from ../src/basic/alloc-util.h:10, from ../src/basic/fs-util.c:11: ../src/basic/fd-util.h: In function ‘format_proc_fd_path’: ../src/fundamental/macro-fundamental.h:74:41: warning: ‘nonnull’ argument ‘buf’ compared to NULL [-Wnonnull-compare] 74 | #define _unlikely_(x) (__builtin_expect(!!(x), 0)) | ^~~~~ ../src/basic/macro.h:150:21: note: in expansion of macro ‘_unlikely_’ 150 | if (_unlikely_(!(expr))) \ | ^~~~~~~~~~ ../src/basic/macro.h:167:22: note: in expansion of macro ‘assert_message_se’ 167 | #define assert(expr) assert_message_se(expr, #expr) | ^~~~~~~~~~~~~~~~~ ../src/basic/fd-util.h:129:9: note: in expansion of macro ‘assert’ 129 | assert(buf); | ^~~~~~ Disabling this selectively only for asserts is a bit painful, since the option is not available in all compilers, and it'd need to be handled in the EFI stuff as well.
2024-01-16 22:25:04 +01:00
'-Wno-nonnull-compare',
meson: apply our -Wno-* options also in c++ calls We compile some c++ code for tests. We would simply use the default options for those. When the previous commit raised the default warning level, we started getting warnings from c++ code. Let's add the most important options to the c++ command, so that we get a compilation without any warnings again. I don't think it makes sense to add *all* the options that we add for c to the c++ flags, because testing them takes quite a while, and the c++ compilations are for small amounts of code, mostly to check that the headers have compatible syntax.
2019-11-10 12:16:41 +01:00
]
meson: enable more warnings when building efi binary
2021-06-02 15:49:44 +09:00
possible_common_cc_flags = [
meson: enable -Warray-bounds and -fstrict-flex-arrays -fstrict-flex-arrays means that the compiler doesn't have to assume that any trailing array is a flex array. I.e. unless the array is declared without a specified size, only indices in the declared range are valid. -Warray-bounds turns on the warnings about out-of-bounds array accesses. -Warray-bounds=2 does some more warnings, with higher false positive rate. But it doesn't seem to yield any false positives in our codebase, so enable it. clang supports -Warray-bounds, but not -Warray-bounds=2. gcc supports both. gcc-13 supports -fstrict-flex-arrays. See https://people.kernel.org/kees/bounded-flexible-arrays-in-c for a long discussion of use in the kernel.
2023-02-05 20:44:19 +01:00
'-Warray-bounds', # clang
'-Warray-bounds=2',
meson: sort compiler flags
2021-06-02 15:53:29 +09:00
'-Wdate-time',
'-Wendif-labels',
meson: handle bool-compare warning as error Then we can detect the error fixed by the previous commit like the following: ../src/libsystemd/sd-device/test-sd-device.c: In function ‘test_sd_device_enumerator_add_all_parents’: ../src/shared/tests.h:225:51: error: comparison of constant ‘0’ with boolean expression is always true [-Werror=bool-compare] 225 | #define ASSERT_OK(expr) __coverity_check__((expr) >= 0) | ^~ ../src/libsystemd/sd-device/test-sd-device.c:547:9: note: in expansion of macro ‘ASSERT_OK’ 547 | ASSERT_OK(sd_device_enumerator_add_all_parents(e) >= 0); | ^~~~~~~~~
2025-03-15 07:21:33 +09:00
'-Werror=bool-compare',
build: fail the build if we accidentally drop a "const" qualifier on a parameter let's be strict here, our codebase allows it.
2025-01-17 14:47:34 +01:00
'-Werror=discarded-qualifiers',
meson: enable -Werror=flex-array-member-not-at-end
2025-04-04 01:18:48 +09:00
'-Werror=flex-array-member-not-at-end',
meson: sort compiler flags
2021-06-02 15:53:29 +09:00
'-Werror=format=2',
meson: Compile with -Werror=format-signedness
2022-08-30 09:57:53 +02:00
'-Werror=format-signedness',
meson: sort compiler flags
2021-06-02 15:53:29 +09:00
'-Werror=implicit-function-declaration',
meson: fail build on implicit int warnings
2023-01-31 16:12:09 +01:00
'-Werror=implicit-int',
meson: sort compiler flags
2021-06-02 15:53:29 +09:00
'-Werror=incompatible-pointer-types',
meson: refuse implicit int <-> pointer conversion Before: ``` Compiling C object src/libsystemd-network/libsystemd-network.a.p/dhcp6-option.c.o ../src/libsystemd-network/dhcp6-option.c: In function ‘dhcp6_option_parse_ia’: ../src/libsystemd-network/dhcp6-option.c:633:70: warning: passing argument 3 of ‘dhcp6_option_parse’ makes pointer from integer without a cast [-Wint-conversion] 633 | r = dhcp6_option_parse(option_data, option_data_len, offset, &subopt, &subdata_len, &subdata); | ^~~~~~ | | | size_t {aka long unsigned int} ../src/libsystemd-network/dhcp6-option.c:358:25: note: expected ‘size_t *’ {aka ‘long unsigned int *’} but argument is of type ‘size_t’ {aka ‘long unsigned int’} 358 | size_t *offset, | ~~~~~~~~^~~~~~ ``` After: ``` ../src/libsystemd-network/dhcp6-option.c: In function ‘dhcp6_option_parse_ia’: ../src/libsystemd-network/dhcp6-option.c:633:70: error: passing argument 3 of ‘dhcp6_option_parse’ makes pointer from integer without a cast [-Werror=int-conversion] 633 | r = dhcp6_option_parse(option_data, option_data_len, offset, &subopt, &subdata_len, &subdata); | ^~~~~~ | | | size_t {aka long unsigned int} ../src/libsystemd-network/dhcp6-option.c:358:25: note: expected ‘size_t *’ {aka ‘long unsigned int *’} but argument is of type ‘size_t’ {aka ‘long unsigned int’} 358 | size_t *offset, | ~~~~~~~~^~~~~~ cc1: some warnings being treated as errors ```
2021-09-25 00:16:20 +09:00
'-Werror=int-conversion',
meson: Share more C flags
2023-03-10 09:57:50 +01:00
'-Werror=missing-declarations',
meson: enable -Werror=missing-parameter-name
2025-06-16 01:44:12 +02:00
'-Werror=missing-parameter-name',
meson: Share more C flags
2023-03-10 09:57:50 +01:00
'-Werror=missing-prototypes',
meson: sort compiler flags
2021-06-02 15:53:29 +09:00
'-Werror=overflow',
meson: upgrade warning about bad init values to an error We never expect to get this warning, it is most likely to occur if somebody messes up enumeration values.
2021-11-26 11:52:04 +01:00
'-Werror=override-init',
meson: make pointer-sign warning critical With the previous commit, no pointer signedness warning should be triggered even built with C23. Let's make the warning critical.
2025-03-16 10:10:28 +09:00
'-Werror=pointer-sign',
meson: sort compiler flags
2021-06-02 15:53:29 +09:00
'-Werror=return-type',
meson: make -Wsequence-point warning critical It detects e.g. multiple evaluations in macro.
2025-07-01 13:00:08 +09:00
'-Werror=sequence-point',
meson: sort compiler flags
2021-06-02 15:53:29 +09:00
'-Werror=shift-count-overflow',
'-Werror=shift-overflow=2',
meson: use -Werror=strict-flex-arrays
2023-05-30 10:30:15 +02:00
'-Werror=strict-flex-arrays',
meson: use get_supported_arguments() This bumps the required minimum version of meson to 0.43, as `get_supported_arguments()` is supported since meson-0.43.
2018-05-10 15:30:42 +09:00
'-Werror=undef',
meson: sort compiler flags
2021-06-02 15:53:29 +09:00
'-Wfloat-equal',
meson: Document why -Wimplicit-fallthrough is not used with clang This changes the macro to ensure proper fallthrough attributes are used with clang in case this option is added in the future.
2022-05-28 11:41:10 +02:00
# gperf prevents us from enabling this because it does not emit fallthrough
# attribute with clang.
#'-Wimplicit-fallthrough',
meson: sort compiler flags
2021-06-02 15:53:29 +09:00
'-Wimplicit-fallthrough=5',
'-Winit-self',
meson: use get_supported_arguments() This bumps the required minimum version of meson to 0.43, as `get_supported_arguments()` is supported since meson-0.43.
2018-05-10 15:30:42 +09:00
'-Wlogical-op',
'-Wmissing-include-dirs',
meson: sort compiler flags
2021-06-02 15:53:29 +09:00
'-Wmissing-noreturn',
'-Wnested-externs',
meson: use get_supported_arguments() This bumps the required minimum version of meson to 0.43, as `get_supported_arguments()` is supported since meson-0.43.
2018-05-10 15:30:42 +09:00
'-Wold-style-definition',
'-Wpointer-arith',
'-Wredundant-decls',
'-Wshadow',
'-Wstrict-aliasing=2',
meson: sort compiler flags
2021-06-02 15:53:29 +09:00
'-Wstrict-prototypes',
'-Wsuggest-attribute=noreturn',
meson: enable -Wunterminated-string-initialization With af1a6db58fde8f64edcf7d27e1f3b636c999934c, now we can build with the option.
2024-08-07 14:03:13 +09:00
'-Wunterminated-string-initialization',
meson: Downgrade unused function from error to warning in local builds When working on systemd, it's often useful to be able to comment out a function to see how a build behaves without it. Currently, when doing this with a static function that's only used once, the build fails because the function then becomes unused. As such, Let's downgrade the unused function error to a warning in local builds.
2021-11-16 09:52:39 +00:00
'-Wunused-function',
meson: use get_supported_arguments() This bumps the required minimum version of meson to 0.43, as `get_supported_arguments()` is supported since meson-0.43.
2018-05-10 15:30:42 +09:00
'-Wwrite-strings',
meson: enable -Wzero-as-null-pointer-constant Support for C added in gcc 15: https://github.com/gcc-mirror/gcc/commit/236c0829ee21a179c81b83f0d7f112ca23c47a4d
2025-01-15 18:41:51 +01:00
'-Wzero-as-null-pointer-constant',
meson: enable -Wzero-length-bounds This will warn if fake flexible arrays are re-introduced. I'm not using -Werror=… because we may still get warnings when compiling against old kernel headers. We can crank this up to error later.
2023-02-06 09:46:55 +01:00
'-Wzero-length-bounds',
build-sys: Use common gcc argument detection for negative warning flags. Starting with meson 0.46, it is able to detect these argument correctly. See this commit in meson codebase for more details: https://github.com/mesonbuild/meson/commit/695b8f3a0377d3e2ce78ba8716adc365b18edea1 We already carry a requirement for meson_version : '>= 0.46', so we can be sure our build system will include this commit. Tested by building systemd using a cloned meson synced to the 0.46.0 tag, confirmed the warnings were detected correctly in that case. The meson messages included this snippet: > Compiler for C supports arguments -Wno-unused-parameter -Wunused-parameter: YES > Compiler for C supports arguments -Wno-missing-field-initializers -Wmissing-field-initializers: YES > Compiler for C supports arguments -Wno-unused-result -Wunused-result: YES > Compiler for C supports arguments -Wno-format-signedness -Wformat-signedness: YES > Compiler for C supports arguments -Wno-error=nonnull -Werror=nonnull: YES > Compiler for C supports arguments -Wno-maybe-uninitialized -Wmaybe-uninitialized: YES
2018-09-08 13:55:09 -07:00
# negative arguments are correctly detected starting with meson 0.46.
build-sys: disable -Wstring-plus-int clang-8 warns about constructs like "foobar"+3 (to get "bar"). We like to use patterns like this, so let's just disable the warning.
2019-03-07 10:56:15 +01:00
'-Wno-error=#warnings', # clang
'-Wno-string-plus-int', # clang
meson: enable -Warray-bounds and -fstrict-flex-arrays -fstrict-flex-arrays means that the compiler doesn't have to assume that any trailing array is a flex array. I.e. unless the array is declared without a specified size, only indices in the declared range are valid. -Warray-bounds turns on the warnings about out-of-bounds array accesses. -Warray-bounds=2 does some more warnings, with higher false positive rate. But it doesn't seem to yield any false positives in our codebase, so enable it. clang supports -Warray-bounds, but not -Warray-bounds=2. gcc supports both. gcc-13 supports -fstrict-flex-arrays. See https://people.kernel.org/kees/bounded-flexible-arrays-in-c for a long discussion of use in the kernel.
2023-02-05 20:44:19 +01:00
meson: Share more C flags
2023-03-10 09:57:50 +01:00
'-fdiagnostics-show-option',
'-fno-common',
boot: Add support for -fstack-protector
2023-03-10 09:01:29 +01:00
'-fstack-protector',
'-fstack-protector-strong',
meson: Use -fstrict-flex-arrays=3 Let's explicitly pass the value to -fstrict-flex-arrays. This does not change behavior but it does (selfishly) make my error not bug out with an error saying -fstrict-flex-arrays does not exist.
2024-08-03 20:10:54 +02:00
'-fstrict-flex-arrays=3',
boot: Add support for -fstack-protector
2023-03-10 09:01:29 +01:00
'--param=ssp-buffer-size=4',
]
possible_common_link_flags = [
'-fstack-protector',
meson: use get_supported_arguments() This bumps the required minimum version of meson to 0.43, as `get_supported_arguments()` is supported since meson-0.43.
2018-05-10 15:30:42 +09:00
]
meson: also check c_args to maybe add -Wno-maybe-uninitialized People (and build systems) sometimes set flags through -Dc_args=… or $CFLAGS. Let's catch this common case too. meson will set c_args from $CFLAGS, so we only need to check the former.
2022-04-29 14:35:20 +02:00
c_args = get_option('c_args')
meson: refuse -ffinite-math-only Closes #23610.
2022-07-18 05:02:48 +09:00
# Our json library does not support -ffinite-math-only, which is enabled by -Ofast or -ffast-math.
meson: check -ffinite-math-only in more detail Even if -fno-finite-math-only is specified, -ffinite-math-only, -ffast-math, or -Ofast may be also specified after that. In that case, -fno-finite-math-only has no effect, and test-json will fail. Replaces #37603.
2025-05-28 05:39:37 +09:00
have = false
foreach arg : c_args
if arg in ['-ffinite-math-only', '-ffast-math', '-Ofast']
have = true
elif arg in ['-fno-finite-math-only', '-fno-fast-math']
have = false
endif
endforeach
if have
error('-ffinite-math-only is enabled (may be implied by -ffast-math or -Ofast) in c_args.')
meson: refuse -ffinite-math-only Closes #23610.
2022-07-18 05:02:48 +09:00
endif
tree-wide: fix typo
2022-03-07 15:32:22 +09:00
# Disable -Wmaybe-uninitialized when compiling with -Os/-O1/-O3/etc. There are
meson: re-enable -Wmaybe-uninitialized on -O[02] with recent gcc versions The warning was disabled in 8794164fed5f0142c34358613f92f4f761af4edd to avoid false positives. But it is useful in finding errors, even if it sometimes results in untrue warnings (c.f. 77fac974fe, da46a1bc3c). After #19168, #19169, and #19175, there are no warnings with -Dbuildtype=debug-optimized/-O2 and gcc-11.0.1-0.3.fc34.x86_64. Warnings are reenabled for -O[23] -O0 is good for development, and -O2 is the default optimization level for Fedora package builds. -Os, -O3, -O1, and -Og still generate some warnings. In fact, with -Os the number of warnings seems completely hopeless. Dozens and dozens.
2021-03-31 17:05:49 +02:00
# too many false positives with gcc >= 8. Effectively, we only test with -O0
# and -O2; this should be enough to catch most important cases without too much
# busywork. See https://github.com/systemd/systemd/pull/19226.
if cc.get_id() == 'gcc' and (not '02'.contains(get_option('optimization')) or
meson: also check c_args to maybe add -Wno-maybe-uninitialized People (and build systems) sometimes set flags through -Dc_args=… or $CFLAGS. Let's catch this common case too. meson will set c_args from $CFLAGS, so we only need to check the former.
2022-04-29 14:35:20 +02:00
cc.version().version_compare('<10') or
'-Os' in c_args or
'-O1' in c_args or
'-O3' in c_args or
meson: also disable -Wmaybe-uninitialized when -Ofast is specified At least with GCC-14, -Ofast produces many false-positive warnings. ``` $ gcc --version gcc (GCC) 14.2.1 20250110 (Red Hat 14.2.1-7) ```
2025-05-28 05:43:48 +09:00
'-Og' in c_args or
'-Ofast' in c_args)
meson: enable more warnings when building efi binary
2021-06-02 15:49:44 +09:00
possible_common_cc_flags += '-Wno-maybe-uninitialized'
meson: re-enable -Wmaybe-uninitialized on -O[02] with recent gcc versions The warning was disabled in 8794164fed5f0142c34358613f92f4f761af4edd to avoid false positives. But it is useful in finding errors, even if it sometimes results in untrue warnings (c.f. 77fac974fe, da46a1bc3c). After #19168, #19169, and #19175, there are no warnings with -Dbuildtype=debug-optimized/-O2 and gcc-11.0.1-0.3.fc34.x86_64. Warnings are reenabled for -O[23] -O0 is good for development, and -O2 is the default optimization level for Fedora package builds. -Os, -O3, -O1, and -Og still generate some warnings. In fact, with -Os the number of warnings seems completely hopeless. Dozens and dozens.
2021-03-31 17:05:49 +02:00
endif
meson: re-enable -Wunused-result with clang We disabled it in f73fb7b742f294b6d2126afa16001bd2ff6ab461 in response to an apparent gcc bug. It seems that depending on the combination of optimization options, gcc still ignores (void). But this seems to work fine with clang, so let's re-enable the warning conditionally.
2021-07-06 15:13:00 +02:00
# Disable -Wno-unused-result with gcc, see
# https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66425.
if cc.get_id() == 'gcc'
possible_common_cc_flags += '-Wno-unused-result'
endif
meson: use get_supported_arguments() This bumps the required minimum version of meson to 0.43, as `get_supported_arguments()` is supported since meson-0.43.
2018-05-10 15:30:42 +09:00
# --as-needed and --no-undefined are provided by meson by default,
meson: require 0.53.2 and drop some workarounds for old meson Ubuntu Bionic 18.04 has 0.45, so it was below the previously required minimum version already. Focal 20.04 has 0.53.2. Let's require that and use various features that are available.
2021-07-27 16:16:07 +02:00
# run 'meson configure' to see what is enabled
meson: use get_supported_arguments() This bumps the required minimum version of meson to 0.43, as `get_supported_arguments()` is supported since meson-0.43.
2018-05-10 15:30:42 +09:00
possible_link_flags = [
meson: Add --warn-common and --fatal-warnings to link flags
2021-11-04 19:03:45 +01:00
'-Wl,--fatal-warnings',
meson: use get_supported_arguments() This bumps the required minimum version of meson to 0.43, as `get_supported_arguments()` is supported since meson-0.43.
2018-05-10 15:30:42 +09:00
'-Wl,-z,now',
meson: Add --warn-common and --fatal-warnings to link flags
2021-11-04 19:03:45 +01:00
'-Wl,-z,relro',
meson: link with -Wl,-z,gcs-report-dynamic=none (#38901) There is a botched arm64 linker transition going on, where a new feature is enabled (GCS) and the linker fails the build unless all object files being linked are built with the new specific feature. This was enabled in the toolchain (GCC 15) _before_ all libraries were rebuilt, including glibc, so everything fails. The toolchain maintainers declined to fix it and instead say that this is a useless warning to have, and to just disable it and ignore it. https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1110461 > systemd fails to build from source on arm64 if built with GCC-15, currently in > experimental. > > GCC-15 includes support for an arm64 security feature called Guarded Control > Stack (GCS). To help with GCS adoption, the linker warns about shared libraries > built without GCS. For example: > > /usr/lib/gcc/aarch64-linux-gnu/15/../../../aarch64-linux-gnu/libncursesw.so.6: > warning: GCS is required by -z gcs, but this shared library lacks the necessary property note. > The dynamic loader might not enable GCS or refuse to load the program unless > all the shared library dependencies have the GCS marking. > > The warning is harmless, and can be ignored. However, systemd is built with > --fatal-warnings, and for this reason will fail to build on arm64 once GCC-15 > becomes the default compiler in Debian. [585/3230] Linking target src/core/libsystemd-core-258.so FAILED: src/core/libsystemd-core-258.so gcc -o src/core/libsystemd-core-258.so -Wl,--as-needed -Wl,--no-undefined -shared -fPIC -Wl,-soname,libsystemd-core-258.so -Wl,--whole-archive -Wl,--start-group src/core/libsystemd-core-258.a -Wl,--no-whole-archive -fstack-protector -Wl,-z,relro -specs=/usr/share/debhelper/dh_package_notes/debian-package-notes.specs -g -O2 -Werror=implicit-function-declaration -ffile-prefix-map=/work/src=. -fstack-protector-strong -fstack-clash-protection -Wformat -Werror=format-security -mbranch-protection=standard -O0 -g -Og -Wdate-time '-Wl,-rpath,$ORIGIN/../shared' src/shared/libsystemd-shared-258.so -shared -Wl,--version-script=/work/src/src/shared/libshared.sym /usr/lib/aarch64-linux-gnu/libacl.so /usr/lib/aarch64-linux-gnu/libaudit.so /usr/lib/aarch64-linux-gnu/libblkid.so -ldl -lm /usr/lib/aarch64-linux-gnu/libmount.so -lrt /usr/lib/aarch64-linux-gnu/libseccomp.so /usr/lib/aarch64-linux-gnu/libselinux.so -Wl,--end-group -pthread -Wl,--fatal-warnings -Wl,-z,now -Wl,-z,relro -Wl,--gc-sections -Wl,--warn-common src/shared/libsystemd-shared-258.so: warning: GCS is required by -z gcs, but this shared library lacks the necessary property note. The dynamic loader might not enable GCS or refuse to load the program unless all the shared library dependencies have the GCS marking. /usr/lib/aarch64-linux-gnu/libacl.so: warning: GCS is required by -z gcs, but this shared library lacks the necessary property note. The dynamic loader might not enable GCS or refuse to load the program unless all the shared library dependencies have the GCS marking. /usr/lib/aarch64-linux-gnu/libaudit.so: warning: GCS is required by -z gcs, but this shared library lacks the necessary property note. The dynamic loader might not enable GCS or refuse to load the program unless all the shared library dependencies have the GCS marking. /usr/lib/aarch64-linux-gnu/libblkid.so: warning: GCS is required by -z gcs, but this shared library lacks the necessary property note. The dynamic loader might not enable GCS or refuse to load the program unless all the shared library dependencies have the GCS marking. /lib/aarch64-linux-gnu/libm.so.6: warning: GCS is required by -z gcs, but this shared library lacks the necessary property note. The dynamic loader might not enable GCS or refuse to load the program unless all the shared library dependencies have the GCS marking. /lib/aarch64-linux-gnu/libmvec.so.1: warning: GCS is required by -z gcs, but this shared library lacks the necessary property note. The dynamic loader might not enable GCS or refuse to load the program unless all the shared library dependencies have the GCS marking. /usr/lib/aarch64-linux-gnu/libmount.so: warning: GCS is required by -z gcs, but this shared library lacks the necessary property note. The dynamic loader might not enable GCS or refuse to load the program unless all the shared library dependencies have the GCS marking. /usr/lib/aarch64-linux-gnu/libseccomp.so: warning: GCS is required by -z gcs, but this shared library lacks the necessary property note. The dynamic loader might not enable GCS or refuse to load the program unless all the shared library dependencies have the GCS marking. /usr/lib/aarch64-linux-gnu/libselinux.so: warning: GCS is required by -z gcs, but this shared library lacks the necessary property note. The dynamic loader might not enable GCS or refuse to load the program unless all the shared library dependencies have the GCS marking. /lib/aarch64-linux-gnu/libc.so.6: warning: GCS is required by -z gcs, but this shared library lacks the necessary property note. The dynamic loader might not enable GCS or refuse to load the program unless all the shared library dependencies have the GCS marking. /lib/ld-linux-aarch64.so.1: warning: GCS is required by -z gcs, but this shared library lacks the necessary property note. The dynamic loader might not enable GCS or refuse to load the program unless all the shared library dependencies have the GCS marking. collect2: error: ld returned 1 exit status
2025-09-15 12:42:48 +01:00
'-Wl,-z,gcs-report-dynamic=none',
meson: drop -ffunction-sections -fdata-sections I added them in 41afb5eb7214727301132aedc381831fbfc78e37 without too much explanation. Most likely the idea was to get rid of unused code in libsystemd.so [1]. But now that I'm testing this, it doesn't seem to have an effect. LTO is needed to get rid of unused functions, and it's enough to have LTO without those options. Those options might have some downsides [2], so let's disable them since there are doubts and no particularly good reason to have them. But keep the -Wl,--gc-sections option. Without this, libsystemd.so grows a little: -rwxr-xr-x 1 zbyszek zbyszek 5532424 07-08 13:24 build/libsystemd.so.0.40.0-orig -rwxr-xr-x 1 zbyszek zbyszek 5614472 07-08 13:26 build/libsystemd.so.0.40.0-no-sections -rwxr-xr-x 1 zbyszek zbyszek 5532392 07-08 13:27 build/libsystemd.so.0.40.0 Let's apply the --gc-sections option always to make the debug and final builds more similar. We need to verify that distro packages don't unexpectedly grow after this. [1] https://unix.stackexchange.com/a/715901 [2] https://stackoverflow.com/a/36033811
2025-07-08 13:18:07 +02:00
'-Wl,--gc-sections',
meson: use get_supported_arguments() This bumps the required minimum version of meson to 0.43, as `get_supported_arguments()` is supported since meson-0.43.
2018-05-10 15:30:42 +09:00
]
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: Add --warn-common and --fatal-warnings to link flags
2021-11-04 19:03:45 +01:00
if get_option('b_sanitize') == 'none'
possible_link_flags += '-Wl,--warn-common'
endif
meson: Compile with -ftrivial-auto-var-init=zero in release mode
2022-05-07 11:31:26 +02:00
if get_option('mode') == 'release'
# We could enable 'pattern' for developer mode, but that can interfere with
# valgrind and sanitizer builds. Also, clang does not zero-initialize unions,
# breaking some of our code (https://reviews.llvm.org/D68115).
possible_common_cc_flags += '-ftrivial-auto-var-init=zero'
endif
meson: Introduce userspace dep This will help in a later commit to separate userspace from EFI builds.
2023-02-26 11:07:24 +01:00
possible_cc_flags = [
meson: do not share compiler flags except for emitting warnings Follow-up for 65267363978dbb298eb4ba9b628d028c969fa616. Prompted by https://github.com/systemd/systemd/issues/19191#issuecomment-856312107.
2021-06-08 09:23:50 +09:00
'-fno-strict-aliasing',
build: Use -fstrict-flex-arrays=1 if supported Due to "historical reasons" both gcc and clang treat *all* trailing arrays members as flexible arrays, this has an evil side effect of inhibiting bounds checks on such members as __builtin_object_size cannot say for sure that: struct { ... type foo[3]; } has a trailing foo member of fixed size rather than unspecified. Ideally we should use -fstrict-flex-arrays as is, but we have to tolerate kernel uapi headers that use [0] and third party libraries written in c89 that may use [1] like curl.
2022-11-09 23:56:46 +00:00
'-fstrict-flex-arrays=1',
meson: do not share compiler flags except for emitting warnings Follow-up for 65267363978dbb298eb4ba9b628d028c969fa616. Prompted by https://github.com/systemd/systemd/issues/19191#issuecomment-856312107.
2021-06-08 09:23:50 +09:00
'-fvisibility=hidden',
]
meson: Build with frame pointers in developer mode Profiling tools tend to work better when binaries and libraries are compiled with frame pointers as without them there's no easy and fast way to get the current stacktrace.
2022-06-08 22:32:49 +02:00
if get_option('mode') == 'developer'
possible_cc_flags += '-fno-omit-frame-pointer'
endif
meson: Introduce userspace dep This will help in a later commit to separate userspace from EFI builds.
2023-02-26 11:07:24 +01:00
add_project_arguments(
cc.get_supported_arguments(
basic_disabled_warnings,
possible_common_cc_flags
),
language : 'c')
boot: Add support for -fstack-protector
2023-03-10 09:01:29 +01:00
add_project_link_arguments(
cc.get_supported_link_arguments(possible_common_link_flags),
language : 'c')
meson: Introduce userspace dep This will help in a later commit to separate userspace from EFI builds.
2023-02-26 11:07:24 +01:00
userspace_c_args += cc.get_supported_arguments(possible_cc_flags)
userspace_c_ld_args += cc.get_supported_link_arguments(possible_link_flags)
meson: use get_supported_arguments() This bumps the required minimum version of meson to 0.43, as `get_supported_arguments()` is supported since meson-0.43.
2018-05-10 15:30:42 +09:00
meson: use multiline string syntax (#8214) The single quote working with multiple lines is likely to be unintended. With current versions of meson, it also causes error messages after it to report the wrong line number. Use the documented syntax instead.
2018-02-19 01:37:19 -08:00
if cc.compiles('''
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
#include <time.h>
#include <inttypes.h>
typedef uint64_t usec_t;
usec_t now(clockid_t clock);
int main(void) {
struct timespec now;
return 0;
}
meson.build: fix detection of -Werror=shadow Pass -Werror=shadow in args of cc.compiles otherwise test will always succeed This fix a build failure with gcc 4.7.3 Fixes: - http://autobuild.buildroot.org/results/ffd71c473d3b29618c18cd2e04705370266696f2 Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
2018-11-29 13:22:08 +01:00
''', args: '-Werror=shadow', name : '-Werror=shadow with local shadowing')
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-17 19:25:00 -04:00
add_project_arguments('-Werror=shadow', language : 'c')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
endif
meson: apply our -Wno-* options also in c++ calls We compile some c++ code for tests. We would simply use the default options for those. When the previous commit raised the default warning level, we started getting warnings from c++ code. Let's add the most important options to the c++ command, so that we get a compilation without any warnings again. I don't think it makes sense to add *all* the options that we add for c to the c++ flags, because testing them takes quite a while, and the c++ compilations are for small amounts of code, mostly to check that the headers have compatible syntax.
2019-11-10 12:16:41 +01:00
if cxx_cmd != ''
add_project_arguments(cxx.get_supported_arguments(basic_disabled_warnings), language : 'cpp')
endif
Revert "Revert "meson: use c_args in generator scripts (#10289)"" This reverts commit 0e3cc902faec4f18d5fa606396f602b08bc94e27. Fixes #10288. I have confirmed that this does now fix cross-compilation. It appears that changes upstream in Meson, probably mesonbuild/meson#5263, have made the original MR, #10289, work now. This needs to be tested to ensure that it doesn't break Travis CI like when it was reverted in #10361.
2023-10-11 13:32:49 -05:00
cpp = ' '.join(cc.cmd_array() + get_option('c_args')) + ' -E'
meson: make cpp invocations cross-compilation friendly This implementation assumes that the arguments in compiler.cmd_array() don't contain any spaces. Since we are only interested in compilation on Linux, I think this is a safe assumption. Solution suggested by Nirbheek Chauhan.
2017-04-11 10:25:34 -04:00
fundamental-macro: conditionalize several gcc warning pragmas This fixes the following error: In file included from ../src/basic/macro.h:13, from ../src/basic/env-util.h:9, from ../src/nss-systemd/nss-systemd.c:7: ../src/fundamental/macro-fundamental.h:61:9: error: option ‘-Wzero-as-null-pointer-constant’ is valid for C++/ObjC++ but not for C [-Werror=pragmas] 61 | _Pragma("GCC diagnostic ignored \"-Wzero-as-null-pointer-constant\"") | ^~~~~~~ ../src/nss-systemd/nss-systemd.c:106:1: note: in expansion of macro ‘DISABLE_WARNING_ZERO_AS_NULL_POINTER_CONSTANT’ 106 | DISABLE_WARNING_ZERO_AS_NULL_POINTER_CONSTANT; | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ cc1: all warnings being treated as errors
2025-01-17 04:27:36 +09:00
# new in GCC 10
have = cc.has_argument('-Wzero-length-bounds')
conf.set10('HAVE_WARNING_ZERO_LENGTH_BOUNDS', have)
# new in GCC 15
have = cc.has_argument('-Wzero-as-null-pointer-constant')
conf.set10('HAVE_WARNING_ZERO_AS_NULL_POINTER_CONSTANT', have)
shared/utmp-wtmp: silence gcc warning about strncpy truncation Unfortunately the warning must be known, or otherwise the pragma generates a warning or an error. So let's do a meson check for it. Is it worth doing this to silence the warning? I think so, because apparently the warning was already emitted by gcc-8.1, and with the recent push in gcc to catch more such cases, we'll most likely only get more of those.
2019-05-07 15:10:58 +02:00
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
#####################################################################
# compilation result tests
meson: define _GNU_SOURCE as '1' This changes the generated config.h file thusly: -#define _GNU_SOURCE +#define _GNU_SOURCE 1 Canonically, _GNU_SOURCE is just defined, without any value, but g++ defines _GNU_SOURCE implicitly [1]. This causes a warning about a redefinition during complilation of C++ programs after '-include config.h'. Our config attempts to inject this (and a bunch of other arguments) into all compliations. But before meson 0.54, flags for dependencies were not propagated correctly (*), and the C++ compilation was done without various flags (**). Once that was fixed, we started getting a warning. [1] http://gcc.gnu.org/onlinedocs/libstdc++/faq.html#faq.predefined (*) Actually, the changelog doesn't say anything. But it mentions various work related to dependency propagation, and apparently this changes as a side effect. (**) -fno-strict-aliasing -fstrict-flex-arrays=1 -fvisibility=hidden -fno-omit-frame-pointer -include config.h This could be solved in various ways, but it'd require either making the compilation command line longer, which we want to avoid for readability of the build logs, or splitting the logic to define the args for C++ progs separately, which would make our meson.build files more complicated. Changing the definition to '1' also solves the issue (because apparently now we match the implicit definition), and shouldn't have other effects. I checked compilation with gcc and clang. Maybe on other systems this could cause problems. We can revisit if people report issues.
2023-06-12 10:45:48 +02:00
conf.set('_GNU_SOURCE', 1)
meson: use booleans for conf.set and drop unecessary conditionals Using conf.set() with a boolean argument does the right thing: either #ifdef or #undef. This means that conf.set can be used unconditionally. Previously I used '1' as the placeholder value, and that needs to be changed to 'true' for consistency (under meson 1 cannot be used in boolean context). All checks need to be adjusted.
2017-04-27 21:13:08 -04:00
conf.set('__SANE_USERSPACE_TYPES__', true)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
conf.set('SIZEOF_DEV_T', cc.sizeof('dev_t', prefix : '#include <sys/types.h>'))
conf.set('SIZEOF_INO_T', cc.sizeof('ino_t', prefix : '#include <sys/types.h>'))
conf.set('SIZEOF_RLIM_T', cc.sizeof('rlim_t', prefix : '#include <sys/resource.h>'))
meson: sort types
2023-06-01 08:18:15 +09:00
conf.set('SIZEOF_TIME_T', cc.sizeof('time_t', prefix : '#include <sys/time.h>'))
meson: fix cross compiling
2021-12-21 20:10:09 +09:00
conf.set('SIZEOF_TIMEX_MEMBER', cc.sizeof('typeof(((struct timex *)0)->freq)', prefix : '#include <sys/timex.h>'))
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: fix cross-compilation of LONG_MAX https://github.com/systemd/systemd/pull/25618#issuecomment-1355019553
2022-12-17 17:00:56 +01:00
long_max = cc.compute_int(
'LONG_MAX',
prefix : '#include <limits.h>',
guess : 0x7FFFFFFFFFFFFFFF,
high : 0x7FFFFFFFFFFFFFFF)
manager: define a string constant for LONG_MAX and use that for sysctl This moves the formatting of the constant to compilation time and let's us avoid asprintf() in the very hot path of initial boot.
2022-12-03 11:55:55 +01:00
assert(long_max > 100000)
meson: using f-strings in meson Those were added before 0.61 which is our minimum version. Dots were dropped from the end of some messages. Most messages did not have the trailing dot. 0.63 added support for multi-line f-strings. We can't use those yet.
2025-10-03 15:26:20 +02:00
conf.set_quoted('LONG_MAX_STR', f'@long_max@')
manager: define a string constant for LONG_MAX and use that for sysctl This moves the formatting of the constant to compilation time and let's us avoid asprintf() in the very hot path of initial boot.
2022-12-03 11:55:55 +01:00
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
foreach ident : [
meson: cleanup glibc header checkers
2025-03-01 01:05:05 +09:00
['set_mempolicy', '''#include <sys/syscall.h>'''], # declared at numaif.h provided by libnuma, which we do not use
['get_mempolicy', '''#include <sys/syscall.h>'''], # declared at numaif.h provided by libnuma, which we do not use
['strerrorname_np', '''#include <string.h>'''], # since glibc-2.32
['mallinfo2', '''#include <malloc.h>'''], # since glibc-2.33
['execveat', '''#include <unistd.h>'''], # since glibc-2.34
['close_range', '''#include <unistd.h>'''], # since glibc-2.34
['epoll_pwait2', '''#include <sys/epoll.h>'''], # since glibc-2.35
['fsconfig', '''#include <sys/mount.h>'''], # since glibc-2.36
['fsmount', '''#include <sys/mount.h>'''], # since glibc-2.36
['fsopen', '''#include <sys/mount.h>'''], # since glibc-2.36
['mount_setattr', '''#include <sys/mount.h>'''], # since glibc-2.36
['move_mount', '''#include <sys/mount.h>'''], # since glibc-2.36
['open_tree', '''#include <sys/mount.h>'''], # since glibc-2.36
['pidfd_open', '''#include <sys/pidfd.h>'''], # since glibc-2.36
['pidfd_send_signal', '''#include <sys/pidfd.h>'''], # since glibc-2.36
['pidfd_spawn', '''#include <spawn.h>'''], # since glibc-2.39
['sched_setattr', '''#include <sched.h>'''], # since glibc-2.41
['ioprio_get', '''#include <sched.h>'''], # no known header declares ioprio_get
['ioprio_set', '''#include <sched.h>'''], # no known header declares ioprio_set
['rt_tgsigqueueinfo', '''#include <signal.h>'''], # no known header declares rt_tgsigqueueinfo
libc-wrapper: introduce a tiny libc wrapper Then, move syscall definitions to the wrapper, and prototypes are moved to relevant headers. This also adds checks for add_key() and request_key(), as one day glibc may be going to add some of them separatedly. The check for fspick in meson.build is dropped, as it is currently unused in our code. This also moves - basic/missing_bpf.h -> include/override/linux/bpf.h, - basic/missing_keyctl.h -> include/override/linux/keyctl.h.
2025-06-21 20:29:51 +09:00
['open_tree_attr', '''#include <sys/mount.h>'''], # no known header declares open_tree_attr
meson: cleanup glibc header checkers
2025-03-01 01:05:05 +09:00
['quotactl_fd', '''#include <sys/quota.h>'''], # no known header declares quotactl_fd
['fchmodat2', '''#include <sys/stat.h>'''], # no known header declares fchmodat2
['bpf', '''#include <sys/syscall.h>'''], # no known header declares bpf
['kcmp', '''#include <sys/syscall.h>'''], # no known header declares kcmp
['keyctl', '''#include <sys/syscall.h>'''], # no known header declares keyctl
libc-wrapper: introduce a tiny libc wrapper Then, move syscall definitions to the wrapper, and prototypes are moved to relevant headers. This also adds checks for add_key() and request_key(), as one day glibc may be going to add some of them separatedly. The check for fspick in meson.build is dropped, as it is currently unused in our code. This also moves - basic/missing_bpf.h -> include/override/linux/bpf.h, - basic/missing_keyctl.h -> include/override/linux/keyctl.h.
2025-06-21 20:29:51 +09:00
['add_key', '''#include <sys/syscall.h>'''], # no known header declares add_key
['request_key', '''#include <sys/syscall.h>'''], # no known header declares request_key
meson: cleanup glibc header checkers
2025-03-01 01:05:05 +09:00
['setxattrat', '''#include <sys/xattr.h>'''], # no known header declares setxattrat
['removexattrat', '''#include <sys/xattr.h>'''], # no known header declares removexattrat
libc-wrapper: introduce a tiny libc wrapper Then, move syscall definitions to the wrapper, and prototypes are moved to relevant headers. This also adds checks for add_key() and request_key(), as one day glibc may be going to add some of them separatedly. The check for fspick in meson.build is dropped, as it is currently unused in our code. This also moves - basic/missing_bpf.h -> include/override/linux/bpf.h, - basic/missing_keyctl.h -> include/override/linux/keyctl.h.
2025-06-21 20:29:51 +09:00
['pivot_root', '''#include <unistd.h>'''], # no known header declares pivot_root
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-17 19:25:00 -04:00
]
meson: use "args" for setting _GNU_SOURCE when checking for functions This reworks how we set _GNU_SOURCE when checking for the availability of functions: 1. We set it for most of the functions we look for. After all we set it for our entire built anyway, and it's usually how Linux-specific definitions in glibc are protected these days. Given that we usually have checks for such modern stuff only anyway, let's just blanket enable it. 2. Use "args" instead of "prefix" to set the macro. This is what is suggested in the meson docs, hence let's do it.
2017-12-25 12:01:14 +01:00
have = cc.has_function(ident[0], prefix : ident[1], args : '-D_GNU_SOURCE')
build-sys: change all HAVE_DECL_ macros to HAVE_ This is a legacy of autotools, where one detection routine used a different prefix then the others. $ git grep -e HAVE_DECL_ -l|xargs sed -i s/HAVE_DECL_/HAVE_/g
2017-10-03 10:32:34 +02:00
conf.set10('HAVE_' + ident[0].to_upper(), have)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
endforeach
#####################################################################
meson: indent find_program() calls for readability
2025-07-25 12:18:40 +02:00
awk = find_program('awk')
diff = find_program('diff')
echo = find_program('echo')
env = find_program('env')
find = find_program('find')
meson: also skip uid/gid check for nobody user/group when id command not found Follow-up for 8b413ae4060b21ed4712fdad7eba195890740756.
2025-02-07 11:36:46 +09:00
getent = find_program('getent', required : false)
meson: indent find_program() calls for readability
2025-07-25 12:18:40 +02:00
git = find_program('git', required : false)
gperf = find_program('gperf')
id = find_program('id', required : false)
ln = find_program('ln')
rsync = find_program('rsync', required : false)
sed = find_program('sed')
sh = find_program('sh')
stat = find_program('stat')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: use install_emptydir() and drop meson-make-symlink.sh The script is mostly equivalent to 'mkdir -p' and 'ln -sfr'. Let's replace it with install_emptydir() builtin function and inline meson call.
2023-08-08 10:38:47 +09:00
ln_s = ln.full_path() + ' -frsT -- "${DESTDIR:-}@0@" "${DESTDIR:-}@1@"'
meson: create dirs and touch /usr This is the equivalent of $(INSTALL_DIRS) and install-touch-usr-hook. I did not bother to create the directories into which we install files, since they will be created anyway. v2: - remove bashism
2017-04-09 23:55:05 -04:00
meson: stop probing for paths of programs in /usr/sbin We dropped support for split-usr a while ago, which means that the programs will be in /usr/sbin, which actually may be the same as /usr/bin on merged-bin systems. So the whole checking is mostly pointless in the usual case. OTOH, on Nix the paths will be totally different and need to be set through the option anyway. So save time during builds by using the "fallback" path unless the option is specified. This avoid some busywork during the slow serial build phase.
2025-10-09 14:53:22 +02:00
# If -Dxxx-path option is found, use that. Otherwise, use the default from the
# middle column; a full path is used directly, a relative path is converted to
# /usr/bin/foo or /usr/sbin/foo, depending on whether split-bin is enabled.
progs = [['quotaon', 'quotaon' ],
['quotacheck', 'quotacheck' ],
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
['kmod', '/usr/bin/kmod' ],
meson: stop probing for paths of programs in /usr/sbin We dropped support for split-usr a while ago, which means that the programs will be in /usr/sbin, which actually may be the same as /usr/bin on merged-bin systems. So the whole checking is mostly pointless in the usual case. OTOH, on Nix the paths will be totally different and need to be set through the option anyway. So save time during builds by using the "fallback" path unless the option is specified. This avoid some busywork during the slow serial build phase.
2025-10-09 14:53:22 +02:00
['kexec', 'kexec' ],
['sulogin', 'sulogin' ],
['swapon', 'swapon' ],
['swapoff', 'swapoff' ],
['agetty', 'agetty' ],
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
['mount', '/usr/bin/mount', 'MOUNT_PATH'],
['umount', '/usr/bin/umount', 'UMOUNT_PATH'],
['loadkeys', '/usr/bin/loadkeys', 'KBD_LOADKEYS'],
['setfont', '/usr/bin/setfont', 'KBD_SETFONT'],
meson: stop probing for paths of programs in /usr/sbin We dropped support for split-usr a while ago, which means that the programs will be in /usr/sbin, which actually may be the same as /usr/bin on merged-bin systems. So the whole checking is mostly pointless in the usual case. OTOH, on Nix the paths will be totally different and need to be set through the option anyway. So save time during builds by using the "fallback" path unless the option is specified. This avoid some busywork during the slow serial build phase.
2025-10-09 14:53:22 +02:00
['nologin', 'nologin', ],
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
]
foreach prog : progs
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-17 19:25:00 -04:00
path = get_option(prog[0] + '-path')
meson: stop probing for paths of programs in /usr/sbin We dropped support for split-usr a while ago, which means that the programs will be in /usr/sbin, which actually may be the same as /usr/bin on merged-bin systems. So the whole checking is mostly pointless in the usual case. OTOH, on Nix the paths will be totally different and need to be set through the option anyway. So save time during builds by using the "fallback" path unless the option is specified. This avoid some busywork during the slow serial build phase.
2025-10-09 14:53:22 +02:00
if path == ''
if prog[1].startswith('/')
path = prog[1]
else
path = '/usr' / (split_bin ? 'sbin' : 'bin') / prog[1]
endif
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-17 19:25:00 -04:00
endif
meson: stop probing for paths of programs in /usr/sbin We dropped support for split-usr a while ago, which means that the programs will be in /usr/sbin, which actually may be the same as /usr/bin on merged-bin systems. So the whole checking is mostly pointless in the usual case. OTOH, on Nix the paths will be totally different and need to be set through the option anyway. So save time during builds by using the "fallback" path unless the option is specified. This avoid some busywork during the slow serial build phase.
2025-10-09 14:53:22 +02:00
message('Using @1@ for @0@'.format(prog[0], path))
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-17 19:25:00 -04:00
name = prog.length() > 2 ? prog[2] : prog[0].to_upper()
conf.set_quoted(name, path)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
endforeach
meson: add check:true/false to all run_command() invocations meson-0.59.4-1.fc35.noarch says: WARNING: You should add the boolean check kwarg to the run_command call. It currently defaults to false, but it will default to true in future releases of meson. See also: https://github.com/mesonbuild/meson/issues/9300
2021-11-17 13:58:53 +01:00
if run_command(ln, '--relative', '--help', check : false).returncode() != 0
meson: add note about coreutils version with ln --relative https://bugs.freedesktop.org/show_bug.cgi?id=90799
2018-03-09 08:56:23 +01:00
error('ln does not support --relative (added in coreutils 8.16)')
meson: add ln --relative check
2017-04-18 19:11:54 -04:00
endif
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: make lines more consistent
2023-12-14 18:35:00 -03:00
#####################################################################
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
gperf_test_format = '''
#include <string.h>
various: move const ptr indicator to return value
2024-05-17 08:54:00 +02:00
const char* in_word_set(const char *, @0@);
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
@1@
'''
meson: use better shellscript argument passing Passing potentially arbitrary data into a shellscript is potentially very broken if you do not correctly quote it for use. This quoting must be done as part of the interpretation of the data itself, e.g. python's shlex.quote; simply formatting it into a string with double quotes is NOT sufficient. An alternative is to communicate the data reliably via argv to the shell process, and allow the shell to internally handle it via `"$1"`, which is quote-safe and will expand the data from argv as a single tokenized word.
2022-05-15 11:11:24 -04:00
gperf_snippet = run_command(sh, '-c', 'echo foo,bar | "$1" -L ANSI-C', '_', gperf,
meson: add check:true/false to all run_command() invocations meson-0.59.4-1.fc35.noarch says: WARNING: You should add the boolean check kwarg to the run_command call. It currently defaults to false, but it will default to true in future releases of meson. See also: https://github.com/mesonbuild/meson/issues/9300
2021-11-17 13:58:53 +01:00
check : true)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
gperf_test = gperf_test_format.format('size_t', gperf_snippet.stdout())
if cc.compiles(gperf_test)
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-17 19:25:00 -04:00
gperf_len_type = 'size_t'
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
else
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-17 19:25:00 -04:00
gperf_test = gperf_test_format.format('unsigned', gperf_snippet.stdout())
if cc.compiles(gperf_test)
gperf_len_type = 'unsigned'
else
error('unable to determine gperf len type')
endif
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
endif
meson: using f-strings in meson Those were added before 0.61 which is our minimum version. Dots were dropped from the end of some messages. Most messages did not have the trailing dot. 0.63 added support for multi-line f-strings. We can't use those yet.
