mir/bind
2
0
Fork 0
mirror of https://gitlab.isc.org/isc-projects/bind9 synced 2025-08-22 18:19:42 +00:00
Code Issues Releases Wiki Activity
bind/configure.ac

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

1773 lines
57 KiB
Plaintext
Raw Normal View History

[master] add uname data to named -V 4308. [func] Added operating system details to "named -V" output. [RT #41452]
2016-01-30 11:03:10 -08:00
# Copyright (C) Internet Systems Consortium, Inc. ("ISC")
update copyright notice
2014-03-10 23:46:12 +00:00
#
update copyright
2000-03-23 19:43:14 +00:00
# SPDX-License-Identifier: MPL-2.0
Update the copyright information in all files in the repository This commit converts the license handling to adhere to the REUSE specification. It specifically: 1. Adds used licnses to LICENSES/ directory 2. Add "isc" template for adding the copyright boilerplate 3. Changes all source files to include copyright and SPDX license header, this includes all the C sources, documentation, zone files, configuration files. There are notes in the doc/dev/copyrights file on how to add correct headers to the new files. 4. Handle the rest that can't be modified via .reuse/dep5 file. The binary (or otherwise unmodifiable) files could have license places next to them in <foo>.license file, but this would lead to cluttered repository and most of the files handled in the .reuse/dep5 file are system test files.
2021-06-03 08:37:05 +02:00
#
update copyright
2000-03-23 19:43:14 +00:00
# This Source Code Form is subject to the terms of the Mozilla Public
# License, v. 2.0. If a copy of the MPL was not distributed with this
# file, you can obtain one at https://mozilla.org/MPL/2.0/.
Update license headers to not include years in copyright in all applicable files
2018-02-23 09:53:12 +01:00
#
update copyright
2000-03-23 19:43:14 +00:00
# See the COPYRIGHT file distributed with this work for additional
Also look in libc for pthread_create before deciding the system doesn't have pthreads support. Not all systems have libpthread (eg, BSD/OS). Many compiler options that were not particular to whether threads were being used were being set in a threads-specific block, meaning they didn't get set at all if threads were not being used. ISC_PLATFORM_USETHREADS could end up still defined even if threading was later disabled, such as if pthread_create was not found.
2000-12-23 18:27:43 +00:00
# information regarding copyright ownership.
[master] use adaptive locks when available 3781. [tuning] Use adaptive mutex locks when available; this has been found to improve performance under load on many systems. "configure --with-locktype=standard" restores conventional mutex locks. [RT #32576]
2014-03-10 12:14:35 -07:00
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
#
# Defining the version in m4 requires a bit of dancing around,
# so the strings are properly concatenated, as you can't use
Update the configure.ac for autoconf >= 2.71 compatibility This mostly removes stuff that's either deprecated, obsolete or not used at all: * Update the minimal autoconf version to 2.69 * AC_PROG_CC_C99 is deprecated, just use AC_PROG_CC as we require C11 anyway * AC_HEADER_TIME is deprecated, both <sys/time.h> and <time.h> can be included at the same time, and we don't use the macros that AC_HEADER_TIME defines anywhere * AC_HEADER_STDC checks for ISO C90 and we require at least C11 * Replace AC_TRY_*([]) with AC_*_IFELSE([AC_LANG_PROGRAM()]) * Update m4/ax_check_openssl.m4 from serial 10 to serial 11 * Update m4/ax_gcc_func_attribute.m4 from serial 10 to serial 13 * Update m4/ax_pthread.m4 from serial 24 to serial 30 * Add early AC_CANONICAL_TARGET call to prevent warning from AX_PTHREAD
2021-04-16 12:25:35 +02:00
# a shell variable in AC_INIT
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
#
m4_define([bind_VERSION_MAJOR], 9)dnl
Update BIND version to 9.19.0
2022-04-11 10:08:24 +02:00
m4_define([bind_VERSION_MINOR], 19)dnl
Update BIND version to 9.19.14-dev
2023-05-10 10:41:42 +02:00
m4_define([bind_VERSION_PATCH], 14)dnl
Update BIND version to 9.19.1-dev
2022-04-12 11:15:13 +02:00
m4_define([bind_VERSION_EXTRA], -dev)dnl
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
m4_define([bind_DESCRIPTION], [(Development Release)])dnl
Revise "srcid" file handling The "srcid" file present in each BIND source tarball contains a shortened hash of the Git commit corresponding to a given BIND release. This allows a Git reference to be included in an archive that otherwise lacks any Git information. Before the move to Automake, if an "srcid" file was present in the root source directory at the time ./configure was run, its contents were used as the value of a compile-time constant which was then baked into BIND binaries; otherwise, "git rev-parse" was used to determine the value of that constant. With Automake, a similar approach was attempted that required the "srcid" file to be present at autoreconf time in order for it to be used. However, note that this means that even if that file is present in a source tarball created using "make dist", its contents are not going to influence the value of the aforementioned compile-time constant because autoreconf hardcodes the output of "git rev-parse" into the configure script at autoreconf time. To make things more clear, always use "git rev-parse" for determining the value of the PACKAGE_SRCID compile-time constant when running autoreconf. This causes "srcid" to be an empty string in source tarballs built from other source tarballs, but that is not deemed to be much of an issue as "make dist" is expected to be run from Git repository clones. Remove stderr redirections to /dev/null to ensure errors caused e.g. by running "make dist" from outside a Git repository clone are not hidden. Trim the Git commit hash to 7 characters for consistency between Unix and Windows systems. Despite the above, ensure the "srcid" file is present in source tarballs created using "make dist" as that file is used by the build process on Windows.
2020-06-09 14:47:06 +02:00
m4_define([bind_SRCID], [m4_esyscmd_s([git rev-parse --short HEAD | cut -b1-7])])dnl
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
m4_define([bind_PKG_VERSION], [[bind_VERSION_MAJOR.bind_VERSION_MINOR.bind_VERSION_PATCH]bind_VERSION_EXTRA])dnl
#
# Autoconf initialization
#
Change the bug reporting address from email to gitlab URI In autoconf, the AC_INIT() accepts bugreport address for reporting issues (f.e. when the test suite fails). Instead of providing generic emails address, change this to the address where to report with the default Bug template applied.
2022-02-02 10:50:45 +01:00
AC_INIT([BIND], bind_PKG_VERSION, [https://gitlab.isc.org/isc-projects/bind9/-/issues/new?issuable_template=Bug], [], [https://www.isc.org/downloads/])
Update the configure.ac for autoconf >= 2.71 compatibility This mostly removes stuff that's either deprecated, obsolete or not used at all: * Update the minimal autoconf version to 2.69 * AC_PROG_CC_C99 is deprecated, just use AC_PROG_CC as we require C11 anyway * AC_HEADER_TIME is deprecated, both <sys/time.h> and <time.h> can be included at the same time, and we don't use the macros that AC_HEADER_TIME defines anywhere * AC_HEADER_STDC checks for ISO C90 and we require at least C11 * Replace AC_TRY_*([]) with AC_*_IFELSE([AC_LANG_PROGRAM()]) * Update m4/ax_check_openssl.m4 from serial 10 to serial 11 * Update m4/ax_gcc_func_attribute.m4 from serial 10 to serial 13 * Update m4/ax_pthread.m4 from serial 24 to serial 30 * Add early AC_CANONICAL_TARGET call to prevent warning from AX_PTHREAD
2021-04-16 12:25:35 +02:00
AC_PREREQ([2.69])
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AC_DEFINE([PACKAGE_VERSION_MAJOR], ["][bind_VERSION_MAJOR]["], [BIND 9 Major part of the version])
AC_DEFINE([PACKAGE_VERSION_MINOR], ["][bind_VERSION_MINOR]["], [BIND 9 Minor part of the version])
AC_DEFINE([PACKAGE_VERSION_PATCH], ["][bind_VERSION_PATCH]["], [BIND 9 Patch part of the version])
AC_DEFINE([PACKAGE_VERSION_EXTRA], ["][bind_VERSION_EXTRA]["], [BIND 9 Extra part of the version])
AC_DEFINE([PACKAGE_DESCRIPTION], [m4_ifnblank(bind_DESCRIPTION, [" ]bind_DESCRIPTION["], [])], [An extra string to print after PACKAGE_STRING])
AC_DEFINE([PACKAGE_SRCID], ["][bind_SRCID]["], [A short hash from git])
bind_CONFIGARGS="${ac_configure_args:-default}"
AC_DEFINE_UNQUOTED([PACKAGE_CONFIGARGS], ["$bind_CONFIGARGS"], [Either 'defaults' or used ./configure options])
AC_DEFINE([PACKAGE_BUILDER], ["make"], [make or Visual Studio])
Update the configure.ac for autoconf >= 2.71 compatibility This mostly removes stuff that's either deprecated, obsolete or not used at all: * Update the minimal autoconf version to 2.69 * AC_PROG_CC_C99 is deprecated, just use AC_PROG_CC as we require C11 anyway * AC_HEADER_TIME is deprecated, both <sys/time.h> and <time.h> can be included at the same time, and we don't use the macros that AC_HEADER_TIME defines anywhere * AC_HEADER_STDC checks for ISO C90 and we require at least C11 * Replace AC_TRY_*([]) with AC_*_IFELSE([AC_LANG_PROGRAM()]) * Update m4/ax_check_openssl.m4 from serial 10 to serial 11 * Update m4/ax_gcc_func_attribute.m4 from serial 10 to serial 13 * Update m4/ax_pthread.m4 from serial 24 to serial 30 * Add early AC_CANONICAL_TARGET call to prevent warning from AX_PTHREAD
2021-04-16 12:25:35 +02:00
#
# Get the Canonical System Type
#
AC_CANONICAL_BUILD
AC_CANONICAL_HOST
AC_CANONICAL_TARGET
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AC_CONFIG_SRCDIR([bin/named/main.c])
Bump the required automake version to 1.14 After the build system refactoring, we no longer call AM_PROG_CC_C_O because it is obsolescent macro. According to the automake manual the `AC_PROG_CC` has been rewritten in automake 1.14 to not required the call, thus we need to require at least automake version 1.14.
2022-02-14 09:17:41 +01:00
AM_INIT_AUTOMAKE([1.14 tar-pax foreign subdir-objects dist-xz no-dist-gzip -Wall -Werror])
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AM_SILENT_RULES([yes])
Convert the documentation to Sphinx documentation format The ARM and the manpages have been converted into Sphinx documentation format. Sphinx uses reStructuredText as its markup language, and many of its strengths come from the power and straightforwardness of reStructuredText and its parsing and translating suite, the Docutils.
2019-03-31 12:49:47 +02:00
AM_EXTRA_RECURSIVE_TARGETS([test unit doc])
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AC_CONFIG_HEADERS([config.h])
AC_CONFIG_MACRO_DIR([m4])
BIND9 Pool Creation
1998-12-11 20:10:26 +00:00
Add AM_MAINTAINER_MODE macro to configure.ac and enable it by default AM_MAINTAINER_MODE macro adds ability to disable rebuilding build file (Makefile.in, configure, ...) when the source file changes. This is important in the CI where the timestamps could get skewed and that triggers the rebuild on every ./configure run.
2019-06-19 14:26:49 +02:00
#
Enable maintainer mode by default only when building from git When maintainer mode is enabled (./configure --enable-maintainer-mode) it enables rebuild of documentation source files that require extra tools to be installed or compiled. For a convenience, those files are already committed into the repository and their rebuild is not required to build BIND 9 from sources.
2020-06-22 14:26:07 +02:00
# Enable maintainer mode by default only when building from git repository
don't enable maintainer mode when cross compiling
2020-12-18 16:37:26 +11:00
# and not cross compling
Add AM_MAINTAINER_MODE macro to configure.ac and enable it by default AM_MAINTAINER_MODE macro adds ability to disable rebuilding build file (Makefile.in, configure, ...) when the source file changes. This is important in the CI where the timestamps could get skewed and that triggers the rebuild on every ./configure run.
2019-06-19 14:26:49 +02:00
#
don't enable maintainer mode when cross compiling
2020-12-18 16:37:26 +11:00
AS_IF([test "$cross_compiling" = no],
[AM_MAINTAINER_MODE(m4_ifset([bind_SRCID],[enable],[disable]))],
[AM_MAINTAINER_MODE([disable])])
Add AM_MAINTAINER_MODE macro to configure.ac and enable it by default AM_MAINTAINER_MODE macro adds ability to disable rebuilding build file (Makefile.in, configure, ...) when the source file changes. This is important in the CI where the timestamps could get skewed and that triggers the rebuild on every ./configure run.
2019-06-19 14:26:49 +02:00
Enable LFS using standard autoconf macros
2019-03-07 13:32:30 +01:00
#
# Enable system extensions to C and POSIX
#
Enable system extensions macros using AC_USE_SYSTEM_EXTENSIONS and remove handcrafted code that enabled __EXTENSIONS__ and _GNU_SOURCE
2018-08-28 21:27:48 +02:00
AC_USE_SYSTEM_EXTENSIONS
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
#
# Compiler compatibility flags
#
Update the configure.ac for autoconf >= 2.71 compatibility This mostly removes stuff that's either deprecated, obsolete or not used at all: * Update the minimal autoconf version to 2.69 * AC_PROG_CC_C99 is deprecated, just use AC_PROG_CC as we require C11 anyway * AC_HEADER_TIME is deprecated, both <sys/time.h> and <time.h> can be included at the same time, and we don't use the macros that AC_HEADER_TIME defines anywhere * AC_HEADER_STDC checks for ISO C90 and we require at least C11 * Replace AC_TRY_*([]) with AC_*_IFELSE([AC_LANG_PROGRAM()]) * Update m4/ax_check_openssl.m4 from serial 10 to serial 11 * Update m4/ax_gcc_func_attribute.m4 from serial 10 to serial 13 * Update m4/ax_pthread.m4 from serial 24 to serial 30 * Add early AC_CANONICAL_TARGET call to prevent warning from AX_PTHREAD
2021-04-16 12:25:35 +02:00
m4_version_prereq([2.70],
[AC_PROG_CC],
[AC_PROG_CC_C99])
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AC_PROG_CPP_WERROR
lib/dns/gen should be built with CC_FOR_BUILD
2020-12-18 15:48:00 +11:00
#
# Find build compiler when cross compiling
#
Replace AX_CC_FOR_BUILD with AX_PROG_CC_FOR_BUILD The AX_PROG_CC_FOR_BUILD implementation to find a native CC compiler is slightly better because it uses AC_PROG_CC and AC_PROG_CPP to find the native compiler instead of just defaulting to `gcc` as AX_CC_FOR_BUILD does. AX_PROG_CC_FOR_BUILD also sets BUILD_EXEEXT that we already use in the Makefile.am for `lib/dns/gen` while AX_CC_FOR_BUILD uses EXEEXT_FOR_BUILD.
2022-01-25 10:47:34 +01:00
AX_PROG_CC_FOR_BUILD
lib/dns/gen should be built with CC_FOR_BUILD
2020-12-18 15:48:00 +11:00
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
#
# Find the machine's endian flavor.
#
AC_C_BIGENDIAN
Enable system extensions macros using AC_USE_SYSTEM_EXTENSIONS and remove handcrafted code that enabled __EXTENSIONS__ and _GNU_SOURCE
2018-08-28 21:27:48 +02:00
Enable LFS using standard autoconf macros
2019-03-07 13:32:30 +01:00
#
# Enable large file support
#
AC_SYS_LARGEFILE
AC_FUNC_FSEEKO
Enable system extensions macros using AC_USE_SYSTEM_EXTENSIONS and remove handcrafted code that enabled __EXTENSIONS__ and _GNU_SOURCE
2018-08-28 21:27:48 +02:00
# Enable RFC 3542 APIs on macOS
AC_DEFINE([__APPLE_USE_RFC_3542], [1], [Select RFC3542 IPv6 API on macOS])
BIND9 Pool Creation
1998-12-11 20:10:26 +00:00
Make this work for NetBSD, and assume ptl2 is installed as the pthreads library
1999-04-29 05:19:29 +00:00
AC_PROG_MAKE_SET
2610. [port] sunos: Change #2363 was not complete. [RT #19796]
2009-06-10 02:23:42 +00:00
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
# Checks for programs.
m4_ifdef([AM_PROG_AR], [AM_PROG_AR]) # call AM_PROG_AR only if available
Make this work for NetBSD, and assume ptl2 is installed as the pthreads library
1999-04-29 05:19:29 +00:00
AC_PROG_INSTALL
1817. [func] add support for additional zone file formats for improving loading performance. The masterfile-format option in named.conf can be used to specify a non-default format. A new separate command named-compilezone was provided to generate zone files in a new format.
2005-06-20 01:05:33 +00:00
AC_PROG_LN_S
Add and use AX_POSIX_SHELL autoconf macro, so we can rely on POSIX shell features
2018-11-06 17:09:08 +07:00
AX_POSIX_SHELL
Convert the documentation to Sphinx documentation format The ARM and the manpages have been converted into Sphinx documentation format. Sphinx uses reStructuredText as its markup language, and many of its strengths come from the power and straightforwardness of reStructuredText and its parsing and translating suite, the Docutils.
2019-03-31 12:49:47 +02:00
AC_PROG_MKDIR_P
Make this work for NetBSD, and assume ptl2 is installed as the pthreads library
1999-04-29 05:19:29 +00:00
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
# Initialize libtool
Disable and disallow static linking Linking BIND 9 programs and libraries statically disables several important features: * dlopen() - relied on by dynamic loading of modules, dlz, and dyndb, * RELRO (read-only relocations) and ASLR (address space layout randomization) - security features which are important for any program interacting with the network and/or user input. Disable and disallow linking BIND 9 binaries statically, thus enforcing dlopen() support and allowing use of RELRO and ASLR by default.
2020-06-16 10:38:46 +02:00
LT_INIT([disable-static dlopen pic-only])
Define and use DLOPEN_LIBS
2023-04-03 16:12:08 +10:00
DLOPEN_LIBS="$lt_cv_dlopen_libs"
AC_SUBST(DLOPEN_LIBS)
Allow static linking in the developer mode Previously, we have disallowed static linking (for good reasons). However, there are legitimate reasons where static linking might be useful, and one of the reasons is the OSS-Fuzz project that doesn't have the libraries used for build, so static linking is the sane option here. The static linking is still disallowed in the "production" builds, but it's not possible to disable shared and enable static libraries when used together with --enable-developer.
2020-08-12 15:19:23 +02:00
AS_IF([test "$enable_static" != "no" && test "$enable_developer" != "yes"],
Disable and disallow static linking Linking BIND 9 programs and libraries statically disables several important features: * dlopen() - relied on by dynamic loading of modules, dlz, and dyndb, * RELRO (read-only relocations) and ASLR (address space layout randomization) - security features which are important for any program interacting with the network and/or user input. Disable and disallow linking BIND 9 binaries statically, thus enforcing dlopen() support and allowing use of RELRO and ASLR by default.
2020-06-16 10:38:46 +02:00
[AC_MSG_ERROR([Static linking is not supported as it disables dlopen() and certain security features (e.g. RELRO, ASLR)])])
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
Define and use DLOPEN_LIBS
2023-04-03 16:12:08 +10:00
Integrate cmocka unit testing framework to kyua
2018-06-04 08:54:58 +02:00
#
Add -Wl,--export-dynamic to standard LDFLAGS if supported From the ld man page: When creating a dynamically linked executable, using the -E option or the --export-dynamic option causes the linker to add all symbols to the dynamic symbol table. The dynamic symbol table is the set of symbols which are visible from dynamic objects at run time. This should allow the backtrace(3) to fully resolve the symbols when creating backtrace on an assertion failure.
2022-04-27 16:41:27 +02:00
# Set the default CFLAGS, CPPFLAGS, and LDFLAGS
Integrate cmocka unit testing framework to kyua
2018-06-04 08:54:58 +02:00
#
nghttp2-based HTTP layer in netmgr This commit includes work-in-progress implementation of DNS-over-HTTP(S). Server-side code remains mostly untested, and there is only support for POST requests.
2020-10-31 20:42:18 +01:00
STD_CFLAGS="-Wall -Wextra -Wwrite-strings -Wpointer-arith -Wno-missing-field-initializers -Wformat -Wshadow"
Integrate cmocka unit testing framework to kyua
2018-06-04 08:54:58 +02:00
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
# These should be always errors
Avoid using C99 variable length arrays From an attacker's point of view, a VLA declaration is essentially a primitive for performing arbitrary arithmetic on the stack pointer. If the attacker can control the size of a VLA they have a very powerful tool for causing memory corruption. To mitigate this kind of attack, and the more general class of stack clash vulnerabilities, C compilers insert extra code when allocating a VLA to probe the growing stack one page at a time. If these probes hit the stack guard page, the program will crash. From the point of view of a C programmer, there are a few things to consider about VLAs: * If it is important to handle allocation failures in a controlled manner, don't use VLAs. You can use VLAs if it is OK for unreasonable inputs to cause an uncontrolled crash. * If the VLA is known to be smaller than some known fixed size, use a fixed size array and a run-time check to ensure it is large enough. This will be more efficient than the compiler's stack probes that need to cope with arbitrary-size VLAs. * If the VLA might be large, allocate it on the heap. The heap allocator can allocate multiple pages in one shot, whereas the stack clash probes work one page at a time. Most of the existing uses of VLAs in BIND are in test code where they are benign, but there was one instance in `named`, in the GSS-TSIG verification code, which has now been removed. This commit adjusts the style guide and the C compiler flags to allow VLAs in test code but not elsewhere.
2022-03-18 14:50:36 +00:00
STD_CFLAGS="$STD_CFLAGS -Werror=implicit-function-declaration -Werror=missing-prototypes -Werror=format-security -Werror=parentheses -Werror=implicit -Werror=strict-prototypes -Werror=vla"
Disable stringop-overread with gcc-11+ Address Sanitizer When Address Sanitizer is enabled in gcc-11+, number of false positives might appear like this: netmgr/udp.c: In function 'isc__nm_udp_send': netmgr/udp.c:729:13: warning: 'uv_udp_send' reading 16 bytes from a region of size 8 [-Wstringop-overread] 729 | r = uv_udp_send(&uvreq->uv_req.udp_send, &sock->uv_handle.udp, | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 730 | &uvreq->uvbuf, 1, sa, udp_send_cb); | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ netmgr/udp.c:729:13: note: referencing argument 3 of type 'const uv_buf_t[0]' In file included from ./include/isc/uv.h:17, from ./include/isc/barrier.h:31, from netmgr/udp.c:17: /usr/include/uv.h:711:15: note: in a call to function 'uv_udp_send' 711 | UV_EXTERN int uv_udp_send(uv_udp_send_t* req, | ^~~~~~~~~~~ Disable the warning globally in the autoconf, instead of just locally in a single CI job, as it might affect people outside our GitLab CI.
2022-09-08 07:33:31 +02:00
# Disable false positives generated by GCC 11+ and ASAN
AC_COMPILE_IFELSE(
[AC_LANG_PROGRAM([[]],
[[#if __GNUC__ >= 11 && defined(__SANITIZE_ADDRESS__)
#error Address Sanitizer enabled
#endif
]])],
[],
[STD_CFLAGS="$STD_CFLAGS -Wno-stringop-overread"])
Add -Wl,--export-dynamic to standard LDFLAGS if supported From the ld man page: When creating a dynamically linked executable, using the -E option or the --export-dynamic option causes the linker to add all symbols to the dynamic symbol table. The dynamic symbol table is the set of symbols which are visible from dynamic objects at run time. This should allow the backtrace(3) to fully resolve the symbols when creating backtrace on an assertion failure.
2022-04-27 16:41:27 +02:00
STD_LDFLAGS=""
Avoid using C99 variable length arrays From an attacker's point of view, a VLA declaration is essentially a primitive for performing arbitrary arithmetic on the stack pointer. If the attacker can control the size of a VLA they have a very powerful tool for causing memory corruption. To mitigate this kind of attack, and the more general class of stack clash vulnerabilities, C compilers insert extra code when allocating a VLA to probe the growing stack one page at a time. If these probes hit the stack guard page, the program will crash. From the point of view of a C programmer, there are a few things to consider about VLAs: * If it is important to handle allocation failures in a controlled manner, don't use VLAs. You can use VLAs if it is OK for unreasonable inputs to cause an uncontrolled crash. * If the VLA is known to be smaller than some known fixed size, use a fixed size array and a run-time check to ensure it is large enough. This will be more efficient than the compiler's stack probes that need to cope with arbitrary-size VLAs. * If the VLA might be large, allocate it on the heap. The heap allocator can allocate multiple pages in one shot, whereas the stack clash probes work one page at a time. Most of the existing uses of VLAs in BIND are in test code where they are benign, but there was one instance in `named`, in the GSS-TSIG verification code, which has now been removed. This commit adjusts the style guide and the C compiler flags to allow VLAs in test code but not elsewhere.
2022-03-18 14:50:36 +00:00
# ... except in test code
TEST_CFLAGS="-Wno-vla"
Integrate cmocka unit testing framework to kyua
2018-06-04 08:54:58 +02:00
Avoid redefining _FORTIFY_SOURCE Some compilers have a built-in definition of the _FORTIFY_SOURCE macro that differs from BIND's preferred setting. This causes errors like the one quoted below. The solution is to undefine the macro before defining it. A similar fix was recently committed to glibc. <command line>: error: '_FORTIFY_SOURCE' macro redefined #define _FORTIFY_SOURCE 2 ^ <built-in>: note: previous definition is here #define _FORTIFY_SOURCE 0 ^ https://sourceware.org/git/glibc.git/commitdiff/35bcb08eaa953c9b
2023-02-15 19:00:37 +00:00
# Fortify the sources by default (we undefine the macro first in case
# the compiler has a different built-in setting)
STD_CPPFLAGS="-U_FORTIFY_SOURCE -D_FORTIFY_SOURCE=2"
4811. [bug] Revert api changes to use <isc/buffer.h> inline macros. Provide a alternative mechanism to turn on the use of inline macros when building BIND. [RT #46520]
2017-11-08 23:28:10 +11:00
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
#
# Additional compiler settings.
#
AX_CHECK_COMPILE_FLAG([-fno-strict-aliasing],
[STD_CFLAGS="$STD_CFLAGS -fno-strict-aliasing"])
Force test for -fno-delete-null-pointer-checks to fail
2020-04-22 13:02:18 +10:00
# Clang only issues a warning so use -Werror to force a error.
AX_CHECK_COMPILE_FLAG([-Werror -fno-delete-null-pointer-checks],
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
[STD_CFLAGS="$STD_CFLAGS -fno-delete-null-pointer-checks"])
AX_CHECK_COMPILE_FLAG([-fdiagnostics-show-option],
[STD_CFLAGS="$STD_CFLAGS -fdiagnostics-show-option"])
[master] clean up gcc -Wshadow warnings 4039. [cleanup] Cleaned up warnings from gcc -Wshadow. [RT #37381]
2015-01-20 13:29:18 -08:00
Add -Wl,--export-dynamic to standard LDFLAGS if supported From the ld man page: When creating a dynamically linked executable, using the -E option or the --export-dynamic option causes the linker to add all symbols to the dynamic symbol table. The dynamic symbol table is the set of symbols which are visible from dynamic objects at run time. This should allow the backtrace(3) to fully resolve the symbols when creating backtrace on an assertion failure.
2022-04-27 16:41:27 +02:00
AX_CHECK_LINK_FLAG([-Wl,--export-dynamic],
[STD_LDFLAGS="$STD_LDFLAGS -Wl,--export-dynamic"])
Fix function overrides in unit tests on macOS Since Mac OS X 10.1, Mach-O object files are by default built with a so-called two-level namespace which prevents symbol lookups in BIND unit tests that attempt to override the implementations of certain library functions from working as intended. This feature can be disabled by passing the "-flat_namespace" flag to the linker. Fix unit tests affected by this issue on macOS by adding "-flat_namespace" to LDFLAGS used for building all object files on that operating system (it is not enough to only set that flag for the unit test executables).
2020-09-28 09:09:21 +02:00
host_macos=no
AS_CASE([$host],[*-darwin*],[host_macos=yes])
AM_CONDITIONAL([HOST_MACOS], [test "$host_macos" = "yes"])
fixed incorrect output from the configuration summary
2020-04-23 14:29:07 -07:00
#
# Change defaults for developers if not explicity set.
# Needs to be before the option is tested.
#
Add pairwise testing Pairwise testing is a test case generation technique based on the observation that most faults are caused by interactions of at most two factors. For BIND, its configure options can be thought of as such factors. Process BIND configure options into a model that is subsequently processed by the PICT tool in order to find an effective test vector. That test vector is then used for configuring and building BIND using various combinations of configure options.
2020-07-01 10:29:36 +02:00
# [pairwise: --enable-developer, --disable-developer]
Use couple more autoconf convenience macros to cleanup configure.ac Pull and use several autoconf archive convenience macros to simplify configure.ac. * AX_CHECK_COMPILE_FLAG(FLAG, ...) - check whether given CFLAG works * AX_CHECK_LINK_FLAG(FLAG, ...) - check whether given LDFLAG works * AX_CHECK_PREPROC_FLAG(FLAG, ...) - check whether give CPPFLAG works * AX_SAVE_FLAGS/AX_RESTORE_FLAGS - save and restore *FLAGS
2019-06-19 11:16:47 +02:00
AC_ARG_ENABLE([developer],
[AS_HELP_STRING([--enable-developer],
[enable developer build settings])])
allow -Werror to be enabled; turn on -Werror by default with --enable-developer (cherry picked from commit 2be0f12aafda4bccd2ffd9cf3dc6fe11b9b20832)
2014-07-08 15:51:40 +10:00
Use couple more autoconf convenience macros to cleanup configure.ac Pull and use several autoconf archive convenience macros to simplify configure.ac. * AX_CHECK_COMPILE_FLAG(FLAG, ...) - check whether given CFLAG works * AX_CHECK_LINK_FLAG(FLAG, ...) - check whether given LDFLAG works * AX_CHECK_PREPROC_FLAG(FLAG, ...) - check whether give CPPFLAG works * AX_SAVE_FLAGS/AX_RESTORE_FLAGS - save and restore *FLAGS
2019-06-19 11:16:47 +02:00
AS_IF([test "$enable_developer" = "yes"],
Only run system tests as root in developer mode Running system tests with root privileges is potentially dangerous. Only allow it when explicitly requested (by building with --enable-developer).
2020-07-30 14:07:49 +02:00
[DEVELOPER_MODE=yes
Add strong rwlock consistency checks to dns_rbtdb The dns_rbtdb unit already tracks the state of the node and tree rwlocks during the top level function and passes the states of the locks to the called functions. Add the tree locking family of macros modeled after node locking macros, and expand both to track the state of the lock in an external variable. Additionally, in developer mode, add precondition to the macros, so the lock is in required state - this should cause an assertion failure on double locking instead of the thread getting stuck.
2022-10-14 14:35:44 +02:00
STD_CPPFLAGS="$STD_CPPFLAGS -DISC_MEM_DEFAULTFILL=1 -DISC_MEM_TRACKLINES=1 -DISC_LIST_CHECKINIT=1 -DISC_STATS_CHECKUNDERFLOW=1 -DDNS_RBTDB_STRONG_RWLOCK_CHECK=1"
Use couple more autoconf convenience macros to cleanup configure.ac Pull and use several autoconf archive convenience macros to simplify configure.ac. * AX_CHECK_COMPILE_FLAG(FLAG, ...) - check whether given CFLAG works * AX_CHECK_LINK_FLAG(FLAG, ...) - check whether given LDFLAG works * AX_CHECK_PREPROC_FLAG(FLAG, ...) - check whether give CPPFLAG works * AX_SAVE_FLAGS/AX_RESTORE_FLAGS - save and restore *FLAGS
2019-06-19 11:16:47 +02:00
test "${enable_fixed_rrset+set}" = set || enable_fixed_rrset=yes
test "${enable_querytrace+set}" = set || enable_querytrace=yes
test "${with_cmocka+set}" = set || with_cmocka=yes
test "${with_zlib+set}" = set || with_zlib=yes
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
test "${enable_warn_error+set}" = set || enable_warn_error=yes
Use couple more autoconf convenience macros to cleanup configure.ac Pull and use several autoconf archive convenience macros to simplify configure.ac. * AX_CHECK_COMPILE_FLAG(FLAG, ...) - check whether given CFLAG works * AX_CHECK_LINK_FLAG(FLAG, ...) - check whether given LDFLAG works * AX_CHECK_PREPROC_FLAG(FLAG, ...) - check whether give CPPFLAG works * AX_SAVE_FLAGS/AX_RESTORE_FLAGS - save and restore *FLAGS
2019-06-19 11:16:47 +02:00
])
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
Only run system tests as root in developer mode Running system tests with root privileges is potentially dangerous. Only allow it when explicitly requested (by building with --enable-developer).
2020-07-30 14:07:49 +02:00
AC_SUBST([DEVELOPER_MODE])
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AC_SUBST([STD_CFLAGS])
AC_SUBST([STD_CPPFLAGS])
Add -Wl,--export-dynamic to standard LDFLAGS if supported From the ld man page: When creating a dynamically linked executable, using the -E option or the --export-dynamic option causes the linker to add all symbols to the dynamic symbol table. The dynamic symbol table is the set of symbols which are visible from dynamic objects at run time. This should allow the backtrace(3) to fully resolve the symbols when creating backtrace on an assertion failure.
2022-04-27 16:41:27 +02:00
AC_SUBST([STD_LDFLAGS])
Avoid using C99 variable length arrays From an attacker's point of view, a VLA declaration is essentially a primitive for performing arbitrary arithmetic on the stack pointer. If the attacker can control the size of a VLA they have a very powerful tool for causing memory corruption. To mitigate this kind of attack, and the more general class of stack clash vulnerabilities, C compilers insert extra code when allocating a VLA to probe the growing stack one page at a time. If these probes hit the stack guard page, the program will crash. From the point of view of a C programmer, there are a few things to consider about VLAs: * If it is important to handle allocation failures in a controlled manner, don't use VLAs. You can use VLAs if it is OK for unreasonable inputs to cause an uncontrolled crash. * If the VLA is known to be smaller than some known fixed size, use a fixed size array and a run-time check to ensure it is large enough. This will be more efficient than the compiler's stack probes that need to cope with arbitrary-size VLAs. * If the VLA might be large, allocate it on the heap. The heap allocator can allocate multiple pages in one shot, whereas the stack clash probes work one page at a time. Most of the existing uses of VLAs in BIND are in test code where they are benign, but there was one instance in `named`, in the GSS-TSIG verification code, which has now been removed. This commit adjusts the style guide and the C compiler flags to allow VLAs in test code but not elsewhere.
2022-03-18 14:50:36 +00:00
AC_SUBST([TEST_CFLAGS])
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
Add pairwise testing Pairwise testing is a test case generation technique based on the observation that most faults are caused by interactions of at most two factors. For BIND, its configure options can be thought of as such factors. Process BIND configure options into a model that is subsequently processed by the PICT tool in order to find an effective test vector. That test vector is then used for configuring and building BIND using various combinations of configure options.
2020-07-01 10:29:36 +02:00
# [pairwise: --enable-warn-error, --disable-warn-error]
fixed incorrect output from the configuration summary
2020-04-23 14:29:07 -07:00
AC_ARG_ENABLE([warn_error],
[AS_HELP_STRING([--enable-warn-error],
[turn on -Werror when compiling])],
[],[enable_warn_error=no])
AS_IF([test "$enable_warn_error" = "yes"],
[STD_CFLAGS="$STD_CFLAGS -Werror"])
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
#
# Use pkg-config
#
PKG_PROG_PKG_CONFIG
AS_IF([test -z "$PKG_CONFIG"],
[AC_MSG_ERROR([The pkg-config script could not be found or is too old.])])
Print out the Userspace-RCU flavor and version The `named -V` now prints out the liburcu flavor used and the compile-time version.
2023-04-24 13:32:34 +02:00
#
# Pick Userspace-RCU flavor (developer only option)
#
Fix the typo in --with-liburcu=mb pairwise definition It was "ucru," but it should be "urcu". The pairwise CI job fails as a result with: configure: error: unrecognized options: --with-libucru
2023-04-28 09:35:33 +02:00
# [pairwise: --with-liburcu=membarrier, --with-liburcu=qsbr, --with-liburcu=mb, --with-liburcu=signal, --with-liburcu=bp]
Improve the Userspace RCU integration This commit allows BIND 9 to be compiled with different flavours of Userspace RCU, and improves the integration between Userspace RCU and our event loop: - In the RCU QSBR, the thread is put offline when polling and online when rcu_dereference, rcu_assign_pointer (or friends) are called. - In other RCU modes, we check that we are not reading when reaching the quiescent callback in the event loop. - We register the thread before uv_work_run() callback is called and after it has finished. The rcu_(un)register_thread() has a large overhead, but that's fine in this case.
2023-03-30 17:04:22 +02:00
AC_ARG_WITH([liburcu],
[AS_HELP_STRING([--with-liburcu=FLAVOR],
[Build with Userspace-RCU variant (membarrier|qsbr|mb|signal|bp) [default=membarrier]]))],
[], [with_liburcu=membarrier])
AC_MSG_CHECKING([liburcu flavor])
AS_CASE([$with_liburcu],
Print out the Userspace-RCU flavor and version The `named -V` now prints out the liburcu flavor used and the compile-time version.
2023-04-24 13:32:34 +02:00
[membarrier],
[RCU_FLAVOR="liburcu"
AC_DEFINE([RCU_MEMBARRIER], [1], [Build with membarrier Userspace-RCU flavor])],
[qsbr],
[RCU_FLAVOR="liburcu-qsbr"
AC_DEFINE([RCU_QSBR], [1], [Build with QSBR Userspace-RCU flavor])],
[mb],
[RCU_FLAVOR="liburcu-mb"
AC_DEFINE([RCU_MB], [1], [Build with mb Userspace-RCU flavor])],
[signal],
[RCU_FLAVOR="liburcu-signal"
AC_DEFINE([RCU_SIGNAL], [1], [Build with signal Userspace-RCU flavor])],
[bp],
[RCU_FLAVOR="liburcu-bp"
AC_DEFINE([RCU_BP], [1], [Build with bulletproof Userspace-RCU flavor])],
Improve the Userspace RCU integration This commit allows BIND 9 to be compiled with different flavours of Userspace RCU, and improves the integration between Userspace RCU and our event loop: - In the RCU QSBR, the thread is put offline when polling and online when rcu_dereference, rcu_assign_pointer (or friends) are called. - In other RCU modes, we check that we are not reading when reaching the quiescent callback in the event loop. - We register the thread before uv_work_run() callback is called and after it has finished. The rcu_(un)register_thread() has a large overhead, but that's fine in this case.
2023-03-30 17:04:22 +02:00
[AC_MSG_ERROR([Invalid Userspace-RCU flavor picked])])
Print out the Userspace-RCU flavor and version The `named -V` now prints out the liburcu flavor used and the compile-time version.
2023-04-24 13:32:34 +02:00
AC_MSG_RESULT([$RCU_FLAVOR])
Replace glue_cache hashtable with direct link in rdatasetheader Instead of having a global hashtable with a global rwlock for the GLUE cache, move the glue_list directly into rdatasetheader and use Userspace-RCU to update the pointer when the glue_list is empty. Additionally, the cached glue_lists needs to be stored in the RBTDB version for early cleaning, otherwise the circular dependencies between nodes and glue_lists will prevent nodes to be ever cleaned up.
2023-05-03 19:38:05 +02:00
#
# liburcu << v0.13.0 didn't add -lurcu-common and some toolchains would
# not add it automatically - we need to add it explicitly in such case.
#
PKG_CHECK_MODULES([LIBURCU], [$RCU_FLAVOR >= 0.13.0 liburcu-cds >= 0.13.0], [],
[PKG_CHECK_MODULES([LIBURCU], [$RCU_FLAVOR liburcu-cds],
[LIBURCU_LIBS="$LIBURCU_LIBS -lurcu-common"])])
Print out the Userspace-RCU flavor and version The `named -V` now prints out the liburcu flavor used and the compile-time version.
2023-04-24 13:32:34 +02:00
AC_DEFINE_UNQUOTED([RCU_FLAVOR], ["$RCU_FLAVOR"], [Chosen Userspace-RCU flavor])
PKG_CHECK_VERSION([RCU_VERSION], [$RCU_FLAVOR])
AC_DEFINE_UNQUOTED([RCU_VERSION], ["$RCU_VERSION"], [Compile-time Userspace-RCU version])
Add Userspace-RCU to global CFLAGS and LIBS The Userspace-RCU headers are now needed for more parts of the libisc and libdns, thus we need to add it globally to prevent compilation failures on systems with non-standard Userspace-RCU installation path.
2023-05-09 13:38:37 +02:00
CFLAGS="$CFLAGS $LIBURCU_CFLAGS"
LIBS="$LIBS $LIBURCU_LIBS"
Add pairwise testing Pairwise testing is a test case generation technique based on the observation that most faults are caused by interactions of at most two factors. For BIND, its configure options can be thought of as such factors. Process BIND configure options into a model that is subsequently processed by the PICT tool in order to find an effective test vector. That test vector is then used for configuring and building BIND using various combinations of configure options.
2020-07-01 10:29:36 +02:00
# Fuzzing is not included in pairwise testing as fuzzing tools are
# not present in the relevant Docker image.
#
# [pairwise: skip]
oss-fuzz: Improve interaction with fuzzing tools and add new tests in fuzz/ directory
2018-07-03 15:45:11 +02:00
AC_ARG_ENABLE([fuzzing],
Add explicit support of OSS-Fuzz The --enable-fuzzing option now allows third choice "ossfuzz" that just adds $LIB_FUZZING_ENGINE to FUZZ_LDFLAGS to make the fuzzer builds compatible with OSS-Fuzz project that has some special quirks (the main() routine is provided in the static library the project provides).
2020-08-12 21:06:26 +02:00
[AS_HELP_STRING([--enable-fuzzing=<afl|libfuzzer|ossfuzz>],
oss-fuzz: Improve interaction with fuzzing tools and add new tests in fuzz/ directory
2018-07-03 15:45:11 +02:00
[Enable fuzzing using American Fuzzy Lop or libFuzzer (default=no)])],
[],
[enable_fuzzing=no])
AC_MSG_CHECKING([whether to enable fuzzing mode])
AS_CASE([$enable_fuzzing],
[no],[AC_MSG_RESULT([no])],
[afl],[
AC_MSG_RESULT([using AFL])
AC_DEFINE([ENABLE_AFL], [1],
[Define to enable American Fuzzy Lop test harness])
Re-enable the fuzzing tests The fuzzing tests were temporarily disabled when the build system has been converted to automake. This commit restores the functionality to run the fuzzing tests as part of the `make check`. When the afl or libfuzzer is enabled via ./configure, it uses a custom LOG_DRIVER (fuzz/<fuzzer.sh>). Currently only libfuzzer.sh has been implemented that runs each fuzz test for 5 seconds each.
2020-07-31 15:20:56 +02:00
STD_CFLAGS="$STD_CFLAGS -DFUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION=1"
FUZZ_LOG_COMPILER="afl.sh"],
oss-fuzz: Improve interaction with fuzzing tools and add new tests in fuzz/ directory
2018-07-03 15:45:11 +02:00
[libfuzzer],[
AC_MSG_RESULT([using libFuzzer])
Re-enable the fuzzing tests The fuzzing tests were temporarily disabled when the build system has been converted to automake. This commit restores the functionality to run the fuzzing tests as part of the `make check`. When the afl or libfuzzer is enabled via ./configure, it uses a custom LOG_DRIVER (fuzz/<fuzzer.sh>). Currently only libfuzzer.sh has been implemented that runs each fuzz test for 5 seconds each.
2020-07-31 15:20:56 +02:00
STD_CFLAGS="$STD_CFLAGS -DFUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION=1 -fsanitize=fuzzer-no-link,address,undefined"
LDFLAGS="$LDFLAGS -fsanitize=address,undefined"
FUZZ_LDFLAGS="-fsanitize=fuzzer,address,undefined"
FUZZ_LOG_COMPILER="libfuzzer.sh"],
Add explicit support of OSS-Fuzz The --enable-fuzzing option now allows third choice "ossfuzz" that just adds $LIB_FUZZING_ENGINE to FUZZ_LDFLAGS to make the fuzzer builds compatible with OSS-Fuzz project that has some special quirks (the main() routine is provided in the static library the project provides).
2020-08-12 21:06:26 +02:00
[ossfuzz],[
AC_MSG_RESULT([using OSS-Fuzz])
FUZZ_LDFLAGS="$LIB_FUZZING_ENGINE"],
oss-fuzz: Improve interaction with fuzzing tools and add new tests in fuzz/ directory
2018-07-03 15:45:11 +02:00
[*],[AC_MSG_ERROR([You need to explicitly select the fuzzer])])
Re-enable the fuzzing tests The fuzzing tests were temporarily disabled when the build system has been converted to automake. This commit restores the functionality to run the fuzzing tests as part of the `make check`. When the afl or libfuzzer is enabled via ./configure, it uses a custom LOG_DRIVER (fuzz/<fuzzer.sh>). Currently only libfuzzer.sh has been implemented that runs each fuzz test for 5 seconds each.