2025-10-03 15:26:20 +02:00
message(f'gperf len type is @gperf_len_type@')
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-17 19:25:00 -04:00
conf.set('GPERF_LEN_TYPE', gperf_len_type,
description : 'The type of gperf "len" parameter')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: make lines more consistent
2023-12-14 18:35:00 -03:00
#####################################################################
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: crypt.h must always exist We require at least crypt_r() exists, and it is provided since glibc-2.0 (and dropped in glibc-2.39) or by libxcrypt, and the function is provided in crypt.h regardless it is provided by glibc or libxcrypt. Hence, we cannot fallback to unistd.h. This makes the condition about crypt.h more strict, and stop compilation earlier when crypt.h does not exist.
2025-07-23 13:13:29 +09:00
foreach header : [
'crypt.h',
'sys/capability.h',
]
if not cc.has_header(header)
meson: using f-strings in meson Those were added before 0.61 which is our minimum version. Dots were dropped from the end of some messages. Most messages did not have the trailing dot. 0.63 added support for multi-line f-strings. We can't use those yet.
2025-10-03 15:26:20 +02:00
error(f'Header file @header@ not found')
meson: crypt.h must always exist We require at least crypt_r() exists, and it is provided since glibc-2.0 (and dropped in glibc-2.39) or by libxcrypt, and the function is provided in crypt.h regardless it is provided by glibc or libxcrypt. Hence, we cannot fallback to unistd.h. This makes the condition about crypt.h more strict, and stop compilation earlier when crypt.h does not exist.
2025-07-23 13:13:29 +09:00
endif
endforeach
foreach header : [
'sys/sdt.h',
'threads.h',
'valgrind/memcheck.h',
'valgrind/valgrind.h',
]
meson: use booleans for conf.set and drop unecessary conditionals Using conf.set() with a boolean argument does the right thing: either #ifdef or #undef. This means that conf.set can be used unconditionally. Previously I used '1' as the placeholder value, and that needs to be changed to 'true' for consistency (under meson 1 cannot be used in boolean context). All checks need to be adjusted.
2017-04-27 21:13:08 -04:00
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 10:41:51 +02:00
conf.set10('HAVE_' + header.underscorify().to_upper(),
cc.has_header(header))
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
endforeach
meson: make lines more consistent
2023-12-14 18:35:00 -03:00
#####################################################################
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
Add a basic test that the configured fallback hostname is OK Ideally, assert_cc() would be used for this, so that it is not possible to even compile systemd with something like '-Dfallback-hostname=.foo'. But to do a proper check we need to call hostname_is_valid(), and we cannot depend on being able to run code (e.g. during cross-compilation). So let's do a very superficial check in meson, and a proper on in test-util.
2020-05-07 17:30:02 +02:00
fallback_hostname = get_option('fallback-hostname')
if fallback_hostname == '' or fallback_hostname[0] == '.' or fallback_hostname[0] == '-'
error('Invalid fallback-hostname configuration')
# A more extensive test is done in test-hostname-util. Let's catch
# the most obvious errors here so we don't fail with an assert later.
endif
conf.set_quoted('FALLBACK_HOSTNAME', fallback_hostname)
meson: allow extra net naming schemes to be defined during configuration In upstream, we have a linearly-growing list of net-naming-scheme defines; we add a new one for every release where we make user-visible changes to the naming scheme. But the general idea was that downstream distributions could define their own combinations (or even just their own names for existing combinations), so provide stability for their users. So far this required patching of the netif-naming-scheme.c and .h files to add the new lines. With this patch, patching is not required: $ meson configure build \ -Dextra-net-naming-schemes=gargoyle=v238+npar_ari+allow_rerenames,gargoyle2=gargoyle+nspawn_long_hash \ -Ddefault-net-naming-scheme=gargoyle2 or even $ meson configure build \ -Dextra-net-naming-schemes=gargoyle=v238+npar_ari+allow_rerenames,gargoyle2=gargoyle+nspawn_long_hash,latest=v249 \ -Ddefault-net-naming-scheme=gargoyle2 The syntax is a comma-separated list of NAME=name+name+… This syntax is a bit scary, but any typos result in compilation errors, so I think it should be OK in practice. With this approach, we don't allow users to define arbitrary combinations: what is allowed is still defined at compilation time, so it's up to the distribution maintainers to provide reasonable combinations. In this regard, the only difference from status quo is that it's much easier to do (and harder to do incorrectly, for example by forgetting to add a name to one of the maps).
2021-09-28 10:12:36 +02:00
extra_net_naming_schemes = []
extra_net_naming_map = []
foreach scheme: get_option('extra-net-naming-schemes').split(',')
if scheme != ''
name = scheme.split('=')[0]
value = scheme.split('=')[1]
NAME = name.underscorify().to_upper()
VALUE = []
foreach field: value.split('+')
VALUE += 'NAMING_' + field.underscorify().to_upper()
endforeach
extra_net_naming_schemes += 'NAMING_@0@ = @1@,'.format(NAME, '|'.join(VALUE))
extra_net_naming_map += '{ "@0@", NAMING_@1@ },'.format(name, NAME)
endif
endforeach
conf.set('EXTRA_NET_NAMING_SCHEMES', ' '.join(extra_net_naming_schemes))
conf.set('EXTRA_NET_NAMING_MAP', ' '.join(extra_net_naming_map))
meson: make net.naming-scheme= default configurable This is useful for distributions, where the stability of interface names should be preseved after an upgrade of systemd. So when some specific release of the distro is made available, systemd defaults to the latest & greatest naming scheme, and subsequent updates set the same default. This default may still be overriden through the kernel and env var options. A special value "latest" is also allowed. Without a specific name, it is harder to verride from meson. In case of 'combo' options, meson reads the default during the initial configuration, and "remembers" this choice. When systemd is updated, old build/ directories could keep the old default, which would be annoying. Hence, "latest" is introduced to make it explicit, yet follow the upstream. This is actually useful for the user too, because it may be used as an override, without having to actually specify a version.
2018-12-11 23:28:29 +01:00
default_net_naming_scheme = get_option('default-net-naming-scheme')
test: make sure that the default naming scheme name maps back to itself We were testing the that C constant is defined, but we weren't actually testing that the string name maps back to itself. This would catch the issue fixed by the grandparent commit. The test for the default name is moved to the test file to keep the tests together. The define is renamed to not have "_TEST" in the name. The issue here is complicated by the fact that we allow downstreams to inject additional fields, so we don't know the name of the default scheme if it not set with -Ddefault-net-naming-scheme=, so _DEFAULT_NET_NAMING_SCHEME[_TEST] is not defined in all cases, but at least in principle it could be used in other places. If it exists, it is fully valid.
2023-10-16 12:53:10 +02:00
conf.set_quoted('DEFAULT_NET_NAMING_SCHEME', default_net_naming_scheme,
description : 'Default naming scheme as a string')
meson: drop the list of valid net naming schemes We used 'combo' type for the scheme list. For a while we forgot to add new names, and recently aa0a23ec86 added v241, v243, v245, and v247. I want to allow defining new values during configuration, which means that we can't use meson to verify the list of options. So any value is allowed, but then two tests are added: one that will fail compilation if some invalid name is given (other than "latest"), and one that converts DEFAULT_NET_NAMING_SCHEME to a NamingScheme pointer.
2021-09-28 09:33:30 +02:00
if default_net_naming_scheme != 'latest'
test: make sure that the default naming scheme name maps back to itself We were testing the that C constant is defined, but we weren't actually testing that the string name maps back to itself. This would catch the issue fixed by the grandparent commit. The test for the default name is moved to the test file to keep the tests together. The define is renamed to not have "_TEST" in the name. The issue here is complicated by the fact that we allow downstreams to inject additional fields, so we don't know the name of the default scheme if it not set with -Ddefault-net-naming-scheme=, so _DEFAULT_NET_NAMING_SCHEME[_TEST] is not defined in all cases, but at least in principle it could be used in other places. If it exists, it is fully valid.
2023-10-16 12:53:10 +02:00
conf.set('_DEFAULT_NET_NAMING_SCHEME',
'NAMING_' + default_net_naming_scheme.underscorify().to_upper(),
description : 'Default naming scheme as a constant')
meson: drop the list of valid net naming schemes We used 'combo' type for the scheme list. For a while we forgot to add new names, and recently aa0a23ec86 added v241, v243, v245, and v247. I want to allow defining new values during configuration, which means that we can't use meson to verify the list of options. So any value is allowed, but then two tests are added: one that will fail compilation if some invalid name is given (other than "latest"), and one that converts DEFAULT_NET_NAMING_SCHEME to a NamingScheme pointer.
2021-09-28 09:33:30 +02:00
endif
meson: make net.naming-scheme= default configurable This is useful for distributions, where the stability of interface names should be preseved after an upgrade of systemd. So when some specific release of the distro is made available, systemd defaults to the latest & greatest naming scheme, and subsequent updates set the same default. This default may still be overriden through the kernel and env var options. A special value "latest" is also allowed. Without a specific name, it is harder to verride from meson. In case of 'combo' options, meson reads the default during the initial configuration, and "remembers" this choice. When systemd is updated, old build/ directories could keep the old default, which would be annoying. Hence, "latest" is introduced to make it explicit, yet follow the upstream. This is actually useful for the user too, because it may be used as an override, without having to actually specify a version.
2018-12-11 23:28:29 +01:00
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
time_epoch = get_option('time-epoch')
meson: use 0 for default uids, gids, and time epoch 0 UID and GID are special, and should not be acceptable for the settings. Hence, we can handle 0 as unset. Strictly speaking, time epoch with 0 is valid, but I guess no one use 0 as a valid value.
2022-07-30 13:44:11 +09:00
if time_epoch <= 0
meson: add check:true/false to all run_command() invocations meson-0.59.4-1.fc35.noarch says: WARNING: You should add the boolean check kwarg to the run_command call. It currently defaults to false, but it will default to true in future releases of meson. See also: https://github.com/mesonbuild/meson/issues/9300
2021-11-17 13:58:53 +01:00
time_epoch = run_command(sh, '-c', 'echo "$SOURCE_DATE_EPOCH"', check : true).stdout().strip()
meson: Copy files with git only in true git repository When mkosi is run from git-worktree(1), the .git is not a repository directory but a textfile pointing to the real git dir (e.g. /home/user/systemd/.git/worktrees/systemd-worktree). This git dir is not bind mounted into build environment and it fails with: > fatal: not a git repository: /home/user/systemd/.git/worktrees/systemd-worktree > test/meson.build:190:16: ERROR: Command `/usr/bin/env -u GIT_WORK_TREE /usr/bin/git --git-dir=/root/src/.git ls-files ':/test/dmidecode-dumps/*.bin'` failed with status 128. There is already a fallback to use shell globbing instead of ls-files, use it with git worktrees as well.
2023-03-01 22:54:06 +01:00
if time_epoch == '' and git.found() and fs.is_dir('.git')
Get SOURCE_EPOCH from the latest git tag instead of NEWS Currently, each change to NEWS triggers a meson reconfigure that changes SOURCE_EPOCH which causes a full rebuild. Since NEWS changes relatively often, we have a full rebuild each time we pull from master even if we pull semi-regularly. This is further compounded when using branches since NEWS has a relatively high chance to differ between branches which causes git to update the modification time, leading to a full rebuild when switching between branches. We fix this by using the creation time of the latest git tag instead.
2020-07-20 20:41:48 +01:00
# If we're in a git repository, use the creation time of the latest git tag.
meson: add check:true/false to all run_command() invocations meson-0.59.4-1.fc35.noarch says: WARNING: You should add the boolean check kwarg to the run_command call. It currently defaults to false, but it will default to true in future releases of meson. See also: https://github.com/mesonbuild/meson/issues/9300
2021-11-17 13:58:53 +01:00
latest_tag = run_command(git, 'describe', '--abbrev=0', '--tags',
check : false)
if latest_tag.returncode() == 0
time_epoch = run_command(
git, 'log', '--no-show-signature', '-1', '--format=%at',
latest_tag.stdout().strip(),
check : false).stdout()
endif
meson: do not choke on time epoch when there are no git tags github ci was failing with: meson.build:685:16: ERROR: String '' cannot be converted to int
2020-07-23 12:23:58 +02:00
endif
if time_epoch == ''
meson: initialize time-epoch to reproducible builds compatible value Debian Policy encourages to preserve timestamps whenever possible in the tarballs, thus stable release updates of systemd usually do not bump NEWS file timestamp. And thus time-epoch remains the same for the lifetime of a release. It would be better, if each new stable release rebuild of systemd would bump the time epoch a bit. But at the same time remain reproducible. SOURCE_DATE_EPOCH is an environmnet variable defined for this purpose. Thus if available, prefer that, instead of the NEWS file modification time. For example, on Debian/Ubuntu under the reproducible builds the SOURCE_DATE_EPOCH is set to the timestamp from the packaging metadata, thus it is incremented on every new stable release update, whilst preserving reproducible builds capability. Reference: https://reproducible-builds.org/docs/timestamps/
2020-05-15 19:16:05 +01:00
NEWS = files('NEWS')
meson: add check:true/false to all run_command() invocations meson-0.59.4-1.fc35.noarch says: WARNING: You should add the boolean check kwarg to the run_command call. It currently defaults to false, but it will default to true in future releases of meson. See also: https://github.com/mesonbuild/meson/issues/9300
2021-11-17 13:58:53 +01:00
time_epoch = run_command(stat, '-c', '%Y', NEWS,
check : true).stdout()
meson: initialize time-epoch to reproducible builds compatible value Debian Policy encourages to preserve timestamps whenever possible in the tarballs, thus stable release updates of systemd usually do not bump NEWS file timestamp. And thus time-epoch remains the same for the lifetime of a release. It would be better, if each new stable release rebuild of systemd would bump the time epoch a bit. But at the same time remain reproducible. SOURCE_DATE_EPOCH is an environmnet variable defined for this purpose. Thus if available, prefer that, instead of the NEWS file modification time. For example, on Debian/Ubuntu under the reproducible builds the SOURCE_DATE_EPOCH is set to the timestamp from the packaging metadata, thus it is incremented on every new stable release update, whilst preserving reproducible builds capability. Reference: https://reproducible-builds.org/docs/timestamps/
2020-05-15 19:16:05 +01:00
endif
meson: strip various strings before converting them to integers "9\n" is not intrinsically a number, although some tools might auto-strip strings before checking if they are a number. It's not guaranteed, anyway.
2022-07-26 21:13:55 -04:00
time_epoch = time_epoch.strip().to_int()
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
endif
conf.set('TIME_EPOCH', time_epoch)
time-set: adjust system clock if rtc is far in future
2021-07-28 12:13:31 +03:00
conf.set('CLOCK_VALID_RANGE_USEC_MAX', get_option('clock-valid-range-usec-max'))
Use new default-user-shell option instead of hard-coding bash in nspawn and user-record Defaults to /bin/bash, no changes in the default configuration The fallback shell for non-root users is as-specified, and the interactive shell for nspawn sessions is started as exec(default-user-shell, "-" + basename(default-user-shell), ...) before falling through to bash and sh
2022-03-24 17:15:39 +01:00
default_user_shell = get_option('default-user-shell')
conf.set_quoted('DEFAULT_USER_SHELL', default_user_shell)
conf.set_quoted('DEFAULT_USER_SHELL_NAME', fs.name(default_user_shell))
Also parse the minimum uid/gid values We don't (and shouldn't I think) look at them when determining the type of the user, but they should be used during user/group allocation. (For example, an admin may specify SYS_UID_MIN==200 to allow statically numbered users that are shared with other systems in the range 1–199.)
2020-09-25 16:50:45 +02:00
foreach tuple : [['system-alloc-uid-min', 'SYS_UID_MIN', 1], # Also see login.defs(5).
['system-uid-max', 'SYS_UID_MAX', 999],
['system-alloc-gid-min', 'SYS_GID_MIN', 1],
['system-gid-max', 'SYS_GID_MAX', 999]]
v = get_option(tuple[0])
meson: use 0 for default uids, gids, and time epoch 0 UID and GID are special, and should not be acceptable for the settings. Hence, we can handle 0 as unset. Strictly speaking, time epoch with 0 is valid, but I guess no one use 0 as a valid value.
2022-07-30 13:44:11 +09:00
if v <= 0
Also parse the minimum uid/gid values We don't (and shouldn't I think) look at them when determining the type of the user, but they should be used during user/group allocation. (For example, an admin may specify SYS_UID_MIN==200 to allow statically numbered users that are shared with other systems in the range 1–199.)
2020-09-25 16:50:45 +02:00
v = run_command(
awk,
'/^\s*@0@\s+/ { uid=$2 } END { print uid }'.format(tuple[1]),
meson: add check:true/false to all run_command() invocations meson-0.59.4-1.fc35.noarch says: WARNING: You should add the boolean check kwarg to the run_command call. It currently defaults to false, but it will default to true in future releases of meson. See also: https://github.com/mesonbuild/meson/issues/9300
2021-11-17 13:58:53 +01:00
'/etc/login.defs',
check : false).stdout().strip()
Also parse the minimum uid/gid values We don't (and shouldn't I think) look at them when determining the type of the user, but they should be used during user/group allocation. (For example, an admin may specify SYS_UID_MIN==200 to allow statically numbered users that are shared with other systems in the range 1–199.)
2020-09-25 16:50:45 +02:00
if v == ''
v = tuple[2]
else
v = v.to_int()
endif
meson: apply defaults if /etc/login.defs doesn't exist Apply defaults for system_{uid,gid}_max even if the /etc/login.defs file doesn't exist (e.g. in Clear Linux with no changes). awk returns an empty string in case the file doesn't exist, causing meson to fail in to_int(). So set the default if output is empty. This makes the BEGIN{} blocks unnecessary, so remove them.
2018-02-18 18:33:16 -08:00
endif
Also parse the minimum uid/gid values We don't (and shouldn't I think) look at them when determining the type of the user, but they should be used during user/group allocation. (For example, an admin may specify SYS_UID_MIN==200 to allow statically numbered users that are shared with other systems in the range 1–199.)
2020-09-25 16:50:45 +02:00
conf.set(tuple[0].underscorify().to_upper(), v)
endforeach
if conf.get('SYSTEM_ALLOC_UID_MIN') >= conf.get('SYSTEM_UID_MAX')
error('Invalid uid allocation range')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
endif
Also parse the minimum uid/gid values We don't (and shouldn't I think) look at them when determining the type of the user, but they should be used during user/group allocation. (For example, an admin may specify SYS_UID_MIN==200 to allow statically numbered users that are shared with other systems in the range 1–199.)
2020-09-25 16:50:45 +02:00
if conf.get('SYSTEM_ALLOC_GID_MIN') >= conf.get('SYSTEM_GID_MAX')
error('Invalid gid allocation range')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
endif
Define uid range for greeter In multi-seat scenarios, a display manager might need to start multiple greeter sessions. But systemd allows at most one graphical session per user. So, display managers now have a range of UIDs to dynamically allocate users for their greeter sessions.
2025-05-21 17:32:03 -04:00
greeter_uid_min = get_option('greeter-uid-min')
greeter_uid_max = get_option('greeter-uid-max')
conf.set('GREETER_UID_MIN', greeter_uid_min)
conf.set('GREETER_UID_MAX', greeter_uid_max)
meson: use integer type in options This bumps the minimum required version of meson to 0.45 and python to 3.5, as integer type option is supported since meson-0.45 and meson-0.45 requires python-3.5.
2018-05-10 16:04:16 +09:00
dynamic_uid_min = get_option('dynamic-uid-min')
dynamic_uid_max = get_option('dynamic-uid-max')
build-sys: make the dynamic UID range, and the container UID range configurable Also, export these ranges in our pkg-config files.
2017-12-02 12:48:31 +01:00
conf.set('DYNAMIC_UID_MIN', dynamic_uid_min)
conf.set('DYNAMIC_UID_MAX', dynamic_uid_max)
meson: use integer type in options This bumps the minimum required version of meson to 0.45 and python to 3.5, as integer type option is supported since meson-0.45 and meson-0.45 requires python-3.5.
2018-05-10 16:04:16 +09:00
container_uid_base_min = get_option('container-uid-base-min')
container_uid_base_max = get_option('container-uid-base-max')
build-sys: make the dynamic UID range, and the container UID range configurable Also, export these ranges in our pkg-config files.
2017-12-02 12:48:31 +01:00
conf.set('CONTAINER_UID_BASE_MIN', container_uid_base_min)
conf.set('CONTAINER_UID_BASE_MAX', container_uid_base_max)
user-classification: add new "foreign" UID range This makes the UID range configurable via build time options, but of course it really shouldn't be changed. The default range I picked is outside even of IPAs current (ridiculously large) allocation ranges, hence hopefully minimizes conflicts.
2024-11-08 12:14:16 +01:00
foreign_uid_base = get_option('foreign-uid-base')
conf.set('FOREIGN_UID_BASE', foreign_uid_base)
meson: print warnings if the "nobody" user/group name is not compatible with the local system At least on Fedora and Debian systems this not obvious to get right, hence warn.
2017-12-05 11:00:24 +01:00
nobody_user = get_option('nobody-user')
nobody_group = get_option('nobody-group')
check nobody user/group validity only when not cross compiling Using `getent' and `id' command in case of cross compiling does not make much sense. This is because it is the host files that are checked. Besides, in some restricted cross compilation environment, these two command may not even be available. This is to avoid host comtamination. So we should only check the validity using getent and id when not cross compiling.
2018-07-23 14:53:09 +08:00
if not meson.is_cross_build()
meson: also skip uid/gid check for nobody user/group when id command not found Follow-up for 8b413ae4060b21ed4712fdad7eba195890740756.
2025-02-07 11:36:46 +09:00
if getent.found()
ret = run_command(getent, 'passwd', '65534', check : false)
if ret.returncode() == 0
name = ret.stdout().split(':')[0]
meson: Skip getent when it's not found
2025-02-05 07:14:20 +00:00
if name != nobody_user
warning('\n' +
meson: using f-strings in meson Those were added before 0.61 which is our minimum version. Dots were dropped from the end of some messages. Most messages did not have the trailing dot. 0.63 added support for multi-line f-strings. We can't use those yet.
2025-10-03 15:26:20 +02:00
f'The local user with the UID 65534 does not match the configured user name "@nobody_user@" of the nobody user (its name is @name@).\n' +
meson: Skip getent when it's not found
2025-02-05 07:14:20 +00:00
'Your build will result in an user table setup that is incompatible with the local system.')
endif
check nobody user/group validity only when not cross compiling Using `getent' and `id' command in case of cross compiling does not make much sense. This is because it is the host files that are checked. Besides, in some restricted cross compilation environment, these two command may not even be available. This is to avoid host comtamination. So we should only check the validity using getent and id when not cross compiling.
2018-07-23 14:53:09 +08:00
endif
meson: print warnings if the "nobody" user/group name is not compatible with the local system At least on Fedora and Debian systems this not obvious to get right, hence warn.
2017-12-05 11:00:24 +01:00
endif
meson: also skip uid/gid check for nobody user/group when id command not found Follow-up for 8b413ae4060b21ed4712fdad7eba195890740756.
2025-02-07 11:36:46 +09:00
if id.found()
ret = run_command(id, '-u', nobody_user, check : false)
if ret.returncode() == 0
uid = ret.stdout().strip().to_int()
if uid != 65534
warning('\n' +
meson: using f-strings in meson Those were added before 0.61 which is our minimum version. Dots were dropped from the end of some messages. Most messages did not have the trailing dot. 0.63 added support for multi-line f-strings. We can't use those yet.
2025-10-03 15:26:20 +02:00
f'The local user with the configured user name "@nobody_user@" of the nobody user does not have UID 65534 (it has @uid@).\n' +
meson: also skip uid/gid check for nobody user/group when id command not found Follow-up for 8b413ae4060b21ed4712fdad7eba195890740756.
2025-02-07 11:36:46 +09:00
'Your build will result in an user table setup that is incompatible with the local system.')
endif
check nobody user/group validity only when not cross compiling Using `getent' and `id' command in case of cross compiling does not make much sense. This is because it is the host files that are checked. Besides, in some restricted cross compilation environment, these two command may not even be available. This is to avoid host comtamination. So we should only check the validity using getent and id when not cross compiling.
2018-07-23 14:53:09 +08:00
endif
meson: print warnings if the "nobody" user/group name is not compatible with the local system At least on Fedora and Debian systems this not obvious to get right, hence warn.
2017-12-05 11:00:24 +01:00
endif
meson: also skip uid/gid check for nobody user/group when id command not found Follow-up for 8b413ae4060b21ed4712fdad7eba195890740756.
2025-02-07 11:36:46 +09:00
if getent.found()
ret = run_command(getent, 'group', '65534', check : false)
if ret.returncode() == 0
name = ret.stdout().split(':')[0]
meson: Skip getent when it's not found
2025-02-05 07:14:20 +00:00
if name != nobody_group
warning('\n' +
meson: using f-strings in meson Those were added before 0.61 which is our minimum version. Dots were dropped from the end of some messages. Most messages did not have the trailing dot. 0.63 added support for multi-line f-strings. We can't use those yet.
2025-10-03 15:26:20 +02:00
f'The local group with the GID 65534 does not match the configured group name "@nobody_group@" of the nobody group (its name is @name@).\n' +
meson: Skip getent when it's not found
2025-02-05 07:14:20 +00:00
'Your build will result in an group table setup that is incompatible with the local system.')
endif
check nobody user/group validity only when not cross compiling Using `getent' and `id' command in case of cross compiling does not make much sense. This is because it is the host files that are checked. Besides, in some restricted cross compilation environment, these two command may not even be available. This is to avoid host comtamination. So we should only check the validity using getent and id when not cross compiling.
2018-07-23 14:53:09 +08:00
endif
meson: print warnings if the "nobody" user/group name is not compatible with the local system At least on Fedora and Debian systems this not obvious to get right, hence warn.
2017-12-05 11:00:24 +01:00
endif
meson: also skip uid/gid check for nobody user/group when id command not found Follow-up for 8b413ae4060b21ed4712fdad7eba195890740756.
2025-02-07 11:36:46 +09:00
if id.found()
ret = run_command(id, '-g', nobody_group, check : false)
if ret.returncode() == 0
gid = ret.stdout().strip().to_int()
if gid != 65534
warning('\n' +
meson: using f-strings in meson Those were added before 0.61 which is our minimum version. Dots were dropped from the end of some messages. Most messages did not have the trailing dot. 0.63 added support for multi-line f-strings. We can't use those yet.
2025-10-03 15:26:20 +02:00
f'The local group with the configured group name "@nobody_group@" of the nobody group does not have GID 65534 (it has @gid@).\n' +
meson: also skip uid/gid check for nobody user/group when id command not found Follow-up for 8b413ae4060b21ed4712fdad7eba195890740756.
2025-02-07 11:36:46 +09:00
'Your build will result in an group table setup that is incompatible with the local system.')
endif
check nobody user/group validity only when not cross compiling Using `getent' and `id' command in case of cross compiling does not make much sense. This is because it is the host files that are checked. Besides, in some restricted cross compilation environment, these two command may not even be available. This is to avoid host comtamination. So we should only check the validity using getent and id when not cross compiling.
2018-07-23 14:53:09 +08:00
endif
meson: print warnings if the "nobody" user/group name is not compatible with the local system At least on Fedora and Debian systems this not obvious to get right, hence warn.
2017-12-05 11:00:24 +01:00
endif
endif
meson: warn if nobody-user and nobody-group are set to different name It may work, but is very strange. So, let's warn about that. v2: Debian uses nobody and nogroup. Do not warn such case.
2017-12-07 17:19:11 +09:00
if nobody_user != nobody_group and not (nobody_user == 'nobody' and nobody_group == 'nogroup')
meson: use warning() method This bumps the required minimum version of meson to 0.44, as `warning()` method is supported since 0.44.
2018-05-10 14:50:52 +09:00
warning('\n' +
meson: using f-strings in meson Those were added before 0.61 which is our minimum version. Dots were dropped from the end of some messages. Most messages did not have the trailing dot. 0.63 added support for multi-line f-strings. We can't use those yet.
2025-10-03 15:26:20 +02:00
f'The configured user name "@nobody_user@" and group name "@nobody_group@" of the nobody user/group are not equivalent.\n' +
meson: use warning() method This bumps the required minimum version of meson to 0.44, as `warning()` method is supported since 0.44.
2018-05-10 14:50:52 +09:00
'Please re-check that both "nobody-user" and "nobody-group" options are correctly set.')
meson: warn if nobody-user and nobody-group are set to different name It may work, but is very strange. So, let's warn about that. v2: Debian uses nobody and nogroup. Do not warn such case.
2017-12-07 17:19:11 +09:00
endif
meson: print warnings if the "nobody" user/group name is not compatible with the local system At least on Fedora and Debian systems this not obvious to get right, hence warn.
2017-12-05 11:00:24 +01:00
conf.set_quoted('NOBODY_USER_NAME', nobody_user)
conf.set_quoted('NOBODY_GROUP_NAME', nobody_group)
build-sys: make the dynamic UID range, and the container UID range configurable Also, export these ranges in our pkg-config files.
2017-12-02 12:48:31 +01:00
meson: allow "soft-static" allocations for uids and gids in the initrd The general idea with users and groups created through sysusers is that an appropriate number is picked when the allocation is made. The number that is selected will be different on each system based on the order of creation of users, installed packages, etc. Since system users and groups are not shared between installations, this generally is not an issue. But it becomes a problem for initrd: some file systems are shared between the initrd and the host (/run and /dev are probably the only ones that matter). If the allocations are different in the host and the initrd, and files survive switch-root, they will have wrong ownership. This makes the gids build-time-configurable for all groups and users where state may survive the switch from initrd to the host. In particular, all "hardware access" groups are like this: files in /dev will be owned by them. Eventually the new udev would change ownership, but there would be a momemnt where the files were owned by the wrong group. The allocations are "soft-static" in the language of Fedora packaging guidelines: the uid/gid will be used if possible, but we'll fall back to a different one. TTY_GID is the exception, because the number is used directly. Similarly, the possibility to configure "soft-static" uids is added for daemons which may usefully run in the initramfs: systemd-network (lease information and interface state is serialized to /run), systemd-resolve (stub files and interface state), systemd-timesync (/run/systemd/timesync). Journal files are owned by the group systemd-journal, and acls are granted for wheel and adm. systemd-oom and systemd-coredump are excluded from this patch: I assume that oomd is not useful in the initrd, and coredump leaves no state (it only creates a pipe in /run?). The defaults are not changed: if nothing is configured, dynamic allocation will be used. I looked at a Debian system, and the numbers are all different than on Fedora. For Fedora, see the list of uids and gids at https://pagure.io/setup/blob/master/f/uidgid. In particular, systemd-network and systemd-resolve got soft-static numbers to make it easy to transition from a non-host-specific initrd to a host system already a few years back (https://bugzilla.redhat.com/show_bug.cgi?id=1102002). I also requested static allocations for sgx, input, render in https://pagure.io/packaging-committee/issue/1078, https://pagure.io/setup/pull-request/27.
2021-05-23 22:00:22 +02:00
static_ugids = []
foreach option : ['adm-gid',
'audio-gid',
'cdrom-gid',
udev: set clock group for PTP and RTC devices Add a new group for clock devices to enable applications like linuxptp to open clocks without root privileges.
2025-01-16 15:34:47 +01:00
'clock-gid',
meson: allow "soft-static" allocations for uids and gids in the initrd The general idea with users and groups created through sysusers is that an appropriate number is picked when the allocation is made. The number that is selected will be different on each system based on the order of creation of users, installed packages, etc. Since system users and groups are not shared between installations, this generally is not an issue. But it becomes a problem for initrd: some file systems are shared between the initrd and the host (/run and /dev are probably the only ones that matter). If the allocations are different in the host and the initrd, and files survive switch-root, they will have wrong ownership. This makes the gids build-time-configurable for all groups and users where state may survive the switch from initrd to the host. In particular, all "hardware access" groups are like this: files in /dev will be owned by them. Eventually the new udev would change ownership, but there would be a momemnt where the files were owned by the wrong group. The allocations are "soft-static" in the language of Fedora packaging guidelines: the uid/gid will be used if possible, but we'll fall back to a different one. TTY_GID is the exception, because the number is used directly. Similarly, the possibility to configure "soft-static" uids is added for daemons which may usefully run in the initramfs: systemd-network (lease information and interface state is serialized to /run), systemd-resolve (stub files and interface state), systemd-timesync (/run/systemd/timesync). Journal files are owned by the group systemd-journal, and acls are granted for wheel and adm. systemd-oom and systemd-coredump are excluded from this patch: I assume that oomd is not useful in the initrd, and coredump leaves no state (it only creates a pipe in /run?). The defaults are not changed: if nothing is configured, dynamic allocation will be used. I looked at a Debian system, and the numbers are all different than on Fedora. For Fedora, see the list of uids and gids at https://pagure.io/setup/blob/master/f/uidgid. In particular, systemd-network and systemd-resolve got soft-static numbers to make it easy to transition from a non-host-specific initrd to a host system already a few years back (https://bugzilla.redhat.com/show_bug.cgi?id=1102002). I also requested static allocations for sgx, input, render in https://pagure.io/packaging-committee/issue/1078, https://pagure.io/setup/pull-request/27.
2021-05-23 22:00:22 +02:00
'dialout-gid',
'disk-gid',
'input-gid',
'kmem-gid',
'kvm-gid',
'lp-gid',
'render-gid',
'sgx-gid',
'tape-gid',
'tty-gid',
'users-gid',
'utmp-gid',
'video-gid',
'wheel-gid',
'systemd-journal-gid',
'systemd-network-uid',
'systemd-resolve-uid',
'systemd-timesync-uid']
name = option.underscorify().to_upper()
val = get_option(option)
meson: fix comment
2024-05-13 15:34:23 +02:00
# Ensure provided GID argument is positive, otherwise fall back to default assignment
meson: use 0 for default uids, gids, and time epoch 0 UID and GID are special, and should not be acceptable for the settings. Hence, we can handle 0 as unset. Strictly speaking, time epoch with 0 is valid, but I guess no one use 0 as a valid value.
2022-07-30 13:44:11 +09:00
conf.set(name, val > 0 ? val : '-')
if val > 0
meson: using f-strings in meson Those were added before 0.61 which is our minimum version. Dots were dropped from the end of some messages. Most messages did not have the trailing dot. 0.63 added support for multi-line f-strings. We can't use those yet.
2025-10-03 15:26:20 +02:00
static_ugids += f'@option@:@val@'
meson: allow "soft-static" allocations for uids and gids in the initrd The general idea with users and groups created through sysusers is that an appropriate number is picked when the allocation is made. The number that is selected will be different on each system based on the order of creation of users, installed packages, etc. Since system users and groups are not shared between installations, this generally is not an issue. But it becomes a problem for initrd: some file systems are shared between the initrd and the host (/run and /dev are probably the only ones that matter). If the allocations are different in the host and the initrd, and files survive switch-root, they will have wrong ownership. This makes the gids build-time-configurable for all groups and users where state may survive the switch from initrd to the host. In particular, all "hardware access" groups are like this: files in /dev will be owned by them. Eventually the new udev would change ownership, but there would be a momemnt where the files were owned by the wrong group. The allocations are "soft-static" in the language of Fedora packaging guidelines: the uid/gid will be used if possible, but we'll fall back to a different one. TTY_GID is the exception, because the number is used directly. Similarly, the possibility to configure "soft-static" uids is added for daemons which may usefully run in the initramfs: systemd-network (lease information and interface state is serialized to /run), systemd-resolve (stub files and interface state), systemd-timesync (/run/systemd/timesync). Journal files are owned by the group systemd-journal, and acls are granted for wheel and adm. systemd-oom and systemd-coredump are excluded from this patch: I assume that oomd is not useful in the initrd, and coredump leaves no state (it only creates a pipe in /run?). The defaults are not changed: if nothing is configured, dynamic allocation will be used. I looked at a Debian system, and the numbers are all different than on Fedora. For Fedora, see the list of uids and gids at https://pagure.io/setup/blob/master/f/uidgid. In particular, systemd-network and systemd-resolve got soft-static numbers to make it easy to transition from a non-host-specific initrd to a host system already a few years back (https://bugzilla.redhat.com/show_bug.cgi?id=1102002). I also requested static allocations for sgx, input, render in https://pagure.io/packaging-committee/issue/1078, https://pagure.io/setup/pull-request/27.
2021-05-23 22:00:22 +02:00
endif
endforeach
sysusers: Provide meson argument to set gid for 'users' group (#7533) To allow better integration with distributions requiring an explicitly set gid for the `users` group, provide the new `-Dusers-gid` option to set to a new numeric value. In the absence of a specified gid, we'll fallback to the default existing behaviour of `-` as the gid value, to automatically assign the next available gid on the system.
2017-12-03 12:28:23 +00:00
meson: generate m4 preprocessor from config.h (#8914)
2018-05-07 18:17:35 +09:00
conf.set10('ENABLE_ADM_GROUP', get_option('adm-group'))
conf.set10('ENABLE_WHEEL_GROUP', get_option('wheel-group'))
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
Do not apply uaccess tag for /dev/kvm if mode is 0666
2018-07-13 23:36:13 +02:00
dev_kvm_mode = get_option('dev-kvm-mode')
meson: use jinja2 for tmpfiles.d templates HAVE_SMACK_RUN_LABEL was dropped back in 348b44372f36010d48d9a7dda14ef67155753a71, so one line in etc.conf was not rendered as expected ;( Checking if names are defined is paying for itself!
2021-05-16 13:35:51 +02:00
conf.set_quoted('DEV_KVM_MODE', dev_kvm_mode) # FIXME: convert to 0o… notation
Do not apply uaccess tag for /dev/kvm if mode is 0666
2018-07-13 23:36:13 +02:00
conf.set10('DEV_KVM_UACCESS', dev_kvm_mode != '0666')
Re-add uaccess tag for /dev/dri/renderD* Setting an access mode != 0666 is explicitly supported via -Dgroup-render-mode In such a case, re-add the uaccess tag. This is basically the same change that was done for /dev/kvm in commit fa53e24130af3a389573acb9585eadbf7192955f and ace5e3111c0b8d8bfd84b32f2c689b0a4d92c061 and partially reverts the changes from 4e15a7343cb389e97f3eb4f49699161862d8b8b2
2019-03-13 23:22:26 +01:00
group_render_mode = get_option('group-render-mode')
meson: use jinja2 for rules.d templates
2021-05-16 15:51:17 +02:00
conf.set_quoted('GROUP_RENDER_MODE', group_render_mode)
Re-add uaccess tag for /dev/dri/renderD* Setting an access mode != 0666 is explicitly supported via -Dgroup-render-mode In such a case, re-add the uaccess tag. This is basically the same change that was done for /dev/kvm in commit fa53e24130af3a389573acb9585eadbf7192955f and ace5e3111c0b8d8bfd84b32f2c689b0a4d92c061 and partially reverts the changes from 4e15a7343cb389e97f3eb4f49699161862d8b8b2
2019-03-13 23:22:26 +01:00
conf.set10('GROUP_RENDER_UACCESS', group_render_mode != '0666')
meson: allow to customize the access mode for tty/pts devices Then, switch the default value to "0600", due to general security concerns about terminals being written to by other users. Closing #35599.
2024-12-16 11:50:53 +09:00
tty_mode = get_option('tty-mode')
# The setting is used as both octal integer and string through STRINGIFY().
# Here, only check if the value starts with '06', and further check will be done in terminal-util.h.
if not tty_mode.startswith('06')
meson: using f-strings in meson Those were added before 0.61 which is our minimum version. Dots were dropped from the end of some messages. Most messages did not have the trailing dot. 0.63 added support for multi-line f-strings. We can't use those yet.
2025-10-03 15:26:20 +02:00
error(f'Unexpected access mode "@tty_mode@" is specified for TTY/PTS device nodes, it must be "06xx"')
meson: allow to customize the access mode for tty/pts devices Then, switch the default value to "0600", due to general security concerns about terminals being written to by other users. Closing #35599.
2024-12-16 11:50:53 +09:00
elif tty_mode != '0600' and tty_mode != '0620'
meson: using f-strings in meson Those were added before 0.61 which is our minimum version. Dots were dropped from the end of some messages. Most messages did not have the trailing dot. 0.63 added support for multi-line f-strings. We can't use those yet.
2025-10-03 15:26:20 +02:00
warning(f'Unexpected access mode "@tty_mode@" is specified for TTY/PTS device nodes, typically it should be "0600" or "0620", proceeding anyway')
meson: allow to customize the access mode for tty/pts devices Then, switch the default value to "0600", due to general security concerns about terminals being written to by other users. Closing #35599.
2024-12-16 11:50:53 +09:00
endif
# Do not use set_quoted() here, so that the value is available as an integer.
conf.set('TTY_MODE', tty_mode)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: fix a bunch of substitutions This makes the diff -r output on autotools and meson installations empty, except for binary files and ## comments.
2017-04-12 19:54:33 -04:00
kill_user_processes = get_option('default-kill-user-processes')
conf.set10('KILL_USER_PROCESSES', kill_user_processes)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-27 20:54:52 -04:00
dns_servers = get_option('dns-servers')
conf.set_quoted('DNS_SERVERS', dns_servers)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-27 20:54:52 -04:00
ntp_servers = get_option('ntp-servers')
conf.set_quoted('NTP_SERVERS', ntp_servers)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
Make default locale a compile time option Default to a locale that's guaranteed to exist everywhere, but let distros override this with something more exotic if they choose to. Closes #11259.
2018-12-28 07:38:36 -05:00
default_locale = get_option('default-locale')
conf.set_quoted('SYSTEMD_DEFAULT_LOCALE', default_locale)
meson: Add nspawn-locale meson option https://github.com/systemd/systemd/pull/23192 caused breakage in Arch Linux's build tooling. Let's give users an opt-out aside from reverting the patch. It's hardly any maintenance work on our side and gives users an easy way to revert the locale change if needed. Of course, by default we still pick C.UTF-8 if the option is not specified.
2022-06-03 13:18:10 +02:00
nspawn_locale = get_option('nspawn-locale')
conf.set_quoted('SYSTEMD_NSPAWN_LOCALE', nspawn_locale)
vconsole: allow setting default keymap through build option Allow defining the default keymap to be used by vconsole-setup through a build option. A template vconsole.conf also gets populated by tmpfiles if it doesn't exist.
2023-01-18 01:21:59 +08:00
default_keymap = get_option('default-keymap')
vconsole: support KEYMAP=kernel for preserving kernel keymap Follow-up for #26089 and #28505 Currently, if default-keymap is not empty, there's no way to ask vconsole-setup to retain the kernel keymap. Let's accept a special value "kernel" for that purpose. Addresses the problem mentioned in https://github.com/systemd/systemd/pull/28505#issuecomment-1663681665
2023-08-03 21:42:00 +08:00
if default_keymap == ''
vconsole-setup: use "@kernel" rather than "kernel" as special string to leave keymap untouched This is a magic string, and we should avoid stepping into the territory of normal keymap names with that, given that users can pick names otherwise freely. Hence, prefix the name with a special char to avoid any namespace issues. Follow-up for: #28660
2023-08-10 15:12:15 +02:00
# We canonicalize empty keymap to '@kernel', as it makes the default value
vconsole: support KEYMAP=kernel for preserving kernel keymap Follow-up for #26089 and #28505 Currently, if default-keymap is not empty, there's no way to ask vconsole-setup to retain the kernel keymap. Let's accept a special value "kernel" for that purpose. Addresses the problem mentioned in https://github.com/systemd/systemd/pull/28505#issuecomment-1663681665
2023-08-03 21:42:00 +08:00
# in the factory provided /etc/vconsole.conf more obvious.
vconsole-setup: use "@kernel" rather than "kernel" as special string to leave keymap untouched This is a magic string, and we should avoid stepping into the territory of normal keymap names with that, given that users can pick names otherwise freely. Hence, prefix the name with a special char to avoid any namespace issues. Follow-up for: #28660
2023-08-10 15:12:15 +02:00
default_keymap = '@kernel'
vconsole: support KEYMAP=kernel for preserving kernel keymap Follow-up for #26089 and #28505 Currently, if default-keymap is not empty, there's no way to ask vconsole-setup to retain the kernel keymap. Let's accept a special value "kernel" for that purpose. Addresses the problem mentioned in https://github.com/systemd/systemd/pull/28505#issuecomment-1663681665
2023-08-03 21:42:00 +08:00
endif
vconsole: allow setting default keymap through build option Allow defining the default keymap to be used by vconsole-setup through a build option. A template vconsole.conf also gets populated by tmpfiles if it doesn't exist.
2023-01-18 01:21:59 +08:00
conf.set_quoted('SYSTEMD_DEFAULT_KEYMAP', default_keymap)
localed: Run locale-gen if available to generate missing locale This change improves integration with distributions using locale-gen to generate missing locale on-demand, like Debian-based distributions (Debian/Ubuntu/PureOS/Tanglu/...) and Arch Linux. We only ever enable new locales for generation, and never disable them. Furthermore, we only generate UTF-8 locale. This feature is only used if explicitly enabled at compile-time, and will also be inert at runtime if the locale-gen binary is missing.
2021-01-08 23:59:38 +01:00
localegen_path = get_option('localegen-path')
if localegen_path != ''
conf.set_quoted('LOCALEGEN_PATH', localegen_path)
endif
units: generate ReadWritePaths= in the template
2021-05-16 20:13:14 +02:00
conf.set10('HAVE_LOCALEGEN', localegen_path != '')
localed: Run locale-gen if available to generate missing locale This change improves integration with distributions using locale-gen to generate missing locale on-demand, like Debian-based distributions (Debian/Ubuntu/PureOS/Tanglu/...) and Arch Linux. We only ever enable new locales for generation, and never disable them. Furthermore, we only generate UTF-8 locale. This feature is only used if explicitly enabled at compile-time, and will also be inert at runtime if the locale-gen binary is missing.
2021-01-08 23:59:38 +01:00
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
conf.set_quoted('GETTEXT_PACKAGE', meson.project_name())
meson: allow WatchdogSec= in services to be configured As discussed on systemd-devel [1], in Fedora we get lots of abrt reports about the watchdog firing [2], but 100% of them seem to be caused by resource starvation in the machine, and never actual deadlocks in the services being monitored. Killing the services not only does not improve anything, but it makes the resource starvation worse, because the service needs cycles to restart, and coredump processing is also fairly expensive. This adds a configuration option to allow the value to be changed. If the setting is not set, there is no change. My plan is to set it to some ridiculusly high value, maybe 1h, to catch cases where a service is actually hanging. [1] https://lists.freedesktop.org/archives/systemd-devel/2019-October/043618.html [2] https://bugzilla.redhat.com/show_bug.cgi?id=1300212
2019-10-25 12:17:24 +02:00
service_watchdog = get_option('service-watchdog')
meson: expand ternary in functions to if statements Per https://github.com/mesonbuild/meson/issues/5003, ternary doesn't always work as function args with older versions of meson. Expand out ternary statements to stay compatible with older versions (< 0.49).
2019-10-25 15:46:21 -07:00
watchdog_value = service_watchdog == '' ? '' : 'WatchdogSec=' + service_watchdog
meson: use jinja2 for unit templates We don't need two (and half) templating systems anymore, yay! I'm keeping the changes minimal, to make the diff manageable. Some enhancements due to a better templating system might be possible in the future. For handling of '## ' — see the next commit.
2021-05-16 11:55:36 +02:00
conf.set_quoted('SERVICE_WATCHDOG', watchdog_value)
meson: allow WatchdogSec= in services to be configured As discussed on systemd-devel [1], in Fedora we get lots of abrt reports about the watchdog firing [2], but 100% of them seem to be caused by resource starvation in the machine, and never actual deadlocks in the services being monitored. Killing the services not only does not improve anything, but it makes the resource starvation worse, because the service needs cycles to restart, and coredump processing is also fairly expensive. This adds a configuration option to allow the value to be changed. If the setting is not set, there is no change. My plan is to set it to some ridiculusly high value, maybe 1h, to catch cases where a service is actually hanging. [1] https://lists.freedesktop.org/archives/systemd-devel/2019-October/043618.html [2] https://bugzilla.redhat.com/show_bug.cgi?id=1300212
2019-10-25 12:17:24 +02:00
meson: use jinja2 for unit templates We don't need two (and half) templating systems anymore, yay! I'm keeping the changes minimal, to make the diff manageable. Some enhancements due to a better templating system might be possible in the future. For handling of '## ' — see the next commit.