2020-07-31 15:20:56 +02:00
AM_CONDITIONAL([HAVE_FUZZ_LOG_COMPILER], [test -n "$FUZZ_LOG_COMPILER"])
AC_SUBST([FUZZ_LOG_COMPILER])
AC_SUBST([FUZZ_LDFLAGS])
oss-fuzz: Improve interaction with fuzzing tools and add new tests in fuzz/ directory
2018-07-03 15:45:11 +02:00
AS_IF([test "$enable_fuzzing" = "afl"],
[AC_MSG_CHECKING("for AFL enabled compiler")
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([],
[#ifndef __AFL_COMPILER
#error AFL compiler required
#endif
])],
[AC_MSG_RESULT([yes])],
[AC_MSG_ERROR([set CC=afl-<gcc|clang> when --enable-fuzzing=afl is used])])
])
4358. [test] Added American Fuzzy Lop harness that allows feeding fuzzed packets into BIND. [RT #41723]
2016-05-05 11:46:11 +02:00
search for Perl at configure time, for use by the system test scripts
2000-06-13 17:34:28 +00:00
#
# Perl is optional; it is used only by some of the system test scripts.
#
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AC_PATH_PROGS([PERL], [perl5 perl])
AC_SUBST([PERL])
AM_CONDITIONAL([HAVE_PERL], [test -n "$PERL"])
search for Perl at configure time, for use by the system test scripts
2000-06-13 17:34:28 +00:00
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AX_PERL_MODULE([Digest::HMAC])
AM_CONDITIONAL([HAVE_PERLMOD_DIGEST_HMAC],
[test "$HAVE_PERLMOD_DIGEST__HMAC" = "yes"])
When --with-python is given to configure, use default search list for python binaries instead of 'yes'
2018-12-18 15:05:52 +01:00
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AX_PERL_MODULE([File::Fetch])
AM_CONDITIONAL([HAVE_PERLMOD_FILE_FETCH],
[test "$HAVE_PERLMOD_FILE__FETCH" = "yes"])
add AC_ARG_VAR([PYTHON], [path to python executable])
2019-12-03 08:38:50 +11:00
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AX_PERL_MODULE([Net::DNS])
AM_CONDITIONAL([HAVE_PERLMOD_NET_DNS],
[test "$HAVE_PERLMOD_NET__DNS" = "yes"])
Fix build-time Python configuration Apply various fixes and tweaks to Python configuration logic implemented in the "configure" script: - Prevent PYTHON_INSTALL_DIR, which holds the value passed to the --with-python-install-dir option, from being set to "unspec" by default as this breaks installing Python modules when the --with-python-install-dir option is not used. - Make the --with-python-install-dir option also work when the Python interpreter is specified explicitly (using --with-python=<...>). - Remove dnspython dependency which was erroneously introduced in commit 31b0dc1f204d8f7520145f21e8ea46d1466412a7: no installed Python module depends on dnspython, it is only used in system tests, for which dedicated scripts exist that check whether dnspython is available and act accordingly. - Improve contents and placement of error messages. - Reduce duplication of code checking Python dependencies. - Use Autoconf macros AS_CASE() and AS_IF() instead of plain shell code. - Update comments. Capitalize the word "Python" when referring to the language itself rather than a specific executable.
2018-11-27 11:43:46 +01:00
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AX_PERL_MODULE([Net::DNS::Nameserver])
AM_CONDITIONAL([HAVE_PERLMOD_NET_DNS_NAMESERVER],
[test "$HAVE_PERLMOD_NET__DNS__NAMESERVER" = "yes"])
When --with-python is given to configure, use default search list for python binaries instead of 'yes'
2018-12-18 15:05:52 +01:00
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AX_PERL_MODULE([Time::HiRes])
AM_CONDITIONAL([HAVE_PERLMOD_TIME_HIRES],
[test "$HAVE_PERLMOD_TIME__HIRES" = "yes"])
Fix build-time Python configuration Apply various fixes and tweaks to Python configuration logic implemented in the "configure" script: - Prevent PYTHON_INSTALL_DIR, which holds the value passed to the --with-python-install-dir option, from being set to "unspec" by default as this breaks installing Python modules when the --with-python-install-dir option is not used. - Make the --with-python-install-dir option also work when the Python interpreter is specified explicitly (using --with-python=<...>). - Remove dnspython dependency which was erroneously introduced in commit 31b0dc1f204d8f7520145f21e8ea46d1466412a7: no installed Python module depends on dnspython, it is only used in system tests, for which dedicated scripts exist that check whether dnspython is available and act accordingly. - Improve contents and placement of error messages. - Reduce duplication of code checking Python dependencies. - Use Autoconf macros AS_CASE() and AS_IF() instead of plain shell code. - Update comments. Capitalize the word "Python" when referring to the language itself rather than a specific executable.
2018-11-27 11:43:46 +01:00
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
#
# Python is optional, it is used only by some of the system test scripts.
#
Require Python 3.6+ for running Python-based tests configure.ac currently requires Python 3.4 for running Python-based system tests. Meanwhile, there are some features in Python 3.6+ that we would like to use for making our Python code cleaner (e.g. f-strings). Update the minimum Python version required for running Python-based system tests to 3.6, noting that: - Python 3.4 has reached end-of-life on March 18th, 2019. - Python 3.5 has reached end-of-life on September 5th, 2020.
2022-04-22 11:25:27 +02:00
AM_PATH_PYTHON([3.6], [], [:])
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AM_CONDITIONAL([HAVE_PYTHON], [test "$PYTHON" != ":"])
Fix build-time Python configuration Apply various fixes and tweaks to Python configuration logic implemented in the "configure" script: - Prevent PYTHON_INSTALL_DIR, which holds the value passed to the --with-python-install-dir option, from being set to "unspec" by default as this breaks installing Python modules when the --with-python-install-dir option is not used. - Make the --with-python-install-dir option also work when the Python interpreter is specified explicitly (using --with-python=<...>). - Remove dnspython dependency which was erroneously introduced in commit 31b0dc1f204d8f7520145f21e8ea46d1466412a7: no installed Python module depends on dnspython, it is only used in system tests, for which dedicated scripts exist that check whether dnspython is available and act accordingly. - Improve contents and placement of error messages. - Reduce duplication of code checking Python dependencies. - Use Autoconf macros AS_CASE() and AS_IF() instead of plain shell code. - Update comments. Capitalize the word "Python" when referring to the language itself rather than a specific executable.
2018-11-27 11:43:46 +01:00
Add py.test to the list of tested pytest names OpenBSD changed the name of the pytest script from py.test-3 in OpenBSD 6.8 to py.test in OpenBSD 6.9. The py.test-3 name which was added in d5562a3e for the sake of OpenBSD and CentOS is still required for CentOS.
2021-05-13 17:24:57 +02:00
AC_PATH_PROGS([PYTEST], [pytest-3 py.test-3 pytest py.test pytest-pypy], [])
Add a pytest runner to run.sh The system tests currently uses patchwork of shell scripts which doesn't offer proper error handling. This commit introduced option to write new tests in pytest framework that also allows easier manipulation of DNS traffic (using dnspython), native XML and JSON manipulation and proper error reporting.
2020-03-27 13:23:24 +01:00
AS_IF([test -z "$PYTEST"],
[AC_MSG_WARN([pytest not found, some system tests will be skipped])])
AC_SUBST([PYTEST])
Added test for rpz-passthru logging channel This test ensures that all rpz-passthru activity will be redirected to a dedicated log file following ns1/named.conf configuration.
2020-04-29 15:55:38 -03:00
AM_CONDITIONAL([HAVE_PYTEST], [test -n "$PYTEST"])
Add a pytest runner to run.sh The system tests currently uses patchwork of shell scripts which doesn't offer proper error handling. This commit introduced option to write new tests in pytest framework that also allows easier manipulation of DNS traffic (using dnspython), native XML and JSON manipulation and proper error reporting.
2020-03-27 13:23:24 +01:00
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AX_PYTHON_MODULE([dns])
AM_CONDITIONAL([HAVE_PYMOD_DNS], [test "$HAVE_PYMOD_DNS" = "yes"])
[RT 17] Special handling of prefix/sysconfdir/localstatedir. If none are given, install to /usr/local but use /etc/named.conf and /var/run/named.pid. If prefix is given, use it in the other two if they are not also provided.
2000-02-10 01:14:06 +00:00
all "dnl" comments are now "#" comments so they appear in configure
2000-02-11 18:35:11 +00:00
#
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
# xsltproc is optional, it is used only by system test scripts.
all "dnl" comments are now "#" comments so they appear in configure
2000-02-11 18:35:11 +00:00
#
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AC_PATH_PROG([XSLTPROC], [xsltproc])
gcc's optimiser is broken at -02 for ultrasparc
2003-01-21 04:11:26 +00:00
Enforce usage -zrelax=transtls when GNU C Compiler is used on Solaris When GNU C Compiler is used on Solaris (11), the Thread Local Storage is completely broken. The behaviour doesn't manifest when GNU ld is used. Thus, we need to enforce usage of GNU ld when GNU C Compiler is the compiler of choice. For more background for this change, see https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90912
2019-06-19 11:21:36 +02:00
#
# Using Solaris linker with gcc on Solaris breaks Thread Local Storage
#
AS_CASE([$host],
[*-solaris*],[
AS_IF([test "$GCC" = "yes"],
[LDFLAGS="$LDFLAGS -zrelax=transtls"
AC_MSG_WARN([When using GNU C Compiler on Solaris, -zrelax=transtls linker flag is used to fix bug in Thread Local Storage])
])
])
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AC_CHECK_HEADERS([fcntl.h regex.h sys/time.h unistd.h sys/mman.h sys/sockio.h sys/select.h sys/param.h sys/sysctl.h net/if6.h sys/socket.h net/route.h linux/netlink.h linux/rtnetlink.h], [], [],
[$ac_includes_default
#ifdef HAVE_SYS_PARAM_H
# include <sys/param.h>
#endif
#ifdef HAVE_SYS_SOCKET_H
# include <sys/socket.h>
#endif
])
test the C compiler with AC_PROG_CC before using it for the endinanness test [RT #742]; it also seems to make sense to test for stuff like openssl after testing the C compiler and basic C language features, not before
2001-01-31 19:22:46 +00:00
Make the xoshiro128plusplus thread-safe
2018-05-29 14:10:32 +02:00
#
Require C11 Atomic Operations via <stdatomic.h> Make the C11 Atomic Operations mandatory and drop the Gcc __atomic builtin shims.
2023-01-10 13:06:13 +01:00
# Check for C11 thread local storage
Make the xoshiro128plusplus thread-safe
2018-05-29 14:10:32 +02:00
#
Require C11 thread_local keyword and <threads.h> header Change the autoconf check to require C11 <threads.h> header and thread_local keyword.
2023-01-10 12:48:03 +01:00
# NOTE: OpenBSD doesn't ship <threads.h>, so we can't require it (yet).
#
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AC_CHECK_HEADERS([threads.h])
Make the xoshiro128plusplus thread-safe
2018-05-29 14:10:32 +02:00
Require C11 Atomic Operations via <stdatomic.h> Make the C11 Atomic Operations mandatory and drop the Gcc __atomic builtin shims.
2023-01-10 13:06:13 +01:00
#
# C11 Atomic Operations
#
AC_CHECK_HEADERS([stdatomic.h],
[],
[AC_MSG_ERROR([C11 Atomic Operations required, update your toolchain to build BIND 9])])
AC_COMPILE_IFELSE(
[AC_LANG_PROGRAM(
[[#include <stdatomic.h>]],
[[atomic_int_fast32_t val = 0; atomic_fetch_add_explicit(&val, 1, memory_order_relaxed);]]
)],
[AC_MSG_CHECKING([whether -latomic is needed for 64-bit stdatomic.h functions])
AC_LINK_IFELSE(
[AC_LANG_PROGRAM(
[[#include <stdatomic.h>]],
[[atomic_int_fast64_t val = 0; atomic_fetch_add_explicit(&val, 1, memory_order_relaxed);]]
)],
[AC_MSG_RESULT([no])],
[LIBS="$LIBS -latomic"
AC_LINK_IFELSE(
[AC_LANG_PROGRAM(
[[#include <stdatomic.h>]],
[[atomic_int_fast64_t val = 0; atomic_fetch_add_explicit(&val, 1, memory_order_relaxed);]]
)],
[AC_MSG_RESULT([yes])],
[AC_MSG_FAILURE([libatomic needed, but linking with -latomic failed, please fix your toolchain.])])
])
],
[AC_MSG_FAILURE([stdatomic.h header found, but compilation failed, please fix your toolchain.])])
AC_CHECK_HEADERS([stdalign.h stdnoreturn.h],
[],
[AC_MSG_ERROR([C11 standard headers not found, update your toolchain.])])
test the C compiler with AC_PROG_CC before using it for the endinanness test [RT #742]; it also seems to make sense to test for stuff like openssl after testing the C compiler and basic C language features, not before
2001-01-31 19:22:46 +00:00
AC_C_CONST
AC_C_INLINE
2287. [bug] Use 'volatile' if the compiler supports it. [RT #17413]
2008-01-04 03:28:39 +00:00
AC_C_VOLATILE
check for systems that support 'inline' but not 'static inline' [RT #1212]
2001-05-03 19:54:44 +00:00
Check for yield support on ARM and print helpfull error message about -march
2018-10-23 09:27:02 +02:00
#
# Check for yield support on ARM processors
#
AS_CASE([$host],
Improve the spinloop pause / yield hint Unfortunately, C still lacks a standard function for pause (x86, sparc) or yeild (arm) instructions, for use in spin lock or CAS loops. BIND has its own based on vendor intrinsics or inline asm. Previously, it was buried in the `isc_rwlock` implementation. This commit renames `isc_rwlock_pause()` to `isc_pause()` and moves it into <isc/pause.h>. This commit also fixes the configure script so that it detects ARM yield support on systems that identify as `aarch*` instead of `arm*`. On 64-bit ARM systems we now use the ISB (instruction synchronization barrier) instruction in preference to yield. The ISB instruction pauses the CPU for longer, several nanoseconds, which is more like the x86 pause instruction. There are more details in a Rust pull request, which also refers to MySQL making the same change: https://github.com/rust-lang/rust/pull/84725
2021-03-24 17:52:56 +01:00
[arm*|aarch64*],
[AC_MSG_CHECKING([for ARM yield instruction support])
arm: just use the compiler's default yield support
2019-05-06 18:06:59 +10:00
AC_COMPILE_IFELSE(
[AC_LANG_PROGRAM([[]],
[[__asm__ __volatile__ ("yield")]])],
[AC_MSG_RESULT([yes])
AC_DEFINE([HAVE_ARM_YIELD], [1],
[define if the ARM yield instruction is available])],
[AC_MSG_RESULT([no])])])
Check for yield support on ARM and print helpfull error message about -march
2018-10-23 09:27:02 +02:00
configure.ac: autodetect 'pause' instruction presence on sparc The change fixes the following build failure on sparc T3 and older CPUs: ``` sparc-unknown-linux-gnu-gcc ... -O2 -mcpu=niagara2 ... -c rwlock.c {standard input}: Assembler messages: {standard input}:398: Error: Architecture mismatch on "pause ". {standard input}:398: (Requires v9e|v9v|v9m|m8; requested architecture is v9b.) make[1]: *** [Makefile:280: rwlock.o] Error 1 ``` `pause` insutruction exists only on `-mcpu=niagara4` (`T4`) and upper. The change adds `pause` configure-time autodetection and uses it if available. config.h.in got new `HAVE_SPARC_PAUSE` knob. Fallback is a fall-through no-op. Build-tested on: - sparc-unknown-linux-gnu-gcc (no `pause`, build succeeds) - sparc-unknown-linux-gnu-gcc -mcpu=niagara4 (`pause`, build succeeds) Reported-by: Rolf Eike Beer Bug: https://bugs.gentoo.org/691708 Signed-off-by: Sergei Trofimovich <slyfox@gentoo.org>
2019-08-08 08:33:10 +01:00
#
# Check for pause support on SPARC processors
#
AS_CASE([$host],
[sparc*],
[AC_MSG_CHECKING([for pause instruction support])
AC_COMPILE_IFELSE(
[AC_LANG_PROGRAM([[]],
[[__asm__ __volatile__ ("pause")]])],
[AC_MSG_RESULT([yes])
AC_DEFINE([HAVE_SPARC_PAUSE], [1],
[define if the SPARC pause instruction is available])],
[AC_MSG_RESULT([no])])])
Replace custom HAVE_SYSCTLBYNAME define with AC_CHECK_FUNCS call
2018-09-05 13:33:59 +02:00
AC_CHECK_FUNCS([sysctlbyname])
2629. [port] Check for seteuid()/setegid(), use setresuid()/ setresgid() if not present. [RT #19932]
2009-07-14 22:39:30 +00:00
#
Refactor the privilege dropping On Linux, the libcap is now mandatory. It makes things simpler for us. System without {set,get}res{uid,gid} now have compatibility shim using setreuid/setregid or seteuid/setegid to setup effective UID/GID, so the same code can be called all the time (including on Linux).
2022-10-06 14:22:20 +02:00
# Check for uid/gid setting variants
2629. [port] Check for seteuid()/setegid(), use setresuid()/ setresgid() if not present. [RT #19932]
2009-07-14 22:39:30 +00:00
#
Refactor the privilege dropping On Linux, the libcap is now mandatory. It makes things simpler for us. System without {set,get}res{uid,gid} now have compatibility shim using setreuid/setregid or seteuid/setegid to setup effective UID/GID, so the same code can be called all the time (including on Linux).
2022-10-06 14:22:20 +02:00
AC_CHECK_FUNCS([setresuid setreuid getresuid])
AC_CHECK_FUNCS([setresgid setregid getresgid])
2629. [port] Check for seteuid()/setegid(), use setresuid()/ setresgid() if not present. [RT #19932]
2009-07-14 22:39:30 +00:00
test the C compiler with AC_PROG_CC before using it for the endinanness test [RT #742]; it also seems to make sense to test for stuff like openssl after testing the C compiler and basic C language features, not before
2001-01-31 19:22:46 +00:00
AC_TYPE_SIZE_T
Check for C99 compatible compiler
2018-07-24 14:42:20 +02:00
AC_TYPE_SSIZE_T
AC_TYPE_UINTPTR_T
[master] add uname data to named -V 4308. [func] Added operating system details to "named -V" output. [RT #41452]
2016-01-30 11:03:10 -08:00
#
# check for uname library routine
#
Remove now dummy NEED_PTHREAD_INIT
2018-09-05 12:58:02 +02:00
AC_MSG_CHECKING([for uname])
AC_COMPILE_IFELSE(
[AC_LANG_PROGRAM(
[[#include <sys/utsname.h>
#include <stdio.h>
]],
[[
struct utsname uts;
uname(&uts);
printf("running on %s %s %s for %s\n",
uts.sysname, uts.release, uts.version, uts.machine);
]])],
[AC_MSG_RESULT(yes)
AC_DEFINE([HAVE_UNAME], [1], [define if uname is available])
],
[AC_MSG_RESULT(no)
AC_MSG_WARN([uname is not correctly supported])
])
[master] add uname data to named -V 4308. [func] Added operating system details to "named -V" output. [RT #41452]
2016-01-30 11:03:10 -08:00
noreturn RT #20257
2009-09-29 15:06:07 +00:00
#
# check for GCC noreturn attribute
#
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AX_GCC_FUNC_ATTRIBUTE([noreturn])
noreturn RT #20257
2009-09-29 15:06:07 +00:00
Add malloc attribute to memory allocation functions The malloc attribute allows compiler to do some optmizations on functions that behave like malloc/calloc, like assuming that the returned pointer do not alias other pointers.
2021-04-01 17:03:59 -03:00
#
# check for GCC malloc attribute
#
AX_GCC_FUNC_ATTRIBUTE([malloc])
AC_MSG_CHECKING([for extended malloc attributes])
AC_COMPILE_IFELSE(
[AC_LANG_PROGRAM(
[[
#include <stddef.h>
#include <stdlib.h>
__attribute__ ((malloc, malloc (free, 1))
void * xmalloc(size_t sz) { return malloc(sz); }
]],
[[
void *p = xmalloc(8);
free(p);
]])],
[AC_MSG_RESULT(yes)
AC_DEFINE([HAVE_MALLOC_EXT_ATTR], [1], [define if extended attributes for malloc are available])
],
[AC_MSG_RESULT(no)])
#
# check for GCC returns_nonnull attribute
#
AX_GCC_FUNC_ATTRIBUTE([returns_nonnull])
Add isc_histo for histogram statistics This is an adaptation of my `hg64` experiments for use in BIND. As well as renaming everything according to ISC style, I have written some more extensive tests that ensure the edge cases are correct and the fenceposts are in the right places. I have added utility functions for working with precision in terms of decimal significant figures as well as this code's native binary.
2023-03-08 09:55:42 +00:00
#
# how to link math functions?
#
AC_SEARCH_LIBS([sqrt],[m])
allow --with-openssl=path
2000-08-02 19:35:35 +00:00
#
[master] native PKCS#11 support 3705. [func] "configure --enable-native-pkcs11" enables BIND to use the PKCS#11 API for all cryptographic functions, so that it can drive a hardware service module directly without the need to use a modified OpenSSL as intermediary (so long as the HSM's vendor provides a complete-enough implementation of the PKCS#11 interface). This has been tested successfully with the Thales nShield HSM and with SoftHSMv2 from the OpenDNSSEC project. [RT #29031]
2014-01-14 15:40:56 -08:00
# GeoIP support?
allow --with-openssl=path
2000-08-02 19:35:35 +00:00
#
use configure --enable-geoip to turn on GeoIP2 - instead of --with-geoip2, use --enable-geoip to turn on the GeoIP feature - --with-maxminddb can be used to set the prefix for maxminddb
2019-06-27 21:19:30 -07:00
# Should be on by default if libmaxminddb exists.
#
Add pairwise testing Pairwise testing is a test case generation technique based on the observation that most faults are caused by interactions of at most two factors. For BIND, its configure options can be thought of as such factors. Process BIND configure options into a model that is subsequently processed by the PICT tool in order to find an effective test vector. That test vector is then used for configuring and building BIND using various combinations of configure options.
2020-07-01 10:29:36 +02:00
# [pairwise: --enable-geoip --with-maxminddb=auto, --enable-geoip --with-maxminddb=yes, --disable-geoip]
use configure --enable-geoip to turn on GeoIP2 - instead of --with-geoip2, use --enable-geoip to turn on the GeoIP feature - --with-maxminddb can be used to set the prefix for maxminddb
2019-06-27 21:19:30 -07:00
AC_ARG_ENABLE([geoip],
[AS_HELP_STRING([--disable-geoip],
[support GeoIP2 geolocation ACLs if available [default=yes]])],
[], [enable_geoip="yes"])
Add pairwise testing Pairwise testing is a test case generation technique based on the observation that most faults are caused by interactions of at most two factors. For BIND, its configure options can be thought of as such factors. Process BIND configure options into a model that is subsequently processed by the PICT tool in order to find an effective test vector. That test vector is then used for configuring and building BIND using various combinations of configure options.
2020-07-01 10:29:36 +02:00
# [pairwise: skip]
use configure --enable-geoip to turn on GeoIP2 - instead of --with-geoip2, use --enable-geoip to turn on the GeoIP feature - --with-maxminddb can be used to set the prefix for maxminddb
2019-06-27 21:19:30 -07:00
AC_ARG_WITH([maxminddb],
[AS_HELP_STRING([--with-maxminddb=PATH],
remove all support for legacy GeoIP
2019-06-27 21:08:20 -07:00
[Build with MaxMind GeoIP2 support (auto|yes|no|path) [default=auto]])],
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
[], [with_maxminddb="auto"])
add a search for GeoIP2 libraries in configure - "--with-geoip" is used to enable the legacy GeoIP library. - "--with-geoip2" is used to enable the new GeoIP2 library (libmaxminddb), and is on by default if the library is found. - using both "--with-geoip" and "--with-geoip2" at the same time is an error. - an attempt is made to determine the default GeoIP2 database path at compile time if pkg-config is able to report the module prefix. if this fails, it will be necessary to set the path in named.conf with geoip-directory - Makefiles have been updated, and a stub lib/dns/geoip2.c has been added for the eventual GeoIP2 search implementation.
2019-06-11 15:59:31 -07:00
use configure --enable-geoip to turn on GeoIP2 - instead of --with-geoip2, use --enable-geoip to turn on the GeoIP feature - --with-maxminddb can be used to set the prefix for maxminddb
2019-06-27 21:19:30 -07:00
AS_IF([test "$enable_geoip" = "yes"],
[AS_CASE([$with_maxminddb],
[no],[AC_MSG_ERROR([Use '--disable-geoip' to disable the GeoIP])],
[auto],[PKG_CHECK_MODULES([MAXMINDDB], [libmaxminddb],
[AC_DEFINE([HAVE_GEOIP2], [1], [Build with GeoIP2 support])
PKG_CHECK_VAR([MAXMINDDB_PREFIX], [libmaxminddb], [prefix], [], [AC_MSG_ERROR([libmaxminddb prefix not found in pkg-config; set MAXMINDDB_PREFIX in the environment])])
],[:])],
[yes],[PKG_CHECK_MODULES([MAXMINDDB], [libmaxminddb],
[AC_DEFINE([HAVE_GEOIP2], [1], [Build with GeoIP2 support])
PKG_CHECK_VAR([MAXMINDDB_PREFIX], [libmaxminddb], [prefix], [], [AC_MSG_ERROR([libmaxminddb prefix not found in pkg-config; set MAXMINDDB_PREFIX in the environment])])
])],
[ # default
AX_SAVE_FLAGS([maxminddb])
MAXMINDDB_CFLAGS="-I$with_maxminddb/include"
MAXMINDDB_LIBS should end with '/lib' not '/libs'
2019-11-05 12:54:35 +11:00
MAXMINDDB_LIBS="-L$with_maxminddb/lib"
use configure --enable-geoip to turn on GeoIP2 - instead of --with-geoip2, use --enable-geoip to turn on the GeoIP feature - --with-maxminddb can be used to set the prefix for maxminddb
2019-06-27 21:19:30 -07:00
CFLAGS="$CFLAGS $MAXMINDDB_CFLAGS"
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
LIBS="$LIBS $MAXMINDDB_LIBS"
use configure --enable-geoip to turn on GeoIP2 - instead of --with-geoip2, use --enable-geoip to turn on the GeoIP feature - --with-maxminddb can be used to set the prefix for maxminddb
2019-06-27 21:19:30 -07:00
AC_SEARCH_LIBS([MMDB_open], [maxminddb],
[AC_DEFINE([HAVE_GEOIP2], [1], [Build with GeoIP2 support])
Fix 'configure --with-maxminddb=<path>'
2021-01-06 18:05:58 +11:00
MAXMINDDB_LIBS="$MAXMINDDB_LIBS $ac_cv_search_MMDB_open"
use configure --enable-geoip to turn on GeoIP2 - instead of --with-geoip2, use --enable-geoip to turn on the GeoIP feature - --with-maxminddb can be used to set the prefix for maxminddb
2019-06-27 21:19:30 -07:00
AC_MSG_NOTICE([GeoIP2 default database path set to $with_maxminddb/share/GeoIP])
Remove extraneous '$' in AS_VAR_COPY([MAXMINDDB_PREFIX], [with_maxminddb])
2022-05-17 19:39:22 +10:00
AS_VAR_COPY([MAXMINDDB_PREFIX], [with_maxminddb])
use configure --enable-geoip to turn on GeoIP2 - instead of --with-geoip2, use --enable-geoip to turn on the GeoIP feature - --with-maxminddb can be used to set the prefix for maxminddb
2019-06-27 21:19:30 -07:00
],
[AC_MSG_ERROR([GeoIP2 requested, but libmaxminddb not found])])
Add missing AX_RESTORE_FLAGS([geoip2]) to configure.ac
2019-07-09 13:35:39 +02:00
AX_RESTORE_FLAGS([maxminddb])
use configure --enable-geoip to turn on GeoIP2 - instead of --with-geoip2, use --enable-geoip to turn on the GeoIP feature - --with-maxminddb can be used to set the prefix for maxminddb
2019-06-27 21:19:30 -07:00
])
AC_ARG_VAR([MAXMINDDB_PREFIX], [value of prefix for MAXMINDDB, overriding pkg-config])
])
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AM_CONDITIONAL([HAVE_GEOIP2], [test -n "$MAXMINDDB_LIBS"])
add a search for GeoIP2 libraries in configure - "--with-geoip" is used to enable the legacy GeoIP library. - "--with-geoip2" is used to enable the new GeoIP2 library (libmaxminddb), and is on by default if the library is found. - using both "--with-geoip" and "--with-geoip2" at the same time is an error. - an attempt is made to determine the default GeoIP2 database path at compile time if pkg-config is able to report the module prefix. if this fails, it will be necessary to set the path in named.conf with geoip-directory - Makefiles have been updated, and a stub lib/dns/geoip2.c has been added for the eventual GeoIP2 search implementation.
2019-06-11 15:59:31 -07:00
AC_SUBST([MAXMINDDB_CFLAGS])
AC_SUBST([MAXMINDDB_LIBS])
add EVP and PKCS11
2008-03-31 14:42:51 +00:00
Improve autoconf pthread detection
2018-08-15 11:12:03 +02:00
AX_PTHREAD
add EVP and PKCS11
2008-03-31 14:42:51 +00:00
Improve autoconf pthread detection
2018-08-15 11:12:03 +02:00
LIBS="$PTHREAD_LIBS $LIBS"
CFLAGS="$CFLAGS $PTHREAD_CFLAGS"
CC="$PTHREAD_CC"
[master] add geoip support 3504. [func] Add support for ACLs based on geographic location, using MaxMind GeoIP databases. Based on code contributed by Ken Brownfield <kb@slide.com>. [RT #30681]
2013-02-27 17:19:39 -08:00
Add isc_spinlock unit with shim pthread_spin implementation The spinlock is small (atomic_uint_fast32_t at most), lightweight synchronization primitive and should only be used for short-lived and most of the time a isc_mutex should be used. Add a isc_spinlock unit which is either (most of the time) a think wrapper around pthread_spin API or an efficient shim implementation of the simple spinlock.
2023-03-03 09:24:13 +01:00
AC_CHECK_FUNCS([pthread_attr_getstacksize pthread_attr_setstacksize pthread_barrier_init pthread_spin_init])
Remove unthreaded support
2018-06-06 11:39:50 +02:00
Add pairwise testing Pairwise testing is a test case generation technique based on the observation that most faults are caused by interactions of at most two factors. For BIND, its configure options can be thought of as such factors. Process BIND configure options into a model that is subsequently processed by the PICT tool in order to find an effective test vector. That test vector is then used for configuring and building BIND using various combinations of configure options.
2020-07-01 10:29:36 +02:00
# [pairwise: --with-locktype=adaptive, --with-locktype=standard]
Improve autoconf pthread detection
2018-08-15 11:12:03 +02:00
AC_ARG_WITH([locktype],
Remove unthreaded support
2018-06-06 11:39:50 +02:00
AS_HELP_STRING([--with-locktype=ARG],
[Specify mutex lock type
Improve autoconf pthread detection
2018-08-15 11:12:03 +02:00
(adaptive or standard)]),
[], [with_locktype="adaptive"])
AS_CASE([$with_locktype],
[adaptive],[
AC_MSG_CHECKING([for PTHREAD_MUTEX_ADAPTIVE_NP])
AC_COMPILE_IFELSE(
[AC_LANG_PROGRAM(
[[
#ifndef _GNU_SOURCE
#define _GNU_SOURCE
#endif
#include <pthread.h>
]],
[[
return (PTHREAD_MUTEX_ADAPTIVE_NP);
]]
)],
[AC_MSG_RESULT([using adaptive lock type])
AC_DEFINE([HAVE_PTHREAD_MUTEX_ADAPTIVE_NP], 1,
[Support for PTHREAD_MUTEX_ADAPTIVE_NP]) ],
[AC_MSG_RESULT([using standard lock type])]
)],
[standard],[AC_MSG_RESULT([using standard lock type])],
[AC_MSG_ERROR([You must specify "adaptive" or "standard" for --with-locktype.])]
)
AC_CHECK_HEADERS([sched.h])
AC_SEARCH_LIBS([sched_yield],[rt])
AC_CHECK_FUNCS([sched_yield pthread_yield pthread_yield_np])
Remove unthreaded support
2018-06-06 11:39:50 +02:00
# Look for functions relating to thread naming
Improve autoconf pthread detection
2018-08-15 11:12:03 +02:00
AC_CHECK_FUNCS([pthread_setname_np pthread_set_name_np])
Remove unthreaded support
2018-06-06 11:39:50 +02:00
AC_CHECK_HEADERS([pthread_np.h], [], [], [#include <pthread.h>])
netmgr: libuv-based network manager This is a replacement for the existing isc_socket and isc_socketmgr implementation. It uses libuv for asynchronous network communication; "networker" objects will be distributed across worker threads reading incoming packets and sending them for processing. UDP listener sockets automatically create an array of "child" sockets so each worker can listen separately. TCP sockets are shared amongst worker threads. A TCPDNS socket is a wrapper around a TCP socket, which handles the the two-byte length field at the beginning of DNS messages over TCP. (Other wrapper socket types can be implemented in the future to handle DNS over TLS, DNS over HTTPS, etc.)
2019-11-05 13:55:54 -08:00
# libuv
Avoid libuv 1.35 and 1.36 that have broken recvmmsg implementation The implementation of UDP recvmmsg in libuv 1.35 and 1.36 is incomplete and could cause assertion failure under certain circumstances. Modify the configure and runtime checks to report a fatal error when trying to compile or run with the affected versions.
2023-02-03 15:47:28 +01:00
PKG_CHECK_MODULES([LIBUV], [libuv >= 1.37.0], [],
[PKG_CHECK_MODULES([LIBUV], [libuv >= 1.34.0 libuv < 1.35.0], [],
[AC_MSG_ERROR([libuv >= 1.34.0 (except 1.35.0 and 1.36.0) not found])])])
Add uv_handle_{get,set}_data functions that's absent in pre-1.19 libuv to make code clearer. This might be removed when we stop supporting older libuv versions.
2019-12-02 11:19:55 +01:00
Use libuv CFLAGS and LIBS when checking for features When autoconf was checking for libuv features, the LIBUV_CFLAGS was not added to CFLAGS and LIBUV_LIBS to LIBS which resulted in false negatives. Use AX_SAVE_FLAGS and AX_RESTORE_FLAGS to temporarily add LIBUV_CFLAGS and LIBUV_LIBS to their respective variables.
2022-05-18 14:10:58 +02:00
AX_SAVE_FLAGS([libuv])
CFLAGS="$CFLAGS $LIBUV_CFLAGS"
LIBS="$LIBS $LIBUV_LIBS"
Fix the UDP recvmmsg support Previously, the netmgr/udp.c tried to detect the recvmmsg detection in libuv with #ifdef UV_UDP_<foo> preprocessor macros. However, because the UV_UDP_<foo> are not preprocessor macros, but enum members, the detection didn't work. Because the detection didn't work, the code didn't have access to the information when we received the final chunk of the recvmmsg and tried to free the uvbuf every time. Fortunately, the isc__nm_free_uvbuf() had a kludge that detected attempt to free in the middle of the receive buffer, so the code worked. However, libuv 1.37.0 changed the way the recvmmsg was enabled from implicit to explicit, and we checked for yet another enum member presence with preprocessor macro, so in fact libuv recvmmsg support was never enabled with libuv >= 1.37.0. This commit changes to the preprocessor macros to autoconf checks for declaration, so the detection now works again. On top of that, it's now possible to cleanup the alloc_cb and free_uvbuf functions because now, the information whether we can or cannot free the buffer is available to us.
2022-01-11 12:14:23 +01:00
# libuv recvmmsg support
Instead of detecting MUSL libc, detect padding in struct msghdr The detection of MUSL libc via autoconf $host turned out to be not reliable. Convert the autoconf check from $host detection to actually detect the padding used in the struct msghdr.
2022-01-13 22:43:20 +01:00
AC_CHECK_DECLS([UV_UDP_MMSG_FREE, UV_UDP_MMSG_CHUNK], [], [], [[#include <uv.h>]])
AC_MSG_CHECKING([whether struct msghdr uses padding for alignment])
AC_COMPILE_IFELSE([AC_LANG_PROGRAM([#include <sys/socket.h>],
[const struct msghdr h = { .__pad1 = 0, .__pad2 = 0 };])],
[AC_MSG_RESULT([yes])
AC_DEFINE([HAVE_DECL_UV_UDP_RECVMMSG],
[0], [Disable recvmmsg support on systems with MUSL glibc])],
[AC_MSG_RESULT([no])
AC_CHECK_DECLS([UV_UDP_RECVMMSG], [], [], [[#include <uv.h>]])])
Fix the UDP recvmmsg support Previously, the netmgr/udp.c tried to detect the recvmmsg detection in libuv with #ifdef UV_UDP_<foo> preprocessor macros. However, because the UV_UDP_<foo> are not preprocessor macros, but enum members, the detection didn't work. Because the detection didn't work, the code didn't have access to the information when we received the final chunk of the recvmmsg and tried to free the uvbuf every time. Fortunately, the isc__nm_free_uvbuf() had a kludge that detected attempt to free in the middle of the receive buffer, so the code worked. However, libuv 1.37.0 changed the way the recvmmsg was enabled from implicit to explicit, and we checked for yet another enum member presence with preprocessor macro, so in fact libuv recvmmsg support was never enabled with libuv >= 1.37.0. This commit changes to the preprocessor macros to autoconf checks for declaration, so the detection now works again. On top of that, it's now possible to cleanup the alloc_cb and free_uvbuf functions because now, the information whether we can or cannot free the buffer is available to us.
2022-01-11 12:14:23 +01:00
Set IP(V6)_RECVERR on connect UDP sockets (via libuv) The connect()ed UDP socket provides feedback on a variety of ICMP errors (eg port unreachable) which bind can then use to decide what to do with errors (report them to the client, try again with a different nameserver etc). However, Linux's implementation does not report what it considers "transient" conditions, which is defined as Destination host Unreachable, Destination network unreachable, Source Route Failed and Message Too Big. Explicitly enable IP_RECVERR / IPV6_RECVERR (via libuv uv_udp_bind() flag) to learn about ICMP destination network/host unreachable.
2022-04-25 14:09:44 +02:00
# libuv recverr support
AC_CHECK_DECLS([UV_UDP_LINUX_RECVERR], [], [], [[#include <uv.h>]])
Use libuv CFLAGS and LIBS when checking for features When autoconf was checking for libuv features, the LIBUV_CFLAGS was not added to CFLAGS and LIBUV_LIBS to LIBS which resulted in false negatives. Use AX_SAVE_FLAGS and AX_RESTORE_FLAGS to temporarily add LIBUV_CFLAGS and LIBUV_LIBS to their respective variables.
2022-05-18 14:10:58 +02:00
AX_RESTORE_FLAGS([libuv])
Make the DNS over HTTPS support optional This commit adds two new autoconf options `--enable-doh` (enabled by default) and `--with-libnghttp2` (mandatory when DoH is enabled). When DoH support is disabled the library is not linked-in and support for http(s) protocol is disabled in the netmgr, named and dig.
2021-04-21 13:52:15 +02:00
# [pairwise: --enable-doh --with-libnghttp2=auto, --enable-doh --with-libnghttp2=yes, --disable-doh]
AC_ARG_ENABLE([doh],
Fix configure --disable-doh help text This is technically fixup for 2bb454182bb499990e191e098a8a33082c672a94. Related: !4926
2022-11-24 10:44:53 +01:00
[AS_HELP_STRING([--disable-doh], [disable DNS over HTTPS, removes dependency on libnghttp2 (default is --enable-doh)])],
Make the DNS over HTTPS support optional This commit adds two new autoconf options `--enable-doh` (enabled by default) and `--with-libnghttp2` (mandatory when DoH is enabled). When DoH support is disabled the library is not linked-in and support for http(s) protocol is disabled in the netmgr, named and dig.
2021-04-21 13:52:15 +02:00
[], [enable_doh=yes])
# [pairwise: skip]
AC_ARG_WITH([libnghttp2],
[AS_HELP_STRING([--with-libnghttp2],
[build with libnghttp2 library [yes|no|auto] (default is auto)])],
[], [with_libnghttp2="auto"])
AS_IF([test "$enable_doh" = "yes"],
[AS_CASE([$with_libnghttp2],
Avoid use of the DoH acronym in ./configure output Replace all uses of the DoH acronym in ./configure output with "DNS-over-HTTPS support" in order to improve error message readability.
2022-01-20 15:40:37 +01:00
[no],[AC_MSG_ERROR([Use '--disable-doh' to disable DNS-over-HTTPS support])],
Make the DNS over HTTPS support optional This commit adds two new autoconf options `--enable-doh` (enabled by default) and `--with-libnghttp2` (mandatory when DoH is enabled). When DoH support is disabled the library is not linked-in and support for http(s) protocol is disabled in the netmgr, named and dig.
2021-04-21 13:52:15 +02:00
[auto|yes],[PKG_CHECK_MODULES([LIBNGHTTP2], [libnghttp2 >= 1.6.0],
Avoid use of the DoH acronym in ./configure output Replace all uses of the DoH acronym in ./configure output with "DNS-over-HTTPS support" in order to improve error message readability.
2022-01-20 15:40:37 +01:00
[AC_DEFINE([HAVE_LIBNGHTTP2], [1], [Build with DNS-over-HTTPS support])],
Suggest --disable-doh when libnghttp2 is not found Extend the error message displayed when support for DNS over HTTPS is requested but libnghttp2 is unavailable at build time, in order to help the user find a way out of such a situation.
2022-01-20 15:40:37 +01:00
[AC_MSG_ERROR(m4_normalize([DNS-over-HTTPS support requested, but libnghttp2 not found.
Either install libnghttp2 or use --disable-doh.]))])],
Make the DNS over HTTPS support optional This commit adds two new autoconf options `--enable-doh` (enabled by default) and `--with-libnghttp2` (mandatory when DoH is enabled). When DoH support is disabled the library is not linked-in and support for http(s) protocol is disabled in the netmgr, named and dig.
2021-04-21 13:52:15 +02:00
[AC_MSG_ERROR([Specifying libnghttp2 installation path is not supported, adjust PKG_CONFIG_PATH instead])])])
AM_CONDITIONAL([HAVE_LIBNGHTTP2], [test -n "$LIBNGHTTP2_LIBS"])
Update BIND version to 9.19.9-dev
2022-12-13 16:44:52 +01:00
nghttp2-based HTTP layer in netmgr This commit includes work-in-progress implementation of DNS-over-HTTP(S). Server-side code remains mostly untested, and there is only support for POST requests.
2020-10-31 20:42:18 +01:00
Restore missing check for flockfile and getc_unlocked
2019-03-07 13:32:30 +01:00
#
# flockfile is usually provided by pthreads
#
AC_CHECK_FUNCS([flockfile getc_unlocked])
Remove unthreaded support
2018-06-06 11:39:50 +02:00
#
# Look for sysconf to allow detection of the number of processors.
#
Replace custom HAVE_SYSCONF with proper AC_CHECK_FUNCS call
2018-09-05 13:29:28 +02:00
AC_CHECK_FUNCS([sysconf])
2105. [func] GSS-TSIG support (RFC 3645).
2006-12-04 01:54:53 +00:00
Add --enable-pthread-rwlock option
2019-01-25 12:29:52 +01:00
#
Use pthread rwlocks by default
2020-02-27 13:21:31 +01:00
# Do we want to use pthread rwlock?
Add --enable-pthread-rwlock option
2019-01-25 12:29:52 +01:00
#
Add pairwise testing Pairwise testing is a test case generation technique based on the observation that most faults are caused by interactions of at most two factors. For BIND, its configure options can be thought of as such factors. Process BIND configure options into a model that is subsequently processed by the PICT tool in order to find an effective test vector. That test vector is then used for configuring and building BIND using various combinations of configure options.
2020-07-01 10:29:36 +02:00
# [pairwise: --enable-pthread-rwlock, --disable-pthread-rwlock]
Add --enable-pthread-rwlock option
2019-01-25 12:29:52 +01:00
AC_ARG_ENABLE([pthread_rwlock],
Make ISC rwlock implementation the default again Revert the change from ad03c22e976411cad743bc02746b803a2f119df7 as further testing has shown that with hyper-threading disabled, named with ISC rwlocks outperforms named with pthread rwlocks in cold cache testing scenarios. Since building named with pthread rwlocks might still be a better choice for some workloads, keep the compile-time option which enables that.
2020-04-17 08:36:24 +02:00
[AS_HELP_STRING([--enable-pthread-rwlock],
[use pthread rwlock instead of internal rwlock implementation])],
[], [enable_pthread_rwlock=no])
Add --enable-pthread-rwlock option
2019-01-25 12:29:52 +01:00
AS_IF([test "$enable_pthread_rwlock" = "yes"],
[AC_CHECK_FUNCS([pthread_rwlock_rdlock], [],
[AC_MSG_ERROR([pthread_rwlock_rdlock requested but not found])])
AC_DEFINE([USE_PTHREAD_RWLOCK],[1],[Define if you want to use pthread rwlock implementation])
])
Make the pthread_rwlock implementation header-only macros [2/2] While using mutrace, the phtread-rwlock based isc_rwlock implementation would be all tracked in the rwlock.c unit losing all useful information as all rwlocks would be traced in a single place. Rewrite the pthread_rwlock based implementation to be header-only macros, so we can use mutrace to properly track the rwlock contention without heavily patching mutrace to understand the libisc synchronization primitives.
2022-10-14 12:54:57 +02:00
AM_CONDITIONAL([USE_ISC_RWLOCK], [test "$enable_pthread_rwlock" != "yes"])
make --without-pic imply --disable-shared and --without-dlopen
2018-08-24 14:17:34 -07:00
Make OpenSSL mandatory
2018-06-12 11:26:04 +02:00
CRYPTO=OpenSSL
Added Ed25519 support (#44696)
2017-07-31 15:26:00 +02:00
Make OpenSSL mandatory
2018-06-12 11:26:04 +02:00
#
Rename OPENSSL_INCLUDES to OPENSSL_CFLAGS in AX_CHECK_OPENSSL() macro The ax_check_openssl m4 macro used OPENSSL_INCLUDES. Rename the subst variable to OPENSSL_CFLAGS and wrap AX_CHECK_OPENSSL() in action-if-not-found part of PKG_CHECK_MODULE check for libcrypto.
2019-06-24 14:06:56 +02:00
# OpenSSL/LibreSSL is mandatory
Make OpenSSL mandatory
2018-06-12 11:26:04 +02:00
#
autoconf: link with libssl This is prerequisite to adding DoT and DoH support to BIND 9.
2020-05-13 17:37:51 +02:00
PKG_CHECK_MODULES([OPENSSL], [libssl libcrypto], [],
Rename OPENSSL_INCLUDES to OPENSSL_CFLAGS in AX_CHECK_OPENSSL() macro The ax_check_openssl m4 macro used OPENSSL_INCLUDES. Rename the subst variable to OPENSSL_CFLAGS and wrap AX_CHECK_OPENSSL() in action-if-not-found part of PKG_CHECK_MODULE check for libcrypto.