2021-05-16 11:55:36 +02:00
conf.set_quoted('SUSHELL', get_option('debug-shell'))
debug-generator: enable custom systemd.debug_shell tty
2019-04-25 12:19:16 +02:00
conf.set_quoted('DEBUGTTY', get_option('debug-tty'))
meson: add a few forgotten substitutions v2: - a few more
2017-04-10 19:06:45 -04:00
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 10:41:51 +02:00
enable_debug_hashmap = false
enable_debug_mmap_cache = false
meson: also add option for debugging siphash
2018-11-23 00:36:35 +09:00
enable_debug_siphash = false
meson: rename -Ddebug to -Ddebug-extra Meson added -Doptimization and -Ddebug options, which obviously causes a conflict with our -Ddebug options. Let's rename it. Fixes #9883.
2018-08-19 19:11:30 +02:00
foreach name : get_option('debug-extra')
meson: use array type option Array type option is supported since 0.44.0.
2018-05-02 13:56:28 +09:00
if name == 'hashmap'
enable_debug_hashmap = true
elif name == 'mmap-cache'
enable_debug_mmap_cache = true
meson: also add option for debugging siphash
2018-11-23 00:36:35 +09:00
elif name == 'siphash'
enable_debug_siphash = true
meson: use array type option Array type option is supported since 0.44.0.
2018-05-02 13:56:28 +09:00
else
meson: using f-strings in meson Those were added before 0.61 which is our minimum version. Dots were dropped from the end of some messages. Most messages did not have the trailing dot. 0.63 added support for multi-line f-strings. We can't use those yet.
2025-10-03 15:26:20 +02:00
message(f'unknown debug option "@name@", ignoring')
meson: use array type option Array type option is supported since 0.44.0.
2018-05-02 13:56:28 +09:00
endif
endforeach
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 10:41:51 +02:00
conf.set10('ENABLE_DEBUG_HASHMAP', enable_debug_hashmap)
conf.set10('ENABLE_DEBUG_MMAP_CACHE', enable_debug_mmap_cache)
meson: also add option for debugging siphash
2018-11-23 00:36:35 +09:00
conf.set10('ENABLE_DEBUG_SIPHASH', enable_debug_siphash)
meson: add -Dlog-trace to set LOG_TRACE The justification is the same as for -Dvalgrind: setting config in meson in this way is easier, because when the value is changed stuff that should be rebuilt is rebuilt.
2018-08-07 17:34:47 +02:00
conf.set10('LOG_TRACE', get_option('log-trace'))
Turn VALGRIND variable into a meson configuration switch Configuration through environment variable is inconvenient with meson, because they cannot be convieniently changed and/or are not preserved during reconfiguration (https://github.com/mesonbuild/meson/issues/1503). This adds -Dvalgrind=true/false, which has the advantage that it can be set at any time with meson configure -Dvalgrind=... and ninja will rebuild targets as necessary. Additional minor advantages are better consistency with the options for hashmap debugging, and typo avoidance with '#if' instead of '#ifdef'.
2018-05-13 22:28:24 +02:00
meson: make user $PATH configurable This partially reverts db11487d1062655f17db54c4d710653f16c87313 (the logic to calculate the correct value is removed, we always use the same setting as for the system manager). Distributions have an easy mechanism to override this if they wish. I think making this configurable is better, because different distros clearly want different defaults here, and making this configurable is nice and clean. If we don't make it configurable, distros which either have to carry patches, or what would be worse, rely on some other configuration mechanism, like /etc/profile. Those other solutions do not apply everywhere (they usually require the shell to be used at some point), so it is better if we provide a nice way to override the default. Fixes #13469.
2019-11-12 15:38:19 +01:00
default_user_path = get_option('user-path')
if default_user_path != ''
conf.set_quoted('DEFAULT_USER_PATH', default_user_path)
endif
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
#####################################################################
threads = dependency('threads')
librt = cc.find_library('rt')
libm = cc.find_library('m')
libdl = cc.find_library('dl')
meson: first try dependency(), then fallback to find_library() This also drops the fallback for libacl, libcap, libcrypt, and libgcrypt, as recent Ubuntu (at least, 20.04 LTS and newer) and Debian (at least, buster and newer) have relevant .pc files. Fixes #28161.
2023-06-27 04:53:13 +09:00
libcap = dependency('libcap')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: always use libatomic if found Semi-quoting https://github.com/systemd/systemd/issues/25057: clang-16 has made the choice to turn on -Werror=implicit-function-declaration,implicit-int. (See Gentoo's tracker bug https://bugs.gentoo.org/870412). Added in commit 132c73b57ad1d363e97e1f4720f0e920826f34e1, systemd now does a check to see if libatomic is needed with some compile/link tests with e.g. __atomic_exchange_1, but the tests don't provide a prototype for __atomic_exchange_1 so with clang-16 the test fails, breaking the build. Let's simplify things by linking to libatomic unconditionally if it is found and seems to work. If actually unneeded, it might be dropped via --as-needed. This seems to work with gcc and clang. declare_dependency() is used instead of cc.find_library(), because the latter picks up a symlink in gcc private directory (e.g. /usr/lib/gcc/x86_64-redhat-linux/12/libatomic.so), and we don't want that. Fixes #25057.
2022-10-19 16:23:41 +02:00
# On some architectures, libatomic is required. But on some installations,
# it is found, but actual linking fails. So let's try to use it opportunistically.
# If it is installed, but not needed, it will be dropped because of --as-needed.
if cc.links('''int main(int argc, char **argv) { return 0; }''',
args : '-latomic',
name : 'libatomic')
libatomic = declare_dependency(link_args : '-latomic')
else
libatomic = []
endif
meson: crypt.h must always exist We require at least crypt_r() exists, and it is provided since glibc-2.0 (and dropped in glibc-2.39) or by libxcrypt, and the function is provided in crypt.h regardless it is provided by glibc or libxcrypt. Hence, we cannot fallback to unistd.h. This makes the condition about crypt.h more strict, and stop compilation earlier when crypt.h does not exist.
2025-07-23 13:13:29 +09:00
libcrypt = dependency('libcrypt', 'libxcrypt', required : false)
if not libcrypt.found()
# fallback to use find_library() if libcrypt is provided by glibc, e.g. for LibreELEC.
libcrypt = cc.find_library('crypt')
endif
meson: test if we have libcrypt_ra We always seem to have either libcrypt_r and not the other two, or all three. So the fallback for libcrypt_ra needs to be based on libcrypt_r.
2020-09-08 20:02:31 +02:00
meson: crypt.h must always exist We require at least crypt_r() exists, and it is provided since glibc-2.0 (and dropped in glibc-2.39) or by libxcrypt, and the function is provided in crypt.h regardless it is provided by glibc or libxcrypt. Hence, we cannot fallback to unistd.h. This makes the condition about crypt.h more strict, and stop compilation earlier when crypt.h does not exist.
2025-07-23 13:13:29 +09:00
foreach func : [
'crypt_ra', # since libxcrypt-4.0.0
'crypt_gensalt_ra', # since libxcrypt-4.0.0
'crypt_preferred_method', # since libxcrypt-4.4.0
]
have = cc.has_function(func, prefix : '''#include <crypt.h>''', args : '-D_GNU_SOURCE',
meson: test if we have libcrypt_ra We always seem to have either libcrypt_r and not the other two, or all three. So the fallback for libcrypt_ra needs to be based on libcrypt_r.
2020-09-08 20:02:31 +02:00
dependencies : libcrypt)
meson: crypt.h must always exist We require at least crypt_r() exists, and it is provided since glibc-2.0 (and dropped in glibc-2.39) or by libxcrypt, and the function is provided in crypt.h regardless it is provided by glibc or libxcrypt. Hence, we cannot fallback to unistd.h. This makes the condition about crypt.h more strict, and stop compilation earlier when crypt.h does not exist.
2025-07-23 13:13:29 +09:00
conf.set10('HAVE_' + func.to_upper(), have)
meson: test if we have libcrypt_ra We always seem to have either libcrypt_r and not the other two, or all three. So the fallback for libcrypt_ra needs to be based on libcrypt_r.
2020-09-08 20:02:31 +02:00
endforeach
meson: Convert bpf-framework to meson feature
2023-08-10 16:33:37 +02:00
bpf_framework = get_option('bpf-framework')
meson: set minimum libbpf/bpftool versions for bpf-gcc We need the following libbpf commits for bpf-gcc compatibility: - https://github.com/libbpf/libbpf/commit/3d484ca47339b23afb2ec7c508ed9a3a5cf1d287 - https://github.com/libbpf/libbpf/commit/b31ca3fa0e62fde6aa66f855136e29e088ad9dde These are first present in libbpf 1.0.0 and bpftool 7.0.0.
2022-10-01 00:09:53 +00:00
bpf_compiler = get_option('bpf-compiler')
meson: Convert bpf-framework to meson feature
2023-08-10 16:33:37 +02:00
libbpf = dependency('libbpf',
required : bpf_framework,
meson: bump libbpf dependency to 1.4.0 when using gcc bpf_core_type_id_kernel() needs libbpf 1.4.0 when building with gcc rather than clang, so bump the dependency accordingly. More precisely, the following change is needed: https://github.com/libbpf/libbpf/commit/b19fdbf1be21a28f88740375a575ebd9dfbea68f Related to: https://github.com/systemd/systemd/issues/31869 Follow-up for 8aee931e7ae1adb01eeac0e1e4c0aef6ed3969ec
2024-04-29 20:25:34 +01:00
version : bpf_compiler == 'gcc' ? '>= 1.4.0' : '>= 0.1.0')
meson, bpf: add HAVE_LIBBPF, BPF_FRAMEWORK options * Add `bpf-framework` feature gate with 'auto', 'true' and 'false' choices * Add libbpf [0] dependency * Search for clang llvm-strip and bpftool binaries in compile time to generate bpf skeleton. For libbpf [0], make 0.2.0 [1] the minimum required version. If libbpf is satisfied, set HAVE_LIBBPF config option to 1. If `bpf-framework` feature gate is set to 'auto', means that whether bpf feature is enabled or now is defined by the presence of all of libbpf, clang, llvm and bpftool in build environment. With 'auto' all dependencies are optional. If the gate is set to `true`, make all of the libbpf, clang and llvm dependencies mandatory. If it's set to `false`, set `BPF_FRAMEWORK` to false and make libbpf dependency optional. libbpf dependency is dynamic followed by the common pattern in systemd. meson, bpf: add build rule for socket_bind program
2020-11-13 17:08:15 -08:00
conf.set10('HAVE_LIBBPF', libbpf.found())
meson: Convert bpf-framework to meson feature
2023-08-10 16:33:37 +02:00
if not libbpf.found()
meson: use boolean for configuration_data.set10() Setting with number is deprecated: ``` meson.build:1008: DEPRECATION: configuration_data.set10 with number. the `set10` method should only be used with booleans ```
2022-05-15 18:10:25 +09:00
conf.set10('BPF_FRAMEWORK', false)
meson, bpf: add HAVE_LIBBPF, BPF_FRAMEWORK options * Add `bpf-framework` feature gate with 'auto', 'true' and 'false' choices * Add libbpf [0] dependency * Search for clang llvm-strip and bpftool binaries in compile time to generate bpf skeleton. For libbpf [0], make 0.2.0 [1] the minimum required version. If libbpf is satisfied, set HAVE_LIBBPF config option to 1. If `bpf-framework` feature gate is set to 'auto', means that whether bpf feature is enabled or now is defined by the presence of all of libbpf, clang, llvm and bpftool in build environment. With 'auto' all dependencies are optional. If the gate is set to `true`, make all of the libbpf, clang and llvm dependencies mandatory. If it's set to `false`, set `BPF_FRAMEWORK` to false and make libbpf dependency optional. libbpf dependency is dynamic followed by the common pattern in systemd. meson, bpf: add build rule for socket_bind program
2020-11-13 17:08:15 -08:00
else
meson: add experimental bpf-gcc compiler support Not fully working but should make it easier to clean up remaining issues.
2022-06-08 10:13:42 +00:00
clang_found = false
clang_supports_bpf = false
bpf_gcc_found = false
meson: set minimum libbpf/bpftool versions for bpf-gcc We need the following libbpf commits for bpf-gcc compatibility: - https://github.com/libbpf/libbpf/commit/3d484ca47339b23afb2ec7c508ed9a3a5cf1d287 - https://github.com/libbpf/libbpf/commit/b31ca3fa0e62fde6aa66f855136e29e088ad9dde These are first present in libbpf 1.0.0 and bpftool 7.0.0.
2022-10-01 00:09:53 +00:00
bpftool_strip = false
meson: add experimental bpf-gcc compiler support Not fully working but should make it easier to clean up remaining issues.
2022-06-08 10:13:42 +00:00
deps_found = false
if bpf_compiler == 'clang'
# Support 'versioned' clang/llvm-strip binaries, as seen on Debian/Ubuntu
# (like clang-10/llvm-strip-10)
if meson.is_cross_build() or cc.get_id() != 'clang' or cc.cmd_array()[0].contains('afl-clang') or cc.cmd_array()[0].contains('hfuzz-clang')
meson: Convert bpf-framework to meson feature
2023-08-10 16:33:37 +02:00
r = find_program('clang',
required : bpf_framework,
version : '>= 10.0.0')
meson: add experimental bpf-gcc compiler support Not fully working but should make it easier to clean up remaining issues.
2022-06-08 10:13:42 +00:00
clang_found = r.found()
if clang_found
meson: bump required version to 0.55.0
2022-05-16 03:22:33 +09:00
clang = r.full_path()
meson: add experimental bpf-gcc compiler support Not fully working but should make it easier to clean up remaining issues.
2022-06-08 10:13:42 +00:00
endif
else
clang_found = true
clang = cc.cmd_array()
meson: use the compiler command array as is Also check if the flags used when building bpf are supported by clang.
2022-01-13 15:43:24 +09:00
endif
meson: check if clang supports bpf
2022-01-13 15:33:46 +09:00
meson: add experimental bpf-gcc compiler support Not fully working but should make it easier to clean up remaining issues.
2022-06-08 10:13:42 +00:00
if clang_found
# Check if 'clang -target bpf' is supported.
clang_supports_bpf = run_command(clang, '-target', 'bpf', '--print-supported-cpus', check : false).returncode() == 0
endif
meson: fix missing failure if bpf-framework was enabled If building with clang and clang does not support bpf, then enabling -Dbpf-framework=enabled would silently drop the feature (even printing bpf-framework: enabled in the meson build recap, and no message anywhere that'd hint at the failure!) This is unexpected, so add check to fail hard in this case. All other code paths (gcc, missing bpftool) properly check for the option, but it is not as easy for a custom command so check explicitly
2024-07-20 20:38:37 +09:00
if bpf_framework.enabled() and not clang_supports_bpf
error('bpf-framework was enabled but clang does not support bpf')
endif
meson: add experimental bpf-gcc compiler support Not fully working but should make it easier to clean up remaining issues.
2022-06-08 10:13:42 +00:00
elif bpf_compiler == 'gcc'
bpf_gcc = find_program('bpf-gcc',
bpf: test with GCC BPF compiler on opensuse
2023-05-29 15:45:13 -06:00
'bpf-none-gcc',
meson: search for 'bpf-unknown-none' too We currently search for 'bpf-gcc' and 'bpf-none-gcc'. Gentoo's sys-devel/bpf-toolchain package uses 'bpf-unknown-none-gcc', as does Fedora's cross-binutils. Search for this name too.
2024-08-24 13:09:47 +01:00
'bpf-unknown-none-gcc',
meson: add experimental bpf-gcc compiler support Not fully working but should make it easier to clean up remaining issues.
2022-06-08 10:13:42 +00:00
required : true,
bpf: stabilize GCC BPF support Now that we have a GCC release which should support our bpf programs lets set the minimum version and stabilize it.
2023-05-29 15:01:34 -06:00
version : '>= 13.1.0')
meson: add experimental bpf-gcc compiler support Not fully working but should make it easier to clean up remaining issues.
2022-06-08 10:13:42 +00:00
bpf_gcc_found = bpf_gcc.found()
meson: check if clang supports bpf
2022-01-13 15:33:46 +09:00
endif
meson: add experimental bpf-gcc compiler support Not fully working but should make it easier to clean up remaining issues.
2022-06-08 10:13:42 +00:00
if clang_supports_bpf or bpf_gcc_found
# Debian installs this in /usr/sbin/ which is not in $PATH.
# We check for 'bpftool' first, honouring $PATH, and in /usr/sbin/ for Debian.
# We use 'bpftool gen object' subcommand for bpftool strip, it was added by d80b2fcbe0a023619e0fc73112f2a02c2662f6ab (v5.13).
meson: use bpftool based strip when available This should be useable in bpftool v5.13 or newer based on: https://github.com/torvalds/linux/commit/d80b2fcbe0a023619e0fc73112f2a02c2662f6ab
2022-01-30 21:47:38 -07:00
bpftool = find_program('bpftool',
'/usr/sbin/bpftool',
meson: Convert bpf-framework to meson feature
2023-08-10 16:33:37 +02:00
required : bpf_framework.enabled() and bpf_compiler == 'gcc',
version : bpf_compiler == 'gcc' ? '>= 7.0.0' : '>= 5.13.0')
meson: use bpftool based strip when available This should be useable in bpftool v5.13 or newer based on: https://github.com/torvalds/linux/commit/d80b2fcbe0a023619e0fc73112f2a02c2662f6ab
2022-01-30 21:47:38 -07:00
meson: add experimental bpf-gcc compiler support Not fully working but should make it easier to clean up remaining issues.
2022-06-08 10:13:42 +00:00
if bpftool.found()
bpftool_strip = true
deps_found = true
meson: set minimum libbpf/bpftool versions for bpf-gcc We need the following libbpf commits for bpf-gcc compatibility: - https://github.com/libbpf/libbpf/commit/3d484ca47339b23afb2ec7c508ed9a3a5cf1d287 - https://github.com/libbpf/libbpf/commit/b31ca3fa0e62fde6aa66f855136e29e088ad9dde These are first present in libbpf 1.0.0 and bpftool 7.0.0.
2022-10-01 00:09:53 +00:00
elif bpf_compiler == 'clang'
meson: add experimental bpf-gcc compiler support Not fully working but should make it easier to clean up remaining issues.
2022-06-08 10:13:42 +00:00
# We require the 'bpftool gen skeleton' subcommand, it was added by 985ead416df39d6fe8e89580cc1db6aa273e0175 (v5.6).
bpftool = find_program('bpftool',
'/usr/sbin/bpftool',
meson: Convert bpf-framework to meson feature
2023-08-10 16:33:37 +02:00
required : bpf_framework,
meson: add experimental bpf-gcc compiler support Not fully working but should make it easier to clean up remaining issues.
2022-06-08 10:13:42 +00:00
version : '>= 5.6.0')
meson: use bpftool based strip when available This should be useable in bpftool v5.13 or newer based on: https://github.com/torvalds/linux/commit/d80b2fcbe0a023619e0fc73112f2a02c2662f6ab
2022-01-30 21:47:38 -07:00
endif
meson: adjust comment about find_program() usage The FIXME was not correct: with 'dirs' we'd put the fixed path first (see the docs for find_program()).
2021-07-27 16:20:40 +02:00
meson: add experimental bpf-gcc compiler support Not fully working but should make it easier to clean up remaining issues.
2022-06-08 10:13:42 +00:00
# We use `llvm-strip` as a fallback if `bpftool gen object` strip support is not available.
if not bpftool_strip and bpftool.found() and clang_supports_bpf
if not meson.is_cross_build()
llvm_strip_bin = run_command(clang, '--print-prog-name', 'llvm-strip',
check : true).stdout().strip()
else
llvm_strip_bin = 'llvm-strip'
endif
meson: Convert bpf-framework to meson feature
2023-08-10 16:33:37 +02:00
llvm_strip = find_program(llvm_strip_bin,
required : bpf_framework,
version : '>= 10.0.0')
meson: add experimental bpf-gcc compiler support Not fully working but should make it easier to clean up remaining issues.
2022-06-08 10:13:42 +00:00
deps_found = llvm_strip.found()
endif
endif
meson: adjust comment about find_program() usage The FIXME was not correct: with 'dirs' we'd put the fixed path first (see the docs for find_program()).
2021-07-27 16:20:40 +02:00
meson, bpf: add HAVE_LIBBPF, BPF_FRAMEWORK options * Add `bpf-framework` feature gate with 'auto', 'true' and 'false' choices * Add libbpf [0] dependency * Search for clang llvm-strip and bpftool binaries in compile time to generate bpf skeleton. For libbpf [0], make 0.2.0 [1] the minimum required version. If libbpf is satisfied, set HAVE_LIBBPF config option to 1. If `bpf-framework` feature gate is set to 'auto', means that whether bpf feature is enabled or now is defined by the presence of all of libbpf, clang, llvm and bpftool in build environment. With 'auto' all dependencies are optional. If the gate is set to `true`, make all of the libbpf, clang and llvm dependencies mandatory. If it's set to `false`, set `BPF_FRAMEWORK` to false and make libbpf dependency optional. libbpf dependency is dynamic followed by the common pattern in systemd. meson, bpf: add build rule for socket_bind program
2020-11-13 17:08:15 -08:00
# Can build BPF program from source code in restricted C
bpf: refactor skeleton generation This should hopefully fix cross compilation for the bpf programs.
2021-08-11 21:59:19 -06:00
conf.set10('BPF_FRAMEWORK', deps_found)
meson, bpf: add HAVE_LIBBPF, BPF_FRAMEWORK options * Add `bpf-framework` feature gate with 'auto', 'true' and 'false' choices * Add libbpf [0] dependency * Search for clang llvm-strip and bpftool binaries in compile time to generate bpf skeleton. For libbpf [0], make 0.2.0 [1] the minimum required version. If libbpf is satisfied, set HAVE_LIBBPF config option to 1. If `bpf-framework` feature gate is set to 'auto', means that whether bpf feature is enabled or now is defined by the presence of all of libbpf, clang, llvm and bpftool in build environment. With 'auto' all dependencies are optional. If the gate is set to `true`, make all of the libbpf, clang and llvm dependencies mandatory. If it's set to `false`, set `BPF_FRAMEWORK` to false and make libbpf dependency optional. libbpf dependency is dynamic followed by the common pattern in systemd. meson, bpf: add build rule for socket_bind program
2020-11-13 17:08:15 -08:00
endif
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
libmount = dependency('mount',
meson: Add libmount feature Let's reduce the dependencies required to build just libsystemd by making libmount optional. The meson disabler feature makes this quite trivial.
2025-06-05 11:05:08 +02:00
version : fuzzer_build ? '>= 0' : '>= 2.30',
disabler : true,
required : get_option('libmount'))
libmount: make dlopen() dependency
2025-09-24 22:36:30 +02:00
libmount_cflags = libmount.partial_dependency(includes: true, compile_args: true)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: Convert fidsk to meson feature By making this a disabler dependency, we can slightly simplify the code and it als fixes the build for -Dfdisk=disabled as we failed to create a fallback empty libshared_fdisk variable.
2023-08-09 18:07:19 +02:00
libfdisk = dependency('fdisk',
version : '>= 2.32',
disabler : true,
required : get_option('fdisk'))
conf.set10('HAVE_LIBFDISK', libfdisk.found())
repart: add new systemd-repart tool Fixes: #14052
2019-12-10 21:31:41 +01:00
meson: Convert more options to meson features The semantics for libidn2 and pwquality have changed slightly: We will pick a preferred one if both are enabled instead of making it an error.
2023-08-10 16:00:55 +02:00
# This prefers pwquality if both are enabled or auto.
feature = get_option('pwquality').disable_auto_if(get_option('passwdqc').enabled())
libpwquality = dependency('pwquality',
version : '>= 1.4.1',
required : feature)
have = libpwquality.found()
if not have
# libpwquality is used for both features for simplicity
libpwquality = dependency('passwdqc',
required : get_option('passwdqc'))
home: add new systemd-homed service that can manage LUKS homes Fixes more or less: https://bugs.freedesktop.org/show_bug.cgi?id=67474
2019-07-04 18:35:39 +02:00
endif
conf.set10('HAVE_PWQUALITY', have)
meson: Convert more options to meson features The semantics for libidn2 and pwquality have changed slightly: We will pick a preferred one if both are enabled instead of making it an error.
2023-08-10 16:00:55 +02:00
conf.set10('HAVE_PASSWDQC', not have and libpwquality.found())
home: add new systemd-homed service that can manage LUKS homes Fixes more or less: https://bugs.freedesktop.org/show_bug.cgi?id=67474
2019-07-04 18:35:39 +02:00
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
libseccomp = dependency('libseccomp',
version : '>= 2.3.1',
required : get_option('seccomp'))
conf.set10('HAVE_SECCOMP', libseccomp.found())
libseccomp: turn into dlopen() dependency
2025-09-08 17:17:05 +02:00
libseccomp_cflags = libseccomp.partial_dependency(includes: true, compile_args: true)
shared: add libpasswdqc support Co-authored-by: Dmitry V. Levin <ldv@altlinux.org> Resolves: #15055
2021-08-06 11:41:01 +03:00
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
libselinux = dependency('libselinux',
version : '>= 2.1.9',
required : get_option('selinux'))
conf.set10('HAVE_SELINUX', libselinux.found())
libselinux: turn into dlopen() dep
2025-09-08 18:31:49 +02:00
libselinux_cflags = libselinux.partial_dependency(includes: true, compile_args: true)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
libapparmor = dependency('libapparmor',
version : '>= 2.13',
required : get_option('apparmor'))
conf.set10('HAVE_APPARMOR', libapparmor.found())
core: dlopen()'ify libapparmor In Arch Linux we currently have a half-baked apparmor support, in particular we cannot link systemd to libapparmor for service context integration, since that will pull apparmor into base system. Hence, let's turn this into a dlopen dep. Ref: https://gitlab.archlinux.org/archlinux/packaging/packages/systemd/-/issues/22
2024-07-07 19:35:40 +02:00
libapparmor_cflags = libapparmor.partial_dependency(includes: true, compile_args: true)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: ignore -Dsmack-run-label= if -Dsmack=false Compilation would fail because we could have HAVE_SMACK_RUN_LABEL without HAVE_SMACK. This doesn't make much sense, so let's just make -Dsmack=false completely disable smack. Also, the logic in smack-setup.c seems dubious: '#ifdef SMACK_RUN_LABEL' would evaluate to true even if -Dsmack-run-label='' is used. I think this was introduced in the conversion to meson: 8b197c3a8a57c3f7c231b39e5660856fd9580c80 added AC_ARG_WITH(smack-run-label, AS_HELP_STRING([--with-smack-run-label=STRING], [run systemd --system with a specific SMACK label]), [AC_DEFINE_UNQUOTED(SMACK_RUN_LABEL, ["$withval"], [Run with a smack label])], []) i.e. it really was undefined if not specified. And it was same still in 72cdb3e783174dcf9223a49f03e3b0e2ca95ddb8 when configure.ac was dropped. So let's use the single conditional HAVE_SMACK_RUN_LABEL everywhere.
2021-09-21 19:53:20 +02:00
have = get_option('smack') and get_option('smack-run-label') != ''
conf.set10('HAVE_SMACK_RUN_LABEL', have)
if have
conf.set_quoted('SMACK_RUN_LABEL', get_option('smack-run-label'))
endif
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
smack: Add DefaultSmackProcessLabel to user.conf and system.conf DefaultSmackProcessLabel tells systemd what label to assign to its child process in case SmackProcessLabel is not set in the service file. By default, when DefaultSmackProcessLabel is not set child processes inherit label from systemd. If DefaultSmackProcessLabel is set to "/" (which is an invalid character for a SMACK label) the DEFAULT_SMACK_PROCESS_LABEL set during compilation is ignored and systemd act as if the option was unset.
2022-07-06 13:09:51 +02:00
have = get_option('smack') and get_option('smack-default-process-label') != ''
if have
conf.set_quoted('SMACK_DEFAULT_PROCESS_LABEL', get_option('smack-default-process-label'))
endif
meson: Convert options to meson features (find_library fallback) This uses a two-step approach to make sure we can fall back to find_library(), while also skipping the detection if the features are explicitly disabled.
2023-08-09 19:52:57 +02:00
feature = get_option('polkit')
libpolkit = dependency('polkit-gobject-1',
required : feature.disabled() ? feature : false)
install_polkit = feature.allowed()
install_polkit_pkla = libpolkit.found() and libpolkit.version().version_compare('< 0.106')
if install_polkit_pkla
message('Old polkit detected, will install pkla files')
meson: properly conditionalize polkit installation ... including pkla installation on Debian. v2: - fix polkit-gobject-1 pkgconfig name
2017-04-12 19:09:26 -04:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 10:41:51 +02:00
conf.set10('ENABLE_POLKIT', install_polkit)
meson: properly conditionalize polkit installation ... including pkla installation on Debian. v2: - fix polkit-gobject-1 pkgconfig name
2017-04-12 19:09:26 -04:00
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
libacl = dependency('libacl',
required : get_option('acl'))
conf.set10('HAVE_ACL', libacl.found())
acl: turn libacl dep into a dlopen() one I initially didn't think it would be worth doing this, but I changed my mind. People out there quite successfully build systemd without ACL support, and that suggests life without it is quite possible. Moreover we only use it as very specific places: 1. in udev/logind for "uaccess" mgmt 2. in tmpfiles to implement explicitly configured acl changes 3. in journald/coredump/pstore to manage access to unpriv users 4. in pid1 to manage access to credential files 5. when shifting UIDs of container trees I specific container environments it should be entirely fine to live without all of these, hence let's pull this in on demand only.
2025-09-23 11:46:49 +02:00
libacl_cflags = libacl.partial_dependency(includes: true, compile_args: true)
meson: add option to disable libacl support
2017-04-21 13:53:59 -04:00
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
libaudit = dependency('audit',
required : get_option('audit'))
conf.set10('HAVE_AUDIT', libaudit.found())
build: make libaudit dep dlopen()
2025-09-08 12:26:35 +02:00
libaudit_cflags = libaudit.partial_dependency(includes: true, compile_args: true)
udev-builtin-blkid: add support for --hint offsets The next libblkid v2.37 is going to support session offsets for multi-session CD/DVDs. This feature is implemented by "hint offsets". These offsets are optional and prober specific (e.g., iso, udf, ...). For this purpose, the library provides a new function blkid_probe_set_hint(), and blkid(8) provides a new command-line option --hint <name>=<offset>. For CD/DVD, the offset name is "session_offset". The difference between classic --offset and the new --hint is that --offset is very restrictive and defines the probing area and the rest of the device is invisible to the library. The new --hint works like a suggestion, it provides a hint where the user assumes the filesystem, but the rest of the device is still readable for the library (for example, to get some additional superblock information etc.). If the --hint is without a value then it defaults to zero. The option --hint implementation in udev-builtin-blkid.c is backwardly compatible. If compiled against old libblkid, then the option is used in the same way as --offset. Addresses: https://github.com/karelzak/util-linux/issues/1161 Addresses: https://github.com/systemd/systemd/pull/17424
2020-11-30 11:38:21 +01:00
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
libblkid = dependency('blkid',
required : get_option('blkid'))
conf.set10('HAVE_BLKID', libblkid.found())
conf.set10('HAVE_BLKID_PROBE_SET_HINT',
libblkid.found() and cc.has_function('blkid_probe_set_hint', dependencies : libblkid))
libblkid: turn into dlopen() based dep
2025-09-08 19:17:40 +02:00
libblkid_cflags = libblkid.partial_dependency(includes: true, compile_args: true)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
libkmod = dependency('libkmod',
version : '>= 15',
required : get_option('kmod'))
conf.set10('HAVE_KMOD', libkmod.found())
libkmod: turn into dlopen() dependency As it turns out libkmod has quite a bunch of deps, including various compressing libs and similar. By turning this into a dlopen() dependency, we can make our depchain during install time quite a bit smaller. In particular as inside of containers kmod doesn't help anyway as CAP_SYS_MODULE is not available anyway. While we are at it, also share the code that sets up logging/kmod context. After: $ lddtree ./build/systemd systemd => ./build/systemd (interpreter => /lib64/ld-linux-x86-64.so.2) libsystemd-core-255.so => ./build/src/core/libsystemd-core-255.so libaudit.so.1 => /lib64/libaudit.so.1 libcap-ng.so.0 => /lib64/libcap-ng.so.0 ld-linux-x86-64.so.2 => /lib64/ld-linux-x86-64.so.2 libm.so.6 => /lib64/libm.so.6 libmount.so.1 => /lib64/libmount.so.1 libblkid.so.1 => /lib64/libblkid.so.1 libseccomp.so.2 => /lib64/libseccomp.so.2 libselinux.so.1 => /lib64/libselinux.so.1 libpcre2-8.so.0 => /lib64/libpcre2-8.so.0 libsystemd-shared-255.so => /home/lennart/projects/systemd/build/src/shared/libsystemd-shared-255.so libacl.so.1 => /lib64/libacl.so.1 libattr.so.1 => /lib64/libattr.so.1 libcap.so.2 => /lib64/libcap.so.2 libcrypt.so.2 => /lib64/libcrypt.so.2 libgcrypt.so.20 => /lib64/libgcrypt.so.20 libgpg-error.so.0 => /lib64/libgpg-error.so.0 liblz4.so.1 => /lib64/liblz4.so.1 libcrypto.so.3 => /lib64/libcrypto.so.3 libz.so.1 => /lib64/libz.so.1 libpam.so.0 => /lib64/libpam.so.0 libeconf.so.0 => /lib64/libeconf.so.0 liblzma.so.5 => /lib64/liblzma.so.5 libzstd.so.1 => /lib64/libzstd.so.1 libc.so.6 => /lib64/libc.so.6 Before: $ lddtree ./build/systemd systemd => ./build/systemd (interpreter => /lib64/ld-linux-x86-64.so.2) libsystemd-core-255.so => ./build/src/core/libsystemd-core-255.so libaudit.so.1 => /lib64/libaudit.so.1 libcap-ng.so.0 => /lib64/libcap-ng.so.0 ld-linux-x86-64.so.2 => /lib64/ld-linux-x86-64.so.2 libkmod.so.2 => /lib64/libkmod.so.2 libzstd.so.1 => /lib64/libzstd.so.1 liblzma.so.5 => /lib64/liblzma.so.5 libz.so.1 => /lib64/libz.so.1 libcrypto.so.3 => /lib64/libcrypto.so.3 libgcc_s.so.1 => /lib64/libgcc_s.so.1 libm.so.6 => /lib64/libm.so.6 libmount.so.1 => /lib64/libmount.so.1 libblkid.so.1 => /lib64/libblkid.so.1 libseccomp.so.2 => /lib64/libseccomp.so.2 libselinux.so.1 => /lib64/libselinux.so.1 libpcre2-8.so.0 => /lib64/libpcre2-8.so.0 libsystemd-shared-255.so => /home/lennart/projects/systemd/build/src/shared/libsystemd-shared-255.so libacl.so.1 => /lib64/libacl.so.1 libattr.so.1 => /lib64/libattr.so.1 libcap.so.2 => /lib64/libcap.so.2 libcrypt.so.2 => /lib64/libcrypt.so.2 libgcrypt.so.20 => /lib64/libgcrypt.so.20 libgpg-error.so.0 => /lib64/libgpg-error.so.0 liblz4.so.1 => /lib64/liblz4.so.1 libpam.so.0 => /lib64/libpam.so.0 libeconf.so.0 => /lib64/libeconf.so.0 libc.so.6 => /lib64/libc.so.6
2024-01-30 14:50:35 +01:00
libkmod_cflags = libkmod.partial_dependency(includes: true, compile_args: true)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
libxenctrl = dependency('xencontrol',
version : '>= 4.9',
required : get_option('xenctrl'))
conf.set10('HAVE_XENCTRL', libxenctrl.found())
meson: make sure we never actually link to libxenctrl We don't make use of any of its symbols, we just want the headers, hence make sure we never try to link against it.
2023-11-02 15:57:34 +01:00
libxenctrl_cflags = libxenctrl.partial_dependency(includes: true, compile_args: true)
shutdown: Add Xen kexec support In the Xen case, it's the hypervisor which manages kexec. We thus have to ask it whether a kernel is loaded, instead of relying on /sys/kernel/kexec_loaded.
2022-11-01 16:53:02 +01:00
meson: Convert options to meson features (find_library fallback) This uses a two-step approach to make sure we can fall back to find_library(), while also skipping the detection if the features are explicitly disabled.
2023-08-09 19:52:57 +02:00
feature = get_option('pam')
libpam = dependency('pam',
required : feature.disabled() ? feature : false)
if not libpam.found()
# Debian older than bookworm and Ubuntu older than 22.10 do not provide the .pc file.
libpam = cc.find_library('pam', required : feature)
endif
libpam_misc = dependency('pam_misc',
required : feature.disabled() ? feature : false)
if not libpam_misc.found()
libpam_misc = cc.find_library('pam_misc', required : feature)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
endif
meson: Convert options to meson features (find_library fallback) This uses a two-step approach to make sure we can fall back to find_library(), while also skipping the detection if the features are explicitly disabled.
2023-08-09 19:52:57 +02:00
conf.set10('HAVE_PAM', libpam.found() and libpam_misc.found())
pam: make libpam a dlopen() based dependency
2025-09-08 16:04:23 +02:00
libpam_cflags = libpam.partial_dependency(includes: true, compile_args: true)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
libmicrohttpd = dependency('libmicrohttpd',
version : '>= 0.9.33',
required : get_option('microhttpd'))
conf.set10('HAVE_MICROHTTPD', libmicrohttpd.found())
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: Convert more options to meson features The semantics for libidn2 and pwquality have changed slightly: We will pick a preferred one if both are enabled instead of making it an error.
2023-08-10 16:00:55 +02:00
libcryptsetup = get_option('libcryptsetup')
libcryptsetup_plugins = get_option('libcryptsetup-plugins')
if libcryptsetup_plugins.enabled()
if libcryptsetup.disabled()
tree-wide: use the usual spelling of "cannot" (There's a bunch more in src/basic/linux/, but those files are copied from the kernel and should not be modified.)
2023-11-08 13:50:39 +01:00
error('libcryptsetup-plugins cannot be requested without libcryptsetup')
meson: first try dependency(), then fallback to find_library() This also drops the fallback for libacl, libcap, libcrypt, and libgcrypt, as recent Ubuntu (at least, 20.04 LTS and newer) and Debian (at least, buster and newer) have relevant .pc files. Fixes #28161.
2023-06-27 04:53:13 +09:00
endif
meson: Convert more options to meson features The semantics for libidn2 and pwquality have changed slightly: We will pick a preferred one if both are enabled instead of making it an error.
2023-08-10 16:00:55 +02:00
libcryptsetup = libcryptsetup_plugins
endif
libcryptsetup = dependency('libcryptsetup',
version : libcryptsetup_plugins.enabled() ? '>= 2.4.0' : '>= 2.0.1',
required : libcryptsetup)
have = libcryptsetup.found()
foreach ident : ['crypt_set_metadata_size',
'crypt_activate_by_signed_key',
'crypt_token_max',
'crypt_reencrypt_init_by_passphrase',
'crypt_reencrypt',
repart: Use crypt_reencrypt_run() if available crypt_reencrypt() is deprecated, so let's look for and prefer crypt_reencrypt_run() if it is available.
2024-06-02 16:24:52 +02:00
'crypt_reencrypt_run',
cryptsetup: Add optional support for linking volume key in keyring. cryptsetup 2.7.0 adds feature to link effective volume key in custom kernel keyring during device activation. It can be used later to pass linked volume key to other services. For example: kdump enabled systems installed on LUKS2 device. This feature allows it to store volume key linked in a kernel keyring to the kdump reserved memory and reuse it to reactivate LUKS2 device in case of kernel crash.
2024-01-31 13:11:21 +01:00
'crypt_set_data_offset',
homework: Always upload volume key to keyring This commit makes homework always upload the LUKS volume key into the kernel keyring. This is different from previous behavior in three notable ways: - Previously, we'd only upload if auto-resize was on. In preparation for upcoming changes, now we always upload - Previously, we'd upload the user's actual password (or a password obtained from a FIDO key or similar). Now, we upload the LUKS volume key itself, to remove a layer of unnecessary indirection. - Previously, Lock() wouldn't remove the key from the kernel keyring. This, of course, defeats the purpose of Lock(), so now it removes the key This commit also allows the LUKS volume to be unlocked using the volume key we obtained from the keyring.
2024-01-31 23:49:24 -05:00
'crypt_set_keyring_to_link',
cryptsetup: use the new crypt_token_set_external_path() API if available Let's make use of libcryptsetup's new crypt_token_set_external_path() API in place of the interposition stuff we have been doing before. Let's kill it entirely, given that this was a developer feature only anyway (and guarded by an appropriate ifdef). Fixes: #30098
2024-09-10 15:40:12 +02:00
'crypt_resume_by_volume_key',
'crypt_token_set_external_path']
meson: Convert more options to meson features The semantics for libidn2 and pwquality have changed slightly: We will pick a preferred one if both are enabled instead of making it an error.
2023-08-10 16:00:55 +02:00
have_ident = have and cc.has_function(
ident,
prefix : '#include <libcryptsetup.h>',
meson: Pass -Wno-deprecated-declarations when detecting libcryptsetup functions Otherwise we fail to detect crypt_reencrypt() if -Werror is used.
2024-06-03 09:31:26 +02:00
# crypt_reencrypt() raises a deprecation warning so make sure -Wno-deprecated-declarations is
# specified otherwise we fail to detect crypt_reencrypt() if -Werror is used.
args : '-Wno-deprecated-declarations',
meson: Convert more options to meson features The semantics for libidn2 and pwquality have changed slightly: We will pick a preferred one if both are enabled instead of making it an error.
2023-08-10 16:00:55 +02:00
dependencies : libcryptsetup)
conf.set10('HAVE_' + ident.to_upper(), have_ident)
endforeach
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 10:41:51 +02:00
conf.set10('HAVE_LIBCRYPTSETUP', have)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: use more features provided by newer meson No functional change. Preparation for future bump of required meson version.
2025-03-05 22:00:41 +09:00
if meson.version().version_compare('>=1.3.0')
meson: fix detection of libcryptsetup functions Meson would generate the following compile test: #define crypt_set_metadata_size meson_disable_define_of_crypt_set_metadata_size #include <limits.h> #undef crypt_set_metadata_size #ifdef __cplusplus extern "C" #endif char crypt_set_metadata_size (void); #if defined __stub_crypt_set_metadata_size || defined __stub___crypt_set_metadata_size fail fail fail this function is not going to work #endif int main(void) { return crypt_set_metadata_size (); } This works fine when the identifier being queried is an actual function. But crypt_token_max() is an inline function, so getting the address would fail, leading to a false negative result. Complation would fail because the function would be defined twice. With this patch, the check is changed to include the header: #include <libcryptsetup.h> #include <limits.h> #if defined __stub_crypt_set_metadata_size || defined __stub___crypt_set_metadata_size fail fail fail this function is not going to work #endif int main(void) { void *a = (void*) &crypt_set_metadata_size; long long b = (long long) a; return (int) b; } which seems to work correctly.
2021-12-30 14:51:44 +01:00
have = (cc.has_function(
'crypt_activate_by_token_pin',
prefix : '#include <libcryptsetup.h>',
meson: use more features provided by newer meson No functional change. Preparation for future bump of required meson version.
2025-03-05 22:00:41 +09:00
dependencies : libcryptsetup,
required : libcryptsetup_plugins) and
meson: fix detection of libcryptsetup functions Meson would generate the following compile test: #define crypt_set_metadata_size meson_disable_define_of_crypt_set_metadata_size #include <limits.h> #undef crypt_set_metadata_size #ifdef __cplusplus extern "C" #endif char crypt_set_metadata_size (void); #if defined __stub_crypt_set_metadata_size || defined __stub___crypt_set_metadata_size fail fail fail this function is not going to work #endif int main(void) { return crypt_set_metadata_size (); } This works fine when the identifier being queried is an actual function. But crypt_token_max() is an inline function, so getting the address would fail, leading to a false negative result. Complation would fail because the function would be defined twice. With this patch, the check is changed to include the header: #include <libcryptsetup.h> #include <limits.h> #if defined __stub_crypt_set_metadata_size || defined __stub___crypt_set_metadata_size fail fail fail this function is not going to work #endif int main(void) { void *a = (void*) &crypt_set_metadata_size; long long b = (long long) a; return (int) b; } which seems to work correctly.
2021-12-30 14:51:44 +01:00
cc.has_function(
'crypt_token_external_path',
prefix : '#include <libcryptsetup.h>',
meson: use more features provided by newer meson No functional change. Preparation for future bump of required meson version.
2025-03-05 22:00:41 +09:00
dependencies : libcryptsetup,
required : libcryptsetup_plugins))
Add support for systemd-tpm2 libcryptsetup plugin. Add support for systemd-tpm2 based LUKS2 device activation via libcryptsetup plugin. This make the feature (tpm2 sealed LUKS2 keyslot passphrase) usable from both systemd utilities and cryptsetup cli. The feature is configured via -Dlibcryptsetup-plugins combo with default value set to 'auto'. It get's enabled automatically when cryptsetup 2.4.0 or later is installed in build system.
2021-03-16 20:13:28 +01:00
else
meson: use more features provided by newer meson No functional change. Preparation for future bump of required meson version.
2025-03-05 22:00:41 +09:00
if libcryptsetup_plugins.allowed()
have = (cc.has_function(
'crypt_activate_by_token_pin',
prefix : '#include <libcryptsetup.h>',
dependencies : libcryptsetup) and
cc.has_function(
'crypt_token_external_path',
prefix : '#include <libcryptsetup.h>',
dependencies : libcryptsetup))
else
have = false
endif
Add support for systemd-tpm2 libcryptsetup plugin. Add support for systemd-tpm2 based LUKS2 device activation via libcryptsetup plugin. This make the feature (tpm2 sealed LUKS2 keyslot passphrase) usable from both systemd utilities and cryptsetup cli. The feature is configured via -Dlibcryptsetup-plugins combo with default value set to 'auto'. It get's enabled automatically when cryptsetup 2.4.0 or later is installed in build system.
2021-03-16 20:13:28 +01:00
endif
conf.set10('HAVE_LIBCRYPTSETUP_PLUGINS', have)
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
libcurl = dependency('libcurl',
version : '>= 7.32.0',
required : get_option('libcurl'))
conf.set10('HAVE_LIBCURL', libcurl.found())
meson: simplify the BUILD_MODE conditional Using a enum is all nice and generic, but at this point it seems unlikely that we'll add further build modes. But having an enum means that we need to include the header file with the enumeration whenerever the conditional is used. I want to use the conditional in log.h, which makes it hard to avoid circular imports.
2021-04-14 13:17:22 +02:00
conf.set10('CURL_NO_OLDIES', conf.get('BUILD_MODE_DEVELOPER') == 1)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: Convert more options to meson features The semantics for libidn2 and pwquality have changed slightly: We will pick a preferred one if both are enabled instead of making it an error.
2023-08-10 16:00:55 +02:00
feature = get_option('libidn2').disable_auto_if(get_option('libidn').enabled())
libidn = dependency('libidn2',
required : feature)
have = libidn.found()
if not have
NEWS: say that libidn2 is experimental (#6335) Handling of "_" and some other details requires more thought: https://gitlab.com/libidn/libidn2/issues/30 Let's switch the default back to libidn and add a note in NEWS.
2017-07-12 03:25:59 -04:00
# libidn is used for both libidn and libidn2 objects
meson: default to use libidn2 if both libidn2 and libidn are installed
2019-06-29 03:13:30 +09:00
libidn = dependency('libidn',
meson: Convert more options to meson features The semantics for libidn2 and pwquality have changed slightly: We will pick a preferred one if both are enabled instead of making it an error.
2023-08-10 16:00:55 +02:00
required : get_option('libidn'))
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
endif
meson: Convert more options to meson features The semantics for libidn2 and pwquality have changed slightly: We will pick a preferred one if both are enabled instead of making it an error.
2023-08-10 16:00:55 +02:00
conf.set10('HAVE_LIBIDN', not have and libidn.found())
conf.set10('HAVE_LIBIDN2', have)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
libqrencode = dependency('libqrencode',
version : '>= 3',
required : get_option('qrencode'))
conf.set10('HAVE_QRENCODE', libqrencode.found())
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: Convert options to meson features (find_library fallback) This uses a two-step approach to make sure we can fall back to find_library(), while also skipping the detection if the features are explicitly disabled.
2023-08-09 19:52:57 +02:00
feature = get_option('gcrypt')
libgcrypt = dependency('libgcrypt',
required : feature)
libgpg_error = dependency('gpg-error',
required : feature.disabled() ? feature : false)
have = libgcrypt.found() and libgpg_error.found()
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 10:41:51 +02:00
if not have
# link to neither of the libs if one is not found
meson.build: reorder tests to match order in meson_options.txt This makes it easier to edit both files.