2019-06-24 14:06:56 +02:00
[AX_CHECK_OPENSSL([:],[AC_MSG_FAILURE([OpenSSL/LibreSSL not found])])])
Make OpenSSL mandatory
2018-06-12 11:26:04 +02:00
Use couple more autoconf convenience macros to cleanup configure.ac Pull and use several autoconf archive convenience macros to simplify configure.ac. * AX_CHECK_COMPILE_FLAG(FLAG, ...) - check whether given CFLAG works * AX_CHECK_LINK_FLAG(FLAG, ...) - check whether given LDFLAG works * AX_CHECK_PREPROC_FLAG(FLAG, ...) - check whether give CPPFLAG works * AX_SAVE_FLAGS/AX_RESTORE_FLAGS - save and restore *FLAGS
2019-06-19 11:16:47 +02:00
AX_SAVE_FLAGS([openssl])
Probe for EVP_default_properties_enable_fips FIPS_mode is not available in OpenSSL 3.0.0
2021-12-16 19:20:40 +11:00
CFLAGS="$OPENSSL_CFLAGS $CFLAGS"
LIBS="$OPENSSL_LIBS $LIBS"
Make OpenSSL mandatory
2018-06-12 11:26:04 +02:00
Refactor TLSDNS module to work with libuv/ssl directly * Following the example set in 634bdfb16d8, the tlsdns netmgr module now uses libuv and SSL primitives directly, rather than opening a TLS socket which opens a TCP socket, as the previous model was difficult to debug. Closes #2335. * Remove the netmgr tls layer (we will have to re-add it for DoH) * Add isc_tls API to wrap the OpenSSL SSL_CTX object into libisc library; move the OpenSSL initialization/deinitialization from dstapi needed for OpenSSL 1.0.x to the isc_tls_{initialize,destroy}() * Add couple of new shims needed for OpenSSL 1.0.x * When LibreSSL is used, require at least version 2.7.0 that has the best OpenSSL 1.1.x compatibility and auto init/deinit * Enforce OpenSSL 1.1.x usage on Windows * Added a TLSDNS unit test and implemented a simple TLSDNS echo server and client.
2020-12-17 11:40:29 +01:00
AC_MSG_CHECKING([for OpenSSL >= 1.0.0 or LibreSSL >= 2.7.0])
Make OpenSSL mandatory
2018-06-12 11:26:04 +02:00
AC_COMPILE_IFELSE(
[AC_LANG_PROGRAM([[#include <openssl/opensslv.h>]],
Refactor TLSDNS module to work with libuv/ssl directly * Following the example set in 634bdfb16d8, the tlsdns netmgr module now uses libuv and SSL primitives directly, rather than opening a TLS socket which opens a TCP socket, as the previous model was difficult to debug. Closes #2335. * Remove the netmgr tls layer (we will have to re-add it for DoH) * Add isc_tls API to wrap the OpenSSL SSL_CTX object into libisc library; move the OpenSSL initialization/deinitialization from dstapi needed for OpenSSL 1.0.x to the isc_tls_{initialize,destroy}() * Add couple of new shims needed for OpenSSL 1.0.x * When LibreSSL is used, require at least version 2.7.0 that has the best OpenSSL 1.1.x compatibility and auto init/deinit * Enforce OpenSSL 1.1.x usage on Windows * Added a TLSDNS unit test and implemented a simple TLSDNS echo server and client.
2020-12-17 11:40:29 +01:00
[[#if (!defined(LIBRESSL_VERSION_NUMBER) && (OPENSSL_VERSION_NUMBER < 0x1000000fL)) || \\
(defined(LIBRESSL_VERSION_NUMBER) && (LIBRESSL_VERSION_NUMBER < 0x02070000fL))
#error OpenSSL >= 1.0.0 or LibreSSL >= 2.7.0 required
Make OpenSSL mandatory
2018-06-12 11:26:04 +02:00
#endif
]])],
[AC_MSG_RESULT([yes])],
[AC_MSG_FAILURE([not found])])
Check for individual OpenSSL functions instead of relying on version number
2018-10-26 05:29:56 +02:00
#
# Check for functions added in OpenSSL or LibreSSL
#
Check for SSL_CTX_set_keylog_callback() support The SSL_CTX_set_keylog_callback() function is a fairly recent OpenSSL addition, having first appeared in version 1.1.1. Add a configure.ac check for the availability of that function to prevent build errors on older platforms. Sort similar checks alphabetically. This makes the SSLKEYLOGFILE mechanism a silent no-op on unsupported platforms, which is considered acceptable for a debugging feature.
2021-12-22 18:17:26 +01:00
AC_CHECK_FUNCS([BIO_read_ex BIO_write_ex])
Use autoconf check for BN_GENCB_new() BIND unconditionally uses shims for BN_GENCB_new(), BN_GENCB_free(), and BN_GENCB_get_arg() for all LibreSSL versions and, correctly, for OpenSSL <1.1.0 versions. This breaks LibreSSL compilation starting with LibreSSL 3.5.0. Use autoconf check instead to check whether the family of the functions are available.
2022-02-25 09:00:52 +00:00
AC_CHECK_FUNCS([BN_GENCB_new])
Check for individual OpenSSL functions instead of relying on version number
2018-10-26 05:29:56 +02:00
AC_CHECK_FUNCS([CRYPTO_zalloc])
Check for SSL_CTX_set_keylog_callback() support The SSL_CTX_set_keylog_callback() function is a fairly recent OpenSSL addition, having first appeared in version 1.1.1. Add a configure.ac check for the availability of that function to prevent build errors on older platforms. Sort similar checks alphabetically. This makes the SSLKEYLOGFILE mechanism a silent no-op on unsupported platforms, which is considered acceptable for a debugging feature.
2021-12-22 18:17:26 +01:00
AC_CHECK_FUNCS([ERR_get_error_all])
Check for individual OpenSSL functions instead of relying on version number
2018-10-26 05:29:56 +02:00
AC_CHECK_FUNCS([EVP_CIPHER_CTX_new EVP_CIPHER_CTX_free])
Use EVP_MD_CTX_get0_md() instead of deprecated EVP_MD_CTX_md() OpenSSL 3.0.0 deprecates the EVP_MD_CTX_md() function. Use EVP_MD_CTX_md() instead of EVP_MD_CTX_get0_md() and create a shim to use the old variant for the older OpenSSL versions which don't have the newer EVP_MD_CTX_get0_md().
2021-09-01 18:50:46 +00:00
AC_CHECK_FUNCS([EVP_MD_CTX_new EVP_MD_CTX_free EVP_MD_CTX_reset EVP_MD_CTX_get0_md])
Check for SSL_CTX_set_keylog_callback() support The SSL_CTX_set_keylog_callback() function is a fairly recent OpenSSL addition, having first appeared in version 1.1.1. Add a configure.ac check for the availability of that function to prevent build errors on older platforms. Sort similar checks alphabetically. This makes the SSLKEYLOGFILE mechanism a silent no-op on unsupported platforms, which is considered acceptable for a debugging feature.
2021-12-22 18:17:26 +01:00
AC_CHECK_FUNCS([EVP_PKEY_new_raw_private_key EVP_PKEY_eq])
AC_CHECK_FUNCS([OPENSSL_init_ssl OPENSSL_init_crypto OPENSSL_cleanup])
AC_CHECK_FUNCS([SSL_CTX_set_keylog_callback])
Refactor TLSDNS module to work with libuv/ssl directly * Following the example set in 634bdfb16d8, the tlsdns netmgr module now uses libuv and SSL primitives directly, rather than opening a TLS socket which opens a TCP socket, as the previous model was difficult to debug. Closes #2335. * Remove the netmgr tls layer (we will have to re-add it for DoH) * Add isc_tls API to wrap the OpenSSL SSL_CTX object into libisc library; move the OpenSSL initialization/deinitialization from dstapi needed for OpenSSL 1.0.x to the isc_tls_{initialize,destroy}() * Add couple of new shims needed for OpenSSL 1.0.x * When LibreSSL is used, require at least version 2.7.0 that has the best OpenSSL 1.1.x compatibility and auto init/deinit * Enforce OpenSSL 1.1.x usage on Windows * Added a TLSDNS unit test and implemented a simple TLSDNS echo server and client.
2020-12-17 11:40:29 +01:00
AC_CHECK_FUNCS([SSL_CTX_set_min_proto_version])
Check for SSL_CTX_set_keylog_callback() support The SSL_CTX_set_keylog_callback() function is a fairly recent OpenSSL addition, having first appeared in version 1.1.1. Add a configure.ac check for the availability of that function to prevent build errors on older platforms. Sort similar checks alphabetically. This makes the SSLKEYLOGFILE mechanism a silent no-op on unsupported platforms, which is considered acceptable for a debugging feature.
2021-12-22 18:17:26 +01:00
AC_CHECK_FUNCS([SSL_CTX_up_ref])
AC_CHECK_FUNCS([SSL_read_ex SSL_peek_ex SSL_write_ex])
Implement shim for SSL_CTX_set1_cert_store() (affects Debian 9) This commit implements a shim for SSL_CTX_set1_cert_store() for OpenSSL/LibreSSL versions where it is not available.
2022-04-01 11:16:44 +03:00
AC_CHECK_FUNCS([SSL_CTX_set1_cert_store X509_STORE_up_ref])
Add isc_tlsctx_attach() The implementation is done on top of the reference counting functionality found in OpenSSL/LibreSSL, which allows for avoiding wrapping the object. Adding this function allows using reference counting for TLS contexts in BIND 9's codebase.
2022-02-22 17:07:57 +02:00
AC_CHECK_FUNCS([SSL_CTX_up_ref])
Add SSL_SESSION_is_resumable() implementation shim This commit adds SSL_SESSION_is_resumable() implementation if it is missing.
2022-05-09 19:08:29 +03:00
AC_CHECK_FUNCS([SSL_SESSION_is_resumable])
Check for individual OpenSSL functions instead of relying on version number
2018-10-26 05:29:56 +02:00
Make OpenSSL mandatory
2018-06-12 11:26:04 +02:00
#
# Check for algorithm support in OpenSSL
#
Make the configure.ac script compatible with OpenSSL 3.0.0 OpenSSL 3.0.0 deprecates many low level API functions. In preparation for the future support of linking BIND with OpenSSL 3.0.0 without the deprecated API functions, change the configure.ac script to use functions which are available on all supported versions of OpenSSL and LibreSSL.
2021-09-01 13:13:24 +00:00
AC_CHECK_FUNCS([EVP_DigestSignInit EVP_DigestVerifyInit], [:],
[AC_MSG_FAILURE([EVP_DigestSignInit/EVP_DigestVerifyInit support in OpenSSL is mandatory.])])
Make OpenSSL mandatory
2018-06-12 11:26:04 +02:00
AC_MSG_CHECKING([for ECDSA P-256 support])
AC_COMPILE_IFELSE(
Make the configure.ac script compatible with OpenSSL 3.0.0 OpenSSL 3.0.0 deprecates many low level API functions. In preparation for the future support of linking BIND with OpenSSL 3.0.0 without the deprecated API functions, change the configure.ac script to use functions which are available on all supported versions of OpenSSL and LibreSSL.
2021-09-01 13:13:24 +00:00
[AC_LANG_PROGRAM([[#include <openssl/evp.h>]],
[[EVP_PKEY_CTX *kctx = EVP_PKEY_CTX_new_id(NID_X9_62_prime256v1, NULL);]])],
Make OpenSSL mandatory
2018-06-12 11:26:04 +02:00
[AC_MSG_RESULT([yes])],
[AC_MSG_FAILURE([not found. ECDSA P-256 support in OpenSSL is mandatory.])])
AC_MSG_CHECKING([for ECDSA P-384 support])
AC_COMPILE_IFELSE(
Make the configure.ac script compatible with OpenSSL 3.0.0 OpenSSL 3.0.0 deprecates many low level API functions. In preparation for the future support of linking BIND with OpenSSL 3.0.0 without the deprecated API functions, change the configure.ac script to use functions which are available on all supported versions of OpenSSL and LibreSSL.
2021-09-01 13:13:24 +00:00
[AC_LANG_PROGRAM([[#include <openssl/evp.h>]],
[[EVP_PKEY_CTX *kctx = EVP_PKEY_CTX_new_id(NID_secp384r1, NULL);]])],
Make OpenSSL mandatory
2018-06-12 11:26:04 +02:00
[AC_MSG_RESULT([yes])],
[AC_MSG_FAILURE([not found. ECDSA P-384 support in OpenSSL is mandatory.])])
AC_MSG_CHECKING([for Ed25519 support])
AC_COMPILE_IFELSE(
Make the configure.ac script compatible with OpenSSL 3.0.0 OpenSSL 3.0.0 deprecates many low level API functions. In preparation for the future support of linking BIND with OpenSSL 3.0.0 without the deprecated API functions, change the configure.ac script to use functions which are available on all supported versions of OpenSSL and LibreSSL.
2021-09-01 13:13:24 +00:00
[AC_LANG_PROGRAM([[#include <openssl/evp.h>]],
[[EVP_PKEY_CTX *kctx = EVP_PKEY_CTX_new_id(NID_ED25519, NULL);]])],
Make OpenSSL mandatory
2018-06-12 11:26:04 +02:00
[AC_DEFINE([HAVE_OPENSSL_ED25519], [1], [define if OpenSSL supports Ed25519])
AC_MSG_RESULT([yes])],
[AC_MSG_RESULT([no])])
Added Ed25519 support (#44696)
2017-07-31 15:26:00 +02:00
Make OpenSSL mandatory
2018-06-12 11:26:04 +02:00
AC_MSG_CHECKING([for Ed448 support])
test Ed448 against test vectors
2019-05-29 15:32:16 +10:00
AC_COMPILE_IFELSE(
Make the configure.ac script compatible with OpenSSL 3.0.0 OpenSSL 3.0.0 deprecates many low level API functions. In preparation for the future support of linking BIND with OpenSSL 3.0.0 without the deprecated API functions, change the configure.ac script to use functions which are available on all supported versions of OpenSSL and LibreSSL.
2021-09-01 13:13:24 +00:00
[AC_LANG_PROGRAM([[#include <openssl/evp.h>]],
[[EVP_PKEY_CTX *kctx = EVP_PKEY_CTX_new_id(NID_ED448, NULL);]])],
test Ed448 against test vectors
2019-05-29 15:32:16 +10:00
[AC_DEFINE([HAVE_OPENSSL_ED448], [1], [define if OpenSSL supports Ed448])
AC_MSG_RESULT([yes])],
configure.ac: Add OpenSSL SipHash support detection Add check for creating new EVP_PKEY with EVP_PKEY_SIPHASH, but disable SipHash on OpenSSL 1.1.1 as the hash length initialization is broken before OpenSSL 1.1.1a release.
2019-08-30 15:32:11 +02:00
[AC_MSG_RESULT([no])])
Make OpenSSL mandatory
2018-06-12 11:26:04 +02:00
#
# Check for OpenSSL SHA-1 support
#
AC_CHECK_FUNCS([EVP_sha1], [:],
[AC_MSG_FAILURE([SHA-1 support in OpenSSL is mandatory.])])
[master] merge several interdependent fixes 3760. [bug] Improve SIT with native PKCS#11 and on Windows. [RT #35433] 3759. [port] Enable delve on Windows. [RT #35441] 3758. [port] Enable export library APIs on windows. [RT #35382]
2014-02-26 19:00:05 -08:00
Make OpenSSL mandatory
2018-06-12 11:26:04 +02:00
#
# Check for OpenSSL SHA-2 support
#
AC_CHECK_FUNCS([EVP_sha224 EVP_sha256 EVP_sha384 EVP_sha512], [:],
[AC_MSG_FAILURE([SHA-2 support in OpenSSL is mandatory.])])
3734. [bug] Improve building with libtool. [RT #35314]
2014-02-10 15:01:06 +11:00
Make OpenSSL mandatory
2018-06-12 11:26:04 +02:00
#
# Check for OpenSSL AES support
#
AC_CHECK_FUNCS([EVP_aes_128_ecb EVP_aes_192_ecb EVP_aes_256_ecb], [:],
[AC_MSG_FAILURE([AES support in OpenSSL is mandatory.])])
2105. [func] GSS-TSIG support (RFC 3645).
2006-12-04 01:54:53 +00:00
[master] native PKCS#11 support 3705. [func] "configure --enable-native-pkcs11" enables BIND to use the PKCS#11 API for all cryptographic functions, so that it can drive a hardware service module directly without the need to use a modified OpenSSL as intermediary (so long as the HSM's vendor provides a complete-enough implementation of the PKCS#11 interface). This has been tested successfully with the Thales nShield HSM and with SoftHSMv2 from the OpenDNSSEC project. [RT #29031]
2014-01-14 15:40:56 -08:00
#
Make OpenSSL mandatory
2018-06-12 11:26:04 +02:00
# Check for OpenSSL 1.1.x/LibreSSL functions
[master] native PKCS#11 support 3705. [func] "configure --enable-native-pkcs11" enables BIND to use the PKCS#11 API for all cryptographic functions, so that it can drive a hardware service module directly without the need to use a modified OpenSSL as intermediary (so long as the HSM's vendor provides a complete-enough implementation of the PKCS#11 interface). This has been tested successfully with the Thales nShield HSM and with SoftHSMv2 from the OpenDNSSEC project. [RT #29031]
2014-01-14 15:40:56 -08:00
#
Remove TKEY Mode 2 (Diffie-Hellman) Completely remove the TKEY Mode 2 (Diffie-Hellman Exchanged Keying) from BIND 9 (from named, named.conf and all the tools). The TKEY usage is fringe at best and in all known cases, GSSAPI is being used as it should. The draft-eastlake-dnsop-rfc2930bis-tkey specifies that: 4.2 Diffie-Hellman Exchanged Keying (Deprecated) The use of this mode (#2) is NOT RECOMMENDED for the following two reasons but the specification is still included in Appendix A in case an implementation is needed for compatibility with old TKEY implementations. See Section 4.6 on ECDH Exchanged Keying. The mixing function used does not meet current cryptographic standards because it uses MD5 [RFC6151]. RSA keys must be excessively long to achieve levels of security required by current standards. We might optionally implement Elliptic Curve Diffie-Hellman (ECDH) key exchange mode 6 if the draft ever reaches the RFC status. Meanwhile the insecure DH mode needs to be removed.
2023-02-28 16:05:34 +01:00
AC_CHECK_FUNCS([ECDSA_SIG_get0 EVP_PKEY_get0_EC_KEY])
Refactor OpenSSL RSA exponent bits checking to a helper function - Make it a separate opensslrsa_check_exponent_bits() function to clean up the code a bit - Always use provider API first if using openssl 3.0, and fallback to EVP API for older openssl or if built with engine support - Use RSA_get0_key() (with shim for openssl 1.0) to avoid memory allocations
2022-12-26 14:25:18 +02:00
AC_CHECK_FUNCS([RSA_set0_key EVP_PKEY_get0_RSA])
2105. [func] GSS-TSIG support (RFC 3645).
2006-12-04 01:54:53 +00:00
Refactor TLSDNS module to work with libuv/ssl directly * Following the example set in 634bdfb16d8, the tlsdns netmgr module now uses libuv and SSL primitives directly, rather than opening a TLS socket which opens a TCP socket, as the previous model was difficult to debug. Closes #2335. * Remove the netmgr tls layer (we will have to re-add it for DoH) * Add isc_tls API to wrap the OpenSSL SSL_CTX object into libisc library; move the OpenSSL initialization/deinitialization from dstapi needed for OpenSSL 1.0.x to the isc_tls_{initialize,destroy}() * Add couple of new shims needed for OpenSSL 1.0.x * When LibreSSL is used, require at least version 2.7.0 that has the best OpenSSL 1.1.x compatibility and auto init/deinit * Enforce OpenSSL 1.1.x usage on Windows * Added a TLSDNS unit test and implemented a simple TLSDNS echo server and client.
2020-12-17 11:40:29 +01:00
AC_CHECK_FUNCS([TLS_server_method TLS_client_method])
autoconf: link with libssl This is prerequisite to adding DoT and DoH support to BIND 9.
2020-05-13 17:37:51 +02:00
Add support for enabling and enforcing FIPS mode in OpenSSL: * Add configure option --enable-fips-mode that detects and enables FIPS mode * Add a function to enable FIPS mode and call it on crypto init * Log an OpenSSL error when FIPS_mode_set() fails and exit * Report FIPS mode status in a separate log message from named
2018-10-14 14:32:02 +02:00
#
# Check whether FIPS mode is available and whether we should enable it
#
Add pairwise testing Pairwise testing is a test case generation technique based on the observation that most faults are caused by interactions of at most two factors. For BIND, its configure options can be thought of as such factors. Process BIND configure options into a model that is subsequently processed by the PICT tool in order to find an effective test vector. That test vector is then used for configuring and building BIND using various combinations of configure options.
2020-07-01 10:29:36 +02:00
# FIPS is not included in pairwise testing as the relevant Docker image
# does not support FIPS mode.
#
Detect if FIPS mode is configured at the OS level Always look for FIPS_mode and EVP_default_properties_enable_fips rather than just when requested by --enable-fips.
2022-06-29 14:10:06 +10:00
AC_CHECK_FUNCS(EVP_default_properties_enable_fips FIPS_mode)
Revert "Enable FIPS testing in Pairwise" This reverts commit e987a0c249882af0395409310b32fce7fc997015. --enable-fips-mode ./configure option does not work on platforms without FIPS-enabled OpenSSL.
2023-04-04 13:55:03 +02:00
# [pairwise: skip]
Add support for enabling and enforcing FIPS mode in OpenSSL: * Add configure option --enable-fips-mode that detects and enables FIPS mode * Add a function to enable FIPS mode and call it on crypto init * Log an OpenSSL error when FIPS_mode_set() fails and exit * Report FIPS mode status in a separate log message from named
2018-10-14 14:32:02 +02:00
AC_ARG_ENABLE([fips-mode],
[AS_HELP_STRING([--enable-fips-mode],
[enable FIPS mode in OpenSSL library [default=no]])],
[], [enable_fips_mode="no"])
AC_MSG_CHECKING([whether to enable FIPS mode in OpenSSL library])
AS_CASE([$enable_fips_mode],
[yes], [AC_MSG_RESULT([yes])
Detect if FIPS mode is configured at the OS level Always look for FIPS_mode and EVP_default_properties_enable_fips rather than just when requested by --enable-fips.
2022-06-29 14:10:06 +10:00
AC_DEFINE([ENABLE_FIPS_MODE], [1], [Define to 1 if you want FIPS mode to be enabled])
AS_IF([test "x$ac_cv_func_FIPS_mode" != xyes -a "x$ac_cv_func_EVP_default_properties_enable_fips" != xyes],
AC_MSG_FAILURE([OpenSSL FIPS mode requested but not available.]))],
Add support for enabling and enforcing FIPS mode in OpenSSL: * Add configure option --enable-fips-mode that detects and enables FIPS mode * Add a function to enable FIPS mode and call it on crypto init * Log an OpenSSL error when FIPS_mode_set() fails and exit * Report FIPS mode status in a separate log message from named
2018-10-14 14:32:02 +02:00
[no], [AC_MSG_RESULT([no])])
Use couple more autoconf convenience macros to cleanup configure.ac Pull and use several autoconf archive convenience macros to simplify configure.ac. * AX_CHECK_COMPILE_FLAG(FLAG, ...) - check whether given CFLAG works * AX_CHECK_LINK_FLAG(FLAG, ...) - check whether given LDFLAG works * AX_CHECK_PREPROC_FLAG(FLAG, ...) - check whether give CPPFLAG works * AX_SAVE_FLAGS/AX_RESTORE_FLAGS - save and restore *FLAGS
2019-06-19 11:16:47 +02:00
AX_RESTORE_FLAGS([openssl])
2105. [func] GSS-TSIG support (RFC 3645).