2017-07-26 14:08:46 -04:00
libgcrypt = []
libgpg_error = []
gcrypt: dlopenify for libsystemd gcrypt is used only for journal sealing operations in libsystemd, so it can be made into a dlopen dependency that is used only on demand. This allows to reduce the footprint of libsystemd in the most common cases. Keep systemd-pull and systemd-resolved with normal linking, as they are executables, and usually built with OpenSSL support anyway.
2024-03-29 23:36:51 +00:00
libgcrypt_cflags = []
else
libgcrypt_cflags = libgcrypt.partial_dependency(includes: true, compile_args: true)
meson.build: reorder tests to match order in meson_options.txt This makes it easier to edit both files.
2017-07-26 14:08:46 -04:00
endif
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 10:41:51 +02:00
conf.set10('HAVE_GCRYPT', have)
meson.build: reorder tests to match order in meson_options.txt This makes it easier to edit both files.
2017-07-26 14:08:46 -04:00
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
libgnutls = dependency('gnutls',
version : '>= 3.1.4',
required : get_option('gnutls'))
conf.set10('HAVE_GNUTLS', libgnutls.found())
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
libopenssl = dependency('openssl',
version : '>= 1.1.0',
required : get_option('openssl'))
conf.set10('HAVE_OPENSSL', libopenssl.found())
resolved: basic OpenSSL support for DNS-over-TLS This provides basic OpenSSL support without optimizations like TCP Fast Open and TLS Session Tickets. Notice only a single SSL library can be enabled at a time and therefore journald functions provided by GnuTLS will be disabled when using OpenSSL. Fixes #9531
2018-07-26 22:47:50 +01:00
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
libp11kit = dependency('p11-kit-1',
version : '>= 0.23.3',
required : get_option('p11kit'))
conf.set10('HAVE_P11KIT', libp11kit.found())
libp11kit_cflags = libp11kit.partial_dependency(includes: true, compile_args: true)
shared: add pkcs11-util.[ch]
2019-11-05 11:49:27 +01:00
meson: Convert options to meson features (require) These options use requre() to conveniently express their dependency requirements.
2023-08-10 15:44:11 +02:00
feature = get_option('libfido2').require(
conf.get('HAVE_OPENSSL') == 1,
error_message : 'openssl required')
libfido2 = dependency('libfido2',
required : feature)
conf.set10('HAVE_LIBFIDO2', libfido2.found())
build-sys: add libfido2 as optional dependency
2020-04-14 15:54:52 +02:00
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
tpm2 = dependency('tss2-esys tss2-rc tss2-mu tss2-tcti-device',
required : get_option('tpm2'))
conf.set10('HAVE_TPM2', tpm2.found())
conf.set10('HAVE_TSS2_ESYS3', tpm2.found() and tpm2.version().version_compare('>= 3.0.0'))
cryptenroll: add support for TPM2 enrolling
2020-11-28 15:27:34 +01:00
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
libdw = dependency('libdw',
required : get_option('elfutils'))
conf.set10('HAVE_ELFUTILS', libdw.found())
# New in elfutils 0.177
conf.set10('HAVE_DWELF_ELF_E_MACHINE_STRING',
libdw.found() and cc.has_function('dwelf_elf_e_machine_string', dependencies : libdw))
coredump: generate properly symbolized stacktrace for containerized processes
2024-07-09 15:21:34 +02:00
# New in elfutils 0.192
conf.set10('HAVE_DWFL_SET_SYSROOT',
libdw.found() and cc.has_function('dwfl_set_sysroot', dependencies : libdw))
elf-util: add function to parse metadata out of ELF objects Parse the packaging metadata from an ELF object, if any, and print a pretty table following the spec defined at: https://systemd.io/COREDUMP_PACKAGE_METADATA/
2021-11-18 00:03:48 +00:00
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
libz = dependency('zlib',
required : get_option('zlib'))
conf.set10('HAVE_ZLIB', libz.found())
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: Convert options to meson features (find_library fallback) This uses a two-step approach to make sure we can fall back to find_library(), while also skipping the detection if the features are explicitly disabled.
2023-08-09 19:52:57 +02:00
feature = get_option('bzip2')
libbzip2 = dependency('bzip2',
required : feature.disabled() ? feature : false)
if not libbzip2.found()
# Debian and Ubuntu do not provide the .pc file.
libbzip2 = cc.find_library('bz2', required : feature)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
endif
meson: Convert options to meson features (find_library fallback) This uses a two-step approach to make sure we can fall back to find_library(), while also skipping the detection if the features are explicitly disabled.
2023-08-09 19:52:57 +02:00
conf.set10('HAVE_BZIP2', libbzip2.found())
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
libxz = dependency('liblzma',
required : get_option('xz'))
conf.set10('HAVE_XZ', libxz.found())
dynamically load compression libraries Dynamically load liblz4, libzstd and liblzma with dlopen(). This helps to reduce the size of the initrd image when these libraries are not really needed.
2024-02-27 21:28:14 +01:00
libxz_cflags = libxz.partial_dependency(includes: true, compile_args: true)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
liblz4 = dependency('liblz4',
version : '>= 1.3.0',
required : get_option('lz4'))
conf.set10('HAVE_LZ4', liblz4.found())
dynamically load compression libraries Dynamically load liblz4, libzstd and liblzma with dlopen(). This helps to reduce the size of the initrd image when these libraries are not really needed.
2024-02-27 21:28:14 +01:00
liblz4_cflags = liblz4.partial_dependency(includes: true, compile_args: true)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
libzstd = dependency('libzstd',
version : '>= 1.4.0',
required : get_option('zstd'))
conf.set10('HAVE_ZSTD', libzstd.found())
dynamically load compression libraries Dynamically load liblz4, libzstd and liblzma with dlopen(). This helps to reduce the size of the initrd image when these libraries are not really needed.
2024-02-27 21:28:14 +01:00
libzstd_cflags = libzstd.partial_dependency(includes: true, compile_args: true)
tree-wide: add new HAVE_COMPRESSION compile time flag let's simplify the checks for ZSTD/LZ4/XZ As suggested: https://github.com/systemd/systemd/pull/16096#discussion_r440705585
2020-06-24 16:33:41 +02:00
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
conf.set10('HAVE_COMPRESSION', libxz.found() or liblz4.found() or libzstd.found())
coredump: add zstandard support for coredumps this will hook libzstd into coredump, using this format as default.
2020-04-12 01:09:05 +02:00
compression: add build-time option to select default Compression and decompression are controlled by the same build flag, so if one wants to use, say, LZ4 to compress, ZSTD has to be disabled, which means one loses the ability to read zstd-compressed journals. Add a default-compression meson option, that allows to select any of the available compression algorithms as the default.
2022-04-09 18:38:06 +01:00
compression = get_option('default-compression')
if compression == 'auto'
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
if libzstd.found()
compression: add build-time option to select default Compression and decompression are controlled by the same build flag, so if one wants to use, say, LZ4 to compress, ZSTD has to be disabled, which means one loses the ability to read zstd-compressed journals. Add a default-compression meson option, that allows to select any of the available compression algorithms as the default.
2022-04-09 18:38:06 +01:00
compression = 'zstd'
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
elif liblz4.found()
compression: add build-time option to select default Compression and decompression are controlled by the same build flag, so if one wants to use, say, LZ4 to compress, ZSTD has to be disabled, which means one loses the ability to read zstd-compressed journals. Add a default-compression meson option, that allows to select any of the available compression algorithms as the default.
2022-04-09 18:38:06 +01:00
compression = 'lz4'
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
elif libxz.found()
compression: add build-time option to select default Compression and decompression are controlled by the same build flag, so if one wants to use, say, LZ4 to compress, ZSTD has to be disabled, which means one loses the ability to read zstd-compressed journals. Add a default-compression meson option, that allows to select any of the available compression algorithms as the default.
2022-04-09 18:38:06 +01:00
compression = 'xz'
else
compression = 'none'
endif
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
elif compression == 'zstd' and not libzstd.found()
compression: add build-time option to select default Compression and decompression are controlled by the same build flag, so if one wants to use, say, LZ4 to compress, ZSTD has to be disabled, which means one loses the ability to read zstd-compressed journals. Add a default-compression meson option, that allows to select any of the available compression algorithms as the default.
2022-04-09 18:38:06 +01:00
error('default-compression=zstd requires zstd')
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
elif compression == 'lz4' and not liblz4.found()
compression: add build-time option to select default Compression and decompression are controlled by the same build flag, so if one wants to use, say, LZ4 to compress, ZSTD has to be disabled, which means one loses the ability to read zstd-compressed journals. Add a default-compression meson option, that allows to select any of the available compression algorithms as the default.
2022-04-09 18:38:06 +01:00
error('default-compression=lz4 requires lz4')
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
elif compression == 'xz' and not libxz.found()
compression: add build-time option to select default Compression and decompression are controlled by the same build flag, so if one wants to use, say, LZ4 to compress, ZSTD has to be disabled, which means one loses the ability to read zstd-compressed journals. Add a default-compression meson option, that allows to select any of the available compression algorithms as the default.
2022-04-09 18:38:06 +01:00
error('default-compression=xz requires xz')
endif
tree-wide: add dlopen ELF notes to all dlopen() deps of ours Use 'recommended' priority for the default compression library, to indicate that it should be prioritized over the other ones, as it will be used to compress journals/core files. Also use 'recommended' for kmod, as systems will likely fail to boot if it's missing from the initrd. Use 'suggested' for everything else. There is one dlopen'ed TPM library that has the name generated at runtime (depending on the driver), so that cannot be added, as it needs to be known at build time. Also when we support multiple ABI versions list them all, as for the same reason we cannot know which one will be used at build time. $ dlopen-notes.py build/libsystemd.so.0.39.0 build/src/shared/libsystemd-shared-256.so libarchive.so.13 suggested libbpf.so.0 suggested libbpf.so.1 suggested libcryptsetup.so.12 suggested libdw.so.1 suggested libelf.so.1 suggested libfido2.so.1 suggested libgcrypt.so.20 suggested libidn2.so.0 suggested libip4tc.so.2 suggested libkmod.so.2 recommended liblz4.so.1 suggested liblzma.so.5 suggested libp11-kit.so.0 suggested libpcre2-8.so.0 suggested libpwquality.so.1 suggested libqrencode.so.3 suggested libqrencode.so.4 suggested libtss2-esys.so.0 suggested libtss2-mu.so.0 suggested libtss2-rc.so.0 suggested libzstd.so.1 recommended Co-authored-by: Luca Boccassi <bluca@debian.org>
2024-04-11 17:50:41 +02:00
# In the dlopen ELF note we save the default compression library with a
# higher priority, so that packages can give it priority over the
# secondary libraries.
conf.set_quoted('COMPRESSION_PRIORITY_ZSTD',
compression == 'zstd' ? 'recommended' : 'suggested')
conf.set_quoted('COMPRESSION_PRIORITY_LZ4',
compression == 'lz4' ? 'recommended' : 'suggested')
conf.set_quoted('COMPRESSION_PRIORITY_XZ',
compression == 'xz' ? 'recommended' : 'suggested')
meson: also use COMPRESSION_NONE for default compression
2022-04-27 20:49:17 +09:00
conf.set('DEFAULT_COMPRESSION', 'COMPRESSION_@0@'.format(compression.to_upper()))
compression: add build-time option to select default Compression and decompression are controlled by the same build flag, so if one wants to use, say, LZ4 to compress, ZSTD has to be disabled, which means one loses the ability to read zstd-compressed journals. Add a default-compression meson option, that allows to select any of the available compression algorithms as the default.
2022-04-09 18:38:06 +01:00
dissect: add --make-archive option to convert DDI to tarball
2024-01-24 12:28:41 +01:00
libarchive = dependency('libarchive',
version : '>= 3.0',
required : get_option('libarchive'))
conf.set10('HAVE_LIBARCHIVE', libarchive.found())
importd: port untarring logic over to libarchive This way we have can expose identical behaviour everywhere, can make use of our atomic replacement calls, and openat() logic, and later apply additional tracks while unpacking, such as putting limits on UID ranges and similar.
2025-07-17 08:27:21 +02:00
conf.set10('HAVE_LIBARCHIVE_UID_IS_SET',
meson: fix HAVE_LIBARCHIVE_* conditions Follow-up for a7c8f92d1f937113a279adbe62399f6f0773473f
2025-10-16 11:10:04 +02:00
libarchive.found() and cc.has_function('archive_entry_uid_is_set', dependencies : libarchive))
importd: port untarring logic over to libarchive This way we have can expose identical behaviour everywhere, can make use of our atomic replacement calls, and openat() logic, and later apply additional tracks while unpacking, such as putting limits on UID ranges and similar.
2025-07-17 08:27:21 +02:00
conf.set10('HAVE_LIBARCHIVE_HARDLINK_IS_SET',
meson: fix HAVE_LIBARCHIVE_* conditions Follow-up for a7c8f92d1f937113a279adbe62399f6f0773473f
2025-10-16 11:10:04 +02:00
libarchive.found() and cc.has_function('archive_entry_hardlink_is_set', dependencies : libarchive))
dissect: add --make-archive option to convert DDI to tarball
2024-01-24 12:28:41 +01:00
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
libxkbcommon = dependency('xkbcommon',
version : '>= 0.3.0',
required : get_option('xkbcommon'))
conf.set10('HAVE_XKBCOMMON', libxkbcommon.found())
meson.build: reorder tests to match order in meson_options.txt This makes it easier to edit both files.
2017-07-26 14:08:46 -04:00
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
libpcre2 = dependency('libpcre2-8',
required : get_option('pcre2'))
conf.set10('HAVE_PCRE2', libpcre2.found())
meson: define tests Tests can be run with 'ninja-build test' or using 'mesontest'. '-Dtests=unsafe' can be used to include the "unsafe" tests in the test suite, same as with autotools. v2: - use more conf.get guards are optional components - declare deps on generated headers for test-{af,arphrd,cap}-list v3: - define environment for tests Most test don't need this, but to be consistent with autotools-based build, and to avoid questions which tests need it and which don't, set the same environment for all tests. v4: - rework test generation Use a list of lists to define each test. This way we can reduce the boilerplate somewhat, although the test listings are still pretty verbose. We can also move the definitions of the tests to the subdirs. Unfortunately some subdirs are included earlier than some of the libraries that test binaries are linked to. So just dump all definitions of all tests that cannot be defined earlier into src/test. The `executable` definitions are still at the top level, so the binaries are compiled into the build root. v5: - tag test-dnssec-complex as manual v6: - fix HAVE_LIBZ typo - add missing libgobject/libgio defs - mark test-qcow2 as manual
2017-04-07 00:19:09 -04:00
meson: Use feature options By using meson features we can replace the handcrafted dependency auto-detection by just passing the value from get_option directly to the required arg for dependency, find_library etc. 'auto' features make the dependency optional, 'enabled' requires it while 'disabled' features will skip detection entirely. Any skipped or not found dependency will just be a no-op when passed to build steps and therefore we can also skip the creation of empty vars. The use of skip_deps for these is dropped here as meson provides a way to disable all optional features in one go by passing '-Dauto_features=disabled'.
2023-08-09 18:15:29 +02:00
libglib = dependency('glib-2.0',
version : '>= 2.22.0',
required : get_option('glib'))
libgobject = dependency('gobject-2.0',
version : '>= 2.22.0',
required : get_option('glib'))
libgio = dependency('gio-2.0',
required : get_option('glib'))
conf.set10('HAVE_GLIB', libglib.found() and libgobject.found() and libgio.found())
meson: define tests Tests can be run with 'ninja-build test' or using 'mesontest'. '-Dtests=unsafe' can be used to include the "unsafe" tests in the test suite, same as with autotools. v2: - use more conf.get guards are optional components - declare deps on generated headers for test-{af,arphrd,cap}-list v3: - define environment for tests Most test don't need this, but to be consistent with autotools-based build, and to avoid questions which tests need it and which don't, set the same environment for all tests. v4: - rework test generation Use a list of lists to define each test. This way we can reduce the boilerplate somewhat, although the test listings are still pretty verbose. We can also move the definitions of the tests to the subdirs. Unfortunately some subdirs are included earlier than some of the libraries that test binaries are linked to. So just dump all definitions of all tests that cannot be defined earlier into src/test. The `executable` definitions are still at the top level, so the binaries are compiled into the build root. v5: - tag test-dnssec-complex as manual v6: - fix HAVE_LIBZ typo - add missing libgobject/libgio defs - mark test-qcow2 as manual
2017-04-07 00:19:09 -04:00
meson: Convert dbus to meson feature Also, there is no need to conditionalize the get_variable() calls because not-found dependencies will just return the passed default value if provided.
2023-08-10 15:16:56 +02:00
libdbus = dependency('dbus-1',
version : '>= 1.3.2',
required : get_option('dbus'))
conf.set10('HAVE_DBUS', libdbus.found())
meson: define tests Tests can be run with 'ninja-build test' or using 'mesontest'. '-Dtests=unsafe' can be used to include the "unsafe" tests in the test suite, same as with autotools. v2: - use more conf.get guards are optional components - declare deps on generated headers for test-{af,arphrd,cap}-list v3: - define environment for tests Most test don't need this, but to be consistent with autotools-based build, and to avoid questions which tests need it and which don't, set the same environment for all tests. v4: - rework test generation Use a list of lists to define each test. This way we can reduce the boilerplate somewhat, although the test listings are still pretty verbose. We can also move the definitions of the tests to the subdirs. Unfortunately some subdirs are included earlier than some of the libraries that test binaries are linked to. So just dump all definitions of all tests that cannot be defined earlier into src/test. The `executable` definitions are still at the top level, so the binaries are compiled into the build root. v5: - tag test-dnssec-complex as manual v6: - fix HAVE_LIBZ typo - add missing libgobject/libgio defs - mark test-qcow2 as manual
2017-04-07 00:19:09 -04:00
meson: Convert dbus to meson feature Also, there is no need to conditionalize the get_variable() calls because not-found dependencies will just return the passed default value if provided.
2023-08-10 15:16:56 +02:00
dbusdatadir = libdbus.get_variable(pkgconfig: 'datadir', default_value: datadir) / 'dbus-1'
meson: obtain dbus directories from pkg-config
2021-12-26 16:19:17 +09:00
dbuspolicydir = get_option('dbuspolicydir')
if dbuspolicydir == ''
dbuspolicydir = dbusdatadir / 'system.d'
endif
dbussessionservicedir = get_option('dbussessionservicedir')
if dbussessionservicedir == ''
meson: Convert dbus to meson feature Also, there is no need to conditionalize the get_variable() calls because not-found dependencies will just return the passed default value if provided.
2023-08-10 15:16:56 +02:00
dbussessionservicedir = libdbus.get_variable(pkgconfig: 'session_bus_services_dir', default_value: dbusdatadir / 'services')
meson: obtain dbus directories from pkg-config
2021-12-26 16:19:17 +09:00
endif
dbussystemservicedir = get_option('dbussystemservicedir')
if dbussystemservicedir == ''
meson: Convert dbus to meson feature Also, there is no need to conditionalize the get_variable() calls because not-found dependencies will just return the passed default value if provided.
2023-08-10 15:16:56 +02:00
dbussystemservicedir = libdbus.get_variable(pkgconfig: 'system_bus_services_dir', default_value: dbusdatadir / 'system-services')
meson: obtain dbus directories from pkg-config
2021-12-26 16:19:17 +09:00
endif
dbus_interfaces_dir = get_option('dbus-interfaces-dir')
if dbus_interfaces_dir == '' or dbus_interfaces_dir == 'yes'
if meson.is_cross_build() and dbus_interfaces_dir != 'yes'
dbus_interfaces_dir = 'no'
warning('Exporting D-Bus interface XML files is disabled during cross build. Pass path or "yes" to force enable.')
else
meson: Convert dbus to meson feature Also, there is no need to conditionalize the get_variable() calls because not-found dependencies will just return the passed default value if provided.
2023-08-10 15:16:56 +02:00
dbus_interfaces_dir = libdbus.get_variable(pkgconfig: 'interfaces_dir', default_value: dbusdatadir / 'interfaces')
meson: obtain dbus directories from pkg-config
2021-12-26 16:19:17 +09:00
endif
endif
dmi: add RISC-V 64bit support
2024-12-04 13:41:14 +01:00
dmi_arches = ['x86', 'x86_64', 'aarch64', 'arm', 'ia64', 'loongarch64', 'mips', 'riscv64']
meson: introduce HAVE_DMI flag The condition is used at several places. Let's introduce a simple flag for that.
2023-07-30 06:32:53 +09:00
conf.set10('HAVE_DMI', host_machine.cpu_family() in dmi_arches)
meson: make DNS-over-TLS support optional This adds dns-over-tls option to meson. If set to 'false', systemd-resolved is not linked with libgnutls.
2018-06-21 01:29:49 +09:00
dns_over_tls = get_option('dns-over-tls')
resolve: always use openssl as backend of DNS-over-TLS This drops support for dns-over-tls=gnutls meson option.
2025-04-02 01:45:51 +09:00
have_openssl = conf.get('HAVE_OPENSSL') == 1
if dns_over_tls == 'false'
meson: fix building with -Ddns-over-tls=false Assigning multiple variables in one line is no valid meson syntax. Without this running meson with -Ddns-over-tls=false fails with: meson.build:1191:8: ERROR: Tried to assign an invalid value to variable.
2018-11-11 11:30:53 +01:00
have = false
resolve: always use openssl as backend of DNS-over-TLS This drops support for dns-over-tls=gnutls meson option.
2025-04-02 01:45:51 +09:00
elif dns_over_tls == 'auto'
have = have_openssl
elif have_openssl
have = true
else
error('DNS-over-TLS support was requested, but OpenSSL support is disabled.')
meson: make DNS-over-TLS support optional This adds dns-over-tls option to meson. If set to 'false', systemd-resolved is not linked with libgnutls.
2018-06-21 01:29:49 +09:00
endif
conf.set10('ENABLE_DNS_OVER_TLS', have)
resolve: rename PrivateDNS to DNSOverTLS PrivateDNS is not considered a good name for this option, so rename it to DNSOverTLS
2018-06-13 20:26:24 +02:00
default_dns_over_tls = get_option('default-dns-over-tls')
meson: make DNS-over-TLS support optional This adds dns-over-tls option to meson. If set to 'false', systemd-resolved is not linked with libgnutls.
2018-06-21 01:29:49 +09:00
if default_dns_over_tls != 'no' and conf.get('ENABLE_DNS_OVER_TLS') == 0
resolved: add strict mode for DNS-over-TLS Add strict mode for DNS-over-TLS, which will require TLS support from the server. Closes #10755
2019-02-18 20:41:46 +01:00
message('default-dns-over-tls cannot be enabled or set to opportunistic when DNS-over-TLS support is disabled. Setting default-dns-over-tls to no.')
resolve: rename PrivateDNS to DNSOverTLS PrivateDNS is not considered a good name for this option, so rename it to DNSOverTLS
2018-06-13 20:26:24 +02:00
default_dns_over_tls = 'no'
resolved: support for DNS-over-TLS Add support for DNS-over-TLS using GnuTLS. To reduce latency also TLS False Start and TLS session resumption is supported.
2018-04-27 17:50:38 +02:00
endif
resolve: rename PrivateDNS to DNSOverTLS PrivateDNS is not considered a good name for this option, so rename it to DNSOverTLS
2018-06-13 20:26:24 +02:00
conf.set('DEFAULT_DNS_OVER_TLS_MODE',
'DNS_OVER_TLS_' + default_dns_over_tls.underscorify().to_upper())
meson: use jinja2 for src/resolve/
2021-05-16 16:55:16 +02:00
conf.set_quoted('DEFAULT_DNS_OVER_TLS_MODE_STR', default_dns_over_tls)
resolved: support for DNS-over-TLS Add support for DNS-over-TLS using GnuTLS. To reduce latency also TLS False Start and TLS session resumption is supported.
2018-04-27 17:50:38 +02:00
meson,resolved: make default LLMNR= and MulticastDNS= values configurable For https://fedoraproject.org/wiki/Changes/systemd-resolved.
2020-04-14 22:18:18 +02:00
default_mdns = get_option('default-mdns')
conf.set('DEFAULT_MDNS_MODE',
'RESOLVE_SUPPORT_' + default_mdns.to_upper())
meson: use jinja2 for src/resolve/
2021-05-16 16:55:16 +02:00
conf.set_quoted('DEFAULT_MDNS_MODE_STR', default_mdns)
meson,resolved: make default LLMNR= and MulticastDNS= values configurable For https://fedoraproject.org/wiki/Changes/systemd-resolved.
2020-04-14 22:18:18 +02:00
default_llmnr = get_option('default-llmnr')
conf.set('DEFAULT_LLMNR_MODE',
'RESOLVE_SUPPORT_' + default_llmnr.to_upper())
meson: use jinja2 for src/resolve/
2021-05-16 16:55:16 +02:00
conf.set_quoted('DEFAULT_LLMNR_MODE_STR', default_llmnr)
meson,resolved: make default LLMNR= and MulticastDNS= values configurable For https://fedoraproject.org/wiki/Changes/systemd-resolved.
2020-04-14 22:18:18 +02:00
meson: Convert options to meson features (require) These options use requre() to conveniently express their dependency requirements.
2023-08-10 15:44:11 +02:00
have = get_option('repart').require(
conf.get('HAVE_LIBFDISK') == 1,
error_message : 'fdisk required').allowed()
repart: add new systemd-repart tool Fixes: #14052
2019-12-10 21:31:41 +01:00
conf.set10('ENABLE_REPART', have)
port string_hashsum from libgcrypt to openssl^gcrypt This allows resolved and importd to be built without libgcrypt. Note that we now say either 'cryptographic library' or 'cryptolib'. Co-authored-by: Zbigniew Jędrzejewski-Szmek <zbyszek@in.waw.pl>
2021-10-27 15:39:48 +02:00
default_dnssec = get_option('default-dnssec')
resolve,import: always use openssl This deprecate cryptolib meson option.
2025-04-02 02:46:06 +09:00
if default_dnssec != 'no' and conf.get('HAVE_OPENSSL') == 0
message('default-dnssec cannot be set to yes or allow-downgrade when openssl is disabled. Setting default-dnssec to no.')
port string_hashsum from libgcrypt to openssl^gcrypt This allows resolved and importd to be built without libgcrypt. Note that we now say either 'cryptographic library' or 'cryptolib'. Co-authored-by: Zbigniew Jędrzejewski-Szmek <zbyszek@in.waw.pl>
2021-10-27 15:39:48 +02:00
default_dnssec = 'no'
endif
conf.set('DEFAULT_DNSSEC_MODE',
'DNSSEC_' + default_dnssec.underscorify().to_upper())
conf.set_quoted('DEFAULT_DNSSEC_MODE_STR', default_dnssec)
meson: Convert options to meson features (require) These options use requre() to conveniently express their dependency requirements.
2023-08-10 15:44:11 +02:00
have = get_option('sysupdate').require(
conf.get('HAVE_OPENSSL') == 1 and
conf.get('HAVE_LIBFDISK') == 1,
error_message : 'fdisk and openssl required').allowed()
sysupdate: add new component "sysupdate"
2020-12-28 15:17:54 +01:00
conf.set10('ENABLE_SYSUPDATE', have)
meson: add separate option for sysupdated, disable in release builds This commit introduces a build-time option to enable/disable sysupdated separately from sysupdate. 'auto' translated to enabled by default in developer builds.
2024-10-30 14:21:09 +01:00
have2 = get_option('sysupdated')
if have2 == 'enabled'
if have
have2 = true
else
error('sysupdated requires sysupdate to be enabled')
endif
elif have2 == 'auto'
have2 = have and conf.get('BUILD_MODE_DEVELOPER') == 1
else
have2 = false
endif
conf.set10('ENABLE_SYSUPDATED', have2)
storagetm: add new systemd-storagetm component This implements a "storage target mode", similar to what MacOS provides since a long time as "Target Disk Mode": https://en.wikipedia.org/wiki/Target_Disk_Mode This implementation is relatively simple: 1. a new generic target "storage-target-mode.target" is added, which when booted into defines the target mode. 2. a small tool and service "systemd-storagetm.service" is added which exposes a specific device or all devices as NVMe-TCP devices over the network. NVMe-TCP appears to be hot shit right now how to expose block devices over the network. And it's really simple to set up via configs, hence our code is relatively short and neat. The idea is that systemd-storagetm.target can be extended sooner or later, for example to expose block devices also as USB mass storage devices and similar, in case the system has "dual mode" USB controller that can also work as device, not just as host. (And people could also plug in sharing as NBD, iSCSI, whatever they want.) How to use this? Boot into your system with a kernel cmdline of "rd.systemd.unit=storage-target-mode.target ip=link-local", and you'll see on screen the precise "nvme connect" command line to make the relevant block devices available locally on some other machine. This all requires that the target mode stuff is included in the initrd of course. And the system will the stay in the initrd forever. Why bother? Primarily three use-cases: 1. Debug a broken system: with very few dependencies during boot get access to the raw block device of a broken machine. 2. Migrate from system to another system, by dd'ing the old to the new directly. 3. Installing an OS remotely on some device (for example via Thunderbolt networking) (And there might be more, for example the ability to boot from a laptop's disk on another system) Limitations: 1. There's no authentication/encryption. Hence: use this on local links only. 2. NVMe target mode on Linux supports r/w operation only. Ideally, we'd have a read-only mode, for security reasons, and default to it. Future love: 1. We should have another mode, where we simply expose the homed LUKS home dirs like that. 2. Some lightweight hookup with plymouth, to display a (shortened) version of the info we write to the console. To test all this, just run: mkosi --kernel-command-line-extra="rd.systemd.unit=storage-target-mode.target" qemu
2023-10-27 14:25:49 +02:00
conf.set10('ENABLE_STORAGETM', get_option('storagetm'))
meson: Convert options to meson features (require) These options use requre() to conveniently express their dependency requirements.
2023-08-10 15:44:11 +02:00
have = get_option('importd').require(
conf.get('HAVE_LIBCURL') == 1 and
resolve,import: always use openssl This deprecate cryptolib meson option.
2025-04-02 02:46:06 +09:00
conf.get('HAVE_OPENSSL') == 1 and
meson: Convert options to meson features (require) These options use requre() to conveniently express their dependency requirements.
2023-08-10 15:44:11 +02:00
conf.get('HAVE_ZLIB') == 1 and
conf.get('HAVE_XZ') == 1,
error_message : 'curl, openssl/grypt, zlib and xz required').allowed()
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 10:41:51 +02:00
conf.set10('ENABLE_IMPORTD', have)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: Convert options to meson features (require) These options use requre() to conveniently express their dependency requirements.
2023-08-10 15:44:11 +02:00
have = get_option('homed').require(
conf.get('HAVE_OPENSSL') == 1 and
conf.get('HAVE_LIBFDISK') == 1 and
homework: Always upload volume key to keyring This commit makes homework always upload the LUKS volume key into the kernel keyring. This is different from previous behavior in three notable ways: - Previously, we'd only upload if auto-resize was on. In preparation for upcoming changes, now we always upload - Previously, we'd upload the user's actual password (or a password obtained from a FIDO key or similar). Now, we upload the LUKS volume key itself, to remove a layer of unnecessary indirection. - Previously, Lock() wouldn't remove the key from the kernel keyring. This, of course, defeats the purpose of Lock(), so now it removes the key This commit also allows the LUKS volume to be unlocked using the volume key we obtained from the keyring.
2024-01-31 23:49:24 -05:00
conf.get('HAVE_LIBCRYPTSETUP') == 1 and
conf.get('HAVE_CRYPT_RESUME_BY_VOLUME_KEY') == 1,
meson: Convert options to meson features (require) These options use requre() to conveniently express their dependency requirements.
2023-08-10 15:44:11 +02:00
error_message : 'openssl, fdisk and libcryptsetup required').allowed()
home: add new systemd-homed service that can manage LUKS homes Fixes more or less: https://bugs.freedesktop.org/show_bug.cgi?id=67474
2019-07-04 18:35:39 +02:00
conf.set10('ENABLE_HOMED', have)
meson, man: do not install pam_systemd_home(8) when pam or homed is disabled Fixes #14725.
2020-02-01 11:01:41 +09:00
have = have and conf.get('HAVE_PAM') == 1
conf.set10('ENABLE_PAM_HOME', have)
meson: Convert more options to meson features The semantics for libidn2 and pwquality have changed slightly: We will pick a preferred one if both are enabled instead of making it an error.
2023-08-10 16:00:55 +02:00
feature = get_option('remote')
journal-remote: Allow building without microhttpd support systemd-journal-remote is useful even if the microhttpd related features are not enabled so let's not skip it entirely if microhttpd is not available.
2025-05-15 13:23:54 +02:00
if feature.enabled()
if conf.get('HAVE_MICROHTTPD') != 1
error('remote support was requested, but microhttpd is not available')
endif
if conf.get('HAVE_LIBCURL') != 1
error('remote support was requested, but libcurl is not available')
endif
endif
# A more minimal version of systemd-journal-remote can always be built, even if neither
# libcurl nor microhttpd are available.
conf.set10('ENABLE_REMOTE', feature.allowed())
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: enable vmspawn by default in developer mode This should also implicitly enabled vmspawn in CI. It wasn't passing even the basic tests, which we didn't see, because it needs to be explicitly enabled.
2023-11-07 21:51:40 +01:00
feature = get_option('vmspawn').disable_auto_if(conf.get('BUILD_MODE_DEVELOPER') == 0)
conf.set10('ENABLE_VMSPAWN', feature.allowed())
meson: add feature flag for nspawn build Other tools have it, nspawn doesn't, add one
2025-03-27 23:51:17 +00:00
feature = get_option('nspawn')
conf.set10('ENABLE_NSPAWN', feature.allowed())
mountfsd: add new systemd-mountfsd component
2023-03-09 12:27:29 +01:00
conf.set10('DEFAULT_MOUNTFSD_TRUSTED_DIRECTORIES', get_option('default-mountfsd-trusted-directories'))
meson: add ENABLE_ANALYZE conditional
2020-08-20 17:35:50 +02:00
foreach term : ['analyze',
'backlight',
meson: generate ENABLE_* names automatically After previous changes, the naming of configuration options and internal defines is consistent.
2017-10-03 13:15:27 +02:00
'binfmt',
tree-wide: Drop gnu-efi This drops all mentions of gnu-efi and its manual build machinery. A future commit will bring bootloader builds back. A new bootloader meson option is now used to control whether to build sd-boot and its userspace tooling.
2023-02-26 14:09:44 +01:00
'compat-mutable-uid-boundaries',
meson: generate ENABLE_* names automatically After previous changes, the naming of configuration options and internal defines is consistent.
2017-10-03 13:15:27 +02:00
'coredump',
meson: add ENABLE_ANALYZE conditional
2020-08-20 17:35:50 +02:00
'efi',
'environment-d',
'firstboot',
'gshadow',
'hibernate',
meson: generate ENABLE_* names automatically After previous changes, the naming of configuration options and internal defines is consistent.
2017-10-03 13:15:27 +02:00
'hostnamed',
meson: add ENABLE_ANALYZE conditional
2020-08-20 17:35:50 +02:00
'hwdb',
'idn',
'ima',
core: load IPE policy on boot IPE is a new LSM being introduced in 6.12. Like IMA, it works based on a policy file that has to be loaded at boot, the earlier the better. So like IMA, if such a policy is present, load it and activate it. If there are any .p7b files in /etc/ipe/, load them as policies. The files have to be inline signed in DER format as per IPE documentation. For more information on the details of IPE: https://microsoft.github.io/ipe/
2024-09-14 14:27:53 +02:00
'ipe',
meson: add ENABLE_ANALYZE conditional
2020-08-20 17:35:50 +02:00
'initrd',
meson: shorten code a bit
2023-06-20 08:06:43 +09:00
'kernel-install',
meson: add ENABLE_ANALYZE conditional
2020-08-20 17:35:50 +02:00
'ldconfig',
meson: generate ENABLE_* names automatically After previous changes, the naming of configuration options and internal defines is consistent.
2017-10-03 13:15:27 +02:00
'localed',
meson: add ENABLE_ANALYZE conditional
2020-08-20 17:35:50 +02:00
'logind',
meson: generate ENABLE_* names automatically After previous changes, the naming of configuration options and internal defines is consistent.
2017-10-03 13:15:27 +02:00
'machined',
mountfsd: add new systemd-mountfsd component
2023-03-09 12:27:29 +01:00
'mountfsd',
meson: generate ENABLE_* names automatically After previous changes, the naming of configuration options and internal defines is consistent.
2017-10-03 13:15:27 +02:00
'networkd',
nsresourced: add new daemon for granting clients user namespaces and assigning resources to them This adds a small, socket-activated Varlink daemon that can delegate UID ranges for user namespaces to clients asking for it. The primary call is AllocateUserRange() where the user passes in an uninitialized userns fd, which is then set up. There are other calls that allow assigning a mount fd to a userns allocated that way, to set up permissions for a cgroup subtree, and to allocate a veth for such a user namespace. Since the UID assignments are supposed to be transitive, i.e. not permanent, care is taken to ensure that users cannot create inodes owned by these UIDs, so that persistancy cannot be acquired. This is implemented via a BPF-LSM module that ensures that any member of a userns allocated that way cannot create files unless the mount it operates on is owned by the userns itself, or is explicitly allowelisted. BPF LSM program with contributions from Alexei Starovoitov.
2023-04-20 19:07:33 +02:00
'nsresourced',
meson: add ENABLE_ANALYZE conditional
2020-08-20 17:35:50 +02:00
'nss-myhostname',
'nss-systemd',
meson: shorten code a bit
2023-06-20 08:06:43 +09:00
'oomd',
meson: add ENABLE_ANALYZE conditional
2020-08-20 17:35:50 +02:00
'portabled',
'pstore',
meson: generate ENABLE_* names automatically After previous changes, the naming of configuration options and internal defines is consistent.
2017-10-03 13:15:27 +02:00
'quotacheck',
meson: add ENABLE_ANALYZE conditional
2020-08-20 17:35:50 +02:00
'randomseed',
'resolve',
'rfkill',
'smack',
tree-wide: Drop gnu-efi This drops all mentions of gnu-efi and its manual build machinery. A future commit will bring bootloader builds back. A new bootloader meson option is now used to control whether to build sd-boot and its userspace tooling.
2023-02-26 14:09:44 +01:00
'sysext',
meson: generate ENABLE_* names automatically After previous changes, the naming of configuration options and internal defines is consistent.
2017-10-03 13:15:27 +02:00
'sysusers',
meson: add ENABLE_ANALYZE conditional
2020-08-20 17:35:50 +02:00
'timedated',
'timesyncd',
meson: generate ENABLE_* names automatically After previous changes, the naming of configuration options and internal defines is consistent.
2017-10-03 13:15:27 +02:00
'tmpfiles',
'tpm',
meson: add ENABLE_ANALYZE conditional
2020-08-20 17:35:50 +02:00
'userdb',
'utmp',
'vconsole',
'xdg-autostart']
meson: generate ENABLE_* names automatically After previous changes, the naming of configuration options and internal defines is consistent.
2017-10-03 13:15:27 +02:00
have = get_option(term)
name = 'ENABLE_' + term.underscorify().to_upper()
conf.set10(name, have)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
endforeach
test/TEST-21-SYSUSERS: turn into a unit test All this test does is manipulate text files in a subdir specified with --testroot. It can be a normal unittest without the overhead of creating a machine image. As a bonus, also test the .standalone version.
2020-09-26 11:58:24 +02:00
enable_sysusers = conf.get('ENABLE_SYSUSERS') == 1
meson: allow building resolved and machined without nss modules This adds -Dnss-resolve= and -Dnss-mymachines= meson options. By using this option, e.g., resolved can be built without nss-resolve. When no nss modules are built, then test-nss is neither built. Also, This changes the option name -Dmyhostname= to -Dnss-myhostname= for consistency to other nss related options. Closes #9596.
2018-07-18 09:25:57 +09:00
foreach tuple : [['nss-mymachines', 'machined'],
['nss-resolve', 'resolve']]
want = get_option(tuple[0])
meson: Convert more options to meson features The semantics for libidn2 and pwquality have changed slightly: We will pick a preferred one if both are enabled instead of making it an error.
2023-08-10 16:00:55 +02:00
if want.allowed()
meson: allow building resolved and machined without nss modules This adds -Dnss-resolve= and -Dnss-mymachines= meson options. By using this option, e.g., resolved can be built without nss-resolve. When no nss modules are built, then test-nss is neither built. Also, This changes the option name -Dmyhostname= to -Dnss-myhostname= for consistency to other nss related options. Closes #9596.
2018-07-18 09:25:57 +09:00
have = get_option(tuple[1])
meson: Convert more options to meson features The semantics for libidn2 and pwquality have changed slightly: We will pick a preferred one if both are enabled instead of making it an error.
2023-08-10 16:00:55 +02:00
if want.enabled() and not have
meson: allow building resolved and machined without nss modules This adds -Dnss-resolve= and -Dnss-mymachines= meson options. By using this option, e.g., resolved can be built without nss-resolve. When no nss modules are built, then test-nss is neither built. Also, This changes the option name -Dmyhostname= to -Dnss-myhostname= for consistency to other nss related options. Closes #9596.
2018-07-18 09:25:57 +09:00
error('@0@ is requested but @1@ is disabled'.format(tuple[0], tuple[1]))
endif
else
have = false
endif
name = 'ENABLE_' + tuple[0].underscorify().to_upper()
conf.set10(name, have)
endforeach
enable_nss = false
foreach term : ['ENABLE_NSS_MYHOSTNAME',
'ENABLE_NSS_MYMACHINES',
'ENABLE_NSS_RESOLVE',
'ENABLE_NSS_SYSTEMD']
if conf.get(term) == 1
enable_nss = true
endif
endforeach
conf.set10('ENABLE_NSS', enable_nss)
meson: generate m4 preprocessor from config.h (#8914)
2018-05-07 18:17:35 +09:00
conf.set10('ENABLE_TIMEDATECTL', get_option('timedated') or get_option('timesyncd'))
timedatectl: add timesync-status and show-timesync commands Closes #1589.
2018-05-03 18:07:43 +09:00
meson: don't install broken tmpfiles config with sshd?confdir == 'no' 20-systemd-ssh-generator.conf expands SSHCONFDIR, which is bogus when we build with -Dsshconfdir=no. Similarly, avoid expanding SSHDCONFDIR in 20-systemd-userdb.conf when building with -Dsshconfdir=no. Follow-up 6c7fc5d5f2.
2024-01-30 11:25:19 +01:00
conf.set10('ENABLE_SSH_PROXY_CONFIG', sshconfdir != 'no')
conf.set10('ENABLE_SSH_USERDB_CONFIG', conf.get('ENABLE_USERDB') == 1 and sshdconfdir != 'no')
test: Only set environment variable if integration tests are enabled. If we set it to '0' if integration tests are not enabled then we can't enable them from the command line since environment from meson takes priority over environment variables from the command line. We also rename the related variables to avoid conflicts with the existing integration_tests variable.
2024-05-02 20:21:29 +02:00
conf.set10('SYSTEMD_SLOW_TESTS_DEFAULT', want_slow_tests)
meson: add option to configure "slow" tests on/off This is just the meson part, no functional change. Use meson -D slow-tests=yes to set the default, or SYSTEMD_SLOW_TESTS=yes build/test-foobar for just one test. Setting the default is more useful for installed tests.
2017-07-12 21:25:17 +00:00
meson: make lines more consistent
2023-12-14 18:35:00 -03:00
#####################################################################
test: Add BCD unit test
2021-12-10 11:55:38 +01:00
ukify: don't install ukify man page if ukify is not enabled Checking for gnu-efi is not enough, since ukify can be explicitly disabled. Resolves: #26274
2023-01-31 21:43:53 +01:00
pymod = import('python')
python = pymod.find_installation('python3', required : true, modules : ['jinja2'])
python_39 = python.language_version().version_compare('>=3.9')
meson: make lines more consistent
2023-12-14 18:35:00 -03:00
#####################################################################
meson: move bpf hookup into main meson build file This way we can use it in systemd-userdbd later on, too.
2023-03-16 15:36:26 +01:00
if conf.get('BPF_FRAMEWORK') == 1
bpf_clang_flags = [
meson: bump C std to gnu17 GCC 8.4 (our baseline) defaults to gnu17 already.
2025-01-16 18:50:54 +01:00
'-std=gnu17',
meson: move bpf hookup into main meson build file This way we can use it in systemd-userdbd later on, too.
2023-03-16 15:36:26 +01:00
'-Wno-compare-distinct-pointer-types',
'-fno-stack-protector',
'-O2',
'-target',
'bpf',
'-g',
'-c',
]
bpf_gcc_flags = [
meson: bump C std to gnu17 GCC 8.4 (our baseline) defaults to gnu17 already.
2025-01-16 18:50:54 +01:00
'-std=gnu17',
meson: move bpf hookup into main meson build file This way we can use it in systemd-userdbd later on, too.
2023-03-16 15:36:26 +01:00
'-fno-stack-protector',
meson: set -fno-ssa-phiopt when building bpf with gcc There are bugs in the kernel verifier that cause legitimate code to be rejected, disabling this optimization makes bpf programs built with a new enough gcc work again. Fixes https://github.com/systemd/systemd/issues/31888
2024-04-04 23:42:45 +01:00
'-fno-ssa-phiopt',
meson: move bpf hookup into main meson build file This way we can use it in systemd-userdbd later on, too.
2023-03-16 15:36:26 +01:00
'-O2',
'-mcpu=v3',
'-mco-re',
'-gbtf',
'-c',
]
meson: copy prefix mapping CFLAGS when building BPF objects Otherwise the filenames will contain variable paths and break reproducibility
2024-04-27 17:55:27 +01:00
# If c_args contains these flags copy them along with the values, in order to avoid breaking
# reproducible builds and other functionality
propagate_cflags = [
'-ffile-prefix-map=',
'-fdebug-prefix-map=',
'-fmacro-prefix-map=',
meson: bpf: propagate 'sysroot' for cross compilation During cross-compilation of systemd, the compiler used to build the bpf's needs to be pointed at the correct include searchpath. Which can be done by passing the corresponding directory in through the cflags; for example in yocto/bitbake this would work: CFLAGS += "--sysroot=${STAGING_DIR_TARGET}" Signed-off-by: Johannes Schneider <johannes.schneider@leica-geosystems.com>
2024-06-20 12:32:18 +02:00
'--sysroot=',
meson: copy prefix mapping CFLAGS when building BPF objects Otherwise the filenames will contain variable paths and break reproducibility
2024-04-27 17:55:27 +01:00
]
foreach opt : c_args
foreach flag : propagate_cflags
if opt.startswith(flag)
bpf_clang_flags += [opt]
bpf_gcc_flags += [opt]
break
endif
endforeach
endforeach
meson: move bpf hookup into main meson build file This way we can use it in systemd-userdbd later on, too.
2023-03-16 15:36:26 +01:00
# Generate defines that are appropriate to tell the compiler what architecture
# we're compiling for. By default we just map meson's cpu_family to __<cpu_family>__.
# This dictionary contains the exceptions where this doesn't work.
#
# C.f. https://mesonbuild.com/Reference-tables.html#cpu-families
# and src/basic/missing_syscall_def.h.
meson: Detect ELF ABI version for bpf build on ppc64 (#38307) On 64-bit POWER, there are multiple versions of the ELF ABI in use. - little-endian powerpc64 is ELFv2 - big-endian powerpc64 is - ELFv2 when using musl - either ELFv1 or ELFv2 when using glibc Previously, the BPF build was hard-coding `-D_CALL_ELF=2`, which is ELFv2. This makes the build fail on ELFv1, similarly to the original issue that necessitated the addition of this flag on powerpc64le. To fix this: 1. Use ELFv1 as the default (when `_CALL_ELF` is not defined, this is the assumption that should be made about the ABI version). 2. Check if the C compiler has `_CALL_ELF` defined, and if it does, override the default with that. That's technically not the *correct* compiler in this situation, but I'm unsure how to get a compiler object for the BPF one from Meson to do the `*_define('_CALL_ELF')` checks with, and they *should* both be targeting the same ABI version anyway. 3. Add the ABI version to the `_CALL_ELF` definition for the BPF compiler flags. This makes a BPF-enabled build succeed on powerpc64 w/ ELFv1 glibc.
2025-07-23 22:42:13 +02:00
# Start with older ABI. When define is missing, we're likely targeting that.
ppc64_elf_version = '1'
if host_machine.cpu_family() == 'ppc64'
# cc doesn't have to be bpf_compiler, but they should be targeting the same ABI
call_elf_value = cc.get_define('_CALL_ELF')
if call_elf_value != ''
ppc64_elf_version = call_elf_value
endif
endif
meson: move bpf hookup into main meson build file This way we can use it in systemd-userdbd later on, too.
2023-03-16 15:36:26 +01:00
cpu_arch_defines = {
meson: add loongarch64's definition to cpu_arch_defines The default definition to add is `-D__loongarch64__`, which is not searched in [bpf_tracing.h](https://github.com/libbpf/libbpf/blob/09b9e83102eb8ab9e540d36b4559c55f3bcdb95d/src/bpf_tracing.h#L68) This may avoid `error: Must specify a BPF target arch via __TARGET_ARCH_xxx` in loongarch64 Signed-off-by: Zhou Qiankang <wszqkzqk@qq.com>
2024-10-28 12:47:20 +08:00
'ppc' : ['-D__powerpc__', '-D__TARGET_ARCH_powerpc'],
meson: Detect ELF ABI version for bpf build on ppc64 (#38307) On 64-bit POWER, there are multiple versions of the ELF ABI in use. - little-endian powerpc64 is ELFv2 - big-endian powerpc64 is - ELFv2 when using musl - either ELFv1 or ELFv2 when using glibc Previously, the BPF build was hard-coding `-D_CALL_ELF=2`, which is ELFv2. This makes the build fail on ELFv1, similarly to the original issue that necessitated the addition of this flag on powerpc64le. To fix this: 1. Use ELFv1 as the default (when `_CALL_ELF` is not defined, this is the assumption that should be made about the ABI version). 2. Check if the C compiler has `_CALL_ELF` defined, and if it does, override the default with that. That's technically not the *correct* compiler in this situation, but I'm unsure how to get a compiler object for the BPF one from Meson to do the `*_define('_CALL_ELF')` checks with, and they *should* both be targeting the same ABI version anyway. 3. Add the ABI version to the `_CALL_ELF` definition for the BPF compiler flags. This makes a BPF-enabled build succeed on powerpc64 w/ ELFv1 glibc.
2025-07-23 22:42:13 +02:00
'ppc64' : ['-D__powerpc64__', '-D__TARGET_ARCH_powerpc', '-D_CALL_ELF=' + ppc64_elf_version],
meson: add loongarch64's definition to cpu_arch_defines The default definition to add is `-D__loongarch64__`, which is not searched in [bpf_tracing.h](https://github.com/libbpf/libbpf/blob/09b9e83102eb8ab9e540d36b4559c55f3bcdb95d/src/bpf_tracing.h#L68) This may avoid `error: Must specify a BPF target arch via __TARGET_ARCH_xxx` in loongarch64 Signed-off-by: Zhou Qiankang <wszqkzqk@qq.com>
2024-10-28 12:47:20 +08:00
'riscv32' : ['-D__riscv', '-D__riscv_xlen=32', '-D__TARGET_ARCH_riscv'],
'riscv64' : ['-D__riscv', '-D__riscv_xlen=64', '-D__TARGET_ARCH_riscv'],
'x86' : ['-D__i386__', '-D__TARGET_ARCH_x86'],
's390x' : ['-D__s390__', '-D__s390x__', '-D__TARGET_ARCH_s390'],
meson: move bpf hookup into main meson build file This way we can use it in systemd-userdbd later on, too.
2023-03-16 15:36:26 +01:00
# For arm, assume hardware fp is available.
meson: add loongarch64's definition to cpu_arch_defines The default definition to add is `-D__loongarch64__`, which is not searched in [bpf_tracing.h](https://github.com/libbpf/libbpf/blob/09b9e83102eb8ab9e540d36b4559c55f3bcdb95d/src/bpf_tracing.h#L68) This may avoid `error: Must specify a BPF target arch via __TARGET_ARCH_xxx` in loongarch64 Signed-off-by: Zhou Qiankang <wszqkzqk@qq.com>
2024-10-28 12:47:20 +08:00
'arm' : ['-D__arm__', '-D__ARM_PCS_VFP', '-D__TARGET_ARCH_arm'],
'loongarch64' : ['-D__loongarch__', '-D__loongarch_grlen=64', '-D__TARGET_ARCH_loongarch']
meson: move bpf hookup into main meson build file This way we can use it in systemd-userdbd later on, too.
2023-03-16 15:36:26 +01:00
}
bpf_arch_flags = cpu_arch_defines.get(host_machine.cpu_family(),
['-D__@0@__'.format(host_machine.cpu_family())])
if bpf_compiler == 'gcc'
bpf_arch_flags += ['-m' + host_machine.endian() + '-endian']
endif
libbpf_include_dir = libbpf.get_variable(pkgconfig : 'includedir')
bpf_o_unstripped_cmd = []
if bpf_compiler == 'clang'
bpf_o_unstripped_cmd += [
clang,
bpf_clang_flags,
bpf_arch_flags,
]
elif bpf_compiler == 'gcc'
bpf_o_unstripped_cmd += [
bpf_gcc,
bpf_gcc_flags,
bpf_arch_flags,
]
endif
bpf_o_unstripped_cmd += ['-I.']
bpf: fix cross build failure on Debian For compiling bpf code, the system include directory needs to be constructed. On Debian-like systems, this requires passing a multiarch directory. Since clang's -dump-machine prints something other that the multiarch triplet, gcc was interrogated earlier, but that also yields a wrong result for cross compilation and was thus skipped resulting in clang not finding asm/types.h. Rather than, -dump-machine we should ask for -print-multiarch (which rarely differs). Whenever gcc is in use, this is right (even for cross building). Since clang does not support -print-multiarch and its -dump-machine never matches Debian's multiarch, we resort to asking gcc when building natively. For cross builds using clang, we are out of luck.
2024-09-30 17:56:18 +02:00
if cc.get_id() == 'gcc' or meson.is_cross_build()
if cc.get_id() != 'gcc'
warning('Cross compiler is not gcc. Guessing the target triplet for bpf likely fails.')
meson: move bpf hookup into main meson build file This way we can use it in systemd-userdbd later on, too.
2023-03-16 15:36:26 +01:00
endif
bpf: fix cross build failure on Debian For compiling bpf code, the system include directory needs to be constructed. On Debian-like systems, this requires passing a multiarch directory. Since clang's -dump-machine prints something other that the multiarch triplet, gcc was interrogated earlier, but that also yields a wrong result for cross compilation and was thus skipped resulting in clang not finding asm/types.h. Rather than, -dump-machine we should ask for -print-multiarch (which rarely differs). Whenever gcc is in use, this is right (even for cross building). Since clang does not support -print-multiarch and its -dump-machine never matches Debian's multiarch, we resort to asking gcc when building natively. For cross builds using clang, we are out of luck.
2024-09-30 17:56:18 +02:00
target_triplet_cmd = run_command(cc.cmd_array(), '-print-multiarch', check: false)
else
# clang does not support -print-multiarch (D133170) and its -dump-machine
# does not match multiarch. Query gcc instead.
target_triplet_cmd = run_command('gcc', '-print-multiarch', check: false)
endif
if target_triplet_cmd.returncode() == 0
meson: prepend sys_root to bpf isystem These sort of absolute include paths are generally unsafe when cross compiling.
2025-07-21 13:04:49 -06:00
sysroot = meson.get_external_property('sys_root', '/')
bpf: fix cross build failure on Debian For compiling bpf code, the system include directory needs to be constructed. On Debian-like systems, this requires passing a multiarch directory. Since clang's -dump-machine prints something other that the multiarch triplet, gcc was interrogated earlier, but that also yields a wrong result for cross compilation and was thus skipped resulting in clang not finding asm/types.h. Rather than, -dump-machine we should ask for -print-multiarch (which rarely differs). Whenever gcc is in use, this is right (even for cross building). Since clang does not support -print-multiarch and its -dump-machine never matches Debian's multiarch, we resort to asking gcc when building natively. For cross builds using clang, we are out of luck.
2024-09-30 17:56:18 +02:00
target_triplet = target_triplet_cmd.stdout().strip()
meson: prepend sys_root to bpf isystem These sort of absolute include paths are generally unsafe when cross compiling.
2025-07-21 13:04:49 -06:00
target_include_dir = sysroot / 'usr' / 'include'
target_triple_include_dir = target_include_dir / target_triplet
isystem_dir = ''
if fs.is_dir(target_triple_include_dir)
isystem_dir = target_triple_include_dir
elif fs.is_dir(target_include_dir)
isystem_dir = target_include_dir
endif
if isystem_dir != ''
bpf_o_unstripped_cmd += [
'-isystem', isystem_dir
]
endif
meson: move bpf hookup into main meson build file This way we can use it in systemd-userdbd later on, too.
2023-03-16 15:36:26 +01:00
endif
bpf_o_unstripped_cmd += [
'-idirafter',
libbpf_include_dir,
'@INPUT@',
'-o',
'@OUTPUT@'
]
if bpftool_strip
bpf_o_cmd = [
bpftool,
'gen',
'object',
'@OUTPUT@',
'@INPUT@'
]
elif bpf_compiler == 'clang'
bpf_o_cmd = [
llvm_strip,
'-g',
'@INPUT@',
'-o',
'@OUTPUT@'
]
endif
skel_h_cmd = [
bpftool,
'gen',
'skeleton',
'@INPUT@'
]
endif
tree-wide: Drop gnu-efi This drops all mentions of gnu-efi and its manual build machinery. A future commit will bring bootloader builds back. A new bootloader meson option is now used to control whether to build sd-boot and its userspace tooling.
2023-02-26 14:09:44 +01:00
#####################################################################
efi_arch = {
boot: Add RISCV32 and LoongArch support This is completely untested, but should work in theory, as it's just adding a couple defines according to the specs.
2023-02-28 18:05:18 +01:00
'aarch64' : 'aa64',
'arm' : 'arm',
'loongarch32' : 'loongarch32',
'loongarch64' : 'loongarch64',
'riscv32' : 'riscv32',
'riscv64' : 'riscv64',
'x86_64' : 'x64',
'x86' : 'ia32',
tree-wide: Drop gnu-efi This drops all mentions of gnu-efi and its manual build machinery. A future commit will bring bootloader builds back. A new bootloader meson option is now used to control whether to build sd-boot and its userspace tooling.
2023-02-26 14:09:44 +01:00
}.get(host_machine.cpu_family(), '')
meson: Convert options to meson features (require) These options use requre() to conveniently express their dependency requirements.
2023-08-10 15:44:11 +02:00
pyelftools = pymod.find_installation('python3',
required : get_option('bootloader'),
modules : ['elftools'])
have = get_option('bootloader').require(
pyelftools.found() and get_option('efi') and efi_arch != '',
error_message : 'unsupported EFI arch or EFI support is disabled').allowed()
conf.set10('ENABLE_BOOTLOADER', have)
conf.set_quoted('EFI_MACHINE_TYPE_NAME', have ? efi_arch : '')
boot: Bring back bootloader builds This adds back sd-boot builds by using meson compile targets directly. We can do this now, because userspace binaries use the special dependency that allows us to easily separate flags, so that we don't pass anything to EFI builds that shouldn't be passed. Additionally, we pass a bunch of flags to hopefully disable/override any distro provided flags that should not be used for EFI binaries. Fixes: #12275
2023-02-27 16:54:48 +01:00
efi_arch_alt = ''
boot: Fix build for x32 When building on a x32 system we need to explicitly pass `-m64` to get the right ABI as the kernel and EFI are still 64bit. For this to actually work, a suitable multilib compiler, 32bit libc headers and libgcc need to be installed (similar to ia32 builds on x86_64).
2023-07-09 19:30:27 +02:00
efi_cpu_family_alt = ''
meson: Convert options to meson features (require) These options use requre() to conveniently express their dependency requirements.
2023-08-10 15:44:11 +02:00
if have and efi_arch == 'x64' and cc.links('''
boot: Bring back bootloader builds This adds back sd-boot builds by using meson compile targets directly. We can do this now, because userspace binaries use the special dependency that allows us to easily separate flags, so that we don't pass anything to EFI builds that shouldn't be passed. Additionally, we pass a bunch of flags to hopefully disable/override any distro provided flags that should not be used for EFI binaries. Fixes: #12275
2023-02-27 16:54:48 +01:00
#include <limits.h>
int main(int argc, char *argv[]) {
return __builtin_popcount(argc - CHAR_MAX);
}''', args : ['-m32', '-march=i686'], name : '32bit build possible')
efi_arch_alt = 'ia32'
boot: Fix build for x32 When building on a x32 system we need to explicitly pass `-m64` to get the right ABI as the kernel and EFI are still 64bit. For this to actually work, a suitable multilib compiler, 32bit libc headers and libgcc need to be installed (similar to ia32 builds on x86_64).
2023-07-09 19:30:27 +02:00
efi_cpu_family_alt = 'x86'
boot: Bring back bootloader builds This adds back sd-boot builds by using meson compile targets directly. We can do this now, because userspace binaries use the special dependency that allows us to easily separate flags, so that we don't pass anything to EFI builds that shouldn't be passed. Additionally, we pass a bunch of flags to hopefully disable/override any distro provided flags that should not be used for EFI binaries. Fixes: #12275
2023-02-27 16:54:48 +01:00
endif
meson: check for pefile dependency before enabling ukify ukify (and all the tests, including the autogenerated check-version-ukify) does not work unless pefile is available, so track it as a dependency in meson to avoid unit test failures later
2023-12-23 09:56:31 +01:00
pefile = pymod.find_installation('python3', required: false, modules : ['pefile'])
want_ukify = get_option('ukify').require(python_39 and (want_tests != 'true' or pefile.found()), error_message : 'Python >= 3.9 and pefile required').allowed()
ukify: don't install ukify man page if ukify is not enabled Checking for gnu-efi is not enough, since ukify can be explicitly disabled. Resolves: #26274
2023-01-31 21:43:53 +01:00
conf.set10('ENABLE_UKIFY', want_ukify)
meson: make lines more consistent
2023-12-14 18:35:00 -03:00
#####################################################################
meson: fix indentation
2023-05-05 09:36:00 +02:00
meson: use files() not find_program() for helper scripts We went back and forth between 'prog.sh', files('prog.sh'), and find_program('prog.sh'). We want to use files() or find_program() so that we get a good error message if the file is missing. Behaviour of meson changed over time, and in the past not all forms could be used in all places. For example 0f4c4f3824891aea6a76b2e861f38a6045b2a246 added find_program() in many places to avoid repeated messages. But it seems that all recent meson versions work fine with files(). find_program prints silly messages: Program tools/make-man-index.py found: YES (/home/zbyszek/src/systemd/tools/make-man-index.py) Program tools/meson-render-jinja2.py found: YES (/home/zbyszek/src/systemd/tools/meson-render-jinja2.py) ... We know that those files will be found, they are part of the git checkout. With files() this is gone and the meson output is easier to read.
2025-07-25 12:18:19 +02:00
check_efi_alignment_py = files('tools/check-efi-alignment.py')
build-sys: pick up vmlinux.h from running kernel BTF or user
2023-12-06 18:37:31 +01:00
#####################################################################
use_provided_vmlinux_h = false
use_generated_vmlinux_h = false
provided_vmlinux_h_path = get_option('vmlinux-h-path')
# For the more complex BPF programs we really want a vmlinux.h (which is arch
# specific, but only somewhat bound to kernel version). Ideally the kernel
# development headers would ship that, but right now they don't. Hence address
# this in two ways:
#
# 1. Provide a vmlinux.h at build time
# 2. Generate the file on the fly where possible (which requires /sys/ to be mounted)
#
# We generally prefer the former (to support reproducible builds), but will
# fallback to the latter.
if conf.get('BPF_FRAMEWORK') == 1
enable_vmlinux_h = get_option('vmlinux-h')
if enable_vmlinux_h == 'auto'
if provided_vmlinux_h_path != ''
use_provided_vmlinux_h = true
elif fs.exists('/sys/kernel/btf/vmlinux') and \
bpftool.found() and \
(host_machine.cpu_family() == build_machine.cpu_family()) and \
host_machine.cpu_family() in ['x86_64', 'aarch64']
# We will only generate a vmlinux.h from the running
# kernel if the host and build machine are of the same
# family. Also for now we focus on x86_64 and aarch64,
# since other archs don't seem to be ready yet.
use_generated_vmlinux_h = true
endif
elif enable_vmlinux_h == 'provided'
use_provided_vmlinux_h = true
elif enable_vmlinux_h == 'generated'
if not fs.exists('/sys/kernel/btf/vmlinux')
error('BTF data from kernel not available (/sys/kernel/btf/vmlinux missing), cannot generate vmlinux.h, but was asked to.')
endif
if not bpftool.found()
error('bpftool not available, cannot generate vmlinux.h, but was asked to.')
endif
use_generated_vmlinux_h = true
endif
endif
meson: Define our own clang-tidy target meson's target has a few issues: - Runs on all source files regardless if they're included in the build or not - Doesn't have any dependencies on generated sources which means we have to do a full build first before we can run clang-tidy - Doesn't allow us to pass any extra arguments To work around these, let's define our own clang-tidy target instead using llvm's run-clang-tidy script. Alongside the clang-tidy target, let's start keeping track of all generated sources which we make the clang-tidy target depend on. We also add a new target which will only generate source files which is useful for setting up the source tree for running code analysis against it.
2025-04-24 11:58:45 +02:00
vmlinux_h_dependency = []
build-sys: pick up vmlinux.h from running kernel BTF or user
2023-12-06 18:37:31 +01:00
if use_provided_vmlinux_h
if not fs.exists(provided_vmlinux_h_path)
error('Path to provided vmlinux.h does not exist.')
endif
bpf_o_unstripped_cmd += ['-I' + fs.parent(provided_vmlinux_h_path)]
meson: using f-strings in meson Those were added before 0.61 which is our minimum version. Dots were dropped from the end of some messages. Most messages did not have the trailing dot. 0.63 added support for multi-line f-strings. We can't use those yet.
2025-10-03 15:26:20 +02:00
message(f'Using provided @provided_vmlinux_h_path@')
build-sys: pick up vmlinux.h from running kernel BTF or user
2023-12-06 18:37:31 +01:00
elif use_generated_vmlinux_h
vmlinux_h_dependency = custom_target(
output: 'vmlinux.h',
command : [ bpftool, 'btf', 'dump', 'file', '/sys/kernel/btf/vmlinux', 'format', 'c' ],
capture : true)
bpf_o_unstripped_cmd += ['-I' + fs.parent(vmlinux_h_dependency.full_path())]
message('Using generated @0@'.format(vmlinux_h_dependency.full_path()))
else
message('Using neither provided nor generated vmlinux.h, some features will not be available.')
endif
conf.set10('HAVE_VMLINUX_H', use_provided_vmlinux_h or use_generated_vmlinux_h)
network: fix build with libbpf < 0.7 bpf_loop() and bpf_strncmp(), used by sysctl-monitor, were introduced in libbpf 0.7, so skip the module if using an older version Follow-up for 6d9ef22acdeac4b429efb75164341233955484af
2025-08-06 12:58:20 +01:00
conf.set10('ENABLE_SYSCTL_BPF', conf.get('HAVE_VMLINUX_H') == 1 and libbpf.version().version_compare('>= 0.7'))
build-sys: pick up vmlinux.h from running kernel BTF or user
2023-12-06 18:37:31 +01:00
#####################################################################
meson: use files() not find_program() for helper scripts We went back and forth between 'prog.sh', files('prog.sh'), and find_program('prog.sh'). We want to use files() or find_program() so that we get a good error message if the file is missing. Behaviour of meson changed over time, and in the past not all forms could be used in all places. For example 0f4c4f3824891aea6a76b2e861f38a6045b2a246 added find_program() in many places to avoid repeated messages. But it seems that all recent meson versions work fine with files(). find_program prints silly messages: Program tools/make-man-index.py found: YES (/home/zbyszek/src/systemd/tools/make-man-index.py) Program tools/meson-render-jinja2.py found: YES (/home/zbyszek/src/systemd/tools/meson-render-jinja2.py) ... We know that those files will be found, they are part of the git checkout. With files() this is gone and the meson output is easier to read.
2025-07-25 12:18:19 +02:00
check_version_history_py = files('tools/check-version-history.py')
elf2efi_py = files('tools/elf2efi.py')
export_dbus_interfaces_py = files('tools/dbus_exporter.py')
generate_gperfs = files('tools/generate-gperfs.py')
make_autosuspend_rules_py = files('tools/make-autosuspend-rules.py')
make_directive_index_py = files('tools/make-directive-index.py')
sync_docs_py = files('tools/sync-docs.py')
make_man_index_py = files('tools/make-man-index.py')
meson_render_jinja2 = files('tools/meson-render-jinja2.py')
update_dbus_docs_py = files('tools/update-dbus-docs.py')
update_hwdb_autosuspend_sh = files('tools/update-hwdb-autosuspend.sh')
update_hwdb_sh = files('tools/update-hwdb.sh')
update_man_rules_py = files('tools/update-man-rules.py')
update_syscall_tables_sh = files('tools/update-syscall-tables.sh')
xml_helper_py = files('tools/xml_helper.py')
meson: move and gather find_program()
2021-01-04 20:47:00 +09:00
meson: make lines more consistent
2023-12-14 18:35:00 -03:00
#####################################################################
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: drop short script to generate vcs tag
2023-07-29 22:41:43 +09:00
version_tag = get_option('version-tag')
meson: Decouple the version tag from the vcs tag Let's split off a new vcs-tag option from version-tag that configures whether the current commit should be appended to the version tag. Doing this saves us from having to fiddle around with generating git versions in packaging specs and instead let's meson do it for us, even if we pass in a custom version tag. With this approach there's no more need for tools/meson-vcs-tag.sh so we remove it.
2024-02-16 16:58:58 +01:00
if version_tag == ''
version_tag = meson.project_version()
endif
conf.set_quoted('VERSION_TAG', version_tag)
meson: Remove unneeded include directories meson by default adds the current source and build directory as include directories. Because we structure our meson code by gathering a giant dict of everything we want to do and then doing all the actual target generation in the top level meson.build, this behavior does not make sense at all because we end up adding the top level repository directory as an include directory which is never what we want. At the same time, let's also make sure the top level directory of the build directory is not an include directory, by moving the version.h generation into the src/version subdirectory and then adding the src/version subdirectory of the build directory as an include directory instead of the top level repository directory. Making this change means that language servers such as clangd can't get confused when they automatically insert an #include line and insert "#include "src/basic/fs-util.h" instead of "#include "fs-util.h".
2025-05-09 20:48:51 +02:00
subdir('src/version')
meson: drop short script to generate vcs tag
2023-07-29 22:41:43 +09:00
shared_lib_tag = get_option('shared-lib-tag')
if shared_lib_tag == ''
meson: Fix various versions Follow up for 8b3b01c4b7e0fde39b4be354990ee68f5e612c52 We switch to PROJECT_VERSION instead of PROJECT_VERSION_FULL where we report our version and which is likely being parsed to avoid breaking compat. If we didn't, the output would change from systemd 255 to systemd 255.1 which could break various tools.
2024-07-04 14:44:21 +02:00
shared_lib_tag = project_major_version
meson: drop short script to generate vcs tag
2023-07-29 22:41:43 +09:00
endif
meson: make lines more consistent
2023-12-14 18:35:00 -03:00
#####################################################################
meson: drop short script to generate vcs tag
2023-07-29 22:41:43 +09:00
meson: fix coverage build This moves coverage.h to src/coverage/, and specifies path to coverage.h with files() directive, to make it can be included even when located outside of the include directories. Otherwise, libc-wrapper cannot be built when -Db_coverage=true option is enabled. Follow-up for 543a48b653ff25cc562ce4dcffec377a875f604e.
2025-07-12 02:14:00 +09:00
subdir('src/coverage')
meson: explicitly include coverage tweaks when built w/ --coverage To make sure we don't miss any _exit() calls let's move the coverage-related tweaks into a separate header file and include it explicitly on the compiler command line using -include when a coverage build is requested. Follow-up to c6552ad381003a23cde7c3228e7071f30465df35.
2022-04-08 13:20:15 +02:00
meson: make lines more consistent
2023-12-14 18:35:00 -03:00
#####################################################################
meson: explicitly include coverage tweaks when built w/ --coverage To make sure we don't miss any _exit() calls let's move the coverage-related tweaks into a separate header file and include it explicitly on the compiler command line using -include when a coverage build is requested. Follow-up to c6552ad381003a23cde7c3228e7071f30465df35.
2022-04-08 13:20:15 +02:00
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
config_h = configure_file(
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-17 19:25:00 -04:00
output : 'config.h',
configuration : conf)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: Introduce userspace dep This will help in a later commit to separate userspace from EFI builds.
2023-02-26 11:07:24 +01:00
userspace_c_args += ['-include', 'config.h']
meson: move and gather find_program()
2021-01-04 20:47:00 +09:00
meson: Remove version_h dependency from jinja2_cmdline version_h includes GIT_VERSION which only makes sense for C files which aren't preprocessed by jinja2 so remove the argument. The end result of this change is that the man pages are not recompiled anymore every time GIT_VERSION changes.
2024-03-06 15:16:52 +01:00
jinja2_cmdline = [meson_render_jinja2, config_h]
meson: also allow setting GIT_VERSION via templates GIT_VERSION is not available as a config.h variable, because it's rendered into version.h during builds. Let's rework jinja2 rendering to also parse version.h. No functional change, the new variable is so far unused. I guess this will make partial rebuilds a bit slower, but it's useful to be able to use the full version string.
2022-04-05 16:52:44 +02:00
meson: Introduce userspace dep This will help in a later commit to separate userspace from EFI builds.
2023-02-26 11:07:24 +01:00
userspace = declare_dependency(
compile_args : userspace_c_args,
link_args : userspace_c_ld_args,
meson: fix coverage build This moves coverage.h to src/coverage/, and specifies path to coverage.h with files() directive, to make it can be included even when located outside of the include directories. Otherwise, libc-wrapper cannot be built when -Db_coverage=true option is enabled. Follow-up for 543a48b653ff25cc562ce4dcffec377a875f604e.
2025-07-12 02:14:00 +09:00
sources : userspace_sources,
meson: Introduce userspace dep This will help in a later commit to separate userspace from EFI builds.
2023-02-26 11:07:24 +01:00
)
man: generate link mode list dynamically The entries are sorted by speed. Some fields are left empty when there is no clear value to use. The table is much longer now, but I think it's better to document the allowed values, even if some are not terribly useful. Fixes #26256.
2023-05-30 13:02:57 +02:00
man_page_depends = []
meson: make lines more consistent
2023-12-14 18:35:00 -03:00
#####################################################################
meson: generate m4 preprocessor from config.h (#8914)
2018-05-07 18:17:35 +09:00
meson: move declarations of arrays
2023-06-23 11:00:17 +09:00
simple_tests = []
meson: merge declarations of normal and test executables
2023-06-26 08:52:26 +09:00
libsystemd_tests = []
meson: move declarations of arrays
2023-06-23 11:00:17 +09:00
simple_fuzzers = []
catalogs = []
meson: introduce infra to build executables and shared libraries by using dictionary That is not used yet. It will be used later.
2023-06-23 11:09:10 +09:00
modules = [] # nss, pam, and other plugins
executables = []
executables_by_name = {}
meson: Extract objects instead of creating intermediate static libraries Currently, when we want to add unit tests for code that is compiled into an executable, we either compile the code at least twice (once for the executable, and once for each test that uses it) or we create a static library which is then used by both the executable and all the tests. Both of these options are not ideal, compiling source files more than once slows down the build for no reason and creating the intermediate static libraries takes a lot of boilerplate. Instead, let's use the extract_objects() method that meson exposes on build targets. This allows us to extract the objects corresponding to specific source files and use them in other executables. Because we define all executables upfront into a dictionary, we integrate this into the dictionary approach by adding two new fields: - 'extract' takes a list of files for which objects should be extracted. The extracted objects are stored in a dict keyed by the executable name from which they were extracted. - 'objects' takes the name of an executable from which the extracted objects should be added to the current executable. One side effect of this approach is that we can't build test executables anymore without building the main executable, so we stop building test executables unless we're also building the main executable. This allows us to switch to using subdir_done() in all of these subdirectories to skip parsing them if the corresponding component is disabled. These changes get me down from 2439 => 2403 ninja targets on a full rebuild from scratch.
2025-05-07 22:55:15 +02:00
objects_by_name = {}
meson: also merge declarations of fuzzers with other executables
2023-06-27 01:07:25 +09:00
fuzzer_exes = []
meson: Define our own clang-tidy target meson's target has a few issues: - Runs on all source files regardless if they're included in the build or not - Doesn't have any dependencies on generated sources which means we have to do a full build first before we can run clang-tidy - Doesn't allow us to pass any extra arguments To work around these, let's define our own clang-tidy target instead using llvm's run-clang-tidy script. Alongside the clang-tidy target, let's start keeping track of all generated sources which we make the clang-tidy target depend on. We also add a new target which will only generate source files which is useful for setting up the source tree for running code analysis against it.
2025-04-24 11:58:45 +02:00
generated_sources = [version_h, vmlinux_h_dependency]
meson: Rework clang-tidy integration to be done via unit tests Instead of using run-clang-tidy.py with its own scheduling, let's just gather a list of source files ourselves and then use that to add a unit test for each source file that runs clang-tidy on the source file. We also add a bit of logic to run clang-tidy on most header files as well for extra coverage. This uncovered various header files that were not standalone so this commit also includes fixes to make sure the clang-tidy tests are all green. We can also use this in a later commit to run clang-include-cleaner on each source file in the same way.
2025-05-15 15:09:27 +02:00
sources = []
meson: move declarations of arrays
2023-06-23 11:00:17 +09:00
meson: sort inclusion of meson.build files in subdirectories
2021-01-04 21:27:00 +09:00
# binaries that have --help and are intended for use by humans,
# usually, but not always, installed in /bin.
public_programs = []
Export D-Bus interfaces to /usr/share/dbus-1/interfaces Pass -Ddbus-interfaces-dir=no to meson to disable export Interfaces from: org.freedesktop.home1 org.freedesktop.hostname1 org.freedesktop.import1 org.freedesktop.locale1 org.freedesktop.LogControl1 org.freedesktop.login1 org.freedesktop.machine1 org.freedesktop.oom1 org.freedesktop.portable1 org.freedesktop.resolve1 org.freedesktop.systemd1 org.freedesktop.timedate1
2021-07-24 10:30:42 +03:00
# D-Bus introspection XML export
dbus_programs = []
test/60-ukify: override stub location in tests Without this, build would fail if the stub is not available in /usr/lib/.
2023-04-14 18:45:24 +02:00
# A list of boot stubs. Required for testing of ukify.
boot_stubs = []
tree-wide: several cleanups for generating symbol lists and gperf files - pass our system include directories to make generators use our libc wrappers and latest kernel headers, - include relevant headers in generated gperf file, - use files() rather than find_program(), as the result of find_program() cannot be passed to 'input' of custom_target(), - move generate-bpf-delegate-configs.py to src/core/, as it is only used by libcore.
2025-07-08 13:07:32 +09:00
# This is similar to system_includes below, but for passing custom_target().
system_include_args = [
'-isystem', meson.project_build_root() / 'src/include/override',
'-isystem', meson.project_source_root() / 'src/include/override',
'-isystem', meson.project_build_root() / 'src/include/uapi',
'-isystem', meson.project_source_root() / 'src/include/uapi',
]
include: move libc header wrappers to src/include/override/, and kernel headers to src/include/uapi/ Preparation for later changes.
2025-07-06 12:19:26 +09:00
system_includes = [
include_directories(
# gcc(1) says
# "Directories specified with -isystem options are scanned in left-to-right order",
# and meson puts the directories in the reversed order. Hence, a directory with a lower
# priority must be listed earlier.
'src/include/uapi',
'src/include/override',
is_system : true,
),
]
tree-wide: Mark linux and sys includes as system includes Even though these are in our tree, we should still treat them as system includes which helps various tools (clangd, iwyu, ...) understand that these are system includes and <> should be used instead of "".
2025-04-12 19:36:48 +02:00
basic_includes = [
include_directories(
'src/basic',
'src/fundamental',
'src/systemd',
),
include: move libc header wrappers to src/include/override/, and kernel headers to src/include/uapi/ Preparation for later changes.
2025-07-06 12:19:26 +09:00
system_includes,
meson: Remove unneeded include directories meson by default adds the current source and build directory as include directories. Because we structure our meson code by gathering a giant dict of everything we want to do and then doing all the actual target generation in the top level meson.build, this behavior does not make sense at all because we end up adding the top level repository directory as an include directory which is never what we want. At the same time, let's also make sure the top level directory of the build directory is not an include directory, by moving the version.h generation into the src/version subdirectory and then adding the src/version subdirectory of the build directory as an include directory instead of the top level repository directory. Making this change means that language servers such as clangd can't get confused when they automatically insert an #include line and insert "#include "src/basic/fs-util.h" instead of "#include "fs-util.h".
2025-05-09 20:48:51 +02:00
version_include,
tree-wide: Mark linux and sys includes as system includes Even though these are in our tree, we should still treat them as system includes which helps various tools (clangd, iwyu, ...) understand that these are system includes and <> should be used instead of "".
2025-04-12 19:36:48 +02:00
]
meson: slightly disentangle code dependencies But, still sd-id128 is used in src/basic.
2021-01-04 23:36:00 +09:00
libsystemd_includes = [basic_includes, include_directories(
'src/libsystemd/sd-bus',
'src/libsystemd/sd-device',
'src/libsystemd/sd-event',
'src/libsystemd/sd-hwdb',
'src/libsystemd/sd-id128',
'src/libsystemd/sd-journal',
libsystemd: turn json.[ch] into a public API This is preparation for making our Varlink API a public API. Since our Varlink API is built on top of our JSON API we need to make that public first (it's a nice API, but JSON APIs there are already enough, this is purely about the Varlink angle). I made most of the json.h APIs public, and just placed them in sd-json.h. Sometimes I wasn't so sure however, since the underlying data structures would have to be made public too. If in doubt I didn#t risk it, and moved the relevant API to src/libsystemd/sd-json/json-util.h instead (without any sd_* symbol prefixes). This is mostly a giant search/replace patch.
2024-05-02 14:26:21 +02:00
'src/libsystemd/sd-json',
meson: slightly disentangle code dependencies But, still sd-id128 is used in src/basic.
2021-01-04 23:36:00 +09:00
'src/libsystemd/sd-netlink',
'src/libsystemd/sd-network',
path-lookup: move from basic/ to libsystemd/ So that sd_path_lookup() can be utilized to replace duplicate functions.
2024-09-15 20:57:03 +02:00
'src/libsystemd/sd-path',
sd-varlink: make our internal Varlink API public as sd-varlink.[ch] It's time. sd-json was already done earlier in this cycle, let's now make sd-varlink public too. This is mostly just a search/replace job of epical proportions. I left some functions internal (mostly IDL handling), and I turned some static inline calls into regular calls.
2024-07-12 11:22:58 +02:00
'src/libsystemd/sd-resolve',
'src/libsystemd/sd-varlink')]
meson: slightly disentangle code dependencies But, still sd-id128 is used in src/basic.
2021-01-04 23:36:00 +09:00
includes = [libsystemd_includes, include_directories('src/shared')]
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
subdir('po')
subdir('catalog')
tree-wide: several cleanups for generating symbol lists and gperf files - pass our system include directories to make generators use our libc wrappers and latest kernel headers, - include relevant headers in generated gperf file, - use files() rather than find_program(), as the result of find_program() cannot be passed to 'input' of custom_target(), - move generate-bpf-delegate-configs.py to src/core/, as it is only used by libcore.
2025-07-08 13:07:32 +09:00
subdir('src/include')
libc-wrapper: introduce a tiny libc wrapper Then, move syscall definitions to the wrapper, and prototypes are moved to relevant headers. This also adds checks for add_key() and request_key(), as one day glibc may be going to add some of them separatedly. The check for fspick in meson.build is dropped, as it is currently unused in our code. This also moves - basic/missing_bpf.h -> include/override/linux/bpf.h, - basic/missing_keyctl.h -> include/override/linux/keyctl.h.
2025-06-21 20:29:51 +09:00
subdir('src/libc')
tree-wide: Drop gnu-efi This drops all mentions of gnu-efi and its manual build machinery. A future commit will bring bootloader builds back. A new bootloader meson option is now used to control whether to build sd-boot and its userspace tooling.
2023-02-26 14:09:44 +01:00
subdir('src/fundamental')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
subdir('src/basic')
subdir('src/libsystemd')
meson: move libudev related definitions to src/libudev/meson.build Also, this makes libudev.so built in build directory.
2021-01-04 21:13:30 +09:00
subdir('src/shared')
subdir('src/libudev')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
libsystemd = shared_library(
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-17 19:25:00 -04:00
'systemd',
meson: move library version defines to the top (#6939)
2017-09-28 19:24:16 +02:00
version : libsystemd_version,
meson: slightly disentangle code dependencies But, still sd-id128 is used in src/basic.
2021-01-04 23:36:00 +09:00
include_directories : libsystemd_includes,
meson: Remove unneeded include directories meson by default adds the current source and build directory as include directories. Because we structure our meson code by gathering a giant dict of everything we want to do and then doing all the actual target generation in the top level meson.build, this behavior does not make sense at all because we end up adding the top level repository directory as an include directory which is never what we want. At the same time, let's also make sure the top level directory of the build directory is not an include directory, by moving the version.h generation into the src/version subdirectory and then adding the src/version subdirectory of the build directory as an include directory instead of the top level repository directory. Making this change means that language servers such as clangd can't get confused when they automatically insert an #include line and insert "#include "src/basic/fs-util.h" instead of "#include "fs-util.h".
2025-05-09 20:48:51 +02:00
implicit_include_directories : false,
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-17 19:25:00 -04:00
link_args : ['-shared',
libsystemd: link with '-z nodelete' We want to avoid reinitialization of our global variables with static storage duration in case we get dlopened multiple times by the same application. This will avoid potential resource leaks that could have happened otherwise (e.g. leaking journal socket fd).
2024-05-22 17:15:07 +02:00
# Make sure our library is never deleted from memory, so that our open logging fds don't leak on dlopen/dlclose cycles.
'-z', 'nodelete',
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-17 19:25:00 -04:00
'-Wl,--version-script=' + libsystemd_sym_path],
libc-wrapper: introduce a tiny libc wrapper Then, move syscall definitions to the wrapper, and prototypes are moved to relevant headers. This also adds checks for add_key() and request_key(), as one day glibc may be going to add some of them separatedly. The check for fspick in meson.build is dropped, as it is currently unused in our code. This also moves - basic/missing_bpf.h -> include/override/linux/bpf.h, - basic/missing_keyctl.h -> include/override/linux/keyctl.h.
2025-06-21 20:29:51 +09:00
link_with : [libc_wrapper_static,
libbasic_static],
sd-journal: move source files for sd-journal to src/libsystemd/sd-journal
2021-01-01 04:30:47 +09:00
link_whole : [libsystemd_static],
meson: Introduce userspace dep This will help in a later commit to separate userspace from EFI builds.
2023-02-26 11:07:24 +01:00
dependencies : [librt,
threads,
userspace],
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-17 19:25:00 -04:00
link_depends : libsystemd_sym,
install : true,
meson: add install_tag to sd-boot, libsystemd and libudev Allows to 'meson install --tags systemd-boot --no-rebuild' to install only the EFI binaries, skipping the rest, for a very quick build: $ ninja src/boot/efi/linuxx64.efi.stub [21/21] Generating src/boot/efi/linuxx64.efi.stub with a custom command $ ninja src/boot/efi/systemd-bootx64.efi [10/10] Generating src/boot/efi/systemd-bootx64.efi with a custom command $ DESTDIR=/tmp/foo meson install --tags systemd-boot --no-rebuild Installing src/boot/efi/systemd-bootx64.efi to /tmp/foo/usr/lib/systemd/boot/efi Requires Meson 0.60 to be used, prints a warning for unknown keyword in earlier versions, but there's no failure https://mesonbuild.com/Installing.html#installation-tags
2022-04-27 10:54:14 +01:00
install_tag: 'libsystemd',
Drop split-usr and unmerged-usr support As previously announced, execute order 66: https://lists.freedesktop.org/archives/systemd-devel/2022-September/048352.html The meson options split-usr, rootlibdir and rootprefix become no-ops that print a warning if they are set to anything other than the default values. We can remove them in a future release.
2023-06-12 02:15:19 +01:00
install_dir : libdir)
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: add static libs to libudev/libsystemd target aliases If static libraries are enabled, then group them in the build target together with the shared libraries, to match the install tags.
2024-05-31 21:44:43 +01:00
if static_libsystemd != 'false'
meson: Don't create static library target unless option is enabled While we don't build these by default, all the source files still get added to the compile_commands.json file by meson, which can confuse tools as they might end up analyzing the source files twice or analyzing the wrong one. To avoid this issue, only define the static library target if the corresponding option is enabled.
2025-05-10 22:19:22 +02:00
install_libsystemd_static = static_library(
'systemd',
libc-wrapper: introduce a tiny libc wrapper Then, move syscall definitions to the wrapper, and prototypes are moved to relevant headers. This also adds checks for add_key() and request_key(), as one day glibc may be going to add some of them separatedly. The check for fspick in meson.build is dropped, as it is currently unused in our code. This also moves - basic/missing_bpf.h -> include/override/linux/bpf.h, - basic/missing_keyctl.h -> include/override/linux/keyctl.h.
2025-06-21 20:29:51 +09:00
libc_wrapper_sources,
meson: Don't create static library target unless option is enabled While we don't build these by default, all the source files still get added to the compile_commands.json file by meson, which can confuse tools as they might end up analyzing the source files twice or analyzing the wrong one. To avoid this issue, only define the static library target if the corresponding option is enabled.
2025-05-10 22:19:22 +02:00
libsystemd_sources,
basic_sources,
fundamental_sources,
include_directories : libsystemd_includes,
implicit_include_directories : false,
install : true,
install_tag: 'libsystemd',
install_dir : libdir,
pic : static_libsystemd_pic,
meson: Remove unnecessary deps from libsystemd-static build blkid, libmount and openssl are not used in src/basic or src/libsystemd, and so shouldn't be required as deps of libsystemd static, so let's drop them.
2025-06-05 11:04:06 +02:00
dependencies : [libcap,
meson: Don't create static library target unless option is enabled While we don't build these by default, all the source files still get added to the compile_commands.json file by meson, which can confuse tools as they might end up analyzing the source files twice or analyzing the wrong one. To avoid this issue, only define the static library target if the corresponding option is enabled.
2025-05-10 22:19:22 +02:00
libdl,
libgcrypt_cflags,
liblz4_cflags,
libm,
librt,
libxz_cflags,
libzstd_cflags,
threads,
userspace],
c_args : libsystemd_c_args + (static_libsystemd_pic ? [] : ['-fno-PIC']))
meson: add static libs to libudev/libsystemd target aliases If static libraries are enabled, then group them in the build target together with the shared libraries, to match the install tags.
2024-05-31 21:44:43 +01:00
alias_target('libsystemd', libsystemd, install_libsystemd_static)
else
alias_target('libsystemd', libsystemd)
endif
meson: move libudev related definitions to src/libudev/meson.build Also, this makes libudev.so built in build directory.
2021-01-04 21:13:30 +09:00
libudev = shared_library(
'udev',
version : libudev_version,
include_directories : includes,
meson: Remove unneeded include directories meson by default adds the current source and build directory as include directories. Because we structure our meson code by gathering a giant dict of everything we want to do and then doing all the actual target generation in the top level meson.build, this behavior does not make sense at all because we end up adding the top level repository directory as an include directory which is never what we want. At the same time, let's also make sure the top level directory of the build directory is not an include directory, by moving the version.h generation into the src/version subdirectory and then adding the src/version subdirectory of the build directory as an include directory instead of the top level repository directory. Making this change means that language servers such as clangd can't get confused when they automatically insert an #include line and insert "#include "src/basic/fs-util.h" instead of "#include "fs-util.h".
2025-05-09 20:48:51 +02:00
implicit_include_directories : false,
meson: move libudev related definitions to src/libudev/meson.build Also, this makes libudev.so built in build directory.
2021-01-04 21:13:30 +09:00
link_args : ['-shared',
'-Wl,--version-script=' + libudev_sym_path],
link_with : [libsystemd_static, libshared_static],
link_whole : libudev_basic,
meson: Introduce userspace dep This will help in a later commit to separate userspace from EFI builds.
2023-02-26 11:07:24 +01:00
dependencies : [threads,
userspace],
meson: move libudev related definitions to src/libudev/meson.build Also, this makes libudev.so built in build directory.
2021-01-04 21:13:30 +09:00
link_depends : libudev_sym,
install : true,
meson: add install_tag to sd-boot, libsystemd and libudev Allows to 'meson install --tags systemd-boot --no-rebuild' to install only the EFI binaries, skipping the rest, for a very quick build: $ ninja src/boot/efi/linuxx64.efi.stub [21/21] Generating src/boot/efi/linuxx64.efi.stub with a custom command $ ninja src/boot/efi/systemd-bootx64.efi [10/10] Generating src/boot/efi/systemd-bootx64.efi with a custom command $ DESTDIR=/tmp/foo meson install --tags systemd-boot --no-rebuild Installing src/boot/efi/systemd-bootx64.efi to /tmp/foo/usr/lib/systemd/boot/efi Requires Meson 0.60 to be used, prints a warning for unknown keyword in earlier versions, but there's no failure https://mesonbuild.com/Installing.html#installation-tags
2022-04-27 10:54:14 +01:00
install_tag: 'libudev',
Drop split-usr and unmerged-usr support As previously announced, execute order 66: https://lists.freedesktop.org/archives/systemd-devel/2022-September/048352.html The meson options split-usr, rootlibdir and rootprefix become no-ops that print a warning if they are set to anything other than the default values. We can remove them in a future release.
2023-06-12 02:15:19 +01:00
install_dir : libdir)
meson: move libudev related definitions to src/libudev/meson.build Also, this makes libudev.so built in build directory.
2021-01-04 21:13:30 +09:00
meson: add static libs to libudev/libsystemd target aliases If static libraries are enabled, then group them in the build target together with the shared libraries, to match the install tags.
2024-05-31 21:44:43 +01:00
if static_libudev != 'false'
meson: Don't create static library target unless option is enabled While we don't build these by default, all the source files still get added to the compile_commands.json file by meson, which can confuse tools as they might end up analyzing the source files twice or analyzing the wrong one. To avoid this issue, only define the static library target if the corresponding option is enabled.
2025-05-10 22:19:22 +02:00
install_libudev_static = static_library(
'udev',
libc-wrapper: introduce a tiny libc wrapper Then, move syscall definitions to the wrapper, and prototypes are moved to relevant headers. This also adds checks for add_key() and request_key(), as one day glibc may be going to add some of them separatedly. The check for fspick in meson.build is dropped, as it is currently unused in our code. This also moves - basic/missing_bpf.h -> include/override/linux/bpf.h, - basic/missing_keyctl.h -> include/override/linux/keyctl.h.
2025-06-21 20:29:51 +09:00
libc_wrapper_sources,
meson: Don't create static library target unless option is enabled While we don't build these by default, all the source files still get added to the compile_commands.json file by meson, which can confuse tools as they might end up analyzing the source files twice or analyzing the wrong one. To avoid this issue, only define the static library target if the corresponding option is enabled.
2025-05-10 22:19:22 +02:00
basic_sources,
fundamental_sources,
shared_sources,
libsystemd_sources,
libudev_sources,
include_directories : includes,
implicit_include_directories : false,
install : true,
install_tag: 'libudev',
install_dir : libdir,
link_depends : libudev_sym,
dependencies : [libmount,
libshared_deps,
userspace],
c_args : static_libudev_pic ? [] : ['-fno-PIC'],
pic : static_libudev_pic)
meson: add static libs to libudev/libsystemd target aliases If static libraries are enabled, then group them in the build target together with the shared libraries, to match the install tags.
2024-05-31 21:44:43 +01:00
alias_target('libudev', libudev, install_libudev_static)
else
alias_target('libudev', libudev)
endif
meson: make lines more consistent
2023-12-14 18:35:00 -03:00
#####################################################################
meson: add syscall-names-update target The calls to find_program("tools/*") are moved earlier so they can be used in libshared/ (and it doesn't make sense to split them).
2020-08-18 16:27:20 +02:00
meson: merge declarations of normal and test executables
2023-06-26 08:52:26 +09:00
runtest_env = custom_target(
output : 'systemd-runtest.env',
meson: quote paths in generated systemd-runtest.env Looking at aecb6eaed7d39d73f296c86a882c644b18b7e634, it seems that we want build paths with spaces to be supported. Let's use 'printf %q' here to make that work. Also, use capture:true to make the command less complicated.
2025-06-28 15:10:00 +02:00
command : ['printf',
'SYSTEMD_TEST_DATA=%q\nSYSTEMD_CATALOG_DIR=%q\n',
meson.project_source_root() / 'test',
meson.project_build_root() / 'catalog'],
capture: true,
meson: merge declarations of normal and test executables
2023-06-26 08:52:26 +09:00
depends : catalogs,
build_by_default : true)
test_cflags = ['-DTEST_CODE=1']
# We intentionally do not do inline initializations with definitions for a
# bunch of _cleanup_ variables in tests, to ensure valgrind is triggered if we
# use the variable unexpectedly. This triggers a lot of maybe-uninitialized
# false positives when the combination of -O2 and -flto is used. Suppress them.
if '-O2' in c_args and '-flto=auto' in c_args
test_cflags += cc.first_supported_argument('-Wno-maybe-uninitialized')
endif
meson: make lines more consistent
2023-12-14 18:35:00 -03:00
#####################################################################
meson: merge declarations of normal and test executables
2023-06-26 08:52:26 +09:00
meson: introduce infra to build executables and shared libraries by using dictionary That is not used yet. It will be used later.