2006-12-04 01:54:53 +00:00
Rename OPENSSL_INCLUDES to OPENSSL_CFLAGS in AX_CHECK_OPENSSL() macro The ax_check_openssl m4 macro used OPENSSL_INCLUDES. Rename the subst variable to OPENSSL_CFLAGS and wrap AX_CHECK_OPENSSL() in action-if-not-found part of PKG_CHECK_MODULE check for libcrypto.
2019-06-24 14:06:56 +02:00
AC_SUBST([OPENSSL_CFLAGS])
Make OpenSSL mandatory
2018-06-12 11:26:04 +02:00
AC_SUBST([OPENSSL_LIBS])
3744. [experimental] SIT: send and process Source Identity Tokens (which are similar to DNS Cookies by Donald Eastlake) and are designed to help clients detect off path spoofed responses and for servers to detect legitimate clients. SIT use a experimental EDNS option code (65001). SIT can be enabled via --enable-developer or --enable-sit. It is on by default in Windows. RRL processing as been updated to know about SIT with legitimate clients not being rate limited. [RT #35389]
2014-02-19 12:53:42 +11:00
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AC_CHECK_FUNCS([clock_gettime])
Make OpenSSL mandatory
2018-06-12 11:26:04 +02:00
Add pairwise testing Pairwise testing is a test case generation technique based on the observation that most faults are caused by interactions of at most two factors. For BIND, its configure options can be thought of as such factors. Process BIND configure options into a model that is subsequently processed by the PICT tool in order to find an effective test vector. That test vector is then used for configuring and building BIND using various combinations of configure options.
2020-07-01 10:29:36 +02:00
# [pairwise: --with-gssapi=yes, --with-gssapi=auto, --without-gssapi]
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AC_ARG_WITH([gssapi],
[AS_HELP_STRING([--with-gssapi=[PATH|[/path/]krb5-config]],
Make OpenSSL mandatory
2018-06-12 11:26:04 +02:00
[Specify path for system-supplied GSSAPI
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
[default=auto]])],
[], [with_gssapi="auto"])
KRB5_CONFIG=
AS_CASE([$with_gssapi],
[no],[AC_MSG_CHECKING([for GSSAPI support])
AC_MSG_RESULT([no])],
[yes],[AC_PATH_PROG([KRB5_CONFIG], [krb5-config])
AS_IF([test -z "$KRB5_CONFIG"],
[AC_MSG_ERROR([krb5-config required but not found])])],
[auto],[AC_PATH_PROG([KRB5_CONFIG], [krb5-config])],
[*krb5-config*],[KRB5_CONFIG="$with_gssapi"],
[AC_MSG_ERROR([--with-gssapi requires yes|no|auto|/path/to/krb5-config])])
GSSAPI_CFLAGS=
GSSAPI_LIBS=
KRB5_CFLAGS=
KRB5_LIBS=
AS_IF([test -n "$KRB5_CONFIG"],
[AC_MSG_CHECKING([for gssapi libraries])
AX_SAVE_FLAGS([gssapi])
GSSAPI_CFLAGS=`"$KRB5_CONFIG" --cflags gssapi`
GSSAPI_LIBS=`"$KRB5_CONFIG" --libs gssapi`
CFLAGS="$CFLAGS $GSSAPI_CFLAGS"
LIBS="$LIBS $GSSAPI_LIBS"
AC_MSG_RESULT([$GSSAPI_CFLAGS $GSSAPI_LIBS])
AC_CHECK_HEADERS([gssapi/gssapi.h], [],
[AC_CHECK_HEADERS([gssapi.h], [],
[AC_MSG_ERROR([neither gssapi/gssapi.h nor gssapi.h found])])])
AC_CHECK_HEADERS([gssapi/gssapi_krb5.h],
[AC_CHECK_HEADERS([gssapi_krb5.h], []
[AC_MSG_ERROR([neither gssapi/gssapi_krb5.h nor gssapi_krb5.h found])])])
AC_CHECK_FUNCS([gss_acquire_cred],[],
[AC_MSG_ERROR([linking with $GSSAPI_LIBS does not work])])
AX_RESTORE_FLAGS([gssapi])
AC_MSG_CHECKING([for krb5 libraries])
AX_SAVE_FLAGS([krb5])
KRB5_CFLAGS=`"$KRB5_CONFIG" --cflags krb5`
KRB5_LIBS=`$KRB5_CONFIG --libs krb5`
CFLAGS="$CFLAGS $KRB5_CFLAGS"
LIBS="$CFLAGS $KRB5_LIBS"
AC_MSG_RESULT([$KRB5_CFLAGS $KRB5_LIBS])
AC_CHECK_HEADERS([krb5/krb5.h], [],
[AC_CHECK_HEADERS([krb5.h], [],
[AC_MSG_ERROR([neither krb5/krb5.h nor krb5 found])])])
AC_CHECK_FUNCS([krb5_init_context], [],
[AC_MSG_ERROR([linking with $KRB5_LIBS failed])])
AX_RESTORE_FLAGS([krb5])
AC_DEFINE([HAVE_GSSAPI], [1], [Define to 1 if you have the Kerberos Framework available])
# kludge to silence compiler warnings which recommend use of GSS.framework on macOS
AS_CASE([$host],[*-darwin*],[CFLAGS="$CFLAGS -Wno-deprecated-declarations"])])
AM_CONDITIONAL([HAVE_GSSAPI], [test -n "$GSSAPI_LIBS"])
AC_SUBST([GSSAPI_CFLAGS])
AC_SUBST([GSSAPI_LIBS])
AC_SUBST([KRB5_CFLAGS])
AC_SUBST([KRB5_LIBS])
[RT#35643] fixed -lrt in LIBS
2014-04-03 09:24:53 +02:00
[master] store "addzone" zone config in a NZD database 4421. [func] When built with LMDB (Lightning Memory-mapped Database), named will now use a database to store the configuration for zones added by "rndc addzone" instead of using a flat NZF file. This improves performance of "rndc delzone" and "rndc modzone" significantly. Existing NZF files will automatically by converted to NZD databases. To view the contents of an NZD or to roll back to NZF format, use "named-nzd2nzf". To disable this feature, use "configure --without-lmdb". [RT #39837]
2016-07-21 11:13:03 -07:00
#
# was --with-lmdb specified?
#
Add pairwise testing Pairwise testing is a test case generation technique based on the observation that most faults are caused by interactions of at most two factors. For BIND, its configure options can be thought of as such factors. Process BIND configure options into a model that is subsequently processed by the PICT tool in order to find an effective test vector. That test vector is then used for configuring and building BIND using various combinations of configure options.
2020-07-01 10:29:36 +02:00
# [pairwise: --with-lmdb=auto, --with-lmdb=yes, --without-lmdb]
Convert lmdb check to m4 macro
2020-03-10 12:37:25 +01:00
AC_ARG_WITH([lmdb],
[AS_HELP_STRING([--with-lmdb=@<:@PATH@:>@],
[use LMDB library @<:@default=auto@:>@, optionally specify the prefix for lmdb library])],
[:],
[with_lmdb="auto"])
[master] store "addzone" zone config in a NZD database 4421. [func] When built with LMDB (Lightning Memory-mapped Database), named will now use a database to store the configuration for zones added by "rndc addzone" instead of using a flat NZF file. This improves performance of "rndc delzone" and "rndc modzone" significantly. Existing NZF files will automatically by converted to NZD databases. To view the contents of an NZD or to roll back to NZF format, use "named-nzd2nzf". To disable this feature, use "configure --without-lmdb". [RT #39837]
2016-07-21 11:13:03 -07:00
Convert lmdb check to m4 macro
2020-03-10 12:37:25 +01:00
ac_lib_lmdb_found=no
AS_CASE([$with_lmdb],
[no],[],
[auto|yes], [PKG_CHECK_MODULES([LMDB], [lmdb],
[ac_lib_lmdb_found=yes],
[for ac_lib_lmdb_path in /usr /usr/local /opt /opt/local; do
AX_LIB_LMDB([$ac_lib_lmdb_path],
[ac_lib_lmdb_found=yes
break])
done
])],
[AX_LIB_LMDB([$with_lmdb],[ac_lib_lmdb_found=yes])])
# don't fail when in automatic mode
AS_IF([test "$with_lmdb" = "auto" && test "$ac_lib_lmdb_found" = "no"],
[with_lmdb=no])
# hard fail when LMDB requested, but not found
AS_IF([test "$with_lmdb" != "no" && test "$ac_lib_lmdb_found" != "yes"],
[AC_MSG_ERROR([LMDB requested, but not found])])
AS_IF([test "$ac_lib_lmdb_found" = "yes"],
[AC_DEFINE([HAVE_LMDB], [1], [Use lmdb library])])
AC_SUBST([LMDB_CFLAGS])
AC_SUBST([LMDB_LIBS])
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AM_CONDITIONAL([HAVE_LMDB], [test -n "$LMDB_LIBS"])
[master] store "addzone" zone config in a NZD database 4421. [func] When built with LMDB (Lightning Memory-mapped Database), named will now use a database to store the configuration for zones added by "rndc addzone" instead of using a flat NZF file. This improves performance of "rndc delzone" and "rndc modzone" significantly. Existing NZF files will automatically by converted to NZD databases. To view the contents of an NZD or to roll back to NZF format, use "named-nzd2nzf". To disable this feature, use "configure --without-lmdb". [RT #39837]
2016-07-21 11:13:03 -07:00
2122. [func] Experimental http server and statistics support for named via xml.
2006-12-21 06:03:37 +00:00
#
# was --with-libxml2 specified?
#
Add pairwise testing Pairwise testing is a test case generation technique based on the observation that most faults are caused by interactions of at most two factors. For BIND, its configure options can be thought of as such factors. Process BIND configure options into a model that is subsequently processed by the PICT tool in order to find an effective test vector. That test vector is then used for configuring and building BIND using various combinations of configure options.
2020-07-01 10:29:36 +02:00
# [pairwise: --with-libxml2=auto, --with-libxml2=yes, --without-libxml2]
Make the usage of libxml2 opaque to the caller The libxml2 have previously leaked into the global namespace leading to forced -I<include_path> for every compilation unit using isc/xml.h header. This MR fixes the usage making the caller object opaque.
2019-06-24 14:25:55 +02:00
AC_ARG_WITH([libxml2],
[AS_HELP_STRING([--with-libxml2],
[build with libxml2 library [yes|no|auto] (default is auto)])],
[], [with_libxml2="auto"])
AS_CASE([$with_libxml2],
[no],[],
[auto],[PKG_CHECK_MODULES([LIBXML2], [libxml-2.0 >= 2.6.0],
[AC_DEFINE([HAVE_LIBXML2], [1], [Use libxml2 library])],
[:])],
[yes],[PKG_CHECK_MODULES([LIBXML2], [libxml-2.0 >= 2.6.0],
[AC_DEFINE([HAVE_LIBXML2], [1], [Use libxml2 library])])],
[AC_MSG_ERROR([Specifying libxml2 installation path is not supported, adjust PKG_CONFIG_PATH instead])])
2122. [func] Experimental http server and statistics support for named via xml.
2006-12-21 06:03:37 +00:00
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AM_CONDITIONAL([HAVE_LIBXML2], [test -n "$LIBXML2_LIBS"])
[master] add JSON statistics channel 3524. [func] Added an alternate statistics channel in JSON format, when the server is built with the json-c library: http://[address]:[port]/json. [RT #32630]
2013-03-13 14:24:50 -07:00
Cleanup the way we detect json-c library to use only pkg-config
2019-02-06 11:56:42 +01:00
#
# was --with-json-c specified?
#
Add pairwise testing Pairwise testing is a test case generation technique based on the observation that most faults are caused by interactions of at most two factors. For BIND, its configure options can be thought of as such factors. Process BIND configure options into a model that is subsequently processed by the PICT tool in order to find an effective test vector. That test vector is then used for configuring and building BIND using various combinations of configure options.
2020-07-01 10:29:36 +02:00
# [pairwise: --with-json-c=detect, --with-json-c=yes, --without-json-c]
Cleanup the way we detect json-c library to use only pkg-config
2019-02-06 11:56:42 +01:00
AC_ARG_WITH([json-c],
[AS_HELP_STRING([--with-json-c],
[build with json-c library [yes|no|detect] (default is detect)])],
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
[], [with_json_c="detect"])
[master] add JSON statistics channel 3524. [func] Added an alternate statistics channel in JSON format, when the server is built with the json-c library: http://[address]:[port]/json. [RT #32630]
2013-03-13 14:24:50 -07:00
Cleanup the way we detect json-c library to use only pkg-config
2019-02-06 11:56:42 +01:00
AS_CASE([$with_json_c],
[no],[],
[detect],[PKG_CHECK_MODULES([JSON_C], [json-c >= 0.11],
[AC_DEFINE([HAVE_JSON_C], [1], [Use json-c library])],
[:])],
[yes],[PKG_CHECK_MODULES([JSON_C], [json-c >= 0.11],
[AC_DEFINE([HAVE_JSON_C], [1], [Use json-c library])])],
[AC_MSG_ERROR([Specifying json-c installation path is not supported, adjust PKG_CONFIG_PATH instead])]
)
[master] add JSON statistics channel 3524. [func] Added an alternate statistics channel in JSON format, when the server is built with the json-c library: http://[address]:[port]/json. [RT #32630]
2013-03-13 14:24:50 -07:00
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AM_CONDITIONAL([HAVE_JSON_C], [test -n "$JSON_C_LIBS"])
Cleanup the way we detect json-c library to use only pkg-config
2019-02-06 11:56:42 +01:00
AC_SUBST([JSON_C_CFLAGS])
AC_SUBST([JSON_C_LIBS])
[master] check for -ljson-c as well as -ljson 3512. [port] Check whether to use -ljson or -ljson-c. [RT #34115]
2013-07-09 16:43:59 -07:00
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
#
# was --with-zlib specified?
#
Add pairwise testing Pairwise testing is a test case generation technique based on the observation that most faults are caused by interactions of at most two factors. For BIND, its configure options can be thought of as such factors. Process BIND configure options into a model that is subsequently processed by the PICT tool in order to find an effective test vector. That test vector is then used for configuring and building BIND using various combinations of configure options.
2020-07-01 10:29:36 +02:00
# [pairwise: --with-zlib=auto, --with-zlib=yes, --without-zlib]
Convert the configure.ac rules for zlib library to use pkg-config
2019-07-31 13:50:15 +02:00
AC_ARG_WITH([zlib],
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
[AS_HELP_STRING([--with-zlib],
[build with zlib for HTTP compression
[default=yes]])],
[], with_zlib="auto")
Convert the configure.ac rules for zlib library to use pkg-config
2019-07-31 13:50:15 +02:00
AS_CASE([$with_zlib],
[no],[],
[auto],[PKG_CHECK_MODULES([ZLIB], [zlib],
add a quick-and-dirty method of debugging a single query when built with "configure --enable-singletrace", named will produce detailed query logging at the highest debug level for any query with query ID zero. this enables monitoring of the progress of a single query by specifying the QID using "dig +qid=0". the "client" logging category should be set to a low severity level to suppress logging of other queries. (the chance of another query using QID=0 at the same time is only 1 in 2^16.) "--enable-singletrace" turns on "--enable-querytrace" as well, so if the logging severity is not lowered, all other queries will be logged verbosely as well. compiling with either of these options will impair query performance; they should only be turned on when testing or troubleshooting.
2020-05-15 16:37:44 -07:00
[AC_DEFINE([HAVE_ZLIB], [1], [Use zlib library])],
[:])],
Convert the configure.ac rules for zlib library to use pkg-config
2019-07-31 13:50:15 +02:00
[yes],[PKG_CHECK_MODULES([ZLIB], [zlib],
add a quick-and-dirty method of debugging a single query when built with "configure --enable-singletrace", named will produce detailed query logging at the highest debug level for any query with query ID zero. this enables monitoring of the progress of a single query by specifying the QID using "dig +qid=0". the "client" logging category should be set to a low severity level to suppress logging of other queries. (the chance of another query using QID=0 at the same time is only 1 in 2^16.) "--enable-singletrace" turns on "--enable-querytrace" as well, so if the logging severity is not lowered, all other queries will be logged verbosely as well. compiling with either of these options will impair query performance; they should only be turned on when testing or troubleshooting.
2020-05-15 16:37:44 -07:00
[AC_DEFINE([HAVE_ZLIB], [1], [Use zlib library])])],
Convert the configure.ac rules for zlib library to use pkg-config
2019-07-31 13:50:15 +02:00
[AC_MSG_ERROR([Specifying zlib installation path is not supported, adjust PKG_CONFIG_PATH instead])])
minor cosmetic fixes - the configuration summary reported zlib compression was not supported even when it was. - when bind.keys.h was regenerated it violated clang-format style.
2020-02-19 00:17:03 -08:00
AC_SUBST([ZLIB_CFLAGS])
AC_SUBST([ZLIB_LIBS])
4234. [func] Add deflate compression in statistics channel HTTP server. [RT #40861]
2015-10-02 10:45:10 +02:00
Add support for reporting status via sd_notify() sd_notify() may be called by a service to notify the service manager about state changes. It can be used to send arbitrary information, encoded in an environment-block-like string. Most importantly, it can be used for start-up completion notification. Add libsystemd check to autoconf script and when the library is detected add calls to sd_notify() around the server->reload_status changes. Co-authored-by: Petr Špaček <pspacek@isc.org>
2021-10-20 18:14:49 +02:00
#
# was --with-libsystemd specified?
#
# [pairwise: --with-libsystemd=auto, --with-libsystemd=yes, --without-libsystemd]
AC_ARG_WITH([libsystemd],
[AS_HELP_STRING([--with-libsystemd],
[build with libsystemd integration [default=auto]])],
[], [with_libsystemd=auto])
AS_CASE([$with_libsystemd],
[no],[],
[auto],[PKG_CHECK_MODULES([LIBSYSTEMD], [libsystemd],
[AC_DEFINE([HAVE_LIBSYSTEMD], [1], [Use libsystemd library])],
[:])],
[yes],[PKG_CHECK_MODULES([LIBSYSTEMD], [libsystemd],
[AC_DEFINE([HAVE_LIBSYSTEMD], [1], [Use libsystemd library])])],
[AC_MSG_ERROR([Specifying libsystemd installation path is not supported, adjust PKG_CONFIG_PATH instead])])
AC_SUBST([LIBSYSTEMD_CFLAGS])
AC_SUBST([LIBSYSTEMD_LIBS])
2667. [func] Add support for logging stack backtrace on assertion failure (not available for all platforms). [RT #19780] 9.7.0
2009-09-01 18:40:25 +00:00
#
Improve the backtrace to print symbols when backtrace_symbols() is available The previous commit removed the code related to the internal symbol table. On platforms where available, we can now use backtrace_symbols() to print more verbose symbols table to the output. As there's now general availability of backtrace() and backtrace_symbols() functions (see below), the commit also removes the usage of glibc internals and the custom stack tracing. * backtrace(), backtrace_symbols(), and backtrace_symbols_fd() are provided in glibc since version 2.1. * backtrace(), backtrace_symbols(), and backtrace_symbols_fd() first appeared in Mac OS X 10.5. * The backtrace() library of functions first appeared in NetBSD 7.0 and FreeBSD 10.0.
2020-03-11 09:55:48 +01:00
# Check if the system supports glibc-compatible backtrace() function.
2667. [func] Add support for logging stack backtrace on assertion failure (not available for all platforms). [RT #19780] 9.7.0
2009-09-01 18:40:25 +00:00
#
Improve the backtrace to print symbols when backtrace_symbols() is available The previous commit removed the code related to the internal symbol table. On platforms where available, we can now use backtrace_symbols() to print more verbose symbols table to the output. As there's now general availability of backtrace() and backtrace_symbols() functions (see below), the commit also removes the usage of glibc internals and the custom stack tracing. * backtrace(), backtrace_symbols(), and backtrace_symbols_fd() are provided in glibc since version 2.1. * backtrace(), backtrace_symbols(), and backtrace_symbols_fd() first appeared in Mac OS X 10.5. * The backtrace() library of functions first appeared in NetBSD 7.0 and FreeBSD 10.0.
2020-03-11 09:55:48 +01:00
AC_CHECK_HEADERS([execinfo.h],
Add support for generating backtraces on Windows This commit adds support for generating backtraces on Windows and refactors the isc_backtrace API to match the Linux/BSD API (without the isc_ prefix) * isc_backtrace_gettrace() was renamed to isc_backtrace(), the third argument was removed and the return type was changed to int * isc_backtrace_symbols() was added * isc_backtrace_symbols_fd() was added and used as appropriate
2021-04-27 16:20:03 +02:00
[AC_SEARCH_LIBS([backtrace_symbols], [execinfo],
[AC_CHECK_FUNCS([backtrace_symbols])])])
[RT #186] check for various networking items (eg, getaddrinfo) after Kame test
2000-07-25 23:54:09 +00:00
all "dnl" comments are now "#" comments so they appear in configure
2000-02-11 18:35:11 +00:00
#
# We do the IPv6 compilation checking after libtool so that we can put
# the right suffix on the files.
#
Require IPv6 support from the OS
2018-08-21 11:50:23 +02:00
AC_MSG_CHECKING([for IPv6 structures])
AC_COMPILE_IFELSE(
[AC_LANG_PROGRAM(
[[
#include <sys/types.h>
#include <sys/socket.h>
#include <netinet/in.h>
]],
[[
Remove support for legacy systems without in6addr_any
2018-08-21 12:54:15 +02:00
struct sockaddr_in6 sin6;
struct in6_addr in6;
Remove support for legacy systems without in6_pktinfo
2018-08-21 13:22:40 +02:00
struct in6_pktinfo in6_pi;
Remove support for legacy systems without inet_{ntop,pton} w/ IPv6 support
2018-08-21 13:48:48 +02:00
struct sockaddr_storage storage;
Remove support for legacy systems without in6addr_any
2018-08-21 12:54:15 +02:00
in6 = in6addr_any;
Remove support for legacy platforms without in6addr_loopback
2018-08-21 13:07:05 +02:00
in6 = in6addr_loopback;
Remove support for legacy systems without sin6_scope_id
2018-08-21 13:17:00 +02:00
sin6.sin6_scope_id = 0;
Remove support for legacy systems without in6addr_any
2018-08-21 12:54:15 +02:00
return (0);
Require IPv6 support from the OS
2018-08-21 11:50:23 +02:00
]])],
[AC_MSG_RESULT([yes])],
[AC_MSG_FAILURE([IPv6 support is mandatory])])
--enable-ipv6/--diable-ipv6 supported. added missing output of "no" when Kame is not found. look for netinet6/in6.h.