2023-06-23 11:09:10 +09:00
executable_template = {
'include_directories' : includes,
'link_with' : libshared,
'install_rpath' : pkglibdir,
'install' : true,
}
generator_template = executable_template + {
'install_dir' : systemgeneratordir,
}
libexec_template = executable_template + {
'install_dir' : libexecdir,
}
executable_additional_kwargs = {
'dependencies' : userspace,
}
meson: merge declarations of normal and test executables
2023-06-26 08:52:26 +09:00
test_template = executable_template + {
'build_by_default' : want_tests != 'false',
'install' : install_tests,
'install_dir' : unittestsdir,
}
test_additional_kwargs = {
'c_args' : test_cflags,
'link_depends' : runtest_env,
}
meson: also merge declarations of fuzzers with other executables
2023-06-27 01:07:25 +09:00
fuzz_template = executable_template + {
meson: Build fuzz executables by default if fuzz-tests option is enabled fuzz-tests is enabled by default now but we still don't build the fuzz executables by default. Let's change that so that we always make sure these still compile when we make changes.
2025-05-15 14:52:48 +02:00
'build_by_default' : want_fuzz_tests,
meson: also merge declarations of fuzzers with other executables
2023-06-27 01:07:25 +09:00
'install' : false,
}
meson: Stop doing nested build when fuzzers are enabled Currently, when fuzzers are enabled, we run meson from within meson to build the fuzzer executables with sanitizers. The idea is that we can build the fuzzers with different kinds of sanitizers independently from the main build. The issue with this setup is that we don't actually make use of it. We only build the fuzzers with one set of sanitizers (address,undefined) so we're adding a bunch of extra complexity without any benefit as we can just setup the top level meson build with these sanitizers and get the same result. The other issue with this setup is that we don't pass on all the options passed to the top level meson build to the nested meson build. The only things we pass on are extra compiler arguments and the value of the auto_features option, but none of the individual feature options if overridden are passed on, which can lead to very hard to debug issues as an option enabled in the top level build is not enabled in the nested build. Since we're not getting anything useful out of this setup, let's simplify and get rid of the nested meson build. Instead, sanitizers should be enabled for the top level meson.build. This currently didn't work as we were overriding the sanitizers passed to the meson build with the fuzzer sanitizer, so we fix that as well by making sure we combine the fuzzer sanitizer with the ones passed in by the user. We also drop support for looking up libFuzzer as a separate library as it has been shipped builtin in clang since clang 6.0, so we can assume that -fsanitize=fuzzer is available. To make sure we still run the fuzzing tests, we enable the fuzz-tests option by default now to make sure they still always run (without instrumentation unless one of llvm-fuzz or oss-fuzz is enabled).
2025-05-11 09:42:28 +02:00
if want_ossfuzz
meson: also merge declarations of fuzzers with other executables
2023-06-27 01:07:25 +09:00
fuzz_additional_kwargs = {
'dependencies' : fuzzing_engine,
}
meson: Stop doing nested build when fuzzers are enabled Currently, when fuzzers are enabled, we run meson from within meson to build the fuzzer executables with sanitizers. The idea is that we can build the fuzzers with different kinds of sanitizers independently from the main build. The issue with this setup is that we don't actually make use of it. We only build the fuzzers with one set of sanitizers (address,undefined) so we're adding a bunch of extra complexity without any benefit as we can just setup the top level meson build with these sanitizers and get the same result. The other issue with this setup is that we don't pass on all the options passed to the top level meson build to the nested meson build. The only things we pass on are extra compiler arguments and the value of the auto_features option, but none of the individual feature options if overridden are passed on, which can lead to very hard to debug issues as an option enabled in the top level build is not enabled in the nested build. Since we're not getting anything useful out of this setup, let's simplify and get rid of the nested meson build. Instead, sanitizers should be enabled for the top level meson.build. This currently didn't work as we were overriding the sanitizers passed to the meson build with the fuzzer sanitizer, so we fix that as well by making sure we combine the fuzzer sanitizer with the ones passed in by the user. We also drop support for looking up libFuzzer as a separate library as it has been shipped builtin in clang since clang 6.0, so we can assume that -fsanitize=fuzzer is available. To make sure we still run the fuzzing tests, we enable the fuzz-tests option by default now to make sure they still always run (without instrumentation unless one of llvm-fuzz or oss-fuzz is enabled).
2025-05-11 09:42:28 +02:00
elif want_libfuzzer
meson: also merge declarations of fuzzers with other executables
2023-06-27 01:07:25 +09:00
fuzz_additional_kwargs = {
'link_args' : ['-fsanitize=fuzzer'],
}
else
fuzz_additional_kwargs = {
'sources' : files('src/fuzz/fuzz-main.c'),
}
endif
fuzz_additional_kwargs += {
'include_directories' : include_directories('src/fuzz'),
'c_args' : test_cflags,
}
meson: introduce infra to build executables and shared libraries by using dictionary That is not used yet. It will be used later.
2023-06-23 11:09:10 +09:00
nss_template = {
'version' : '2',
'include_directories' : includes,
# Note that we link NSS modules with '-z nodelete' so that mempools never get orphaned
'link_args' : ['-z', 'nodelete'],
'link_with' : [
libc-wrapper: introduce a tiny libc wrapper Then, move syscall definitions to the wrapper, and prototypes are moved to relevant headers. This also adds checks for add_key() and request_key(), as one day glibc may be going to add some of them separatedly. The check for fspick in meson.build is dropped, as it is currently unused in our code. This also moves - basic/missing_bpf.h -> include/override/linux/bpf.h, - basic/missing_keyctl.h -> include/override/linux/keyctl.h.
2025-06-21 20:29:51 +09:00
libc_wrapper_static,
meson: introduce infra to build executables and shared libraries by using dictionary That is not used yet. It will be used later.
2023-06-23 11:09:10 +09:00
libsystemd_static,
libshared_static,
meson: rename libbasic to libbasic_static Our variables for internal libraries are named 'libfoo' for the shared lib variant, and 'libfoo_static' for the static lib variant. The only exception was libbasic, because we didn't have a shared variant for it. But let's rename it for consitency. This makes the build config easier to understand.
2024-07-03 16:51:05 +02:00
libbasic_static,
meson: introduce infra to build executables and shared libraries by using dictionary That is not used yet. It will be used later.
2023-06-23 11:09:10 +09:00
],
'dependencies' : [
librt,
threads,
],
'install' : true,
'install_tag' : 'nss',
'install_dir' : libdir,
}
pam_template = {
'name_prefix' : '',
'include_directories' : includes,
'link_with' : [
libsystemd_static,
libshared_static,
],
'dependencies' : [
pam: make libpam a dlopen() based dependency
2025-09-08 16:04:23 +02:00
# Note: our PAM modules also call dlopen_libpam() and use
meson: fix type in comment As per https://github.com/systemd/systemd/pull/38999#discussion_r2363704203
2025-09-20 08:41:56 +02:00
# symbols acquired through that, hence the explicit dep here is
pam: make libpam a dlopen() based dependency
2025-09-08 16:04:23 +02:00
# strictly speaking unnecessary. We put it in place anyway,
# since for the PAM modules we cannot avoid libpam anyway,
# after all they are loaded *by* libpam, and hence there's no
# loss in having explicit deps here, but there's a win: it
# makes the deps more visible.
#
# (In case you wonder why we do dlopen_libpam() from the PAM
# modules in the first place: that's mostly so that all our PAM
# code (regardless if our PAM modules or our PAM consuming
# programs) can use the same helpers, which hence go via
# dlopen_libpam().
meson: introduce infra to build executables and shared libraries by using dictionary That is not used yet. It will be used later.
2023-06-23 11:09:10 +09:00
libpam_misc,
libpam,
threads,
],
'install' : true,
'install_tag' : 'pam',
'install_dir' : pamlibdir,
}
module_additional_kwargs = {
'link_args' : ['-shared'],
'dependencies' : userspace,
}
meson: make lines more consistent
2023-12-14 18:35:00 -03:00
#####################################################################
meson: introduce infra to build executables and shared libraries by using dictionary That is not used yet. It will be used later.
2023-06-23 11:09:10 +09:00
meson: sort inclusion of meson.build files in subdirectories
2021-01-04 21:27:00 +09:00
# systemd-analyze requires 'libcore'
meson: add test-dlopen test-dlopen is a very simple binary that is only linked with libc and libdl. From it we do dlopen() on the nss and pam modules to check that they are linked to all necessary libs. (meson-compiled nss modules are linked to less libraries, for whatever reason. I suspected that some deps are missing, but it turns out that my suspicions weren't justified, and the modules load just fine. Let's keep the test though, it is very quick, and might detect missing linkage in the future.)
2017-04-14 20:10:28 -04:00
subdir('src/core')
meson: sort inclusion of meson.build files in subdirectories
2021-01-04 21:27:00 +09:00
# systemd-networkd requires 'libsystemd_network'
subdir('src/libsystemd-network')
meson: move declarations of socket-proxy, udevadm, quotacheck, and shutdown
2023-06-26 03:35:47 +09:00
# hwdb requires 'udev_link_with' and 'udev_rpath'
subdir('src/udev')
meson: add test-dlopen test-dlopen is a very simple binary that is only linked with libc and libdl. From it we do dlopen() on the nss and pam modules to check that they are linked to all necessary libs. (meson-compiled nss modules are linked to less libraries, for whatever reason. I suspected that some deps are missing, but it turns out that my suspicions weren't justified, and the modules load just fine. Let's keep the test though, it is very quick, and might detect missing linkage in the future.)
2017-04-14 20:10:28 -04:00
meson: move declarations of ac-power, detect-virt, sysctl, and sleep
2023-06-25 16:28:21 +09:00
subdir('src/ac-power')
meson: add test-dlopen test-dlopen is a very simple binary that is only linked with libc and libdl. From it we do dlopen() on the nss and pam modules to check that they are linked to all necessary libs. (meson-compiled nss modules are linked to less libraries, for whatever reason. I suspected that some deps are missing, but it turns out that my suspicions weren't justified, and the modules load just fine. Let's keep the test though, it is very quick, and might detect missing linkage in the future.)
2017-04-14 20:10:28 -04:00
subdir('src/analyze')
meson: move declarations of ask-password and friends
2023-06-26 03:18:38 +09:00
subdir('src/ask-password')
meson: move declarations of backlight, rfkill, and sysupdate-generator
2023-06-25 15:55:33 +09:00
subdir('src/backlight')
meson: move declarations of delta, escape, notify, creds, and battery-check
2023-06-25 16:31:06 +09:00
subdir('src/battery-check')
meson: move declarations of pstore, oomd, and binfmt
2023-06-25 16:10:00 +09:00
subdir('src/binfmt')
Move bless-boot components to their own source subdirectory
2024-11-07 12:28:42 +01:00
subdir('src/bless-boot')
Rename src/boot/efi to just src/boot I very much dislike the approach in which we were mixing Linux and UEFI C code in the same subdirectory. No code was shared between two environments. This layout was created in e7dd673d1e0acfe5420599588c559fd85a3a9e8f, with the justification of "being more consistent with the rest of systemd", but I don't see how it's supposed to be so. Originally, when the C code was just a single bootctl.c file, this wasn't so bad. But over time the userspace code grew quite a bit. With the moves done in previuos commits, the intermediate subdirectory is now empty except for the efi/ subdir, and this additional subdirectory level doesn't have a good justification. The components is called "systemd-boot", not "systemd-efi", and we can remove one level of indentation.
2024-11-07 12:45:28 +01:00
subdir('src/boot')
Move bootctl to its own source subdirectory It's been split into a bunch of files and deserves its own subdirectory similarly to systemctl.
2024-11-07 12:24:40 +01:00
subdir('src/bootctl')
meson: move source file list for busctl
2021-01-04 20:35:33 +09:00
subdir('src/busctl')
meson: move declarations of cgls, cgtop, initctl, and systemd-mount
2023-06-26 03:21:42 +09:00
subdir('src/cgls')
subdir('src/cgtop')
meson: sort inclusion of meson.build files in subdirectories
2021-01-04 21:27:00 +09:00
subdir('src/coredump')
meson: move declarations of delta, escape, notify, creds, and battery-check
2023-06-25 16:31:06 +09:00
subdir('src/creds')
meson: move source file list for systemd-cryptenroll
2021-01-04 20:29:01 +09:00
subdir('src/cryptenroll')
meson: move source file list for systemd-cryptsetup
2021-01-04 20:26:20 +09:00
subdir('src/cryptsetup')
meson: move declarations of several generators
2023-06-25 15:20:29 +09:00
subdir('src/debug-generator')
meson: move declarations of delta, escape, notify, creds, and battery-check
2023-06-25 16:31:06 +09:00
subdir('src/delta')
meson: move declarations of ac-power, detect-virt, sysctl, and sleep
2023-06-25 16:28:21 +09:00
subdir('src/detect-virt')
meson: move declarations of dissect and friends
2023-06-25 15:34:09 +09:00
subdir('src/dissect')
meson: move declarations of several generators
2023-06-25 15:20:29 +09:00
subdir('src/environment-d-generator')
meson: move declarations of delta, escape, notify, creds, and battery-check
2023-06-25 16:31:06 +09:00
subdir('src/escape')
factory-reset: revamp infrastructure This introduces a bunch of facilities: 1. The factory-reset.target unit that requests a factory reset is now complemented by factory-reset-now.target that executes it at next boot. 2. This latter is added to the initial transaction via the new trivial systemd-factory-reset-generator. 3. A tool systemd-factory-reset has been added to query, request, cancel, complete factory reset operations (via EFI variables). Two of these are wrapped into units that are plugged into factory-reset.target and factory-reset-now.target respectively. The tool also provides a simple Varlink API. This should make things a lot cleaner, and both be useful as explicit implementation on UEFI, and as template + hookpoints for alternative implementations on non-UEFI.
2025-02-20 23:19:01 +01:00
subdir('src/factory-reset')
meson: move declarations of fsck, firstboot, machine-id-setup, and remount-fs
2023-06-25 16:21:17 +09:00
subdir('src/firstboot')
subdir('src/fsck')
meson: move declarations of several generators
2023-06-25 15:20:29 +09:00
subdir('src/fstab-generator')
subdir('src/getty-generator')
meson: move declarations of dissect and friends
2023-06-25 15:34:09 +09:00
subdir('src/gpt-auto-generator')
Move growfs+makefs to src/growfs/ Those two programs are used together and it makes sense to keep them together. makefs is smaller, so name the directory after growfs.
2024-11-10 14:01:34 +01:00
subdir('src/growfs')
meson: move declarations of hibernate-resume
2023-06-25 15:31:38 +09:00
subdir('src/hibernate-resume')
meson: sort inclusion of meson.build files in subdirectories
2021-01-04 21:27:00 +09:00
subdir('src/home')
meson: add test-dlopen test-dlopen is a very simple binary that is only linked with libc and libdl. From it we do dlopen() on the nss and pam modules to check that they are linked to all necessary libs. (meson-compiled nss modules are linked to less libraries, for whatever reason. I suspected that some deps are missing, but it turns out that my suspicions weren't justified, and the modules load just fine. Let's keep the test though, it is very quick, and might detect missing linkage in the future.)
2017-04-14 20:10:28 -04:00
subdir('src/hostname')
meson: move declarations of hwdb, sysusers, and tmpfiles
2023-06-26 03:28:32 +09:00
subdir('src/hwdb')
meson: move declarations of cgroups-agent, id128, volatile-root, and path
2023-06-25 16:34:27 +09:00
subdir('src/id128')
meson: add test-dlopen test-dlopen is a very simple binary that is only linked with libc and libdl. From it we do dlopen() on the nss and pam modules to check that they are linked to all necessary libs. (meson-compiled nss modules are linked to less libraries, for whatever reason. I suspected that some deps are missing, but it turns out that my suspicions weren't justified, and the modules load just fine. Let's keep the test though, it is very quick, and might detect missing linkage in the future.)
2017-04-14 20:10:28 -04:00
subdir('src/import')
meson: move declarations of cryptsetup and friends
2023-06-25 15:57:47 +09:00
subdir('src/integritysetup')
journal-remote: drop dependencies to journald Now journal-remote can be built without libjournal_core.
2023-09-30 19:08:23 +09:00
subdir('src/journal')
meson: sort inclusion of meson.build files in subdirectories
2021-01-04 21:27:00 +09:00
subdir('src/journal-remote')
meson: add test-dlopen test-dlopen is a very simple binary that is only linked with libc and libdl. From it we do dlopen() on the nss and pam modules to check that they are linked to all necessary libs. (meson-compiled nss modules are linked to less libraries, for whatever reason. I suspected that some deps are missing, but it turns out that my suspicions weren't justified, and the modules load just fine. Let's keep the test though, it is very quick, and might detect missing linkage in the future.)
2017-04-14 20:10:28 -04:00
subdir('src/kernel-install')
Introduce systemd-keyutil to do various key/certificate operations Let's gather generic key/certificate operations in a new tool systemd-keyutil instead of spreading them across various special purpose tools. Fixes #35087
2024-11-08 11:34:21 +01:00
subdir('src/keyutil')
meson: add test-dlopen test-dlopen is a very simple binary that is only linked with libc and libdl. From it we do dlopen() on the nss and pam modules to check that they are linked to all necessary libs. (meson-compiled nss modules are linked to less libraries, for whatever reason. I suspected that some deps are missing, but it turns out that my suspicions weren't justified, and the modules load just fine. Let's keep the test though, it is very quick, and might detect missing linkage in the future.)
2017-04-14 20:10:28 -04:00
subdir('src/locale')
meson: sort inclusion of meson.build files in subdirectories
2021-01-04 21:27:00 +09:00
subdir('src/login')
meson: add test-dlopen test-dlopen is a very simple binary that is only linked with libc and libdl. From it we do dlopen() on the nss and pam modules to check that they are linked to all necessary libs. (meson-compiled nss modules are linked to less libraries, for whatever reason. I suspected that some deps are missing, but it turns out that my suspicions weren't justified, and the modules load just fine. Let's keep the test though, it is very quick, and might detect missing linkage in the future.)
2017-04-14 20:10:28 -04:00
subdir('src/machine')
meson: move declarations of fsck, firstboot, machine-id-setup, and remount-fs
2023-06-25 16:21:17 +09:00
subdir('src/machine-id-setup')
Move systemd-measure to its own source subdirectory We have other subdirectories with just a single C file. And I expect that systemd-measure will only grow over time, adding new functionality. It's nicer to give its own subdirectory to maintain consistent structure.
2024-11-07 12:37:45 +01:00
subdir('src/measure')
meson: move declarations of modules-load, nspawn, update-done, and update-utmp
2023-06-26 03:38:48 +09:00
subdir('src/modules-load')
meson: move declarations of cgls, cgtop, initctl, and systemd-mount
2023-06-26 03:21:42 +09:00
subdir('src/mount')
meson: sort components
2025-08-28 11:50:39 +02:00
subdir('src/mountfsd')
mute-console: add simple varlink service that can disable log/status spew to kernel console For "wizard" style interactive tools it's very annoying if they are interrupted by kernel log output or PID1's status output. let's add some infra to disable this temporarily. I decided to implement this as an IPC service so that we can make this robust: if the client request the muting dies we can automatically unmute again. This is hence a tiny varlink service, but it can also be started directly from the cmdline.
2025-09-15 18:19:22 +02:00
subdir('src/mute-console')
meson: sort inclusion of meson.build files in subdirectories
2021-01-04 21:27:00 +09:00
subdir('src/network')
meson: move declarations of delta, escape, notify, creds, and battery-check
2023-06-25 16:31:06 +09:00
subdir('src/notify')
meson: add test-dlopen test-dlopen is a very simple binary that is only linked with libc and libdl. From it we do dlopen() on the nss and pam modules to check that they are linked to all necessary libs. (meson-compiled nss modules are linked to less libraries, for whatever reason. I suspected that some deps are missing, but it turns out that my suspicions weren't justified, and the modules load just fine. Let's keep the test though, it is very quick, and might detect missing linkage in the future.)
2017-04-14 20:10:28 -04:00
subdir('src/nspawn')
nsresourced: add new daemon for granting clients user namespaces and assigning resources to them This adds a small, socket-activated Varlink daemon that can delegate UID ranges for user namespaces to clients asking for it. The primary call is AllocateUserRange() where the user passes in an uninitialized userns fd, which is then set up. There are other calls that allow assigning a mount fd to a userns allocated that way, to set up permissions for a cgroup subtree, and to allocate a veth for such a user namespace. Since the UID assignments are supposed to be transitive, i.e. not permanent, care is taken to ensure that users cannot create inodes owned by these UIDs, so that persistancy cannot be acquired. This is implemented via a BPF-LSM module that ensures that any member of a userns allocated that way cannot create files unless the mount it operates on is owned by the userns itself, or is explicitly allowelisted. BPF LSM program with contributions from Alexei Starovoitov.
2023-04-20 19:07:33 +02:00
subdir('src/nsresourced')
meson: move declarations of nss modules
2023-06-23 11:28:22 +09:00
subdir('src/nss-myhostname')
subdir('src/nss-mymachines')
subdir('src/nss-resolve')
subdir('src/nss-systemd')
meson: sort inclusion of meson.build files in subdirectories
2021-01-04 21:27:00 +09:00
subdir('src/oom')
meson: move declarations of cgroups-agent, id128, volatile-root, and path
2023-06-25 16:34:27 +09:00
subdir('src/path')
pcrphase: rename binary to pcrextend The tool initially just measured the boot phase, but was subsequently extended to measure file system and machine IDs, too. At AllSystemsGo there were request to add more, and make the tool generically accessible. Hence, let's rename the binary (but not the pcrphase services), to make clear the tool is not just measureing the boot phase, but a lot of other things too. The tool is located in /usr/lib/ and still relatively new, hence let's just rename the binary and be done with it, while keeping the unit names stable. While we are at it, also move the tool out of src/boot/ and into its own src/pcrextend/ dir, since it's not really doing boot related stuff anymore.
2023-09-25 10:38:01 +02:00
subdir('src/pcrextend')
pcrlock: add new pcrlock tool
2023-08-16 11:39:23 +02:00
subdir('src/pcrlock')
meson: sort inclusion of meson.build files in subdirectories
2021-01-04 21:27:00 +09:00
subdir('src/portable')
subdir('src/pstore')
Introduce systemd-pty-forward This allows running a command with tinted terminal background.
2024-12-26 22:58:33 +01:00
subdir('src/ptyfwd')
meson: move declarations of socket-proxy, udevadm, quotacheck, and shutdown
2023-06-26 03:35:47 +09:00
subdir('src/quotacheck')
meson: move declarations of random-seed, vconsole, and sysupdate
2023-06-25 16:11:55 +09:00
subdir('src/random-seed')
meson: move declarations of several generators
2023-06-25 15:59:49 +09:00
subdir('src/rc-local-generator')
meson: move declarations of fsck, firstboot, machine-id-setup, and remount-fs
2023-06-25 16:21:17 +09:00
subdir('src/remount-fs')
Rename src/partition to src/repart
2024-11-10 14:05:08 +01:00
subdir('src/repart')
meson: move declarations of ask-password and friends
2023-06-26 03:18:38 +09:00
subdir('src/reply-password')
meson: add test-dlopen test-dlopen is a very simple binary that is only linked with libc and libdl. From it we do dlopen() on the nss and pam modules to check that they are linked to all necessary libs. (meson-compiled nss modules are linked to less libraries, for whatever reason. I suspected that some deps are missing, but it turns out that my suspicions weren't justified, and the modules load just fine. Let's keep the test though, it is very quick, and might detect missing linkage in the future.)
2017-04-14 20:10:28 -04:00
subdir('src/resolve')
meson: move declarations of backlight, rfkill, and sysupdate-generator
2023-06-25 15:55:33 +09:00
subdir('src/rfkill')
Move rpm stuff into a separate src/rpm/ directory It is only of interest to rpm-based distros, we can move it out of src/core/ which is pretty busy.
2021-01-31 17:16:32 +01:00
subdir('src/rpm')
meson: move declarations of busctl, stdio-bridge, and run
2023-06-26 03:23:59 +09:00
subdir('src/run')
meson: move declarations of several generators
2023-06-25 15:20:29 +09:00
subdir('src/run-generator')
Move systemd-sbsign to its own source subdirectory It's already two files, and I expect that more will come. It's nicer to give its own subdirectory to maintain consistent structure.
2024-11-07 12:37:45 +01:00
subdir('src/sbsign')
meson: sort inclusion of meson.build files in subdirectories
2021-01-04 21:27:00 +09:00
subdir('src/shutdown')
meson: move declarations of ac-power, detect-virt, sysctl, and sleep
2023-06-25 16:28:21 +09:00
subdir('src/sleep')
meson: move declaration of socket-activate
2023-06-25 15:42:23 +09:00
subdir('src/socket-activate')
meson: move declarations of socket-proxy, udevadm, quotacheck, and shutdown
2023-06-26 03:35:47 +09:00
subdir('src/socket-proxy')
ssh-generator: add simple new generator
2024-01-04 18:39:03 +01:00
subdir('src/ssh-generator')
meson: move declarations of busctl, stdio-bridge, and run
2023-06-26 03:23:59 +09:00
subdir('src/stdio-bridge')
meson: sort components
2025-08-28 11:50:39 +02:00
subdir('src/storagetm')
meson: move declarations of kernel-install and sulogin-shell
2023-06-26 03:44:39 +09:00
subdir('src/sulogin-shell')
meson: move declarations of ac-power, detect-virt, sysctl, and sleep
2023-06-25 16:28:21 +09:00
subdir('src/sysctl')
sysext: new tool for managing "system extensions" for /usr/ + /opt/
2021-01-08 16:57:27 +01:00
subdir('src/sysext')
meson: move declarations of backlight, rfkill, and sysupdate-generator
2023-06-25 15:55:33 +09:00
subdir('src/system-update-generator')
meson: move definition of systemctl source files
2021-01-04 20:12:28 +09:00
subdir('src/systemctl')
sysupdate: add new component "sysupdate"
2020-12-28 15:17:54 +01:00
subdir('src/sysupdate')
meson: move declarations of hwdb, sysusers, and tmpfiles
2023-06-26 03:28:32 +09:00
subdir('src/sysusers')
meson: move declarations of several generators
2023-06-25 15:59:49 +09:00
subdir('src/sysv-generator')
meson: add test-dlopen test-dlopen is a very simple binary that is only linked with libc and libdl. From it we do dlopen() on the nss and pam modules to check that they are linked to all necessary libs. (meson-compiled nss modules are linked to less libraries, for whatever reason. I suspected that some deps are missing, but it turns out that my suspicions weren't justified, and the modules load just fine. Let's keep the test though, it is very quick, and might detect missing linkage in the future.)
2017-04-14 20:10:28 -04:00
subdir('src/timedate')
subdir('src/timesync')
meson: build standalone version of systemd-tmpfiles Use -Dstandalone-binaries=yes to enable building and installing this standalone version of the binary without a dependency on the systemd-shared solib. Also move the list of sources for systemd-tmpfiles to its own meson.build file.
2020-06-03 14:49:53 -07:00
subdir('src/tmpfiles')
storagetm: add new systemd-storagetm component This implements a "storage target mode", similar to what MacOS provides since a long time as "Target Disk Mode": https://en.wikipedia.org/wiki/Target_Disk_Mode This implementation is relatively simple: 1. a new generic target "storage-target-mode.target" is added, which when booted into defines the target mode. 2. a small tool and service "systemd-storagetm.service" is added which exposes a specific device or all devices as NVMe-TCP devices over the network. NVMe-TCP appears to be hot shit right now how to expose block devices over the network. And it's really simple to set up via configs, hence our code is relatively short and neat. The idea is that systemd-storagetm.target can be extended sooner or later, for example to expose block devices also as USB mass storage devices and similar, in case the system has "dual mode" USB controller that can also work as device, not just as host. (And people could also plug in sharing as NBD, iSCSI, whatever they want.) How to use this? Boot into your system with a kernel cmdline of "rd.systemd.unit=storage-target-mode.target ip=link-local", and you'll see on screen the precise "nvme connect" command line to make the relevant block devices available locally on some other machine. This all requires that the target mode stuff is included in the initrd of course. And the system will the stay in the initrd forever. Why bother? Primarily three use-cases: 1. Debug a broken system: with very few dependencies during boot get access to the raw block device of a broken machine. 2. Migrate from system to another system, by dd'ing the old to the new directly. 3. Installing an OS remotely on some device (for example via Thunderbolt networking) (And there might be more, for example the ability to boot from a laptop's disk on another system) Limitations: 1. There's no authentication/encryption. Hence: use this on local links only. 2. NVMe target mode on Linux supports r/w operation only. Ideally, we'd have a read-only mode, for security reasons, and default to it. Future love: 1. We should have another mode, where we simply expose the homed LUKS home dirs like that. 2. Some lightweight hookup with plymouth, to display a (shortened) version of the info we write to the console. To test all this, just run: mkosi --kernel-command-line-extra="rd.systemd.unit=storage-target-mode.target" qemu
2023-10-27 14:25:49 +02:00
subdir('src/tpm2-setup')
meson: move declarations of ask-password and friends
2023-06-26 03:18:38 +09:00
subdir('src/tty-ask-password-agent')
meson: move declarations of modules-load, nspawn, update-done, and update-utmp
2023-06-26 03:38:48 +09:00
subdir('src/update-done')
subdir('src/update-utmp')
meson: move declaration of user-sessions
2023-06-25 15:39:45 +09:00
subdir('src/user-sessions')
meson: sort inclusion of meson.build files in subdirectories
2021-01-04 21:27:00 +09:00
subdir('src/userdb')
validatefs: add new tool that enforces mount constraints This new tool looks for a three xattr on the root inode of a file system that encode mount constraints of the file system. The tool is supposed to be hooke into the mount logic and is supposed to protect against misappropriating trusted file systems in unintended ways. Consider the following scenario: we boot up on first boot and create a tpm-locked pair of /var/ and /srv/ partitions via systemd-repart. An attacker then offline modifies the partition table, exchanging the metadata of the /var/ and /srv/ partition. So far we'd happily accept that, honour the modified metadata and boot up. This could be used to revert changes to /var/ or similar. And all that even though both partitions are encrypted and locked to TPM! With this new mechanism we can encode in the protected contents of the file systems the ways it can be used: the partition type uuid, the partition label and the intended mount point can be stored in xattrs, and we can check them automatically on mount, and take action on mismatch. (action would typically be immediate reboot).
2025-03-12 08:37:05 +01:00
subdir('src/validatefs')
varlinkctl: add new varlinkctl tool
2023-09-22 22:44:28 +02:00
subdir('src/varlinkctl')
meson: move declarations of random-seed, vconsole, and sysupdate
2023-06-25 16:11:55 +09:00
subdir('src/vconsole')
meson: move declarations of cryptsetup and friends
2023-06-25 15:57:47 +09:00
subdir('src/veritysetup')
vmspawn: added initial code for vmspawn vmspawn-settings.c is currently empty but this will be used in future to house code for parsing settings from a file
2023-10-12 13:13:19 +01:00
subdir('src/vmspawn')
meson: move declarations of cgroups-agent, id128, volatile-root, and path
2023-06-25 16:34:27 +09:00
subdir('src/volatile-root')
vpick: add new tool "systemd-vpick" which exposes vpick on the command line Usecase: $ du $(systemd-vpick /srv/myimages.v/foo___.raw) In order to determine size of newest image in /srv/myimages.v/
2023-03-03 18:26:45 +01:00
subdir('src/vpick')
meson: move source file list for systemd-xdg-autostart-generator and its tests
2021-01-04 20:41:37 +09:00
subdir('src/xdg-autostart-generator')
meson: add test-dlopen test-dlopen is a very simple binary that is only linked with libc and libdl. From it we do dlopen() on the nss and pam modules to check that they are linked to all necessary libs. (meson-compiled nss modules are linked to less libraries, for whatever reason. I suspected that some deps are missing, but it turns out that my suspicions weren't justified, and the modules load just fine. Let's keep the test though, it is very quick, and might detect missing linkage in the future.)
2017-04-14 20:10:28 -04:00
meson: move libudev related definitions to src/libudev/meson.build Also, this makes libudev.so built in build directory.
2021-01-04 21:13:30 +09:00
subdir('src/systemd')
meson: add test-dlopen test-dlopen is a very simple binary that is only linked with libc and libdl. From it we do dlopen() on the nss and pam modules to check that they are linked to all necessary libs. (meson-compiled nss modules are linked to less libraries, for whatever reason. I suspected that some deps are missing, but it turns out that my suspicions weren't justified, and the modules load just fine. Let's keep the test though, it is very quick, and might detect missing linkage in the future.)
2017-04-14 20:10:28 -04:00
subdir('src/test')
fuzz: add initial fuzzing infrastructure The fuzzers will be used by oss-fuzz to automatically and continuously fuzz systemd. This commit includes the build tooling necessary to build fuzz targets, and a fuzzer for the DNS packet parser.
2018-01-13 19:51:07 -05:00
subdir('src/fuzz')
tests: add pytest tests for ukify Some gymnastics were needed to import ukify as a module. Before the file was templated, this was trivial: insert the directory in sys.path, call import. But it's a real pain to import the unsuffixed file after processing. Instead, the untemplated file is imported, which works well enough for tests and is very simple. The tests can be called via pytest: PATH=build/:$PATH pytest -v src/ukify/test/test_ukify.py or directly: PATH=build/:$PATH src/ukify/test/test_ukify.py or via the meson test machinery output: meson test -C build test-ukify -v or without verbose output: meson test -C build test-ukify Zekret files are obfuscated using base64.
2022-11-25 17:21:40 +01:00
subdir('src/ukify/test') # needs to be last for test_env variable
meson: also merge declarations of fuzzers with other executables
2023-06-27 01:07:25 +09:00
subdir('test/fuzz')
meson: add test-dlopen test-dlopen is a very simple binary that is only linked with libc and libdl. From it we do dlopen() on the nss and pam modules to check that they are linked to all necessary libs. (meson-compiled nss modules are linked to less libraries, for whatever reason. I suspected that some deps are missing, but it turns out that my suspicions weren't justified, and the modules load just fine. Let's keep the test though, it is very quick, and might detect missing linkage in the future.)
2017-04-14 20:10:28 -04:00
mime: register confext/sysext images in shared-mime-info This make them recognized by file managers and stuff. Maybe one day we should properly register mime types in the "vnd." namespace with IANA, but I am too lazy to deal with the bureaucracy for that, hence let's stick with the x. namespace for now. This defines confext/sysext DDIs as subtype of: https://www.iana.org/assignments/media-types/application/vnd.efi.img Which is what everyone appears to use for raw disk images, in particular if they contain a GPT partition table.
2023-11-20 11:51:47 +01:00
subdir('mime')
meson: add systemd.pc and udev.pc to 'devel' target/tag Same as the other pkg-config files, it is useful to install these in a specific target that doesn't require compiling anything
2023-06-12 00:41:49 +01:00
alias_target('devel', libsystemd_pc, libudev_pc, systemd_pc, udev_pc)
meson: make lines more consistent
2023-12-14 18:35:00 -03:00
#####################################################################
meson: add test-dlopen test-dlopen is a very simple binary that is only linked with libc and libdl. From it we do dlopen() on the nss and pam modules to check that they are linked to all necessary libs. (meson-compiled nss modules are linked to less libraries, for whatever reason. I suspected that some deps are missing, but it turns out that my suspicions weren't justified, and the modules load just fine. Let's keep the test though, it is very quick, and might detect missing linkage in the future.)
2017-04-14 20:10:28 -04:00
meson: merge declarations of normal and test executables
2023-06-26 08:52:26 +09:00
foreach test : simple_tests
executables += test_template + { 'sources' : [test] }
endforeach
foreach test : libsystemd_tests
executables += test_template + test
endforeach
meson: also merge declarations of fuzzers with other executables
2023-06-27 01:07:25 +09:00
foreach fuzzer : simple_fuzzers
executables += fuzz_template + { 'sources' : [fuzzer] }
endforeach
meson: introduce infra to build executables and shared libraries by using dictionary That is not used yet. It will be used later.
2023-06-23 11:09:10 +09:00
foreach dict : executables
meson: merge declarations of normal and test executables
2023-06-26 08:52:26 +09:00
name = dict.get('name', '')
if name == ''
name = fs.stem(dict.get('sources')[0])
meson: also merge declarations of fuzzers with other executables
2023-06-27 01:07:25 +09:00
assert(name.split('-')[0] in ['test', 'fuzz'])
meson: merge declarations of normal and test executables
2023-06-26 08:52:26 +09:00
endif
is_test = name.startswith('test-')
meson: also merge declarations of fuzzers with other executables
2023-06-27 01:07:25 +09:00
is_fuzz = name.startswith('fuzz-')
meson: introduce infra to build executables and shared libraries by using dictionary That is not used yet. It will be used later.
2023-06-23 11:09:10 +09:00
build = true
foreach cond : dict.get('conditions', [])
if conf.get(cond) != 1
build = false
break
endif
endforeach
if not build
continue
endif
meson: Don't define targets for standalone binaries if option is not enabled These are built from the same sources as the regular binaries, so we end up with the same sources multiple times in the compilation database but with different command line arguments, which trips up tooling that uses the compilation database. Let's not define the standalone targets if the option is not enabled to avoid this problem.
2025-05-21 13:08:50 +02:00
if name.endswith('.standalone') and not have_standalone_binaries
continue
endif
meson: introduce infra to build executables and shared libraries by using dictionary That is not used yet. It will be used later.
2023-06-23 11:09:10 +09:00
kwargs = {}
foreach key, val : dict
meson: merge declarations of normal and test executables
2023-06-26 08:52:26 +09:00
if key in ['name', 'dbus', 'public', 'conditions',
meson: Extract objects instead of creating intermediate static libraries Currently, when we want to add unit tests for code that is compiled into an executable, we either compile the code at least twice (once for the executable, and once for each test that uses it) or we create a static library which is then used by both the executable and all the tests. Both of these options are not ideal, compiling source files more than once slows down the build for no reason and creating the intermediate static libraries takes a lot of boilerplate. Instead, let's use the extract_objects() method that meson exposes on build targets. This allows us to extract the objects corresponding to specific source files and use them in other executables. Because we define all executables upfront into a dictionary, we integrate this into the dictionary approach by adding two new fields: - 'extract' takes a list of files for which objects should be extracted. The extracted objects are stored in a dict keyed by the executable name from which they were extracted. - 'objects' takes the name of an executable from which the extracted objects should be added to the current executable. One side effect of this approach is that we can't build test executables anymore without building the main executable, so we stop building test executables unless we're also building the main executable. This allows us to switch to using subdir_done() in all of these subdirectories to skip parsing them if the corresponding component is disabled. These changes get me down from 2439 => 2403 ninja targets on a full rebuild from scratch.
2025-05-07 22:55:15 +02:00
'type', 'suite', 'timeout', 'parallel',
'objects', 'extract']
meson: introduce infra to build executables and shared libraries by using dictionary That is not used yet. It will be used later.
2023-06-23 11:09:10 +09:00
continue
endif
meson: merge declarations of normal and test executables
2023-06-26 08:52:26 +09:00
meson: introduce infra to build executables and shared libraries by using dictionary That is not used yet. It will be used later.
2023-06-23 11:09:10 +09:00
kwargs += { key : val }
endforeach
foreach key, val : executable_additional_kwargs
kwargs += { key : [ kwargs.get(key, []), val ]}
endforeach
meson: Extract objects instead of creating intermediate static libraries Currently, when we want to add unit tests for code that is compiled into an executable, we either compile the code at least twice (once for the executable, and once for each test that uses it) or we create a static library which is then used by both the executable and all the tests. Both of these options are not ideal, compiling source files more than once slows down the build for no reason and creating the intermediate static libraries takes a lot of boilerplate. Instead, let's use the extract_objects() method that meson exposes on build targets. This allows us to extract the objects corresponding to specific source files and use them in other executables. Because we define all executables upfront into a dictionary, we integrate this into the dictionary approach by adding two new fields: - 'extract' takes a list of files for which objects should be extracted. The extracted objects are stored in a dict keyed by the executable name from which they were extracted. - 'objects' takes the name of an executable from which the extracted objects should be added to the current executable. One side effect of this approach is that we can't build test executables anymore without building the main executable, so we stop building test executables unless we're also building the main executable. This allows us to switch to using subdir_done() in all of these subdirectories to skip parsing them if the corresponding component is disabled. These changes get me down from 2439 => 2403 ninja targets on a full rebuild from scratch.
2025-05-07 22:55:15 +02:00
foreach val : dict.get('objects', [])
obj = objects_by_name[val]
kwargs += {
'objects' : obj['objects'],
'include_directories' : [
kwargs.get('include_directories', []),
obj['include_directories'],
],
}
endforeach
meson: merge declarations of normal and test executables
2023-06-26 08:52:26 +09:00
if is_test
kwargs += { 'install_dir' : kwargs.get('install_dir') / dict.get('type', '') }
foreach key, val : test_additional_kwargs
kwargs += { key : [ kwargs.get(key, []), val ] }
endforeach
endif
meson: also merge declarations of fuzzers with other executables
2023-06-27 01:07:25 +09:00
if is_fuzz
foreach key, val : fuzz_additional_kwargs
kwargs += { key : [ kwargs.get(key, []), val ] }
endforeach
endif
meson: introduce infra to build executables and shared libraries by using dictionary That is not used yet. It will be used later.
2023-06-23 11:09:10 +09:00
exe = executable(
name,
kwargs : kwargs,
meson: Remove unneeded include directories meson by default adds the current source and build directory as include directories. Because we structure our meson code by gathering a giant dict of everything we want to do and then doing all the actual target generation in the top level meson.build, this behavior does not make sense at all because we end up adding the top level repository directory as an include directory which is never what we want. At the same time, let's also make sure the top level directory of the build directory is not an include directory, by moving the version.h generation into the src/version subdirectory and then adding the src/version subdirectory of the build directory as an include directory instead of the top level repository directory. Making this change means that language servers such as clangd can't get confused when they automatically insert an #include line and insert "#include "src/basic/fs-util.h" instead of "#include "fs-util.h".
2025-05-09 20:48:51 +02:00
implicit_include_directories : false,
meson: introduce infra to build executables and shared libraries by using dictionary That is not used yet. It will be used later.
2023-06-23 11:09:10 +09:00
)
executables_by_name += { name : exe }
meson: Rework clang-tidy integration to be done via unit tests Instead of using run-clang-tidy.py with its own scheduling, let's just gather a list of source files ourselves and then use that to add a unit test for each source file that runs clang-tidy on the source file. We also add a bit of logic to run clang-tidy on most header files as well for extra coverage. This uncovered various header files that were not standalone so this commit also includes fixes to make sure the clang-tidy tests are all green. We can also use this in a later commit to run clang-include-cleaner on each source file in the same way.
2025-05-15 15:09:27 +02:00
if not name.endswith('.standalone')
sources += dict.get('sources', [])
endif
meson: Extract objects instead of creating intermediate static libraries Currently, when we want to add unit tests for code that is compiled into an executable, we either compile the code at least twice (once for the executable, and once for each test that uses it) or we create a static library which is then used by both the executable and all the tests. Both of these options are not ideal, compiling source files more than once slows down the build for no reason and creating the intermediate static libraries takes a lot of boilerplate. Instead, let's use the extract_objects() method that meson exposes on build targets. This allows us to extract the objects corresponding to specific source files and use them in other executables. Because we define all executables upfront into a dictionary, we integrate this into the dictionary approach by adding two new fields: - 'extract' takes a list of files for which objects should be extracted. The extracted objects are stored in a dict keyed by the executable name from which they were extracted. - 'objects' takes the name of an executable from which the extracted objects should be added to the current executable. One side effect of this approach is that we can't build test executables anymore without building the main executable, so we stop building test executables unless we're also building the main executable. This allows us to switch to using subdir_done() in all of these subdirectories to skip parsing them if the corresponding component is disabled. These changes get me down from 2439 => 2403 ninja targets on a full rebuild from scratch.
2025-05-07 22:55:15 +02:00
if dict.has_key('extract')
objects_by_name += {
name : {
'objects' : exe.extract_objects(dict['extract']),
'include_directories' : fs.parent(dict['extract'][0]),
}
}
endif
meson: introduce infra to build executables and shared libraries by using dictionary That is not used yet. It will be used later.
2023-06-23 11:09:10 +09:00
if dict.get('build_by_default', true)
if dict.get('dbus', false)
dbus_programs += exe
endif
if dict.get('public', false)
public_programs += exe
endif
endif
meson: merge declarations of normal and test executables
2023-06-26 08:52:26 +09:00
if is_test
type = dict.get('type', '')
suite = dict.get('suite', '')
if suite == ''
suite = fs.name(fs.parent(dict.get('sources')[0]))
if suite.startswith('sd-')
suite = 'libsystemd'
endif
endif
if type == 'manual'
meson: using f-strings in meson Those were added before 0.61 which is our minimum version. Dots were dropped from the end of some messages. Most messages did not have the trailing dot. 0.63 added support for multi-line f-strings. We can't use those yet.
2025-10-03 15:26:20 +02:00
message(f'@suite@/@name@ is a manual test')
meson: merge declarations of normal and test executables
2023-06-26 08:52:26 +09:00
elif type == 'unsafe' and want_tests != 'unsafe'
meson: using f-strings in meson Those were added before 0.61 which is our minimum version. Dots were dropped from the end of some messages. Most messages did not have the trailing dot. 0.63 added support for multi-line f-strings. We can't use those yet.
2025-10-03 15:26:20 +02:00
message(f'@suite@/@name@ is an unsafe test')
meson: merge declarations of normal and test executables
2023-06-26 08:52:26 +09:00
elif dict.get('build_by_default')
test(name, exe,
env : test_env,
timeout : dict.get('timeout', 30),
suite : suite,
is_parallel : dict.get('parallel', true))
endif
endif
meson: also merge declarations of fuzzers with other executables
2023-06-27 01:07:25 +09:00
if is_fuzz
fuzzer_exes += exe
meson: Stop doing nested build when fuzzers are enabled Currently, when fuzzers are enabled, we run meson from within meson to build the fuzzer executables with sanitizers. The idea is that we can build the fuzzers with different kinds of sanitizers independently from the main build. The issue with this setup is that we don't actually make use of it. We only build the fuzzers with one set of sanitizers (address,undefined) so we're adding a bunch of extra complexity without any benefit as we can just setup the top level meson build with these sanitizers and get the same result. The other issue with this setup is that we don't pass on all the options passed to the top level meson build to the nested meson build. The only things we pass on are extra compiler arguments and the value of the auto_features option, but none of the individual feature options if overridden are passed on, which can lead to very hard to debug issues as an option enabled in the top level build is not enabled in the nested build. Since we're not getting anything useful out of this setup, let's simplify and get rid of the nested meson build. Instead, sanitizers should be enabled for the top level meson.build. This currently didn't work as we were overriding the sanitizers passed to the meson build with the fuzzer sanitizer, so we fix that as well by making sure we combine the fuzzer sanitizer with the ones passed in by the user. We also drop support for looking up libFuzzer as a separate library as it has been shipped builtin in clang since clang 6.0, so we can assume that -fsanitize=fuzzer is available. To make sure we still run the fuzzing tests, we enable the fuzz-tests option by default now to make sure they still always run (without instrumentation unless one of llvm-fuzz or oss-fuzz is enabled).
2025-05-11 09:42:28 +02:00
if want_tests != 'false' and want_fuzz_tests
meson: also merge declarations of fuzzers with other executables
2023-06-27 01:07:25 +09:00
fuzz_ins = fuzz_regression_tests.get(name, {})
foreach directive : fuzz_ins.get('directives', [])
meson: disallow fuzz test names above 60 characters The name is created as "systemd:fuzz / fuzz-<fuzzer_name>_<sample_name>" and if that's very long, output gets wrapped when 'meson test' is run, and this is rather annoying. Disallow filenames above 45 characters, which leads a 60 char names.
2023-09-02 17:23:22 +03:00
tt = '@0@_@1@'.format(name, fs.name(directive.full_path()))
if tt.substring(45) != ''
error('Directive sample name is too long:', directive.full_path())
endif
test(tt,
meson: also merge declarations of fuzzers with other executables
2023-06-27 01:07:25 +09:00
exe,
suite : 'fuzz',
args : directive.full_path(),
meson: Stop doing nested build when fuzzers are enabled Currently, when fuzzers are enabled, we run meson from within meson to build the fuzzer executables with sanitizers. The idea is that we can build the fuzzers with different kinds of sanitizers independently from the main build. The issue with this setup is that we don't actually make use of it. We only build the fuzzers with one set of sanitizers (address,undefined) so we're adding a bunch of extra complexity without any benefit as we can just setup the top level meson build with these sanitizers and get the same result. The other issue with this setup is that we don't pass on all the options passed to the top level meson build to the nested meson build. The only things we pass on are extra compiler arguments and the value of the auto_features option, but none of the individual feature options if overridden are passed on, which can lead to very hard to debug issues as an option enabled in the top level build is not enabled in the nested build. Since we're not getting anything useful out of this setup, let's simplify and get rid of the nested meson build. Instead, sanitizers should be enabled for the top level meson.build. This currently didn't work as we were overriding the sanitizers passed to the meson build with the fuzzer sanitizer, so we fix that as well by making sure we combine the fuzzer sanitizer with the ones passed in by the user. We also drop support for looking up libFuzzer as a separate library as it has been shipped builtin in clang since clang 6.0, so we can assume that -fsanitize=fuzzer is available. To make sure we still run the fuzzing tests, we enable the fuzz-tests option by default now to make sure they still always run (without instrumentation unless one of llvm-fuzz or oss-fuzz is enabled).
2025-05-11 09:42:28 +02:00
env : ['UBSAN_OPTIONS=print_stacktrace=1:print_summary=1:halt_on_error=1'],
timeout : 60,
meson: also merge declarations of fuzzers with other executables
2023-06-27 01:07:25 +09:00
depends : directive)
endforeach
foreach file : fuzz_ins.get('files', [])
meson: disallow fuzz test names above 60 characters The name is created as "systemd:fuzz / fuzz-<fuzzer_name>_<sample_name>" and if that's very long, output gets wrapped when 'meson test' is run, and this is rather annoying. Disallow filenames above 45 characters, which leads a 60 char names.
2023-09-02 17:23:22 +03:00
tt = '@0@_@1@'.format(name, fs.name(file))
if tt.substring(45) != ''
error('Fuzz sample name is too long:', fs.name(file))
endif
test(tt,
meson: also merge declarations of fuzzers with other executables
2023-06-27 01:07:25 +09:00
exe,
suite : 'fuzz',
meson: Stop doing nested build when fuzzers are enabled Currently, when fuzzers are enabled, we run meson from within meson to build the fuzzer executables with sanitizers. The idea is that we can build the fuzzers with different kinds of sanitizers independently from the main build. The issue with this setup is that we don't actually make use of it. We only build the fuzzers with one set of sanitizers (address,undefined) so we're adding a bunch of extra complexity without any benefit as we can just setup the top level meson build with these sanitizers and get the same result. The other issue with this setup is that we don't pass on all the options passed to the top level meson build to the nested meson build. The only things we pass on are extra compiler arguments and the value of the auto_features option, but none of the individual feature options if overridden are passed on, which can lead to very hard to debug issues as an option enabled in the top level build is not enabled in the nested build. Since we're not getting anything useful out of this setup, let's simplify and get rid of the nested meson build. Instead, sanitizers should be enabled for the top level meson.build. This currently didn't work as we were overriding the sanitizers passed to the meson build with the fuzzer sanitizer, so we fix that as well by making sure we combine the fuzzer sanitizer with the ones passed in by the user. We also drop support for looking up libFuzzer as a separate library as it has been shipped builtin in clang since clang 6.0, so we can assume that -fsanitize=fuzzer is available. To make sure we still run the fuzzing tests, we enable the fuzz-tests option by default now to make sure they still always run (without instrumentation unless one of llvm-fuzz or oss-fuzz is enabled).
2025-05-11 09:42:28 +02:00
env : ['UBSAN_OPTIONS=print_stacktrace=1:print_summary=1:halt_on_error=1'],
timeout : 60,
meson: also merge declarations of fuzzers with other executables
2023-06-27 01:07:25 +09:00
args : file)
endforeach
endif
endif
meson: introduce infra to build executables and shared libraries by using dictionary That is not used yet. It will be used later.
2023-06-23 11:09:10 +09:00
endforeach
meson: also merge declarations of fuzzers with other executables
2023-06-27 01:07:25 +09:00
alias_target('fuzzers', fuzzer_exes)
meson: make lines more consistent
2023-12-14 18:35:00 -03:00
#####################################################################
meson: introduce infra to build executables and shared libraries by using dictionary That is not used yet. It will be used later.
2023-06-23 11:09:10 +09:00
meson: merge declarations of normal and test executables
2023-06-26 08:52:26 +09:00
test_dlopen = executables_by_name.get('test-dlopen')
meson: add test-dlopen test-dlopen is a very simple binary that is only linked with libc and libdl. From it we do dlopen() on the nss and pam modules to check that they are linked to all necessary libs. (meson-compiled nss modules are linked to less libraries, for whatever reason. I suspected that some deps are missing, but it turns out that my suspicions weren't justified, and the modules load just fine. Let's keep the test though, it is very quick, and might detect missing linkage in the future.)
2017-04-14 20:10:28 -04:00
meson: add alias targets to group nss and pam modules Add aliases grouping these modules, so that they can be built without knowing the SONAME version in advance. Match the install tag names.
2024-05-31 21:02:31 +01:00
nss_targets = []
pam_targets = []
meson: introduce infra to build executables and shared libraries by using dictionary That is not used yet. It will be used later.
2023-06-23 11:09:10 +09:00
foreach dict : modules
name = dict.get('name')
is_nss = name.startswith('nss_')
is_pam = name.startswith('pam_')
build = true
foreach cond : dict.get('conditions', [])
if conf.get(cond) != 1
build = false
break
endif
endforeach
if not build
continue
endif
kwargs = {}
foreach key, val : dict
meson: Fix version script handling Build targets should have a link dependency on the version scripts they use. This also uses absolute paths in anticipation for meson 1.3 needlessly deprecating file to string conversions.
2023-09-26 15:14:38 +02:00
if key in ['name', 'conditions', 'version-script']
meson: introduce infra to build executables and shared libraries by using dictionary That is not used yet. It will be used later.
2023-06-23 11:09:10 +09:00
continue
endif
kwargs += { key : val }
endforeach
kwargs += {
'link_args' : [
kwargs.get('link_args', []),
meson: Fix version script handling Build targets should have a link dependency on the version scripts they use. This also uses absolute paths in anticipation for meson 1.3 needlessly deprecating file to string conversions.
2023-09-26 15:14:38 +02:00
'-Wl,--version-script=' + dict.get('version-script'),
],
'link_depends' : [
kwargs.get('link_depends', []),
dict.get('version-script'),
meson: introduce infra to build executables and shared libraries by using dictionary That is not used yet. It will be used later.
2023-06-23 11:09:10 +09:00
],
}
foreach key, val : module_additional_kwargs
kwargs += { key : [ kwargs.get(key, []), val ]}
endforeach
lib = shared_library(
name,
kwargs : kwargs,
meson: Remove unneeded include directories meson by default adds the current source and build directory as include directories. Because we structure our meson code by gathering a giant dict of everything we want to do and then doing all the actual target generation in the top level meson.build, this behavior does not make sense at all because we end up adding the top level repository directory as an include directory which is never what we want. At the same time, let's also make sure the top level directory of the build directory is not an include directory, by moving the version.h generation into the src/version subdirectory and then adding the src/version subdirectory of the build directory as an include directory instead of the top level repository directory. Making this change means that language servers such as clangd can't get confused when they automatically insert an #include line and insert "#include "src/basic/fs-util.h" instead of "#include "fs-util.h".
2025-05-09 20:48:51 +02:00
implicit_include_directories : false,
meson: introduce infra to build executables and shared libraries by using dictionary That is not used yet. It will be used later.
2023-06-23 11:09:10 +09:00
)
if is_nss
# We cannot use shared_module because it does not support version suffix.
# Unfortunately shared_library insists on creating the symlink…
meson: using f-strings in meson Those were added before 0.61 which is our minimum version. Dots were dropped from the end of some messages. Most messages did not have the trailing dot. 0.63 added support for multi-line f-strings. We can't use those yet.
2025-10-03 15:26:20 +02:00
meson.add_install_script(sh, '-c', f'rm -f $DESTDIR@libdir@/lib@name@.so',
meson: introduce infra to build executables and shared libraries by using dictionary That is not used yet. It will be used later.
2023-06-23 11:09:10 +09:00
install_tag : 'nss')
meson: add alias targets to group nss and pam modules Add aliases grouping these modules, so that they can be built without knowing the SONAME version in advance. Match the install tag names.
2024-05-31 21:02:31 +01:00
nss_targets += lib
endif
if is_pam
pam_targets += lib
meson: introduce infra to build executables and shared libraries by using dictionary That is not used yet. It will be used later.
2023-06-23 11:09:10 +09:00
endif
if want_tests != 'false' and (is_nss or is_pam)
test('dlopen-' + name,
test_dlopen,
# path to dlopen must include a slash
args : lib.full_path(),
meson: set suite for all tests, and adjust suite for some tests
2023-06-27 01:28:32 +09:00
depends : lib,
suite : is_nss ? 'nss' : 'pam')
meson: introduce infra to build executables and shared libraries by using dictionary That is not used yet. It will be used later.
2023-06-23 11:09:10 +09:00
endif
endforeach
meson: add alias targets to group nss and pam modules Add aliases grouping these modules, so that they can be built without knowing the SONAME version in advance. Match the install tag names.
2024-05-31 21:02:31 +01:00
# We need the actual targets to build aliases
if nss_targets.length() > 0
alias_target('nss', nss_targets)
endif
if pam_targets.length() > 0
alias_target('pam', pam_targets)
endif
meson: make lines more consistent
2023-12-14 18:35:00 -03:00
#####################################################################
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
Add proper dependencies to ukify target Also remove the systemd-measure dependency from the mkosi target as mkosi doesn't invoke systemd-measure itself.
2024-11-15 10:34:27 +01:00
ukify_depends = []
foreach executable : ['systemd-measure', 'systemd-sbsign', 'systemd-keyutil']
if executable in executables_by_name
ukify_depends += [executables_by_name[executable]]
endif
endforeach
test-kernel-install: test 60-ukify.install and 90-uki-copy.install We install a kernel with layout=uki and uki_generator=ukify, and test that a UKI gets installed in the expected place. The two plugins cooperate, so it's easiest to test them together.
2023-04-13 18:11:39 +02:00
ukify = custom_target(
meson,ukify: hook up ukify, add --version option The option is added because we have a similar one for kernel-install. This program requires python, and some people might want to skip it because of this. The tool is installed in /usr/lib/systemd for now, since the interface might change. A template file is used, but there is no .in suffix. The problem is that we'll later want to import the file as a module for tests, but recent Python versions make it annoyingly hard to import a module from a file without a .py suffix. imp.load_sources() works, but it is deprecated and throws warnings. importlib.machinery.SourceFileLoader().load_module() works, but is also deprecated. And the documented replacements are a maze of twisted little callbacks that result in an empty module. So let's take the easy way out, and skip the suffix which makes it easy to import the template as a module after adding the directory to sys.path.
2022-11-16 15:52:47 +01:00
input : 'src/ukify/ukify.py',
output : 'ukify',
command : [jinja2_cmdline, '@INPUT@', '@OUTPUT@'],
meson: avoid building executables that won't be installed When executable() or custom_target() has install: that is conditional as is false (i.e. not install:true), it won't be built by default. (build_by_default: defaults to install:). But if that program is added to public_programs, it will be build by default because it is pulled in by the test, effectively defeating the disablement. While at it, make 'ukify' follow the same pattern as 'kernel-install'. They will be used later together.
2023-04-14 18:10:58 +02:00
install : want_ukify,
meson,ukify: hook up ukify, add --version option The option is added because we have a similar one for kernel-install. This program requires python, and some people might want to skip it because of this. The tool is installed in /usr/lib/systemd for now, since the interface might change. A template file is used, but there is no .in suffix. The problem is that we'll later want to import the file as a module for tests, but recent Python versions make it annoyingly hard to import a module from a file without a .py suffix. imp.load_sources() works, but it is deprecated and throws warnings. importlib.machinery.SourceFileLoader().load_module() works, but is also deprecated. And the documented replacements are a maze of twisted little callbacks that result in an empty module. So let's take the easy way out, and skip the suffix which makes it easy to import the template as a module after adding the directory to sys.path.
2022-11-16 15:52:47 +01:00
install_mode : 'rwxr-xr-x',
Add proper dependencies to ukify target Also remove the systemd-measure dependency from the mkosi target as mkosi doesn't invoke systemd-measure itself.
2024-11-15 10:34:27 +01:00
depends : ukify_depends,
ukify: move to /usr/bin and mark as non non-experimental The notice in the man page is removed and the tool is moved into the $PATH. A compat symlink is provided. It is fairly widely used now, and realistically we need to keep backwards compat or people will be very unhappy.
2023-08-28 18:22:43 +03:00
install_dir : bindir)
meson: avoid building executables that won't be installed When executable() or custom_target() has install: that is conditional as is false (i.e. not install:true), it won't be built by default. (build_by_default: defaults to install:). But if that program is added to public_programs, it will be build by default because it is pulled in by the test, effectively defeating the disablement. While at it, make 'ukify' follow the same pattern as 'kernel-install'. They will be used later together.
2023-04-14 18:10:58 +02:00
if want_ukify
kernel-install: rewrite in C This is mostly a one-to-one translation of kernel-install.sh, except for the followings: - BOOT_ROOT is searched with find_{esp,xbootldr}_and_warn(). - entry token is searched with boot_entry_token_ensure(). - inspect command verboses more information, e.g. found plugins, environment variables explicitly passed to plugins, arguments passed to plugins. - paths specified in $KERNEL_INSTALL_PLUGINS must be absolute. - LC_COLLATE is set to C.UTF-8 (or any specified on build time). By writing kernel-install C, we can share the code used by bootctl or so, and can introduce --root and/or --image options later.
2023-03-14 23:55:18 +09:00
public_programs += ukify
ukify: move to /usr/bin and mark as non non-experimental The notice in the man page is removed and the tool is moved into the $PATH. A compat symlink is provided. It is fairly widely used now, and realistically we need to keep backwards compat or people will be very unhappy.
2023-08-28 18:22:43 +03:00
meson: add comments to compat symlinks
2023-09-26 17:03:15 +02:00
# symlink for backwards compatibility after rename
meson: use install_symlink() where applicable Now our baseline of meson is 0.62, hence install_symlink() can be used. Note, install_symlink() implies install_emptydir() for specified install_dir. Hence, this also drops several unnecessary install_emptydir() calls. Note, the function currently does not support 'relative' and 'force' flags, so several 'ln -frsT' inline calls cannot be replaced.
2025-03-05 06:27:08 +09:00
install_symlink('ukify',
pointing_to : libexecdir_to_bin / 'ukify',
install_dir : libexecdir)
test-kernel-install: test 60-ukify.install and 90-uki-copy.install We install a kernel with layout=uki and uki_generator=ukify, and test that a UKI gets installed in the expected place. The two plugins cooperate, so it's easiest to test them together.
2023-04-13 18:11:39 +02:00
endif
meson: make lines more consistent
2023-12-14 18:35:00 -03:00
#####################################################################
meson: define tests Tests can be run with 'ninja-build test' or using 'mesontest'. '-Dtests=unsafe' can be used to include the "unsafe" tests in the test suite, same as with autotools. v2: - use more conf.get guards are optional components - declare deps on generated headers for test-{af,arphrd,cap}-list v3: - define environment for tests Most test don't need this, but to be consistent with autotools-based build, and to avoid questions which tests need it and which don't, set the same environment for all tests. v4: - rework test generation Use a list of lists to define each test. This way we can reduce the boilerplate somewhat, although the test listings are still pretty verbose. We can also move the definitions of the tests to the subdirs. Unfortunately some subdirs are included earlier than some of the libraries that test binaries are linked to. So just dump all definitions of all tests that cannot be defined earlier into src/test. The `executable` definitions are still at the top level, so the binaries are compiled into the build root. v5: - tag test-dnssec-complex as manual v6: - fix HAVE_LIBZ typo - add missing libgobject/libgio defs - mark test-qcow2 as manual
2017-04-07 00:19:09 -04:00
test: Various mkosi integration test improvements - Stop using logging module since the default output formatting is pretty bad. Prefer print() for now. - Log less, logging the full mkosi command line is rather verbose, especially when it contains multi-line dropins. - Streamline the journalctl command we output for debugging failed tests. - Don't force usage of the disk image format. - Don't force running without unit tests. - Don't force disabling RuntimeBuildSources. - Update documentation to streamline the command for running a single test and remove sudo as it's not required anymore. - Improve the console output by having the test unit's output logged to both the journal and the console. - Disable journal console log forwarding as we have journal forwarding as a better alternative. - Delete existing journal file before running test. - Delete journal files of succeeded tests to reduce disk usage. - Rename system_mkosi target to just mkosi - Pass in mkosi source directory explicitly to accomodate arbitrary build directory locations. - Add test interactive debugging if stdout is connected to a tty - Stop explicitly using the 'system' image since it'll likely be dropped soon. - Only forward journal if we're not running in debugging mode. - Stop using testsuite.target and instead just add the necessary extras to the main testsuite unit via the credential dropin. - Override type to idle so test output is not interleaved with status output. - Don't build mkosi target by default - Always add the mkosi target if mkosi is found - Remove dependency of the integration tests on the mkosi target as otherwise the image is always built, even though we configure it to not be built by default. - Move mkosi output, cache and build directory into build/ so that invocations from meson and regular invocations share the same directories. - Various aesthetic cleanups.
2024-04-19 09:17:58 +02:00
mkosi = find_program('mkosi', required : false)
meson: Properly check dependencies of mkosi target Let's insist on mkosi being found if the integration-tests option is enabled and let's only add dependencies on systemd-journal-remote and systemd-measure if they're being built. Drop ukify from the list as its part of public_programs.
2024-04-24 21:17:06 +02:00
mkosi_depends = public_programs
Add proper dependencies to ukify target Also remove the systemd-measure dependency from the mkosi target as mkosi doesn't invoke systemd-measure itself.
2024-11-15 10:34:27 +01:00
foreach executable : ['systemd-journal-remote', 'systemd-sbsign', 'systemd-keyutil']
meson: Properly check dependencies of mkosi target Let's insist on mkosi being found if the integration-tests option is enabled and let's only add dependencies on systemd-journal-remote and systemd-measure if they're being built. Drop ukify from the list as its part of public_programs.
2024-04-24 21:17:06 +02:00
if executable in executables_by_name
mkosi_depends += [executables_by_name[executable]]
endif
endforeach
test: Various mkosi integration test improvements - Stop using logging module since the default output formatting is pretty bad. Prefer print() for now. - Log less, logging the full mkosi command line is rather verbose, especially when it contains multi-line dropins. - Streamline the journalctl command we output for debugging failed tests. - Don't force usage of the disk image format. - Don't force running without unit tests. - Don't force disabling RuntimeBuildSources. - Update documentation to streamline the command for running a single test and remove sudo as it's not required anymore. - Improve the console output by having the test unit's output logged to both the journal and the console. - Disable journal console log forwarding as we have journal forwarding as a better alternative. - Delete existing journal file before running test. - Delete journal files of succeeded tests to reduce disk usage. - Rename system_mkosi target to just mkosi - Pass in mkosi source directory explicitly to accomodate arbitrary build directory locations. - Add test interactive debugging if stdout is connected to a tty - Stop explicitly using the 'system' image since it'll likely be dropped soon. - Only forward journal if we're not running in debugging mode. - Stop using testsuite.target and instead just add the necessary extras to the main testsuite unit via the credential dropin. - Override type to idle so test output is not interleaved with status output. - Don't build mkosi target by default - Always add the mkosi target if mkosi is found - Remove dependency of the integration tests on the mkosi target as otherwise the image is always built, even though we configure it to not be built by default. - Move mkosi output, cache and build directory into build/ so that invocations from meson and regular invocations share the same directories. - Various aesthetic cleanups.
2024-04-19 09:17:58 +02:00
if mkosi.found()
meson: drop remaining target names Most of those were dropped in 7d247d3cb875e1778e6cc81dbf255b13092b29e0. I left behind the few cases where the 'output' name was different from the target name. But we actually don't need those either, so get rid of those. (And one case where I missed the name argument.) No functional change, except that e.g. 'ninja -C build update-man-rules-impl' doesn't work, but I hope nobody was using that.
2025-10-04 18:05:47 +02:00
custom_target(
test: Add mkosi-based integration test runner The first two tests are included to ensure parallel test execution is demonstrable.
2024-04-05 17:19:59 +01:00
build_always_stale : true,
test: Various mkosi integration test improvements - Stop using logging module since the default output formatting is pretty bad. Prefer print() for now. - Log less, logging the full mkosi command line is rather verbose, especially when it contains multi-line dropins. - Streamline the journalctl command we output for debugging failed tests. - Don't force usage of the disk image format. - Don't force running without unit tests. - Don't force disabling RuntimeBuildSources. - Update documentation to streamline the command for running a single test and remove sudo as it's not required anymore. - Improve the console output by having the test unit's output logged to both the journal and the console. - Disable journal console log forwarding as we have journal forwarding as a better alternative. - Delete existing journal file before running test. - Delete journal files of succeeded tests to reduce disk usage. - Rename system_mkosi target to just mkosi - Pass in mkosi source directory explicitly to accomodate arbitrary build directory locations. - Add test interactive debugging if stdout is connected to a tty - Stop explicitly using the 'system' image since it'll likely be dropped soon. - Only forward journal if we're not running in debugging mode. - Stop using testsuite.target and instead just add the necessary extras to the main testsuite unit via the credential dropin. - Override type to idle so test output is not interleaved with status output. - Don't build mkosi target by default - Always add the mkosi target if mkosi is found - Remove dependency of the integration tests on the mkosi target as otherwise the image is always built, even though we configure it to not be built by default. - Move mkosi output, cache and build directory into build/ so that invocations from meson and regular invocations share the same directories. - Various aesthetic cleanups.
2024-04-19 09:17:58 +02:00
build_by_default: false,
test: Add mkosi-based integration test runner The first two tests are included to ensure parallel test execution is demonstrable.
2024-04-05 17:19:59 +01:00
console : true,
meson: drop remaining target names Most of those were dropped in 7d247d3cb875e1778e6cc81dbf255b13092b29e0. I left behind the few cases where the 'output' name was different from the target name. But we actually don't need those either, so get rid of those. (And one case where I missed the name argument.) No functional change, except that e.g. 'ninja -C build update-man-rules-impl' doesn't work, but I hope nobody was using that.
2025-10-04 18:05:47 +02:00
output : 'mkosi',
test: Various mkosi integration test improvements - Stop using logging module since the default output formatting is pretty bad. Prefer print() for now. - Log less, logging the full mkosi command line is rather verbose, especially when it contains multi-line dropins. - Streamline the journalctl command we output for debugging failed tests. - Don't force usage of the disk image format. - Don't force running without unit tests. - Don't force disabling RuntimeBuildSources. - Update documentation to streamline the command for running a single test and remove sudo as it's not required anymore. - Improve the console output by having the test unit's output logged to both the journal and the console. - Disable journal console log forwarding as we have journal forwarding as a better alternative. - Delete existing journal file before running test. - Delete journal files of succeeded tests to reduce disk usage. - Rename system_mkosi target to just mkosi - Pass in mkosi source directory explicitly to accomodate arbitrary build directory locations. - Add test interactive debugging if stdout is connected to a tty - Stop explicitly using the 'system' image since it'll likely be dropped soon. - Only forward journal if we're not running in debugging mode. - Stop using testsuite.target and instead just add the necessary extras to the main testsuite unit via the credential dropin. - Override type to idle so test output is not interleaved with status output. - Don't build mkosi target by default - Always add the mkosi target if mkosi is found - Remove dependency of the integration tests on the mkosi target as otherwise the image is always built, even though we configure it to not be built by default. - Move mkosi output, cache and build directory into build/ so that invocations from meson and regular invocations share the same directories. - Various aesthetic cleanups.
2024-04-19 09:17:58 +02:00
command : [
meson: Look up mkosi once
2024-05-27 11:15:02 +02:00
mkosi,
test: Various mkosi integration test improvements - Stop using logging module since the default output formatting is pretty bad. Prefer print() for now. - Log less, logging the full mkosi command line is rather verbose, especially when it contains multi-line dropins. - Streamline the journalctl command we output for debugging failed tests. - Don't force usage of the disk image format. - Don't force running without unit tests. - Don't force disabling RuntimeBuildSources. - Update documentation to streamline the command for running a single test and remove sudo as it's not required anymore. - Improve the console output by having the test unit's output logged to both the journal and the console. - Disable journal console log forwarding as we have journal forwarding as a better alternative. - Delete existing journal file before running test. - Delete journal files of succeeded tests to reduce disk usage. - Rename system_mkosi target to just mkosi - Pass in mkosi source directory explicitly to accomodate arbitrary build directory locations. - Add test interactive debugging if stdout is connected to a tty - Stop explicitly using the 'system' image since it'll likely be dropped soon. - Only forward journal if we're not running in debugging mode. - Stop using testsuite.target and instead just add the necessary extras to the main testsuite unit via the credential dropin. - Override type to idle so test output is not interleaved with status output. - Don't build mkosi target by default - Always add the mkosi target if mkosi is found - Remove dependency of the integration tests on the mkosi target as otherwise the image is always built, even though we configure it to not be built by default. - Move mkosi output, cache and build directory into build/ so that invocations from meson and regular invocations share the same directories. - Various aesthetic cleanups.
2024-04-19 09:17:58 +02:00
'--directory', meson.current_source_dir(),
'--output-dir', meson.current_build_dir() / 'mkosi.output',
'--cache-dir', meson.current_build_dir() / 'mkosi.cache',
'--build-dir', meson.current_build_dir() / 'mkosi.builddir',
mkosi: Streamline running the integration tests without building systemd Let's document in detail how to build the integration test image and run the integration tests without building systemd. To streamline the process, we stop automatically using binaries from build/ when invoking mkosi directly and don't automatically use a tools tree anymore if systemd on the host is too old. Instead, we document these options in HACKING.md and change the mkosi meson target to automatically use the current build directory as an extra binary search path for mkosi.
2024-07-11 16:09:47 +02:00
'--extra-search-path', meson.current_build_dir(),
test: Various mkosi integration test improvements - Stop using logging module since the default output formatting is pretty bad. Prefer print() for now. - Log less, logging the full mkosi command line is rather verbose, especially when it contains multi-line dropins. - Streamline the journalctl command we output for debugging failed tests. - Don't force usage of the disk image format. - Don't force running without unit tests. - Don't force disabling RuntimeBuildSources. - Update documentation to streamline the command for running a single test and remove sudo as it's not required anymore. - Improve the console output by having the test unit's output logged to both the journal and the console. - Disable journal console log forwarding as we have journal forwarding as a better alternative. - Delete existing journal file before running test. - Delete journal files of succeeded tests to reduce disk usage. - Rename system_mkosi target to just mkosi - Pass in mkosi source directory explicitly to accomodate arbitrary build directory locations. - Add test interactive debugging if stdout is connected to a tty - Stop explicitly using the 'system' image since it'll likely be dropped soon. - Only forward journal if we're not running in debugging mode. - Stop using testsuite.target and instead just add the necessary extras to the main testsuite unit via the credential dropin. - Override type to idle so test output is not interleaved with status output. - Don't build mkosi target by default - Always add the mkosi target if mkosi is found - Remove dependency of the integration tests on the mkosi target as otherwise the image is always built, even though we configure it to not be built by default. - Move mkosi output, cache and build directory into build/ so that invocations from meson and regular invocations share the same directories. - Various aesthetic cleanups.
2024-04-19 09:17:58 +02:00
'--force',
meson: Properly check dependencies of mkosi target Let's insist on mkosi being found if the integration-tests option is enabled and let's only add dependencies on systemd-journal-remote and systemd-measure if they're being built. Drop ukify from the list as its part of public_programs.
2024-04-24 21:17:06 +02:00
'build',
test: Various mkosi integration test improvements - Stop using logging module since the default output formatting is pretty bad. Prefer print() for now. - Log less, logging the full mkosi command line is rather verbose, especially when it contains multi-line dropins. - Streamline the journalctl command we output for debugging failed tests. - Don't force usage of the disk image format. - Don't force running without unit tests. - Don't force disabling RuntimeBuildSources. - Update documentation to streamline the command for running a single test and remove sudo as it's not required anymore. - Improve the console output by having the test unit's output logged to both the journal and the console. - Disable journal console log forwarding as we have journal forwarding as a better alternative. - Delete existing journal file before running test. - Delete journal files of succeeded tests to reduce disk usage. - Rename system_mkosi target to just mkosi - Pass in mkosi source directory explicitly to accomodate arbitrary build directory locations. - Add test interactive debugging if stdout is connected to a tty - Stop explicitly using the 'system' image since it'll likely be dropped soon. - Only forward journal if we're not running in debugging mode. - Stop using testsuite.target and instead just add the necessary extras to the main testsuite unit via the credential dropin. - Override type to idle so test output is not interleaved with status output. - Don't build mkosi target by default - Always add the mkosi target if mkosi is found - Remove dependency of the integration tests on the mkosi target as otherwise the image is always built, even though we configure it to not be built by default. - Move mkosi output, cache and build directory into build/ so that invocations from meson and regular invocations share the same directories. - Various aesthetic cleanups.
2024-04-19 09:17:58 +02:00
],
meson: Drop genkey target In mkosi.images/system/mkosi.conf, we configure the certificate as an extra tree so it's available inside the image. However, we pick up the certificate from the top level repository directory and not from the build directory where it is generated by the genkey meson target. We currently have no way to access the build directory that mkosi was invoked from when parsing the configuration file. Thus we have no way to specify the correct location to the certificate when it's located in the build directory. For now, let's look for the key and certificate in the top level repository root directory and drop the genkey target. We don't have to change the Github Actions CI because it already runs genkey manually before the image build (which is something we forgot to remove when introducing the genkey target and is the reason this didn't cause issues before).
2024-06-28 09:58:56 +02:00
depends : mkosi_depends,
test: Add mkosi-based integration test runner The first two tests are included to ensure parallel test execution is demonstrable.
2024-04-05 17:19:59 +01:00
)
endif
test: Install files required to run integration tests standalone Let's make it possible to run the integration tests without requiring the sources by installing the mkosi and integration-tests directories.
2025-04-08 11:00:20 +02:00
if install_tests
install_subdir('mkosi',
install_dir : testsdir,
exclude_files : ['mkosi.local.conf', 'mkosi.key', 'mkosi.crt'],
exclude_directories : ['mkosi.local'])
endif
test: Add mkosi-based integration test runner The first two tests are included to ensure parallel test execution is demonstrable.
2024-04-05 17:19:59 +01:00
############################################################
meson: move several test declarations One of the notable change is that previously test-sysusers.sh was installed unconditionally, but now it is installed only when sysusers is enabled. Another change is that test-sysv-generator is now re-introduced which was mistakenly dropped by 6c713961ab0831fe744a2df9c4e9e258b6ba3105.
2023-06-26 03:46:05 +09:00
subdir('rules.d')
subdir('test')
meson: make lines more consistent
2023-12-14 18:35:00 -03:00
#####################################################################
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: resort imports
2022-11-16 15:52:15 +01:00
subdir('docs/sysvinit')
subdir('docs/var-log')
subdir('hwdb.d')
subdir('man')
meson: install the right README file in modprobe.d We put the "global" README file there. Introduced in d83e90c73cf25a839f5e60f355baa0d38364ff41.
2021-07-07 12:39:33 +02:00
subdir('modprobe.d')
meson: resort imports
2022-11-16 15:52:15 +01:00
subdir('network')
subdir('presets')
shell: define three system credentials we can propagate into shell prompts and welcome messages
2024-09-06 16:30:54 +02:00
subdir('profile.d')
meson: resort imports
2022-11-16 15:52:15 +01:00
subdir('shell-completion/bash')
subdir('shell-completion/zsh')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
subdir('sysctl.d')
subdir('sysusers.d')
subdir('tmpfiles.d')
Rename udev's hwdb/ to hwdb.d/ As in the parent commit, this makes the name consistent with the rest of the source tree and the actuall installation path.
2019-10-08 16:52:10 +02:00
subdir('units')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
install_subdir('factory/etc',
install_dir : factorydir)
factory: populate /etc/locale.conf with systemd build-time setting This adds /etc/locale.conf to the set of configuration files populated by tmpfiles.d factory /etc handling. In particular, the build-time locale configuration in systemd is now wired to a /usr factory file, and installed to the system. On boot, if other locale customization tools did not write /etc/locale.conf on the system, the factory default file gets copied to /etc by systemd-tmpfiles. This is done in order to avoid skews between different system components when no locale settings are configured. At that point, systemd can safely act as the fallback owner of /etc/locale.conf.
2021-12-23 11:01:07 +00:00
subdir('factory/templates')
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
meson: add option to skip installing to $sysconfdir This is useful for development where overwriting files out side the configured prefix will affect the host as well as stateless systems such as NixOS that don't let packages install to /etc but handle configuration on their own. Alternative to https://github.com/systemd/systemd/pull/17501 tested with: $ mkdir inst build && cd build $ meson \ -Dcreate-log-dirs=false \ -Dsysvrcnd-path=$(realpath ../inst)/etc/rc.d \ -Dsysvinit-path=$(realpath ../inst)/etc/init.d \ -Drootprefix=$(realpath ../inst) \ -Dinstall-sysconfdir=false \ --prefix=$(realpath ../inst) .. $ ninja install
2020-11-12 08:01:39 +01:00
if install_sysconfdir
install_data('xorg/50-systemd-user.sh',
install_dir : xinitrcdir)
endif
doc: add a new doc/ directory, and move two markdown docs into them I figure sooneror later we'll have more of these docs, hence let's give them a clean place to be. This leaves NEWS and README/README.md as well as the LICENSE texts in the root directory of the project since that appears to be customary for Free Software projects.
2018-02-26 11:48:46 +01:00
install_data('LICENSE.GPL2',
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
'LICENSE.LGPL2.1',
doc: add a new doc/ directory, and move two markdown docs into them I figure sooneror later we'll have more of these docs, hence let's give them a clean place to be. This leaves NEWS and README/README.md as well as the LICENSE texts in the root directory of the project since that appears to be customary for Free Software projects.
2018-02-26 11:48:46 +01:00
'NEWS',
'README',
Revert "docs: use collections to structure the data" This reverts commit 5e8ff010a1436d33bbf3c108335af6e0b4ff7a2a. This broke all the URLs, we can't have that. (And actually, we probably don't _want_ to make the change either. It's nicer to have all the pages in one directory, so one doesn't have to figure out to which collection the page belongs.)
2024-02-23 09:48:47 +01:00
'docs/ENVIRONMENT.md',
'docs/TRANSIENT-SETTINGS.md',
'docs/UIDS-GIDS.md',
meson: build systemd using meson It's crucial that we can build systemd using VS2010! ... er, wait, no, that's not the official reason. We need to shed old systems by requring python 3! Oh, no, it's something else. Maybe we need to throw out 345 years of knowlege accumulated in autotools? Whatever, this new thing is cool and shiny, let's use it. This is not complete, I'm throwing it out here for your amusement and critique. - rules for sd-boot are missing. Those might be quite complicated. - rules for tests are missing too. Those are probably quite simple and repetitive, but there's lots of them. - it's likely that I didn't get all the conditions right, I only tested "full" compilation where most deps are provided and nothing is disabled. - busname.target and all .busname units are skipped on purpose. Otherwise, installation into $DESTDIR has the same list of files and the autoconf install, except for .la files. It'd be great if people had a careful look at all the library linking options. I added stuff until things compiled, and in the end there's much less linking then in the old system. But it seems that there's still a lot of unnecessary deps. meson has a `shared_module` statement, which sounds like something appropriate for our nss and pam modules. Unfortunately, I couldn't get it to work. For the nss modules, we need an .so version of '2', but `shared_module` disallows the version argument. For the pam module, it also didn't work, I forgot the reason. The handling of .m4 and .in and .m4.in files is rather awkward. It's likely that this could be simplified. If make support is ever dropped, I think it'd make sense to switch to a different templating system so that two different languages and not required, which would make everything simpler yet. v2: - use get_pkgconfig_variable - use sh not bash - use add_project_arguments v3: - drop required:true and fix progs/prog typo v4: - use find_library('bz2') - add TTY_GID definition - define __SANE_USERSPACE_TYPES__ - use join_paths(prefix, ...) is used on all paths to make them all absolute v5: - replace all declare_dependency's with [] - add more conf.get guards around optional components v6: - drop -pipe, -Wall which are the default in meson - use compiler.has_function() and compiler.has_header_symbol instead of the hand-rolled checks. - fix duplication in 'liblibsystemd' library name - use the right .sym file for pam_systemd - rename 'compiler' to 'cc': shorter, and more idiomatic. v7: - use ENABLE_ENVIRONMENT_D not HAVE_ENVIRONMENT_D - rename prefix to prefixdir, rootprefix to rootprefixdir ("prefix" is too common of a name and too easy to overwrite by mistake) - wrap more stuff with conf.get('ENABLE...') == 1 - use rootprefix=='/' and rootbindir as install_dir, to fix paths under split-usr==true. v8: - use .split() also for src/coredump. Now everything is consistent ;) - add rootlibdir option and use it on the libraries that require it v9: - indentation v10: - fix check for qrencode and libaudit v11: - unify handling of executable paths, provide options for all progs This makes the meson build behave slightly differently than the autoconf-based one, because we always first try to find the executable in the filesystem, and fall back to the default. I think different handling of loadkeys, setfont, and telinit was just a historical accident. In addition to checking in $PATH, also check /usr/sbin/, /sbin for programs. In Fedora $PATH includes /usr/sbin, (and /sbin is is a symlink to /usr/sbin), but in Debian, those directories are not included in the path. C.f. https://github.com/mesonbuild/meson/issues/1576. - call all the options 'xxx-path' for clarity. - sort man/rules/meson.build properly so it's stable
2017-04-04 23:03:47 -04:00
install_dir : docdir)
meson: add TAGS target Quite necessary for all the emacs users out there ;) v2: - fix error with etags or git are not found v3: - move the definition of git and etags up so it's available for man/
2017-04-06 14:33:27 -04:00
Add all other applicable licenses under LICENSES/ License texts copied from: https://github.com/spdx/license-list-data/blob/master/text/
2021-09-29 22:10:34 +01:00
install_subdir('LICENSES',
install_dir : docdir)
meson: use install_emptydir() and drop meson-make-symlink.sh The script is mostly equivalent to 'mkdir -p' and 'ln -sfr'. Let's replace it with install_emptydir() builtin function and inline meson call.
2023-08-08 10:38:47 +09:00
install_emptydir(systemdstatedir)
meson: create dirs and touch /usr This is the equivalent of $(INSTALL_DIRS) and install-touch-usr-hook. I did not bother to create the directories into which we install files, since they will be created anyway. v2: - remove bashism
2017-04-09 23:55:05 -04:00
meson: make lines more consistent
2023-12-14 18:35:00 -03:00
#####################################################################
meson: add TAGS target Quite necessary for all the emacs users out there ;) v2: - fix error with etags or git are not found v3: - move the definition of git and etags up so it's available for man/
2017-04-06 14:33:27 -04:00
build tests: check that Github pages can be built successfully
2021-05-19 14:45:47 +01:00
# Ensure that changes to the docs/ directory do not break the
# basic Github pages build. But only run it in developer mode,
# as it might be fragile due to changes in the tooling, and it is
# not generally useful for users.
jekyll = find_program('jekyll', required : false)
if get_option('mode') == 'developer' and want_tests != 'false' and jekyll.found()
test('github-pages',
jekyll,
meson: rename "dist-check" suite to "dist" *Now* one of those tests is the longest, let's make them a bit shorter.
2023-05-18 17:16:33 +02:00
suite : 'dist',
build tests: check that Github pages can be built successfully
2021-05-19 14:45:47 +01:00
args : ['build',
meson: Drop project_source_root and project_build_root variables We can now rely on meson.project_source_root() and meson.project_build_root() which do the same thing.
2025-03-26 13:57:33 +01:00
'--source', meson.project_source_root() / 'docs',
'--destination', meson.project_build_root() / '_site'])
build tests: check that Github pages can be built successfully
2021-05-19 14:45:47 +01:00
endif
meson: make lines more consistent
2023-12-14 18:35:00 -03:00
#####################################################################
build tests: check that Github pages can be built successfully
2021-05-19 14:45:47 +01:00
meson: use files() not find_program() for helper scripts We went back and forth between 'prog.sh', files('prog.sh'), and find_program('prog.sh'). We want to use files() or find_program() so that we get a good error message if the file is missing. Behaviour of meson changed over time, and in the past not all forms could be used in all places. For example 0f4c4f3824891aea6a76b2e861f38a6045b2a246 added find_program() in many places to avoid repeated messages. But it seems that all recent meson versions work fine with files(). find_program prints silly messages: Program tools/make-man-index.py found: YES (/home/zbyszek/src/systemd/tools/make-man-index.py) Program tools/meson-render-jinja2.py found: YES (/home/zbyszek/src/systemd/tools/meson-render-jinja2.py) ... We know that those files will be found, they are part of the git checkout. With files() this is gone and the meson output is easier to read.
2025-07-25 12:18:19 +02:00
check_help = files('tools/check-help.sh')
check_version = files('tools/check-version.sh')
meson: recreate dist-check-help
2017-04-13 11:52:05 -04:00
foreach exec : public_programs
meson: Use fs module in more places
2023-06-14 16:05:52 +02:00
name = fs.name(exec.full_path())
meson: disable _all_ tests when -Dtests=false Back in 08318a2c5acb3d0e4243c36461e69a3e45482441, value "false" was enabled for '-Dtests=', but various tests were not conditionalized properly. So even with -Dtests=false -Dslow-tests=false we'd run 120 tests. Let's make this consistent.
2018-09-12 11:08:49 +02:00
if want_tests != 'false'
test('check-help-' + name,
meson: drop "meson-" prefix from various helper script filenames In a few cases, the prefix was originally necessary because a different helper script was used for automake, and a different one for meson. But now we use meson exclusively, and the prefix isn't useful. This also synchronizes the target name, file name, and variable name in meson.build. The targets exposed by meson didn't have the prefix, so the user interface is unchanged. (The prefix is retained in the few tools that are used for meson itself, e.g. meosn-vcs-tag.sh, meson-make-symlink.sh, etc.)
2020-05-07 13:54:10 +02:00
check_help,
meson: rename "dist-check" suite to "dist" *Now* one of those tests is the longest, let's make them a bit shorter.
2023-05-18 17:16:33 +02:00
suite : 'dist',
meson: Add missing test dependencies Currently, running "meson build" followed by "meson test -C build" will result in many failed tests due to missing dependencies. This commit adds the missing dependencies to make sure no tests fail.
2022-01-21 14:28:23 +00:00
args : exec.full_path(),
depends: exec)
tests: add a smoke test for --version option in binaries This is very similar to (and directly based on) the test for --help. I think it's nice to do this: the test is very quick, but it'll catch cases where we forgot to hook up the option, or forgot to exit after printing --version, and it'll also increase our test coverage a bit.
2022-04-05 10:51:21 +02:00
test('check-version-' + name,
check_version,
meson: rename "dist-check" suite to "dist" *Now* one of those tests is the longest, let's make them a bit shorter.
2023-05-18 17:16:33 +02:00
suite : 'dist',
tests: add a smoke test for --version option in binaries This is very similar to (and directly based on) the test for --help. I think it's nice to do this: the test is very quick, but it'll catch cases where we forgot to hook up the option, or forgot to exit after printing --version, and it'll also increase our test coverage a bit.
2022-04-05 10:51:21 +02:00
args : [exec.full_path(),
meson: Fix various versions Follow up for 8b3b01c4b7e0fde39b4be354990ee68f5e612c52 We switch to PROJECT_VERSION instead of PROJECT_VERSION_FULL where we report our version and which is likely being parsed to avoid breaking compat. If we didn't, the output would change from systemd 255 to systemd 255.1 which could break various tools.
2024-07-04 14:44:21 +02:00
project_major_version],
tests: add a smoke test for --version option in binaries This is very similar to (and directly based on) the test for --help. I think it's nice to do this: the test is very quick, but it'll catch cases where we forgot to hook up the option, or forgot to exit after printing --version, and it'll also increase our test coverage a bit.
2022-04-05 10:51:21 +02:00
depends: exec)
meson: disable _all_ tests when -Dtests=false Back in 08318a2c5acb3d0e4243c36461e69a3e45482441, value "false" was enabled for '-Dtests=', but various tests were not conditionalized properly. So even with -Dtests=false -Dslow-tests=false we'd run 120 tests. Let's make this consistent.
2018-09-12 11:08:49 +02:00
endif
meson: recreate dist-check-help
2017-04-13 11:52:05 -04:00
endforeach
meson: make lines more consistent
2023-12-14 18:35:00 -03:00
#####################################################################
Hook up oss-fuzz test cases as tests This is a bit painful because a separate build of systemd is necessary. The tests are guarded by tests!=false and slow-tests==true. Running them is not slow, but compilation certainly is. If this proves unwieldy, we can add a separate option controlling those builds later. The build for each sanitizer has its own directory, and we build all fuzzer tests there, and then pull them out one-by-one by linking into the target position as necessary. It would be nicer to just build the desired fuzzer, but we need to build the whole nested build as one unit. [I also tried making systemd and nested meson subproject. This would work nicely, but meson does not allow that because the nested target names are the same as the outer project names. If that is ever fixed, that would be the way to go.] v2: - make sure things still work if memory sanitizer is not available v3: - switch to syntax which works with meson 0.42.1 found in Ubuntu
2018-01-19 17:54:30 +11:00
meson: add support for ctags This is a squash of casync commits https://github.com/systemd/casync/commit/02fbbdb2b9a926a695a3ede7f3e3c17b9779db1a (by Silvio Fricke) and https://github.com/systemd/casync/commit/b687a94b1e24df73d32f8cdcee29f3c00eae69c9. Instead of checking during every meson config whether etags are available, just try to call them and error out if not. This has the advantage that the target is always available (if git is installed), and the error message gives a hint what needs to be installed. The naming is confusing, but etags(1) is pretty clear: - emacs expects TAGS file in etags format - vi expects tags file in ctags format and automake docs are pretty clear too: - tags target generates TAGS file - ctags target generates tags file
2017-07-03 12:42:29 -04:00
if git.found()
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-17 19:25:00 -04:00
all_files = run_command(
meson: call find_program() once and reuse the variable everywhere Meson 0.58 has gotten quite bad with emitting a message every time a quoted command is used: Program /home/zbyszek/src/systemd-work/tools/meson-make-symlink.sh found: YES (/home/zbyszek/src/systemd-work/tools/meson-make-symlink.sh) Program sh found: YES (/usr/bin/sh) Program sh found: YES (/usr/bin/sh) Program sh found: YES (/usr/bin/sh) Program sh found: YES (/usr/bin/sh) Program sh found: YES (/usr/bin/sh) Program sh found: YES (/usr/bin/sh) Program xsltproc found: YES (/usr/bin/xsltproc) Configuring custom-entities.ent using configuration Message: Skipping bootctl.1 because ENABLE_EFI is false Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Message: Skipping journal-remote.conf.5 because HAVE_MICROHTTPD is false Message: Skipping journal-upload.conf.5 because HAVE_MICROHTTPD is false Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Message: Skipping loader.conf.5 because ENABLE_EFI is false Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) ... Let's suffer one message only for each command. Hopefully we can silence even this when https://github.com/mesonbuild/meson/issues/8642 is resolved.
2021-05-14 14:16:17 +02:00
env, '-u', 'GIT_WORK_TREE',
meson: Drop project_source_root and project_build_root variables We can now rely on meson.project_source_root() and meson.project_build_root() which do the same thing.
2025-03-26 13:57:33 +01:00
git, '--git-dir=@0@/.git'.format(meson.project_source_root()),
meson: include .cc files in tags too We only have one, but it seems reasonable to not exclude it. Result tested with emacs.
2023-05-05 09:33:12 +02:00
'ls-files', ':/*.[ch]', ':/*.cc',
meson: add check:true/false to all run_command() invocations meson-0.59.4-1.fc35.noarch says: WARNING: You should add the boolean check kwarg to the run_command call. It currently defaults to false, but it will default to true in future releases of meson. See also: https://github.com/mesonbuild/meson/issues/9300
2021-11-17 13:58:53 +01:00
check : false)
if all_files.returncode() == 0
all_files = files(all_files.stdout().split())
custom_target(
output : 'tags',
meson: Drop project_source_root and project_build_root variables We can now rely on meson.project_source_root() and meson.project_build_root() which do the same thing.