2000-01-07 02:44:13 +00:00
4440. [func] Enable TCP fast open support when available on the server side. [RT #42866]
2016-08-12 15:31:33 +10:00
#
[master] Improve handling of TCP_FASTOPEN on FreeBSD 4726. [port] Prevent setsockopt() errors related to TCP_FASTOPEN from being logged on FreeBSD if the kernel does not support it. Notify the user when the kernel does support TCP_FASTOPEN, but it is disabled by sysctl. Add a new configure option, --disable-tcp-fastopen, to disable use of TCP_FASTOPEN altogether. [RT #44754]
2017-09-18 08:34:41 +02:00
# Allow forcibly disabling TCP Fast Open support as autodetection might yield
# confusing results on some systems (e.g. FreeBSD; see set_tcp_fastopen()
Completely remove BIND 9 Windows support The Windows support has been completely removed from the source tree and BIND 9 now no longer supports native compilation on Windows. We might consider reviewing mingw-w64 port if contributed by external party, but no development efforts will be put into making BIND 9 compile and run on Windows again.
2021-05-20 15:53:50 +02:00
# comment in lib/isc/socket.c).
4440. [func] Enable TCP fast open support when available on the server side. [RT #42866]
2016-08-12 15:31:33 +10:00
#
Add pairwise testing Pairwise testing is a test case generation technique based on the observation that most faults are caused by interactions of at most two factors. For BIND, its configure options can be thought of as such factors. Process BIND configure options into a model that is subsequently processed by the PICT tool in order to find an effective test vector. That test vector is then used for configuring and building BIND using various combinations of configure options.
2020-07-01 10:29:36 +02:00
# [pairwise: --enable-tcp-fastopen, --disable-tcp-fastopen]
Replace platform ISC_PLATFORM_HAVETFO with config ENABLE_TCP_FASTOPEN
2018-09-05 12:34:10 +02:00
AC_ARG_ENABLE([tcp_fastopen],
[AS_HELP_STRING([--disable-tcp-fastopen],
[disable TCP Fast Open support [default=yes]])],
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
[], [enable_tcp_fastopen="yes"])
[master] Improve handling of TCP_FASTOPEN on FreeBSD 4726. [port] Prevent setsockopt() errors related to TCP_FASTOPEN from being logged on FreeBSD if the kernel does not support it. Notify the user when the kernel does support TCP_FASTOPEN, but it is disabled by sysctl. Add a new configure option, --disable-tcp-fastopen, to disable use of TCP_FASTOPEN altogether. [RT #44754]
2017-09-18 08:34:41 +02:00
Replace platform ISC_PLATFORM_HAVETFO with config ENABLE_TCP_FASTOPEN
2018-09-05 12:34:10 +02:00
AS_IF([test "$enable_tcp_fastopen" = "yes"],
[AC_DEFINE([ENABLE_TCP_FASTOPEN], [1], [define if you want TCP_FASTOPEN enabled if available])])
4440. [func] Enable TCP fast open support when available on the server side. [RT #42866]
2016-08-12 15:31:33 +10:00
[RT #186] check for various networking items (eg, getaddrinfo) after Kame test
2000-07-25 23:54:09 +00:00
#
Correctly detect inet_aton, inet_pton and inet_ptop on BSD/OS 4.1.
2000-04-26 21:50:49 +00:00
# Check for some other useful functions that are not ever-present.
[RT #186] check for various networking items (eg, getaddrinfo) after Kame test
2000-07-25 23:54:09 +00:00
#
Use strnstr implementation from FreeBSD if not provided by OS
2022-09-15 17:03:50 +10:00
AC_CHECK_FUNCS([strlcpy strlcat strnstr])
test for strsep() using AC_TRY_LINK() instead of AC_CHECK_FUNC() to deal with AIX braindamage [RT #2190]
2001-12-03 21:52:07 +00:00
Update and cleanup the readline library support This commit updates and simplifies the checks for the readline support in nslookup and nsupdate: * Change the autoconf checks to pkg-config only, all supported libraries have accompanying .pc files now. * Add editline support in addition to libedit and GNU readline * Add isc/readline.h shim header that defines dummy readline() function when no readline library is available
2020-08-11 16:14:36 +02:00
#
# Check for readline support
#
Add pairwise testing Pairwise testing is a test case generation technique based on the observation that most faults are caused by interactions of at most two factors. For BIND, its configure options can be thought of as such factors. Process BIND configure options into a model that is subsequently processed by the PICT tool in order to find an effective test vector. That test vector is then used for configuring and building BIND using various combinations of configure options.
2020-07-01 10:29:36 +02:00
# editline is not included in pairwise testing as the library
# is not present in the relevant Docker image.
#
# GNU Readline is not included in pairwise testing as the library
# in the relevant Docker image does not support pkg-config.
#
# [pairwise: --with-readline=auto, --with-readline=yes, --with-readline=libedit, --without-readline]
Update and cleanup the readline library support This commit updates and simplifies the checks for the readline support in nslookup and nsupdate: * Change the autoconf checks to pkg-config only, all supported libraries have accompanying .pc files now. * Add editline support in addition to libedit and GNU readline * Add isc/readline.h shim header that defines dummy readline() function when no readline library is available
2020-08-11 16:14:36 +02:00
AC_ARG_WITH([readline],
[AS_HELP_STRING([--with-readline=yes|no|libedit|readline],
[specify readline library [default auto]])],
[], [with_readline="auto"])
AS_CASE([$with_readline],
[no],[],
Add missing support for --with-readline=yes By error, ./configure --with-readline would fail, because 'yes' wasn't handled properly in the top level case.
2020-08-18 16:31:20 +02:00
[auto|yes],
Update and cleanup the readline library support This commit updates and simplifies the checks for the readline support in nslookup and nsupdate: * Change the autoconf checks to pkg-config only, all supported libraries have accompanying .pc files now. * Add editline support in addition to libedit and GNU readline * Add isc/readline.h shim header that defines dummy readline() function when no readline library is available
2020-08-11 16:14:36 +02:00
[PKG_CHECK_MODULES([READLINE], [libedit],
[AC_DEFINE([HAVE_READLINE_LIBEDIT], [1], [Build with libedit support])],
editline's pkg-config name is libeditline
2020-08-19 17:12:25 +02:00
[PKG_CHECK_MODULES([READLINE], [libeditline],
Update and cleanup the readline library support This commit updates and simplifies the checks for the readline support in nslookup and nsupdate: * Change the autoconf checks to pkg-config only, all supported libraries have accompanying .pc files now. * Add editline support in addition to libedit and GNU readline * Add isc/readline.h shim header that defines dummy readline() function when no readline library is available
2020-08-11 16:14:36 +02:00
[AC_DEFINE([HAVE_READLINE_EDITLINE], [1], [Build with editline support.])],
[PKG_CHECK_MODULES([READLINE], [readline],
[AC_DEFINE([HAVE_READLINE_READLINE], [1], [Build with readline support.])],
[AS_IF([test "$with_readline" = "yes"],
[AC_MSG_ERROR([readline support requested, but none of the libraries have been found.])])])])])],
[libedit],
[PKG_CHECK_MODULES([READLINE], [libedit],
[AC_DEFINE([HAVE_READLINE_LIBEDIT], [1], [Build with libedit support])])],
[editline],
editline's pkg-config name is libeditline
2020-08-19 17:12:25 +02:00
[PKG_CHECK_MODULES([READLINE], [libeditline],
Update and cleanup the readline library support This commit updates and simplifies the checks for the readline support in nslookup and nsupdate: * Change the autoconf checks to pkg-config only, all supported libraries have accompanying .pc files now. * Add editline support in addition to libedit and GNU readline * Add isc/readline.h shim header that defines dummy readline() function when no readline library is available
2020-08-11 16:14:36 +02:00
[AC_DEFINE([HAVE_READLINE_EDITLINE], [1], [Build with editline support])])],
[readline],
[PKG_CHECK_MODULES([READLINE], [readline],
[AC_DEFINE([HAVE_READLINE_READLINE], [1], [Build with readline support])])],
[AC_MSG_ERROR([Unknown readline '$with_readline' library requested.])])
AM_CONDITIONAL([HAVE_READLINE], [test -n "$READLINE_LIBS"])
AC_SUBST([READLINE_CFLAGS])
AC_SUBST([READLINE_LIBS])
3244. [func] Added readline support to nslookup and nsupdate. Also simplified nsupdate syntax to make "update" and "prereq" optional. [RT #24659]
2011-12-16 23:01:17 +00:00
all "dnl" comments are now "#" comments so they appear in configure
2000-02-11 18:35:11 +00:00
#
# Security Stuff
#
2503. [port] linux: improve compatibility with Linux Standard Base. [RT #18793]
2008-12-01 03:53:32 +00:00
# Note it is very recommended to *not* disable chroot(),
# this is only because chroot() was made obsolete by Posix.
Add pairwise testing Pairwise testing is a test case generation technique based on the observation that most faults are caused by interactions of at most two factors. For BIND, its configure options can be thought of as such factors. Process BIND configure options into a model that is subsequently processed by the PICT tool in order to find an effective test vector. That test vector is then used for configuring and building BIND using various combinations of configure options.
2020-07-01 10:29:36 +02:00
#
# [pairwise: --enable-chroot, --disable-chroot]
4814. [cleanup] Use AS_HELP_STRING for consistent help text. [RT #46521]
2017-11-13 11:36:45 +11:00
AC_ARG_ENABLE(chroot, AS_HELP_STRING([--disable-chroot], [disable chroot]))
2503. [port] linux: improve compatibility with Linux Standard Base. [RT #18793]
2008-12-01 03:53:32 +00:00
case "$enable_chroot" in
yes|'')
AC_CHECK_FUNCS(chroot)
;;
no)
;;
esac
Make libcap mandatory when Linux capabilities are enabled
2018-06-26 14:52:11 +02:00
Refactor the privilege dropping On Linux, the libcap is now mandatory. It makes things simpler for us. System without {set,get}res{uid,gid} now have compatibility shim using setreuid/setregid or seteuid/setegid to setup effective UID/GID, so the same code can be called all the time (including on Linux).
2022-10-06 14:22:20 +02:00
AS_CASE([$host],
[*-linux*],
[PKG_CHECK_MODULES([LIBCAP], [libcap],
[AC_DEFINE([HAVE_LIBCAP], [1], [Define to 1 if libcap was found])])])
Make libcap mandatory when Linux capabilities are enabled
2018-06-26 14:52:11 +02:00
1814. [func] UNIX domain controls are now supported.
2005-02-23 01:09:23 +00:00
case "$host" in
*-solaris*)
AC_DEFINE(NEED_SECURE_DIRECTORY, 1,
[Define if connect does not honour the permission on the UNIX domain socket.])
;;
esac
1550. [port] Call tzset(), if available, before calling chroot().
2004-01-07 05:48:15 +00:00
#
# Time Zone Stuff
#
Remove now dummy NEED_PTHREAD_INIT
2018-09-05 12:58:02 +02:00
AC_CHECK_FUNCS([tzset])
1550. [port] Call tzset(), if available, before calling chroot().
2004-01-07 05:48:15 +00:00
spelling
2009-01-17 09:35:58 +00:00
AC_MSG_CHECKING(for optarg declaration)
Update the configure.ac for autoconf >= 2.71 compatibility This mostly removes stuff that's either deprecated, obsolete or not used at all: * Update the minimal autoconf version to 2.69 * AC_PROG_CC_C99 is deprecated, just use AC_PROG_CC as we require C11 anyway * AC_HEADER_TIME is deprecated, both <sys/time.h> and <time.h> can be included at the same time, and we don't use the macros that AC_HEADER_TIME defines anywhere * AC_HEADER_STDC checks for ISO C90 and we require at least C11 * Replace AC_TRY_*([]) with AC_*_IFELSE([AC_LANG_PROGRAM()]) * Update m4/ax_check_openssl.m4 from serial 10 to serial 11 * Update m4/ax_gcc_func_attribute.m4 from serial 10 to serial 13 * Update m4/ax_pthread.m4 from serial 24 to serial 30 * Add early AC_CANONICAL_TARGET call to prevent warning from AX_PTHREAD
2021-04-16 12:25:35 +02:00
AC_COMPILE_IFELSE(
[AC_LANG_PROGRAM(
[[#include <unistd.h>]],
[[optarg = 0;]]
)],
[AC_MSG_RESULT(yes)],
[AC_MSG_RESULT(no)
GEN_NEED_OPTARG="-DNEED_OPTARG=1"
AC_DEFINE(NEED_OPTARG, 1, [Defined if extern char *optarg is not declared.])]
)
1883. [port] sunos: portability fixes. [RT #14814]
2005-06-08 02:09:18 +00:00
[master] nanosecond-accurate file modification times 4361. [cleanup] Where supported, file modification times returned by isc_file_getmodtime() are now accurate to the nanosecond. [RT #41968]
2016-05-05 12:22:26 -07:00
#
# Check for nanoseconds in file stats
#
Replace platform ISC_PLATFORM_HAVESTATNSEC with config ENABLE_STAT_NSEC
2018-09-05 12:39:51 +02:00
AC_MSG_CHECKING([for st_mtim.tv_nsec])
AC_COMPILE_IFELSE(
[AC_LANG_PROGRAM(
Use '#include <sys/stat.h>' for the stat structure definition POSIX.1-2008 changed the st_atim, st_mtim, and st_ctime members of the struct stat from time_t to struct timespec and because not all operating systems already implemented this version of the standard or historically deviated to include own nanosecond precision in the structure. The autoconf script used to include <sys/fcntl.h> which contradicts POSIX.1 as it mandates <sys/stat.h> inclusion. Change the autoconf check to include <sys/stat.h>. Also fix the missing AC_MSG_RESULT([yes/no]) in the check.
2021-09-20 17:33:56 +10:00
[[#include <sys/stat.h>]],
Replace platform ISC_PLATFORM_HAVESTATNSEC with config ENABLE_STAT_NSEC
2018-09-05 12:39:51 +02:00
[[struct stat s;
return(s.st_mtim.tv_nsec);
]])],
Use '#include <sys/stat.h>' for the stat structure definition POSIX.1-2008 changed the st_atim, st_mtim, and st_ctime members of the struct stat from time_t to struct timespec and because not all operating systems already implemented this version of the standard or historically deviated to include own nanosecond precision in the structure. The autoconf script used to include <sys/fcntl.h> which contradicts POSIX.1 as it mandates <sys/stat.h> inclusion. Change the autoconf check to include <sys/stat.h>. Also fix the missing AC_MSG_RESULT([yes/no]) in the check.
2021-09-20 17:33:56 +10:00
[AC_MSG_RESULT([yes])
AC_DEFINE([HAVE_STAT_NSEC], [1], [define if struct stat has st_mtim.tv_nsec field])],
[AC_MSG_RESULT([no])])
[master] nanosecond-accurate file modification times 4361. [cleanup] Where supported, file modification times returned by isc_file_getmodtime() are now accurate to the nanosecond. [RT #41968]
2016-05-05 12:22:26 -07:00
check for the existance of if_nametoindex(). developer: jinmei reviewer: marka
2002-11-26 03:53:42 +00:00
#
# Check for if_nametoindex() for IPv6 scoped addresses support
#
Replace platform.h ISC_PLATFORM_HAVEIFNAMETOINDEX with config HAVE_IF_NAMETOINDEX
2018-09-04 12:35:42 +02:00
AC_CHECK_FUNCS([if_nametoindex])
check for the existance of if_nametoindex(). developer: jinmei reviewer: marka
2002-11-26 03:53:42 +00:00
Replace arch specific atomic.h with global atomic.h header using either stdatomic, __atomic or __sync primitives
2018-08-14 10:23:50 +02:00
AC_CHECK_HEADERS([uchar.h])
Add small tweaks to the code to fix compilation when ISC assertions are disabled While implementing the new unit testing framework cmocka, it was found that the BIND 9 code doesn't compile when assertions are disabled or replaced with any function (such as mock_assert() from cmocka unit testing framework) that's not directly recognized as assertion by the compiler. This made the compiler to complain about blocks of code that was recognized as unreachable before, but now it isn't. The changes in this commit include: * assigns default values to couple of local variables, * moves some return statements around INSIST assertions, * adds __builtin_unreachable(); annotations after some INSIST assertions, * fixes one broken assertion (= instead of ==)
2018-06-04 13:41:09 +02:00
#
# Check for __builtin_unreachable
#
AC_MSG_CHECKING([compiler support for __builtin_unreachable()])
AC_LINK_IFELSE(
[AC_LANG_PROGRAM(
[[]],
[[__builtin_unreachable();]]
)],
[AC_MSG_RESULT([yes])
AC_DEFINE([HAVE_BUILTIN_UNREACHABLE], [1], [define if the compiler supports __builtin_unreachable().])
],
[AC_MSG_RESULT([no])
])
Use GCC builtin for clz in RPZ lookup code (#42818)
2016-07-10 19:46:17 +05:30
#
# Check for __builtin_clz
#
AC_MSG_CHECKING([compiler support for __builtin_clz])
Update the configure.ac for autoconf >= 2.71 compatibility This mostly removes stuff that's either deprecated, obsolete or not used at all: * Update the minimal autoconf version to 2.69 * AC_PROG_CC_C99 is deprecated, just use AC_PROG_CC as we require C11 anyway * AC_HEADER_TIME is deprecated, both <sys/time.h> and <time.h> can be included at the same time, and we don't use the macros that AC_HEADER_TIME defines anywhere * AC_HEADER_STDC checks for ISO C90 and we require at least C11 * Replace AC_TRY_*([]) with AC_*_IFELSE([AC_LANG_PROGRAM()]) * Update m4/ax_check_openssl.m4 from serial 10 to serial 11 * Update m4/ax_gcc_func_attribute.m4 from serial 10 to serial 13 * Update m4/ax_pthread.m4 from serial 24 to serial 30 * Add early AC_CANONICAL_TARGET call to prevent warning from AX_PTHREAD
2021-04-16 12:25:35 +02:00
AC_LINK_IFELSE(
[AC_LANG_PROGRAM(
[[]],
[[return (__builtin_clz(0xff) == 24 ? 1 : 0);]]
)],
Use __builtin_*_overflow for isc_time_{add,subtract}() Use the __builtin_uadd_overflow() and __builtin_usub_overflow() for overflow checks in isc_time_add() and isc_time_subtract(). This generates more efficient and safe code.
2021-10-20 11:22:52 +02:00
[AC_MSG_RESULT([yes])
AC_DEFINE(HAVE_BUILTIN_CLZ, 1, [Define to 1 if the compiler supports __builtin_clz.])
],
[AC_MSG_RESULT([no])]
Update the configure.ac for autoconf >= 2.71 compatibility This mostly removes stuff that's either deprecated, obsolete or not used at all: * Update the minimal autoconf version to 2.69 * AC_PROG_CC_C99 is deprecated, just use AC_PROG_CC as we require C11 anyway * AC_HEADER_TIME is deprecated, both <sys/time.h> and <time.h> can be included at the same time, and we don't use the macros that AC_HEADER_TIME defines anywhere * AC_HEADER_STDC checks for ISO C90 and we require at least C11 * Replace AC_TRY_*([]) with AC_*_IFELSE([AC_LANG_PROGRAM()]) * Update m4/ax_check_openssl.m4 from serial 10 to serial 11 * Update m4/ax_gcc_func_attribute.m4 from serial 10 to serial 13 * Update m4/ax_pthread.m4 from serial 24 to serial 30 * Add early AC_CANONICAL_TARGET call to prevent warning from AX_PTHREAD
2021-04-16 12:25:35 +02:00
)
Use __builtin_*_overflow for isc_time_{add,subtract}() Use the __builtin_uadd_overflow() and __builtin_usub_overflow() for overflow checks in isc_time_add() and isc_time_subtract(). This generates more efficient and safe code.
2021-10-20 11:22:52 +02:00
#
Check for working __builtin_mul_overflow() implementation Instead of using generic HAVE_BUILTIN_OVERFLOW, we need to check whether the overflow functions actually work as there was a bug in GCC that it would not detect mul overflow when compiled with `-m32` option without optimizations and the bug was fixed only for GCC 6.5+ and 7.3+/8+. For further details see: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82274
2022-09-23 15:35:58 +02:00
# Check for __builtin_*_overflow
Use __builtin_*_overflow for isc_time_{add,subtract}() Use the __builtin_uadd_overflow() and __builtin_usub_overflow() for overflow checks in isc_time_add() and isc_time_subtract(). This generates more efficient and safe code.
2021-10-20 11:22:52 +02:00
#
Check for working __builtin_mul_overflow() implementation Instead of using generic HAVE_BUILTIN_OVERFLOW, we need to check whether the overflow functions actually work as there was a bug in GCC that it would not detect mul overflow when compiled with `-m32` option without optimizations and the bug was fixed only for GCC 6.5+ and 7.3+/8+. For further details see: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82274
2022-09-23 15:35:58 +02:00
AC_MSG_CHECKING([compiler support for __builtin_add_overflow()])
Use __builtin_*_overflow for isc_time_{add,subtract}() Use the __builtin_uadd_overflow() and __builtin_usub_overflow() for overflow checks in isc_time_add() and isc_time_subtract(). This generates more efficient and safe code.
2021-10-20 11:22:52 +02:00
AC_LINK_IFELSE(
[AC_LANG_PROGRAM(
[[#include <limits.h>]],
Check for working __builtin_mul_overflow() implementation Instead of using generic HAVE_BUILTIN_OVERFLOW, we need to check whether the overflow functions actually work as there was a bug in GCC that it would not detect mul overflow when compiled with `-m32` option without optimizations and the bug was fixed only for GCC 6.5+ and 7.3+/8+. For further details see: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82274
2022-09-23 15:35:58 +02:00
[[return (__builtin_add_overflow((unsigned int)UINT_MAX, (unsigned int)UINT_MAX, &(unsigned int){ 0 }));]]
Use __builtin_*_overflow for isc_time_{add,subtract}() Use the __builtin_uadd_overflow() and __builtin_usub_overflow() for overflow checks in isc_time_add() and isc_time_subtract(). This generates more efficient and safe code.
2021-10-20 11:22:52 +02:00
)],
[AC_MSG_RESULT([yes])
Check for working __builtin_mul_overflow() implementation Instead of using generic HAVE_BUILTIN_OVERFLOW, we need to check whether the overflow functions actually work as there was a bug in GCC that it would not detect mul overflow when compiled with `-m32` option without optimizations and the bug was fixed only for GCC 6.5+ and 7.3+/8+. For further details see: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82274
2022-09-23 15:35:58 +02:00
AC_DEFINE([HAVE_BUILTIN_ADD_OVERFLOW], [1], [define if the compiler supports __builtin_add_overflow().])
],
[AC_MSG_RESULT([no])
])
AC_MSG_CHECKING([compiler support for __builtin_sub_overflow()])
AC_LINK_IFELSE(
[AC_LANG_PROGRAM(
[[#include <limits.h>]],
[[return (__builtin_sub_overflow((unsigned int)0, (unsigned int)UINT_MAX, &(unsigned int){ 0 }));]]
)],
[AC_MSG_RESULT([yes])
AC_DEFINE([HAVE_BUILTIN_SUB_OVERFLOW], [1], [define if the compiler supports __builtin_sub_overflow().])