2025-03-26 13:57:33 +01:00
command : [env, 'etags', '-o', '@0@/TAGS'.format(meson.project_source_root())] + all_files)
meson: add check:true/false to all run_command() invocations meson-0.59.4-1.fc35.noarch says: WARNING: You should add the boolean check kwarg to the run_command call. It currently defaults to false, but it will default to true in future releases of meson. See also: https://github.com/mesonbuild/meson/issues/9300
2021-11-17 13:58:53 +01:00
run_target(
'ctags',
meson: Drop project_source_root and project_build_root variables We can now rely on meson.project_source_root() and meson.project_build_root() which do the same thing.
2025-03-26 13:57:33 +01:00
command : [env, 'ctags', '--tag-relative=never', '-o', '@0@/tags'.format(meson.project_source_root())] + all_files)
meson: add check:true/false to all run_command() invocations meson-0.59.4-1.fc35.noarch says: WARNING: You should add the boolean check kwarg to the run_command call. It currently defaults to false, but it will default to true in future releases of meson. See also: https://github.com/mesonbuild/meson/issues/9300
2021-11-17 13:58:53 +01:00
endif
meson: add git-contrib target
2017-04-15 00:16:23 -04:00
meson: add check-includes test to the test suite Let's just call it always. It is quite fast (meson says 0.12 s).
2023-05-05 09:35:54 +02:00
####################################################
meson: use run_target instead of custom_target where appropriate This way a fake output does not need to be specified.
2017-04-17 19:48:20 -04:00
run_target(
meson: reindent all files with 8 spaces The indentation for emacs'es meson-mode is added .dir-locals. All files are reindented automatically, using the lasest meson-mode from git. Indentation should now be fairly consistent.
2017-04-17 19:25:00 -04:00
'git-contrib',
meson: use files() not find_program() for helper scripts We went back and forth between 'prog.sh', files('prog.sh'), and find_program('prog.sh'). We want to use files() or find_program() so that we get a good error message if the file is missing. Behaviour of meson changed over time, and in the past not all forms could be used in all places. For example 0f4c4f3824891aea6a76b2e861f38a6045b2a246 added find_program() in many places to avoid repeated messages. But it seems that all recent meson versions work fine with files(). find_program prints silly messages: Program tools/make-man-index.py found: YES (/home/zbyszek/src/systemd/tools/make-man-index.py) Program tools/meson-render-jinja2.py found: YES (/home/zbyszek/src/systemd/tools/meson-render-jinja2.py) ... We know that those files will be found, they are part of the git checkout. With files() this is gone and the meson output is easier to read.
2025-07-25 12:18:19 +02:00
command : files('tools/git-contrib.sh'))
meson: add git-snapshot target The output is systemd-<sha1abbrev>.tar.gz in source root.
2017-04-24 19:28:05 -04:00
meson: add check-includes test to the test suite Let's just call it always. It is quite fast (meson says 0.12 s).
2023-05-05 09:35:54 +02:00
####################################################
meson: add git-snapshot target The output is systemd-<sha1abbrev>.tar.gz in source root.
2017-04-24 19:28:05 -04:00
git_head = run_command(
meson: Drop project_source_root and project_build_root variables We can now rely on meson.project_source_root() and meson.project_build_root() which do the same thing.
2025-03-26 13:57:33 +01:00
git, '--git-dir=@0@/.git'.format(meson.project_source_root()),
meson: add check:true/false to all run_command() invocations meson-0.59.4-1.fc35.noarch says: WARNING: You should add the boolean check kwarg to the run_command call. It currently defaults to false, but it will default to true in future releases of meson. See also: https://github.com/mesonbuild/meson/issues/9300
2021-11-17 13:58:53 +01:00
'rev-parse', 'HEAD',
check : false).stdout().strip()
meson: add git-snapshot target The output is systemd-<sha1abbrev>.tar.gz in source root.
2017-04-24 19:28:05 -04:00
git_head_short = run_command(
meson: Drop project_source_root and project_build_root variables We can now rely on meson.project_source_root() and meson.project_build_root() which do the same thing.
2025-03-26 13:57:33 +01:00
git, '--git-dir=@0@/.git'.format(meson.project_source_root()),
meson: add check:true/false to all run_command() invocations meson-0.59.4-1.fc35.noarch says: WARNING: You should add the boolean check kwarg to the run_command call. It currently defaults to false, but it will default to true in future releases of meson. See also: https://github.com/mesonbuild/meson/issues/9300
2021-11-17 13:58:53 +01:00
'rev-parse', '--short=7', 'HEAD',
check : false).stdout().strip()
meson: add git-snapshot target The output is systemd-<sha1abbrev>.tar.gz in source root.
2017-04-24 19:28:05 -04:00
run_target(
'git-snapshot',
meson: call find_program() once and reuse the variable everywhere Meson 0.58 has gotten quite bad with emitting a message every time a quoted command is used: Program /home/zbyszek/src/systemd-work/tools/meson-make-symlink.sh found: YES (/home/zbyszek/src/systemd-work/tools/meson-make-symlink.sh) Program sh found: YES (/usr/bin/sh) Program sh found: YES (/usr/bin/sh) Program sh found: YES (/usr/bin/sh) Program sh found: YES (/usr/bin/sh) Program sh found: YES (/usr/bin/sh) Program sh found: YES (/usr/bin/sh) Program xsltproc found: YES (/usr/bin/xsltproc) Configuring custom-entities.ent using configuration Message: Skipping bootctl.1 because ENABLE_EFI is false Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Message: Skipping journal-remote.conf.5 because HAVE_MICROHTTPD is false Message: Skipping journal-upload.conf.5 because HAVE_MICROHTTPD is false Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Message: Skipping loader.conf.5 because ENABLE_EFI is false Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) Program ln found: YES (/usr/bin/ln) ... Let's suffer one message only for each command. Hopefully we can silence even this when https://github.com/mesonbuild/meson/issues/8642 is resolved.
2021-05-14 14:16:17 +02:00
command : [git, 'archive',
meson: Drop project_source_root and project_build_root variables We can now rely on meson.project_source_root() and meson.project_build_root() which do the same thing.
2025-03-26 13:57:33 +01:00
'-o', '@0@/systemd-@1@.tar.gz'.format(meson.project_source_root(),
meson: add git-snapshot target The output is systemd-<sha1abbrev>.tar.gz in source root.
2017-04-24 19:28:05 -04:00
git_head_short),
meson: using f-strings in meson Those were added before 0.61 which is our minimum version. Dots were dropped from the end of some messages. Most messages did not have the trailing dot. 0.63 added support for multi-line f-strings. We can't use those yet.
2025-10-03 15:26:20 +02:00
'--prefix', f'systemd-@git_head@/',
meson: add git-snapshot target The output is systemd-<sha1abbrev>.tar.gz in source root.
2017-04-24 19:28:05 -04:00
'HEAD'])
endif
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-27 20:54:52 -04:00
meson: make lines more consistent
2023-12-14 18:35:00 -03:00
#####################################################################
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-27 20:54:52 -04:00
meson: Define our own clang-tidy target meson's target has a few issues: - Runs on all source files regardless if they're included in the build or not - Doesn't have any dependencies on generated sources which means we have to do a full build first before we can run clang-tidy - Doesn't allow us to pass any extra arguments To work around these, let's define our own clang-tidy target instead using llvm's run-clang-tidy script. Alongside the clang-tidy target, let's start keeping track of all generated sources which we make the clang-tidy target depend on. We also add a new target which will only generate source files which is useful for setting up the source tree for running code analysis against it.
2025-04-24 11:58:45 +02:00
alias_target('gensources', generated_sources)
meson: Rework clang-tidy integration to be done via unit tests Instead of using run-clang-tidy.py with its own scheduling, let's just gather a list of source files ourselves and then use that to add a unit test for each source file that runs clang-tidy on the source file. We also add a bit of logic to run clang-tidy on most header files as well for extra coverage. This uncovered various header files that were not standalone so this commit also includes fixes to make sure the clang-tidy tests are all green. We can also use this in a later commit to run clang-include-cleaner on each source file in the same way.
2025-05-15 15:09:27 +02:00
clang_tidy = find_program('clang-tidy', required : false)
if meson.version().version_compare('>=1.4.0')
foreach source : sources
clang-tidy: Skip public headers We're very limited in our ability to change these due to backwards compat, so let's skip them from analysis since we won't be able to fix the errors anyway.
2025-05-23 13:03:57 +02:00
if systemd_headers.contains(source)
continue
endif
meson: Run clang-tidy on libsystemd headers as well
2025-05-22 11:06:54 +02:00
if not source.full_path().endswith('.c') and not source.full_path().endswith('.h')
meson: Rework clang-tidy integration to be done via unit tests Instead of using run-clang-tidy.py with its own scheduling, let's just gather a list of source files ourselves and then use that to add a unit test for each source file that runs clang-tidy on the source file. We also add a bit of logic to run clang-tidy on most header files as well for extra coverage. This uncovered various header files that were not standalone so this commit also includes fixes to make sure the clang-tidy tests are all green. We can also use this in a later commit to run clang-include-cleaner on each source file in the same way.
2025-05-15 15:09:27 +02:00
continue
endif
inputs = [source]
header = source.full_path().replace('.c', '.h')
clang-tidy: Skip public headers We're very limited in our ability to change these due to backwards compat, so let's skip them from analysis since we won't be able to fix the errors anyway.
2025-05-23 13:03:57 +02:00
if fs.exists(header) and header != source.full_path() and header != libudev_h_path
meson: Rework clang-tidy integration to be done via unit tests Instead of using run-clang-tidy.py with its own scheduling, let's just gather a list of source files ourselves and then use that to add a unit test for each source file that runs clang-tidy on the source file. We also add a bit of logic to run clang-tidy on most header files as well for extra coverage. This uncovered various header files that were not standalone so this commit also includes fixes to make sure the clang-tidy tests are all green. We can also use this in a later commit to run clang-include-cleaner on each source file in the same way.
2025-05-15 15:09:27 +02:00
inputs += header
endif
foreach input : inputs
if clang_tidy.found()
test(
'clang-tidy-@0@'.format(fs.name(input)),
clang_tidy,
args : ['-p', meson.project_build_root(), input],
suite : 'clang-tidy',
depends : generated_sources,
)
endif
endforeach
endforeach
meson: Define our own clang-tidy target meson's target has a few issues: - Runs on all source files regardless if they're included in the build or not - Doesn't have any dependencies on generated sources which means we have to do a full build first before we can run clang-tidy - Doesn't allow us to pass any extra arguments To work around these, let's define our own clang-tidy target instead using llvm's run-clang-tidy script. Alongside the clang-tidy target, let's start keeping track of all generated sources which we make the clang-tidy target depend on. We also add a new target which will only generate source files which is useful for setting up the source tree for running code analysis against it.
2025-04-24 11:58:45 +02:00
endif
meson: resurrect API documentation target We had this functionality back in Automake times, let's resurrect it.
2017-12-20 12:51:14 +01:00
run_target(
'check-api-docs',
depends : [man, libsystemd, libudev],
meson: use files() not find_program() for helper scripts We went back and forth between 'prog.sh', files('prog.sh'), and find_program('prog.sh'). We want to use files() or find_program() so that we get a good error message if the file is missing. Behaviour of meson changed over time, and in the past not all forms could be used in all places. For example 0f4c4f3824891aea6a76b2e861f38a6045b2a246 added find_program() in many places to avoid repeated messages. But it seems that all recent meson versions work fine with files(). find_program prints silly messages: Program tools/make-man-index.py found: YES (/home/zbyszek/src/systemd/tools/make-man-index.py) Program tools/meson-render-jinja2.py found: YES (/home/zbyszek/src/systemd/tools/meson-render-jinja2.py) ... We know that those files will be found, they are part of the git checkout. With files() this is gone and the meson output is easier to read.
2025-07-25 12:18:19 +02:00
command : [files('tools/check-api-docs.sh'),
libsystemd.full_path(),
libudev.full_path()])
meson: resurrect API documentation target We had this functionality back in Automake times, let's resurrect it.
2017-12-20 12:51:14 +01:00
meson: add target aliases 'man' and 'html' Previously we would need to say 'ninja -C build man/man man/html', now just 'ninja -C build man html'.
2025-10-04 18:09:20 +02:00
alias_target('man', man)
alias_target('html', html)
meson: use alias_target for doc update commands This undoes part of 4c890ad3cc7b3445683d7b52bc00e4a58bef5e94: the implementations of update-dbus-docs and update-man-rules are moved back to man/meson.build, and alias_target() is used to keep the visible target names unchanged. The rules for man pages are reworked so that it's possible to invoke the targets even if xstlproc is not available. After all, xsltproc is only needed for the final formatted output, and not other processing.
2021-07-27 20:11:41 +02:00
alias_target('update-dbus-docs', update_dbus_docs)
alias_target('update-man-rules', update_man_rules)
meson: rename target to update-man-rules Same justification as for update-dbus-docs.
2021-01-27 09:10:25 +01:00
meson: disable export-dbus-interfaces target when cross-compiling ERROR: Cannot use target systemd as a generator because it is built for the host machine and no exe wrapper is defined or needs_exe_wrapper is true. You might want to set `native: true` instead to build it for the build machine.
2022-02-08 13:19:52 +00:00
if not meson.is_cross_build()
custom_target(
'export-dbus-interfaces',
meson: use fs.name() and fs.parent() Follow-up for 1bd0cc452ce1cbfb330db6b0855e178b4814e7de.
2022-05-16 04:05:51 +09:00
output : fs.name(dbus_interfaces_dir),
meson: disable export-dbus-interfaces target when cross-compiling ERROR: Cannot use target systemd as a generator because it is built for the host machine and no exe wrapper is defined or needs_exe_wrapper is true. You might want to set `native: true` instead to build it for the build machine.
2022-02-08 13:19:52 +00:00
install : dbus_interfaces_dir != 'no',
meson: use fs.name() and fs.parent() Follow-up for 1bd0cc452ce1cbfb330db6b0855e178b4814e7de.
2022-05-16 04:05:51 +09:00
install_dir : fs.parent(dbus_interfaces_dir),
meson: disable export-dbus-interfaces target when cross-compiling ERROR: Cannot use target systemd as a generator because it is built for the host machine and no exe wrapper is defined or needs_exe_wrapper is true. You might want to set `native: true` instead to build it for the build machine.
2022-02-08 13:19:52 +00:00
command : [export_dbus_interfaces_py, '@OUTPUT@', dbus_programs])
endif
Export D-Bus interfaces to /usr/share/dbus-1/interfaces Pass -Ddbus-interfaces-dir=no to meson to disable export Interfaces from: org.freedesktop.home1 org.freedesktop.hostname1 org.freedesktop.import1 org.freedesktop.locale1 org.freedesktop.LogControl1 org.freedesktop.login1 org.freedesktop.machine1 org.freedesktop.oom1 org.freedesktop.portable1 org.freedesktop.resolve1 org.freedesktop.systemd1 org.freedesktop.timedate1
2021-07-24 10:30:42 +03:00
meson: drop explicit custom_target names [1] says: > Since 0.60.0 the name argument is optional and defaults to the basename of > the first output We specify >= 0.62 as the supported version, so drop the duplicate name in all cases where it is the same as outputs[0], i.e. almost all cases. [1] https://mesonbuild.com/Reference-manual_functions.html#custom_target
2025-06-28 17:14:50 +02:00
custom_target(
meson: adjust indentation
2025-06-28 17:18:35 +02:00
output : 'installed-unit-files.txt',
capture : true,
install : want_tests != 'no' and install_tests,
install_dir : testdata_dir,
meson: use files() not find_program() for helper scripts We went back and forth between 'prog.sh', files('prog.sh'), and find_program('prog.sh'). We want to use files() or find_program() so that we get a good error message if the file is missing. Behaviour of meson changed over time, and in the past not all forms could be used in all places. For example 0f4c4f3824891aea6a76b2e861f38a6045b2a246 added find_program() in many places to avoid repeated messages. But it seems that all recent meson versions work fine with files(). find_program prints silly messages: Program tools/make-man-index.py found: YES (/home/zbyszek/src/systemd/tools/make-man-index.py) Program tools/meson-render-jinja2.py found: YES (/home/zbyszek/src/systemd/tools/meson-render-jinja2.py) ... We know that those files will be found, they are part of the git checkout. With files() this is gone and the meson output is easier to read.
2025-07-25 12:18:19 +02:00
command : [files('tools/meson-extract-unit-files.py'),
meson.project_build_root()])
test: verify our own units (where applicable) This is inspired by one of our internal tests that does pretty much the same thing. However, it is slightly more convoluted than I'd like it to be, since I really don't want to duplicate the list of our units in another place, so we need to, somehow, pass the list from the meson file to the test script. I originally envisioned this to be a part of the unit test suite, but this doesn't work for unit files with absolute paths to binaries, as we'd have to install the build first (maybe using a chroot would work?). It doesn't check man pages (since they might not be installed on the test machine) and also skip recursive dependencies (as that would trip over issues in files that are not under our direct control), but it should still cover typos and such. There are currently two units for which the check had to be disabled - syslog.socket, as the corresponding syslog.service might not be installed, and rc-local.service as that's a compat API and the necessary /etc/rc.d/rc.local file may not (and most likely won't be) present.
2024-02-19 20:41:49 +01:00
meson: make lines more consistent
2023-12-14 18:35:00 -03:00
#####################################################################
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-27 20:54:52 -04:00
meson: add udev/hwdb build aliases Allows to do: meson compile libudev udev hwdb meson install --no-rebuild --tags libudev,udev,hwdb
2025-01-14 23:22:04 +00:00
udev_targets = []
foreach bin : udev_binaries
udev_targets += executables_by_name.get(bin, [])
endforeach
alias_target('udev', buildable_rules, udev_targets, udev_units)
if conf.get('ENABLE_HWDB') == 1
alias_target('hwdb', auto_suspend_rules, executables_by_name.get('systemd-hwdb'), hwdb_units)
endif
meson: using f-strings in meson Those were added before 0.61 which is our minimum version. Dots were dropped from the end of some messages. Most messages did not have the trailing dot. 0.63 added support for multi-line f-strings. We can't use those yet.
2025-10-03 15:26:20 +02:00
alt_time_epoch = run_command('date', '-Is', '-u', '-d', f'@@time_epoch@',
meson: add check:true/false to all run_command() invocations meson-0.59.4-1.fc35.noarch says: WARNING: You should add the boolean check kwarg to the run_command call. It currently defaults to false, but it will default to true in future releases of meson. See also: https://github.com/mesonbuild/meson/issues/9300
2021-11-17 13:58:53 +01:00
check : true).stdout().strip()
meson: use meson's summary() The output is similar to our hand-crafted status message, but it's nice to use the built-in functionality. After all, it was amended during development to support our use case.
2021-07-28 10:04:01 +02:00
summary({
meson: realign summary
2025-08-14 21:48:43 +09:00
'split bin-sbin' : split_bin,
'prefix directory' : prefixdir,
'sysconf directory' : sysconfdir,
'include directory' : includedir,
'lib directory' : libdir,
'SysV init scripts' : sysvinit_path,
'SysV rc?.d directories' : sysvrcnd_path,
'SysV rc.local script' : sysvrclocal_path,
'PAM modules directory' : pamlibdir,
'PAM configuration directory' : pamconfdir,
'ssh server configuration directory' : sshdconfdir,
ssh-generator: create privsep dir via tmpfiles.d/ if we are told to To make it easy to have a workable ssh-generator on various distros, let's optionally generate the ssh privsep dir via tmpfiles.d/ drop-in. This enables the concept with a path of /run/sshd/ as default. This is the path Debian/Ubuntu uses, and means that we just work on those distros. Debian/Ubuntu is the only distro (apparently?) that puts the privsep dir under /run/, hence always needs the dir to be created manually. Other distros don't need it that much, because they place the dir in /usr/ (fedora, best choice!) or /var/ (others, not ideal, because still mutable). Also adds a longer explanation about this in NEWS, in the hope that distro maintaines read that and maybe start cleaning this up. Alternative to: #31543
2024-03-01 11:25:52 +01:00
'ssh server privilege separation directory' : sshdprivsepdir,
meson: realign summary
2025-08-14 21:48:43 +09:00
'ssh client configuration directory' : sshconfdir,
'libcryptsetup plugins directory' : libcryptsetup_plugins_dir,
'Shell profile directory' : shellprofiledir,
'RPM macros directory' : rpmmacrosdir,
'modprobe.d directory' : modprobedir,
'D-Bus policy directory' : dbuspolicydir,
'D-Bus session directory' : dbussessionservicedir,
'D-Bus system directory' : dbussystemservicedir,
'D-Bus interfaces directory' : dbus_interfaces_dir,
'bash completions directory' : bashcompletiondir,
'zsh completions directory' : zshcompletiondir,
'private shared lib version tag' : shared_lib_tag,
'debug shell' : '@0@ @ @1@'.format(get_option('debug-shell'),
meson: using f-strings in meson Those were added before 0.61 which is our minimum version. Dots were dropped from the end of some messages. Most messages did not have the trailing dot. 0.63 added support for multi-line f-strings. We can't use those yet.
2025-10-03 15:26:20 +02:00
get_option('debug-tty')),
meson: realign summary
2025-08-14 21:48:43 +09:00
'system UIDs' : '<=@0@ (alloc >=@1@)'.format(conf.get('SYSTEM_UID_MAX'),
conf.get('SYSTEM_ALLOC_UID_MIN')),
'system GIDs' : '<=@0@ (alloc >=@1@)'.format(conf.get('SYSTEM_GID_MAX'),
conf.get('SYSTEM_ALLOC_GID_MIN')),
meson: using f-strings in meson Those were added before 0.61 which is our minimum version. Dots were dropped from the end of some messages. Most messages did not have the trailing dot. 0.63 added support for multi-line f-strings. We can't use those yet.
2025-10-03 15:26:20 +02:00
'greeter UIDs' : f'@greeter_uid_min@…@greeter_uid_max@',
'dynamic UIDs' : f'@dynamic_uid_min@…@dynamic_uid_max@',
'container UID bases' : f'@container_uid_base_min@…@container_uid_base_max@',
'foreign UID base' : f'@foreign_uid_base@',
meson: realign summary
2025-08-14 21:48:43 +09:00
'static UID/GID allocations' : ' '.join(static_ugids),
'/dev/kvm access mode' : get_option('dev-kvm-mode'),
'render group access mode' : get_option('group-render-mode'),
'certificate root directory' : get_option('certificate-root'),
'support URL' : support_url,
'nobody user name' : nobody_user,
'nobody group name' : nobody_group,
'fallback hostname' : get_option('fallback-hostname'),
'default compression method' : compression,
'default DNSSEC mode' : default_dnssec,
'default DNS-over-TLS mode' : default_dns_over_tls,
'default mDNS mode' : default_mdns,
'default LLMNR mode' : default_llmnr,
'default DNS servers' : dns_servers.split(' '),
'default NTP servers' : ntp_servers.split(' '),
'default net.naming_scheme= value' : default_net_naming_scheme,
'default KillUserProcesses= value' : kill_user_processes,
'default locale' : default_locale,
'default nspawn locale' : nspawn_locale,
'default status unit format' : status_unit_format_default,
'default user $PATH' : default_user_path != '' ? default_user_path : '(same as system services)',
'systemd service watchdog' : service_watchdog == '' ? 'disabled' : service_watchdog,
meson: using f-strings in meson Those were added before 0.61 which is our minimum version. Dots were dropped from the end of some messages. Most messages did not have the trailing dot. 0.63 added support for multi-line f-strings. We can't use those yet.
2025-10-03 15:26:20 +02:00
'time epoch' : f'@time_epoch@ (@alt_time_epoch@)',
meson: realign summary
2025-08-14 21:48:43 +09:00
})
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-27 20:54:52 -04:00
# TODO:
# CFLAGS: ${OUR_CFLAGS} ${CFLAGS}
# CPPFLAGS: ${OUR_CPPFLAGS} ${CPPFLAGS}
# LDFLAGS: ${OUR_LDFLAGS} ${LDFLAGS}
found = []
missing = []
foreach tuple : [
meson: reorder the "features" list Try to make this more manageable by reording: - dependencies / inputs (with subcategory of compression libraries) - major components / outputs - optional features / conditionals that don't fit into the two above categories The division isn't well defined, because libraries often correspond one-to-one to feature, but not always.
2021-04-15 10:59:03 +02:00
# dependencies
['ACL'],
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-27 20:54:52 -04:00
['AUDIT'],
['AppArmor'],
meson: reorder the "features" list Try to make this more manageable by reording: - dependencies / inputs (with subcategory of compression libraries) - major components / outputs - optional features / conditionals that don't fit into the two above categories The division isn't well defined, because libraries often correspond one-to-one to feature, but not always.
2021-04-15 10:59:03 +02:00
['IMA'],
meson.build: add a few features to summary
2024-11-08 19:28:24 +01:00
['IPE'],
meson: reorder the "features" list Try to make this more manageable by reording: - dependencies / inputs (with subcategory of compression libraries) - major components / outputs - optional features / conditionals that don't fit into the two above categories The division isn't well defined, because libraries often correspond one-to-one to feature, but not always.
2021-04-15 10:59:03 +02:00
['PAM'],
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-27 20:54:52 -04:00
['SECCOMP'],
meson: reorder the "features" list Try to make this more manageable by reording: - dependencies / inputs (with subcategory of compression libraries) - major components / outputs - optional features / conditionals that don't fit into the two above categories The division isn't well defined, because libraries often correspond one-to-one to feature, but not always.
2021-04-15 10:59:03 +02:00
['SELinux'],
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-27 20:54:52 -04:00
['SMACK'],
meson: reorder the "features" list Try to make this more manageable by reording: - dependencies / inputs (with subcategory of compression libraries) - major components / outputs - optional features / conditionals that don't fit into the two above categories The division isn't well defined, because libraries often correspond one-to-one to feature, but not always.
2021-04-15 10:59:03 +02:00
['blkid'],
['elfutils'],
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-27 20:54:52 -04:00
['gcrypt'],
['gnutls'],
meson.build: add a few features to summary
2024-11-08 19:28:24 +01:00
['libarchive'],
meson, bpf: add HAVE_LIBBPF, BPF_FRAMEWORK options * Add `bpf-framework` feature gate with 'auto', 'true' and 'false' choices * Add libbpf [0] dependency * Search for clang llvm-strip and bpftool binaries in compile time to generate bpf skeleton. For libbpf [0], make 0.2.0 [1] the minimum required version. If libbpf is satisfied, set HAVE_LIBBPF config option to 1. If `bpf-framework` feature gate is set to 'auto', means that whether bpf feature is enabled or now is defined by the presence of all of libbpf, clang, llvm and bpftool in build environment. With 'auto' all dependencies are optional. If the gate is set to `true`, make all of the libbpf, clang and llvm dependencies mandatory. If it's set to `false`, set `BPF_FRAMEWORK` to false and make libbpf dependency optional. libbpf dependency is dynamic followed by the common pattern in systemd. meson, bpf: add build rule for socket_bind program
2020-11-13 17:08:15 -08:00
['libbpf'],
meson: reorder the "features" list Try to make this more manageable by reording: - dependencies / inputs (with subcategory of compression libraries) - major components / outputs - optional features / conditionals that don't fit into the two above categories The division isn't well defined, because libraries often correspond one-to-one to feature, but not always.
2021-04-15 10:59:03 +02:00
['libcryptsetup'],
Add support for systemd-tpm2 libcryptsetup plugin. Add support for systemd-tpm2 based LUKS2 device activation via libcryptsetup plugin. This make the feature (tpm2 sealed LUKS2 keyslot passphrase) usable from both systemd utilities and cryptsetup cli. The feature is configured via -Dlibcryptsetup-plugins combo with default value set to 'auto'. It get's enabled automatically when cryptsetup 2.4.0 or later is installed in build system.
2021-03-16 20:13:28 +01:00
['libcryptsetup-plugins'],
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-27 20:54:52 -04:00
['libcurl'],
meson: reorder the "features" list Try to make this more manageable by reording: - dependencies / inputs (with subcategory of compression libraries) - major components / outputs - optional features / conditionals that don't fit into the two above categories The division isn't well defined, because libraries often correspond one-to-one to feature, but not always.
2021-04-15 10:59:03 +02:00
['libfdisk'],
['libfido2'],
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-27 20:54:52 -04:00
['libidn'],
meson: reorder the "features" list Try to make this more manageable by reording: - dependencies / inputs (with subcategory of compression libraries) - major components / outputs - optional features / conditionals that don't fit into the two above categories The division isn't well defined, because libraries often correspond one-to-one to feature, but not always.
2021-04-15 10:59:03 +02:00
['libidn2'],
['microhttpd'],
['openssl'],
['p11kit'],
shared: add libpasswdqc support Co-authored-by: Dmitry V. Levin <ldv@altlinux.org> Resolves: #15055
2021-08-06 11:41:01 +03:00
['passwdqc'],
meson: reorder the "features" list Try to make this more manageable by reording: - dependencies / inputs (with subcategory of compression libraries) - major components / outputs - optional features / conditionals that don't fit into the two above categories The division isn't well defined, because libraries often correspond one-to-one to feature, but not always.
2021-04-15 10:59:03 +02:00
['pcre2'],
['pwquality'],
['qrencode'],
['tpm2'],
['xkbcommon'],
# compression libs
['zstd'],
['lz4'],
['xz'],
['zlib'],
['bzip2'],
# components
['backlight'],
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-27 20:54:52 -04:00
['binfmt'],
tree-wide: Drop gnu-efi This drops all mentions of gnu-efi and its manual build machinery. A future commit will bring bootloader builds back. A new bootloader meson option is now used to control whether to build sd-boot and its userspace tooling.
2023-02-26 14:09:44 +01:00
['bootloader'],
journalctl: allow statically linked build The journalctl tool may be needed on cross compilation hosts in order to run --update-catalog against a target rootfs. To avoid reliability issues caused by shared linking allow journalctl to be linked statically.
2022-07-27 15:28:09 -06:00
['bpf-framework', conf.get('BPF_FRAMEWORK') == 1],
meson: reorder the "features" list Try to make this more manageable by reording: - dependencies / inputs (with subcategory of compression libraries) - major components / outputs - optional features / conditionals that don't fit into the two above categories The division isn't well defined, because libraries often correspond one-to-one to feature, but not always.
2021-04-15 10:59:03 +02:00
['coredump'],
['efi'],
tree-wide: Drop gnu-efi This drops all mentions of gnu-efi and its manual build machinery. A future commit will bring bootloader builds back. A new bootloader meson option is now used to control whether to build sd-boot and its userspace tooling.
2023-02-26 14:09:44 +01:00
['environment.d'],
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-27 20:54:52 -04:00
['firstboot'],
meson: reorder the "features" list Try to make this more manageable by reording: - dependencies / inputs (with subcategory of compression libraries) - major components / outputs - optional features / conditionals that don't fit into the two above categories The division isn't well defined, because libraries often correspond one-to-one to feature, but not always.
2021-04-15 10:59:03 +02:00
['hibernate'],
['homed'],
['hostnamed'],
['hwdb'],
['importd'],
['initrd'],
meson: build kernel-install man page when necessary
2022-03-31 11:17:10 +02:00
['kernel-install'],
meson: reorder the "features" list Try to make this more manageable by reording: - dependencies / inputs (with subcategory of compression libraries) - major components / outputs - optional features / conditionals that don't fit into the two above categories The division isn't well defined, because libraries often correspond one-to-one to feature, but not always.
2021-04-15 10:59:03 +02:00
['localed'],
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-27 20:54:52 -04:00
['logind'],
['machined'],
meson.build: add a few features to summary
2024-11-08 19:28:24 +01:00
['mountfsd'],
meson: reorder the "features" list Try to make this more manageable by reording: - dependencies / inputs (with subcategory of compression libraries) - major components / outputs - optional features / conditionals that don't fit into the two above categories The division isn't well defined, because libraries often correspond one-to-one to feature, but not always.
2021-04-15 10:59:03 +02:00
['networkd'],
meson: add feature flag for nspawn build Other tools have it, nspawn doesn't, add one
2025-03-27 23:51:17 +00:00
['nspawn'],
meson.build: add a few features to summary
2024-11-08 19:28:24 +01:00
['nsresourced'],
meson: reorder the "features" list Try to make this more manageable by reording: - dependencies / inputs (with subcategory of compression libraries) - major components / outputs - optional features / conditionals that don't fit into the two above categories The division isn't well defined, because libraries often correspond one-to-one to feature, but not always.
2021-04-15 10:59:03 +02:00
['nss-myhostname'],
['nss-mymachines'],
['nss-resolve'],
['nss-systemd'],
['oomd'],
add new portable service framework This adds a small service "systemd-portabled" and a matching client "portablectl", which implement the "portable service" concept. The daemon implements the actual operations, is PolicyKit-enabled and is activated on demand with exit-on-idle. Both the daemon and the client are an optional build artifact, enabled by default rhough.
2018-04-16 21:41:40 +02:00
['portabled'],
meson: reorder the "features" list Try to make this more manageable by reording: - dependencies / inputs (with subcategory of compression libraries) - major components / outputs - optional features / conditionals that don't fit into the two above categories The division isn't well defined, because libraries often correspond one-to-one to feature, but not always.
2021-04-15 10:59:03 +02:00
['pstore'],
['quotacheck'],
['randomseed'],
['repart'],
['resolve'],
['rfkill'],
sysext: new tool for managing "system extensions" for /usr/ + /opt/
2021-01-08 16:57:27 +01:00
['sysext'],
journalctl: allow statically linked build The journalctl tool may be needed on cross compilation hosts in order to run --update-catalog against a target rootfs. To avoid reliability issues caused by shared linking allow journalctl to be linked statically.
2022-07-27 15:28:09 -06:00
['systemd-analyze', conf.get('ENABLE_ANALYZE') == 1],
sysupdate: add new component "sysupdate"
2020-12-28 15:17:54 +01:00
['sysupdate'],
meson: add separate option for sysupdated, disable in release builds This commit introduces a build-time option to enable/disable sysupdated separately from sysupdate. 'auto' translated to enabled by default in developer builds.
2024-10-30 14:21:09 +01:00
['sysupdated'],
meson: reorder the "features" list Try to make this more manageable by reording: - dependencies / inputs (with subcategory of compression libraries) - major components / outputs - optional features / conditionals that don't fit into the two above categories The division isn't well defined, because libraries often correspond one-to-one to feature, but not always.
2021-04-15 10:59:03 +02:00
['sysusers'],
storagetm: add new systemd-storagetm component This implements a "storage target mode", similar to what MacOS provides since a long time as "Target Disk Mode": https://en.wikipedia.org/wiki/Target_Disk_Mode This implementation is relatively simple: 1. a new generic target "storage-target-mode.target" is added, which when booted into defines the target mode. 2. a small tool and service "systemd-storagetm.service" is added which exposes a specific device or all devices as NVMe-TCP devices over the network. NVMe-TCP appears to be hot shit right now how to expose block devices over the network. And it's really simple to set up via configs, hence our code is relatively short and neat. The idea is that systemd-storagetm.target can be extended sooner or later, for example to expose block devices also as USB mass storage devices and similar, in case the system has "dual mode" USB controller that can also work as device, not just as host. (And people could also plug in sharing as NBD, iSCSI, whatever they want.) How to use this? Boot into your system with a kernel cmdline of "rd.systemd.unit=storage-target-mode.target ip=link-local", and you'll see on screen the precise "nvme connect" command line to make the relevant block devices available locally on some other machine. This all requires that the target mode stuff is included in the initrd of course. And the system will the stay in the initrd forever. Why bother? Primarily three use-cases: 1. Debug a broken system: with very few dependencies during boot get access to the raw block device of a broken machine. 2. Migrate from system to another system, by dd'ing the old to the new directly. 3. Installing an OS remotely on some device (for example via Thunderbolt networking) (And there might be more, for example the ability to boot from a laptop's disk on another system) Limitations: 1. There's no authentication/encryption. Hence: use this on local links only. 2. NVMe target mode on Linux supports r/w operation only. Ideally, we'd have a read-only mode, for security reasons, and default to it. Future love: 1. We should have another mode, where we simply expose the homed LUKS home dirs like that. 2. Some lightweight hookup with plymouth, to display a (shortened) version of the info we write to the console. To test all this, just run: mkosi --kernel-command-line-extra="rd.systemd.unit=storage-target-mode.target" qemu
2023-10-27 14:25:49 +02:00
['storagetm'],
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-27 20:54:52 -04:00
['timedated'],
['timesyncd'],
meson: reorder the "features" list Try to make this more manageable by reording: - dependencies / inputs (with subcategory of compression libraries) - major components / outputs - optional features / conditionals that don't fit into the two above categories The division isn't well defined, because libraries often correspond one-to-one to feature, but not always.
2021-04-15 10:59:03 +02:00
['tmpfiles'],
['userdb'],
['vconsole'],
vmspawn: added initial code for vmspawn vmspawn-settings.c is currently empty but this will be used in future to house code for parsing settings from a file
2023-10-12 13:13:19 +01:00
['vmspawn'],
meson: reorder the "features" list Try to make this more manageable by reording: - dependencies / inputs (with subcategory of compression libraries) - major components / outputs - optional features / conditionals that don't fit into the two above categories The division isn't well defined, because libraries often correspond one-to-one to feature, but not always.
2021-04-15 10:59:03 +02:00
['xdg-autostart'],
# optional features
meson: introduce HAVE_DMI flag The condition is used at several places. Let's introduce a simple flag for that.
2023-07-30 06:32:53 +09:00
['dmi'],
resolve: always use openssl as backend of DNS-over-TLS This drops support for dns-over-tls=gnutls meson option.
2025-04-02 01:45:51 +09:00
['DNS-over-TLS'],
meson: reorder the "features" list Try to make this more manageable by reording: - dependencies / inputs (with subcategory of compression libraries) - major components / outputs - optional features / conditionals that don't fit into the two above categories The division isn't well defined, because libraries often correspond one-to-one to feature, but not always.
2021-04-15 10:59:03 +02:00
['idn'],
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-27 20:54:52 -04:00
['polkit'],
journalctl: allow statically linked build The journalctl tool may be needed on cross compilation hosts in order to run --update-catalog against a target rootfs. To avoid reliability issues caused by shared linking allow journalctl to be linked statically.
2022-07-27 15:28:09 -06:00
['legacy-pkla', install_polkit_pkla],
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-27 20:54:52 -04:00
['kmod'],
shutdown: Add Xen kexec support In the Xen case, it's the hypervisor which manages kexec. We thus have to ask it whether a kernel is loaded, instead of relying on /sys/kernel/kexec_loaded.
2022-11-01 16:53:02 +01:00
['xenctrl'],
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-27 20:54:52 -04:00
['dbus'],
['glib'],
['tpm'],
journalctl: allow statically linked build The journalctl tool may be needed on cross compilation hosts in order to run --update-catalog against a target rootfs. To avoid reliability issues caused by shared linking allow journalctl to be linked statically.
2022-07-27 15:28:09 -06:00
['man pages', want_man],
['html pages', want_html],
['man page indices', want_man and have_lxml],
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-27 20:54:52 -04:00
['SysV compat'],
meson: reorder the "features" list Try to make this more manageable by reording: - dependencies / inputs (with subcategory of compression libraries) - major components / outputs - optional features / conditionals that don't fit into the two above categories The division isn't well defined, because libraries often correspond one-to-one to feature, but not always.
2021-04-15 10:59:03 +02:00
['compat-mutable-uid-boundaries'],
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-27 20:54:52 -04:00
['utmp'],
['ldconfig'],
journalctl: allow statically linked build The journalctl tool may be needed on cross compilation hosts in order to run --update-catalog against a target rootfs. To avoid reliability issues caused by shared linking allow journalctl to be linked statically.
2022-07-27 15:28:09 -06:00
['adm group', get_option('adm-group')],
['wheel group', get_option('wheel-group')],
sysusers: make group shadow support configurable Some distros (openSUSE) don't have group shadow support enabled. This can lead to the following error: # systemd-sysusers Creating group foofoo with gid 478. # systemd-sysusers # groupdel foofoo # systemd-sysusers Creating group foofoo with gid 478. Failed to write files: File exists This patch adds --disable-gshadow option to configure. If used, systemd-sysvusers won't consider /etc/gshadow.
2017-05-09 14:02:37 +02:00
['gshadow'],
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-27 20:54:52 -04:00
['debug hashmap'],
['debug mmap cache'],
meson: also add option for debugging siphash
2018-11-23 00:36:35 +09:00
['debug siphash'],
journalctl: allow statically linked build The journalctl tool may be needed on cross compilation hosts in order to run --update-catalog against a target rootfs. To avoid reliability issues caused by shared linking allow journalctl to be linked statically.
2022-07-27 15:28:09 -06:00
['trace logging', conf.get('LOG_TRACE') == 1],
test: Only set environment variable if integration tests are enabled. If we set it to '0' if integration tests are not enabled then we can't enable them from the command line since environment from meson takes priority over environment variables from the command line. We also rename the related variables to avoid conflicts with the existing integration_tests variable.
2024-05-02 20:21:29 +02:00
['slow tests', want_slow_tests],
['fuzz tests', want_fuzz_tests],
journalctl: allow statically linked build The journalctl tool may be needed on cross compilation hosts in order to run --update-catalog against a target rootfs. To avoid reliability issues caused by shared linking allow journalctl to be linked statically.
2022-07-27 15:28:09 -06:00
['install tests', install_tests],
['link-udev-shared', get_option('link-udev-shared')],
['link-systemctl-shared', get_option('link-systemctl-shared')],
['link-networkd-shared', get_option('link-networkd-shared')],
['link-timesyncd-shared', get_option('link-timesyncd-shared')],
['link-journalctl-shared', get_option('link-journalctl-shared')],
['link-boot-shared', get_option('link-boot-shared')],
portable, meson: allow statically linked build Build option "link-portabled-shared" to build a statically linked systemd-portabled by using -Dlink-portabled-shared=false on systems with full systemd stack except systemd-portabled, such as CentOS/RHEL 9.
2023-05-14 03:17:47 +02:00
['link-portabled-shared', get_option('link-portabled-shared')],
manager: optionally, do a full preset on first boot A compile time option is added to select behaviour: by default UNIT_FILE_PRESET_ENABLE_ONLY is still used, but the intent is to change to UNIT_FILE_PRESET_FULL at some point in the future. Distros that want to opt-in can use the config option to change the behaviour. (The option is just a boolean: it would be possible to make it multi-valued, and allow full, enable-only, disable-only, none. But so far nobody has asked for this, and it's better not to complicate things needlessly.) With the configuration option flipped, instead of only doing enablements, perform a full preset on first boot. The reason is that although `/etc/machine-id` might be missing, there may be other files provisioned in `/etc` (in fact, this use case is mentioned in `log_execution_mode`). Some of those possible files include enablement symlinks even if presets dictate it should be disabled. Such a seemingly contradictory situation occurs in {RHEL,Fedora} CoreOS, where we ship `/etc` as if `preset-all` were called. However, we want to allow users to disable default-enabled services via Ignition, which does this by creating preset dropins before switchroot. (For why we do `preset-all` at compose time, see: https://github.com/coreos/fedora-coreos-config/pull/77). For example, the composed FCOS image has a `enable zincati.service` preset and an enablement for that in `/etc`, while at boot time when we switch root, there may be a `disable zincati.service` preset with higher precedence. In that case, we want systemd to disable the service. This is essentially a revert of 304b3079a203. It seems like systemd *used* to do this, but it was changed to try to make the container workflow a bit faster. Resolves: https://github.com/coreos/fedora-coreos-tracker/issues/392 Co-authored-by: Zbigniew Jędrzejewski-Szmek <zbyszek@in.waw.pl>
2020-03-23 12:25:19 -04:00
['first-boot-full-preset'],
meson: add option for fexecve use There are downsides to using fexecve: when fexecve is used (for normal executables), /proc/pid/status shows Name: 3, which means that ps -C foobar doesn't work. pidof works, because it checks /proc/self/cmdline. /proc/self/exe also shows the correct link, but requires privileges to read. /proc/self/comm also shows "3". I think this can be considered a kernel deficiency: when O_CLOEXEC is used, this "3" is completely meaningless. It could be any number. The kernel should use argv[0] instead, which at least has *some* meaning. I think the approach with fexecve/execveat is instersting, so let's provide it as opt-in.
2020-11-06 15:01:13 +01:00
['fexecve'],
journalctl: allow statically linked build The journalctl tool may be needed on cross compilation hosts in order to run --update-catalog against a target rootfs. To avoid reliability issues caused by shared linking allow journalctl to be linked statically.
2022-07-27 15:28:09 -06:00
['standalone-binaries', get_option('standalone-binaries')],
['coverage', get_option('b_coverage')],
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-27 20:54:52 -04:00
]
meson: avoid warning about comparison of bool and string meson.build:2907: WARNING: Trying to compare values of different types (bool, str) using ==. The result of this is undefined and will become a hard error in a future Meson release.
2018-03-09 14:21:08 +01:00
if tuple.length() >= 2
cond = tuple[1]
else
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-27 20:54:52 -04:00
ident1 = 'HAVE_' + tuple[0].underscorify().to_upper()
ident2 = 'ENABLE_' + tuple[0].underscorify().to_upper()
build-sys: use #if Y instead of #ifdef Y everywhere The advantage is that is the name is mispellt, cpp will warn us. $ git grep -Ee "conf.set\('(HAVE|ENABLE)_" -l|xargs sed -r -i "s/conf.set\('(HAVE|ENABLE)_/conf.set10('\1_/" $ git grep -Ee '#ifn?def (HAVE|ENABLE)' -l|xargs sed -r -i 's/#ifdef (HAVE|ENABLE)/#if \1/; s/#ifndef (HAVE|ENABLE)/#if ! \1/;' $ git grep -Ee 'if.*defined\(HAVE' -l|xargs sed -i -r 's/defined\((HAVE_[A-Z0-9_]*)\)/\1/g' $ git grep -Ee 'if.*defined\(ENABLE' -l|xargs sed -i -r 's/defined\((ENABLE_[A-Z0-9_]*)\)/\1/g' + manual changes to meson.build squash! build-sys: use #if Y instead of #ifdef Y everywhere v2: - fix incorrect setting of HAVE_LIBIDN2
2017-10-03 10:41:51 +02:00
cond = conf.get(ident1, 0) == 1 or conf.get(ident2, 0) == 1
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-27 20:54:52 -04:00
endif
if cond
meson: drop parens when appending to list Meson does not care either way, so let's use the simpler syntax. And files() already gives a list, so nesting this in a list wouldn't be necessary even if meson did not flatten everything.
2018-07-31 09:44:11 +02:00
found += tuple[0]
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-27 20:54:52 -04:00
else
meson: drop parens when appending to list Meson does not care either way, so let's use the simpler syntax. And files() already gives a list, so nesting this in a list wouldn't be necessary even if meson did not flatten everything.
2018-07-31 09:44:11 +02:00
missing += tuple[0]
meson: add status report This is similar to what ./configure prints. Instead of a long list of yes/no lines, I added two lines at the end with "enabled features" and "disabled features". This is what the mplayer/mencoder ./configure script did back in the day. The advantage is that it's easy to look at the list of disabled features and check for any unexpected entries.
2017-04-27 20:54:52 -04:00
endif
endforeach
meson: fold static libsystemd and libudev into the generic status line It's a very specialized feature and we don't need to devote a separate line to it.
2021-04-15 09:56:06 +02:00
if static_libsystemd == 'false'
missing += 'static-libsystemd'
else
meson: using f-strings in meson Those were added before 0.61 which is our minimum version. Dots were dropped from the end of some messages. Most messages did not have the trailing dot. 0.63 added support for multi-line f-strings. We can't use those yet.
2025-10-03 15:26:20 +02:00
found += f'static-libsystemd(@static_libsystemd@)'
meson: fold static libsystemd and libudev into the generic status line It's a very specialized feature and we don't need to devote a separate line to it.
2021-04-15 09:56:06 +02:00
endif
if static_libudev == 'false'
missing += 'static-libudev'
else
meson: using f-strings in meson Those were added before 0.61 which is our minimum version. Dots were dropped from the end of some messages. Most messages did not have the trailing dot. 0.63 added support for multi-line f-strings. We can't use those yet.
2025-10-03 15:26:20 +02:00
found += f'static-libudev(@static_libudev@)'
meson: fold static libsystemd and libudev into the generic status line It's a very specialized feature and we don't need to devote a separate line to it.
2021-04-15 09:56:06 +02:00
endif
meson: use meson's summary() The output is similar to our hand-crafted status message, but it's nice to use the built-in functionality. After all, it was amended during development to support our use case.
2021-07-28 10:04:01 +02:00
summary({
meson: realign summary
2025-08-14 21:48:43 +09:00
'enabled' : ', '.join(found),
'disabled' : ', '.join(missing)
},
meson: use meson's summary() The output is similar to our hand-crafted status message, but it's nice to use the built-in functionality. After all, it was amended during development to support our use case.
2021-07-28 10:04:01 +02:00
section : 'Features')
Reference in New Issue Copy Permalink