],
[AC_MSG_RESULT([no])
])
AC_MSG_CHECKING([compiler support for __builtin_mul_overflow()])
AC_LINK_IFELSE(
[AC_LANG_PROGRAM(
configure: Fix __builtin_mul_overflow() compiler support check `UINT64_C(UINT64_MAX)' is redundant and leads to a compilation error since UINT64_C involves token concatenation, causing the check to fail. This change fixes that by using UINT64_MAX directly, and including the appropriate header. Signed-off-by: Arjun Shankar <arjun@redhat.com>
2023-05-02 13:58:18 +02:00
[[#include <limits.h>
#include <stdint.h>
]],
[[return (__builtin_mul_overflow(UINT64_MAX, UINT64_MAX, &(uint64_t){ 0 }));]]
Check for working __builtin_mul_overflow() implementation Instead of using generic HAVE_BUILTIN_OVERFLOW, we need to check whether the overflow functions actually work as there was a bug in GCC that it would not detect mul overflow when compiled with `-m32` option without optimizations and the bug was fixed only for GCC 6.5+ and 7.3+/8+. For further details see: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82274
2022-09-23 15:35:58 +02:00
)],
[AC_MSG_RESULT([yes])
AC_DEFINE([HAVE_BUILTIN_MUL_OVERFLOW], [1], [define if the compiler supports __builtin_mul_overflow().])
Use __builtin_*_overflow for isc_time_{add,subtract}() Use the __builtin_uadd_overflow() and __builtin_usub_overflow() for overflow checks in isc_time_add() and isc_time_subtract(). This generates more efficient and safe code.
2021-10-20 11:22:52 +02:00
],
[AC_MSG_RESULT([no])
])
Use GCC builtin for clz in RPZ lookup code (#42818)
2016-07-10 19:46:17 +05:30
Make "rrset-order fixed" a compile-time option. settable by "./configure --enable-fixed-rrset". Disabled by default. [rt17977]
2008-04-23 21:32:57 +00:00
#
# Activate "rrset-order fixed" or not?
#
Add pairwise testing Pairwise testing is a test case generation technique based on the observation that most faults are caused by interactions of at most two factors. For BIND, its configure options can be thought of as such factors. Process BIND configure options into a model that is subsequently processed by the PICT tool in order to find an effective test vector. That test vector is then used for configuring and building BIND using various combinations of configure options.
2020-07-01 10:29:36 +02:00
# [pairwise: --enable-fixed-rrset, --disable-fixed-rrset]
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AC_ARG_ENABLE([fixed-rrset],
[AS_HELP_STRING([--enable-fixed-rrset],
[enable fixed rrset ordering [default=no]])],
[], [enable_fixed_rrset="no"])
AS_IF([test "$enable_fixed_rrset" = "yes"],
[AC_DEFINE([DNS_RDATASET_FIXED], [1],
[Define to enable "rrset-order fixed" syntax.])])
Make "rrset-order fixed" a compile-time option. settable by "./configure --enable-fixed-rrset". Disabled by default. [rt17977]
2008-04-23 21:32:57 +00:00
[master] dnstap 4235. [func] Added support in named for "dnstap", a fast method of capturing and logging DNS traffic, and a new command "dnstap-read" to read a dnstap log file. Use "configure --enable-dnstap" to enable this feature (note that this requires libprotobuf-c and libfstrm). See the ARM for configuration details. Thanks to Robert Edmonds of Farsight Security. [RT #40211]
2015-10-02 12:32:42 -07:00
#
# Activate dnstap?
#
Add pairwise testing Pairwise testing is a test case generation technique based on the observation that most faults are caused by interactions of at most two factors. For BIND, its configure options can be thought of as such factors. Process BIND configure options into a model that is subsequently processed by the PICT tool in order to find an effective test vector. That test vector is then used for configuring and building BIND using various combinations of configure options.
2020-07-01 10:29:36 +02:00
# [pairwise: --enable-dnstap, --disable-dnstap]
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AC_ARG_ENABLE([dnstap],
[AS_HELP_STRING([--enable-dnstap],
[enable dnstap support
(requires fstrm, protobuf-c)])],
[], [enable_dnstap=no])
AS_IF([test "$enable_dnstap" != "no"],
[PKG_CHECK_MODULES([DNSTAP], [libfstrm libprotobuf-c], [],
[AC_MSG_FAILURE([Required libraries (fstrm, protobuf-c) were not found, please install them.])])
Ensure fstrm_capture is looked for fstrm_capture is not an essential utility, but its corresponding Makefile token needs to substituted even if it is not found in PATH or else the "dnstap" system test will consistently fail.
2020-04-23 08:34:11 +02:00
AC_PATH_PROG([FSTRM_CAPTURE], [fstrm_capture])
AC_PATH_PROG([PROTOC_C], [protoc-c])
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AS_IF([test -z "$PROTOC_C"],
[AC_MSG_ERROR([protoc-c compiler not found])])
AC_DEFINE([HAVE_DNSTAP], 1, [Define to 1 to enable dnstap support])
])
2123. [func] Use Doxygen to generate internal documention. [RT #11398]
2006-12-22 01:46:19 +00:00
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AC_SUBST([DNSTAP_CFLAGS])
AC_SUBST([DNSTAP_LIBS])
AM_CONDITIONAL([HAVE_DNSTAP], [test "$enable_dnstap" != "no"])
Convert the documentation to Sphinx documentation format The ARM and the manpages have been converted into Sphinx documentation format. Sphinx uses reStructuredText as its markup language, and many of its strengths come from the power and straightforwardness of reStructuredText and its parsing and translating suite, the Docutils.
2019-03-31 12:49:47 +02:00
#
# The following sections deal with tools used for formatting
# the documentation. They are all optional, unless you are
# a developer editing the documentation source.
#
#
# Look for sphinx-build
#
AC_ARG_VAR([SPHINX_BUILD], [path to sphinx-build binary used to build the documentation])
Remove pregenerated manpages from the repo We don't need them in the repo, it's sufficient if we pregenerate them while preparing the tarball. That way we don't have overhead while modifying them but they are still available for installations without Sphinx. I assume that this will make rebases and cherry-picks across branches easier, with less trial and error churn required in the CI. It's implemented in the way that we build the manpages only when we either have pregenerated pages available at the configure time or sphinx-build is installed and working.
2022-07-04 17:25:11 +02:00
AC_PATH_PROG([SPHINX_BUILD], [sphinx-build], [])
AM_CONDITIONAL([HAVE_SPHINX_BUILD], [test -n "$SPHINX_BUILD"])
Convert the documentation to Sphinx documentation format The ARM and the manpages have been converted into Sphinx documentation format. Sphinx uses reStructuredText as its markup language, and many of its strengths come from the power and straightforwardness of reStructuredText and its parsing and translating suite, the Docutils.
2019-03-31 12:49:47 +02:00
Remove pregenerated manpages from the repo We don't need them in the repo, it's sufficient if we pregenerate them while preparing the tarball. That way we don't have overhead while modifying them but they are still available for installations without Sphinx. I assume that this will make rebases and cherry-picks across branches easier, with less trial and error churn required in the CI. It's implemented in the way that we build the manpages only when we either have pregenerated pages available at the configure time or sphinx-build is installed and working.
2022-07-04 17:25:11 +02:00
AC_PATH_PROG([XELATEX], [xelatex], [])
AC_PATH_PROG([LATEXMK], [latexmk], [])
AM_CONDITIONAL([HAVE_XELATEX], [test -n "$XELATEX" && test -n "$LATEXMK"])
#
# Build the man pages only if we have prebuilt manpages or we can build them from RST sources
#
AM_CONDITIONAL([BUILD_MANPAGES], [test -e doc/man/named.conf.5in || test -n "$SPHINX_BUILD"])
Convert the documentation to Sphinx documentation format The ARM and the manpages have been converted into Sphinx documentation format. Sphinx uses reStructuredText as its markup language, and many of its strengths come from the power and straightforwardness of reStructuredText and its parsing and translating suite, the Docutils.
2019-03-31 12:49:47 +02:00
#
# Pull release date from CHANGES file last modification date
# for reproducible builds
#
Use UTC datetime for reproducible builds For reproducible builds, we use last modification time of the CHANGES file. This works pretty well, unless the builds are made in different timezones. Use UTC option to date command to make the builds reproducible.
2022-01-24 16:09:48 +01:00
release_date=`date -u -r CHANGES +%Y-%m-%d`
Convert the documentation to Sphinx documentation format The ARM and the manpages have been converted into Sphinx documentation format. Sphinx uses reStructuredText as its markup language, and many of its strengths come from the power and straightforwardness of reStructuredText and its parsing and translating suite, the Docutils.
2019-03-31 12:49:47 +02:00
AC_SUBST([RELEASE_DATE], $release_date)
#
# Look for Doxygen
#
AC_PATH_PROGS([DOXYGEN], [doxygen])
AC_CONFIG_FILES([doc/doxygen/doxygen-input-filter],
[chmod +x doc/doxygen/doxygen-input-filter])
2123. [func] Use Doxygen to generate internal documention. [RT #11398]
2006-12-22 01:46:19 +00:00
Fix a bug printing zone names with '/' character in XML and JSON stats (#39873)
2015-06-29 18:33:18 +05:30
#
# Look for curl
#
AC_PATH_PROG(CURL, curl, curl)
AC_SUBST(CURL)
Handle HTTP/1.1 pipelined requests Check to see whether there are outstanding requests in the httpd receive buffer after sending the response, and if so, process them. Test that pipelined requests are handled by sending multiple minimal HTTP/1.1 using netcat (nc) and checking that we get back the same number of responses.
2021-10-26 15:28:36 +11:00
#
# Look for nc
#
AC_PATH_PROGS(NC, nc, nc)
AC_SUBST(NC)
Add support for libidn2 Added two new configure options: --with-libidn2 - to enable IDN using GNU libidn2 idnkit, libidn and libidn2 support can not be used at the same time. NOTE: libidn2 does not support punycode back to Unicode characters, so support for this is missing. Signed-off-by: Tomas Hozza <thozza@redhat.com> Removed iconv, convert directly from locale to ACE Fix libidn2 and idnkit origin appending Make IDN options in help less different Signed-off-by: Petr Menšík <pemensik@redhat.com>
2015-09-23 16:47:37 +02:00
#
# IDN support using libidn2
#
Simplify the libidn2 configure checks
2018-03-14 11:30:47 +01:00
LIBIDN2_CFLAGS=
Remove conversion from locale to utf8 from public API Emit fatal failures on locale to ACE encoding Separate idnout support, disable it for libidn2 < 2.0 Add custom path to libidn. Leave default path for multilib support. Allow turning off IDN input processing by dig option Improve documentation, fix support in host Fix configure changes to adjust help text Use strlcpy with size guard Improve IDN variants choosing. Fix idn2 function name. Remove immediate idn_locale_to_ace and idn_ace_to_locale. Signed-off-by: Petr Menšík <pemensik@redhat.com>
2017-08-15 14:36:59 +02:00
LIBIDN2_LIBS=
Add pairwise testing Pairwise testing is a test case generation technique based on the observation that most faults are caused by interactions of at most two factors. For BIND, its configure options can be thought of as such factors. Process BIND configure options into a model that is subsequently processed by the PICT tool in order to find an effective test vector. That test vector is then used for configuring and building BIND using various combinations of configure options.
2020-07-01 10:29:36 +02:00
# [pairwise: --with-libidn2=yes, --without-libidn2]
Rework libidn2 detection Clean up the parts of configure.in responsible for handling libidn2 detection and adjust other pieces of the build system to match these cleanups: - use pkg-config when --with-libidn2 is used without an explicit path, - look for idn2_to_ascii_lz() rather than idn2_to_ascii_8z() as the former is used in BIND while the latter is not, - do not look for idn2_to_unicode_8zlz() as it is present in all libidn2 versions which have idn2_to_ascii_lz(), - check whether the <idn2.h> header is usable, - set LDFLAGS in the Makefile for dig so that, if specified, the requested libidn2 path is used when linking with libidn2, - override CPPFLAGS when looking for libidn2 components so that the configure script does not produce warnings when libidn2 is not installed system-wide, - merge the AS_CASE() call into the AS_IF() call below it to simplify code, - indicate the default value of --with-libidn2 in "./configure --help" output, - use $with_libidn2 rather than $use_libidn2 to better match the name of the configure script argument, - stop differentiating between IDN "in" and "out" support, i.e. make dig either support libidn2 or not; remove WITH_* Autoconf macros and use a new one, HAVE_LIBIDN2, to determine whether libidn2 support should be enabled.
2018-07-10 14:34:35 +02:00
AC_ARG_WITH([libidn2],
[AS_HELP_STRING([--with-libidn2[=PATH]], [enable IDN support using GNU libidn2 [yes|no(default)|path]])],
[with_libidn2="$withval"], [with_libidn2="no"])
AS_CASE([$with_libidn2],
[yes], [PKG_CHECK_MODULES([LIBIDN2], [libidn2],
[AC_DEFINE([HAVE_LIBIDN2], [1], [Define if libidn2 was found])])],
[no], [],
Use couple more autoconf convenience macros to cleanup configure.ac Pull and use several autoconf archive convenience macros to simplify configure.ac. * AX_CHECK_COMPILE_FLAG(FLAG, ...) - check whether given CFLAG works * AX_CHECK_LINK_FLAG(FLAG, ...) - check whether given LDFLAG works * AX_CHECK_PREPROC_FLAG(FLAG, ...) - check whether give CPPFLAG works * AX_SAVE_FLAGS/AX_RESTORE_FLAGS - save and restore *FLAGS
2019-06-19 11:16:47 +02:00
[*], [AX_SAVE_FLAGS([libidn2])
Rework libidn2 detection Clean up the parts of configure.in responsible for handling libidn2 detection and adjust other pieces of the build system to match these cleanups: - use pkg-config when --with-libidn2 is used without an explicit path, - look for idn2_to_ascii_lz() rather than idn2_to_ascii_8z() as the former is used in BIND while the latter is not, - do not look for idn2_to_unicode_8zlz() as it is present in all libidn2 versions which have idn2_to_ascii_lz(), - check whether the <idn2.h> header is usable, - set LDFLAGS in the Makefile for dig so that, if specified, the requested libidn2 path is used when linking with libidn2, - override CPPFLAGS when looking for libidn2 components so that the configure script does not produce warnings when libidn2 is not installed system-wide, - merge the AS_CASE() call into the AS_IF() call below it to simplify code, - indicate the default value of --with-libidn2 in "./configure --help" output, - use $with_libidn2 rather than $use_libidn2 to better match the name of the configure script argument, - stop differentiating between IDN "in" and "out" support, i.e. make dig either support libidn2 or not; remove WITH_* Autoconf macros and use a new one, HAVE_LIBIDN2, to determine whether libidn2 support should be enabled.
2018-07-10 14:34:35 +02:00
LIBIDN2_CFLAGS="-I$with_libidn2/include"
Refactor libidn2 detection code Make the code block handling the --with-libidn2=/path/to/libidn2 form of the --with-libidn2 build-time option behave more similarly to the PKG_CHECK_MODULES() macro.
2020-11-24 14:51:51 +01:00
LIBIDN2_LIBS="-L$with_libidn2/lib"
CFLAGS="$CFLAGS $LIBIDN2_CFLAGS"
LIBS="$LIBS $LIBIDN2_LIBS"
Rework libidn2 detection Clean up the parts of configure.in responsible for handling libidn2 detection and adjust other pieces of the build system to match these cleanups: - use pkg-config when --with-libidn2 is used without an explicit path, - look for idn2_to_ascii_lz() rather than idn2_to_ascii_8z() as the former is used in BIND while the latter is not, - do not look for idn2_to_unicode_8zlz() as it is present in all libidn2 versions which have idn2_to_ascii_lz(), - check whether the <idn2.h> header is usable, - set LDFLAGS in the Makefile for dig so that, if specified, the requested libidn2 path is used when linking with libidn2, - override CPPFLAGS when looking for libidn2 components so that the configure script does not produce warnings when libidn2 is not installed system-wide, - merge the AS_CASE() call into the AS_IF() call below it to simplify code, - indicate the default value of --with-libidn2 in "./configure --help" output, - use $with_libidn2 rather than $use_libidn2 to better match the name of the configure script argument, - stop differentiating between IDN "in" and "out" support, i.e. make dig either support libidn2 or not; remove WITH_* Autoconf macros and use a new one, HAVE_LIBIDN2, to determine whether libidn2 support should be enabled.
2018-07-10 14:34:35 +02:00
AC_CHECK_HEADERS([idn2.h],
[],
[AC_MSG_ERROR([idn2.h not found])])
AC_SEARCH_LIBS([idn2_to_ascii_lz], [idn2],
Refactor libidn2 detection code Make the code block handling the --with-libidn2=/path/to/libidn2 form of the --with-libidn2 build-time option behave more similarly to the PKG_CHECK_MODULES() macro.
2020-11-24 14:51:51 +01:00
[LIBIDN2_LIBS="$LIBIDN2_LIBS $ac_cv_search_idn2_to_ascii_lz"
Rework libidn2 detection Clean up the parts of configure.in responsible for handling libidn2 detection and adjust other pieces of the build system to match these cleanups: - use pkg-config when --with-libidn2 is used without an explicit path, - look for idn2_to_ascii_lz() rather than idn2_to_ascii_8z() as the former is used in BIND while the latter is not, - do not look for idn2_to_unicode_8zlz() as it is present in all libidn2 versions which have idn2_to_ascii_lz(), - check whether the <idn2.h> header is usable, - set LDFLAGS in the Makefile for dig so that, if specified, the requested libidn2 path is used when linking with libidn2, - override CPPFLAGS when looking for libidn2 components so that the configure script does not produce warnings when libidn2 is not installed system-wide, - merge the AS_CASE() call into the AS_IF() call below it to simplify code, - indicate the default value of --with-libidn2 in "./configure --help" output, - use $with_libidn2 rather than $use_libidn2 to better match the name of the configure script argument, - stop differentiating between IDN "in" and "out" support, i.e. make dig either support libidn2 or not; remove WITH_* Autoconf macros and use a new one, HAVE_LIBIDN2, to determine whether libidn2 support should be enabled.
2018-07-10 14:34:35 +02:00
AC_DEFINE([HAVE_LIBIDN2], [1], [Define if libidn2 was found])],
[AC_MSG_ERROR([libidn2 requested, but not found])])
Use couple more autoconf convenience macros to cleanup configure.ac Pull and use several autoconf archive convenience macros to simplify configure.ac. * AX_CHECK_COMPILE_FLAG(FLAG, ...) - check whether given CFLAG works * AX_CHECK_LINK_FLAG(FLAG, ...) - check whether given LDFLAG works * AX_CHECK_PREPROC_FLAG(FLAG, ...) - check whether give CPPFLAG works * AX_SAVE_FLAGS/AX_RESTORE_FLAGS - save and restore *FLAGS
2019-06-19 11:16:47 +02:00
AX_RESTORE_FLAGS([libidn2])])
Simplify the libidn2 configure checks
2018-03-14 11:30:47 +01:00
AC_SUBST([LIBIDN2_CFLAGS])
AC_SUBST([LIBIDN2_LIBS])
Add support for libidn2 Added two new configure options: --with-libidn2 - to enable IDN using GNU libidn2 idnkit, libidn and libidn2 support can not be used at the same time. NOTE: libidn2 does not support punycode back to Unicode characters, so support for this is missing. Signed-off-by: Tomas Hozza <thozza@redhat.com> Removed iconv, convert directly from locale to ACE Fix libidn2 and idnkit origin appending Make IDN options in help less different Signed-off-by: Petr Menšík <pemensik@redhat.com>
2015-09-23 16:47:37 +02:00
Integrate cmocka unit testing framework to kyua
2018-06-04 08:54:58 +02:00
#
# Check whether to build with cmocka unit testing framework
#
Add pairwise testing Pairwise testing is a test case generation technique based on the observation that most faults are caused by interactions of at most two factors. For BIND, its configure options can be thought of as such factors. Process BIND configure options into a model that is subsequently processed by the PICT tool in order to find an effective test vector. That test vector is then used for configuring and building BIND using various combinations of configure options.
2020-07-01 10:29:36 +02:00
# [pairwise: --with-cmocka=detect, --with-cmocka=yes, --without-cmocka]
Integrate cmocka unit testing framework to kyua
2018-06-04 08:54:58 +02:00
AC_ARG_WITH([cmocka],
configure could fail if cmocka was not detected
2019-02-06 15:26:19 -08:00
[AS_HELP_STRING([--with-cmocka=detect],[enable CMocka based tests (default is detect)])],
Simplify cmocka detection to use only pkg-config and always use cmocka if available
2019-02-06 01:30:57 +01:00
[],[with_cmocka=detect])
Integrate cmocka unit testing framework to kyua
2018-06-04 08:54:58 +02:00
AS_CASE([$with_cmocka],
Simplify cmocka detection to use only pkg-config and always use cmocka if available
2019-02-06 01:30:57 +01:00
[no],[],
Require CMocka >= 1.1.3 to run the unit tests In CMocka versions << 1.1.3, the skip() function would cause the whole unit test to abort when CMOCKA_TEST_ABORT is set. As this is problem only in Debian 9 Stretch and Ubuntu 16.04 Xenial, we just require the CMocka >= 1.1.3 and disable the unit testing on Debian 9 Stretch until we can pull the libcmocka-dev from stretch-backports and remove the Ubuntu 16.04 Xenial from the CI as it is reaching End of Standard Support at the end of April 2021.
2021-03-15 10:22:15 +01:00
[detect],[PKG_CHECK_MODULES([CMOCKA], [cmocka >= 1.1.3],
configure could fail if cmocka was not detected
2019-02-06 15:26:19 -08:00
[AC_DEFINE([HAVE_CMOCKA], [1], [Use CMocka])
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
with_cmocka=yes],[with_cmocka=no])],
Require CMocka >= 1.1.3 to run the unit tests In CMocka versions << 1.1.3, the skip() function would cause the whole unit test to abort when CMOCKA_TEST_ABORT is set. As this is problem only in Debian 9 Stretch and Ubuntu 16.04 Xenial, we just require the CMocka >= 1.1.3 and disable the unit testing on Debian 9 Stretch until we can pull the libcmocka-dev from stretch-backports and remove the Ubuntu 16.04 Xenial from the CI as it is reaching End of Standard Support at the end of April 2021.
2021-03-15 10:22:15 +01:00
[yes],[PKG_CHECK_MODULES([CMOCKA], [cmocka >= 1.1.3],
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
[AC_DEFINE([HAVE_CMOCKA], [1], [Use CMocka])])],
configure could fail if cmocka was not detected
2019-02-06 15:26:19 -08:00
[AC_MSG_ERROR([Use PKG_CONFIG_PATH to specify path to CMocka library])]
Simplify cmocka detection to use only pkg-config and always use cmocka if available
2019-02-06 01:30:57 +01:00
)
Integrate cmocka unit testing framework to kyua
2018-06-04 08:54:58 +02:00
AC_SUBST([CMOCKA_CFLAGS])
AC_SUBST([CMOCKA_LIBS])
Remove embedded atf-src and enforce atf >= 0.20 + kyua as execution engine
2018-10-23 11:44:33 +02:00
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AM_CONDITIONAL([HAVE_CMOCKA], [test "$with_cmocka" = "yes"])
Remove embedded atf-src and enforce atf >= 0.20 + kyua as execution engine
2018-10-23 11:44:33 +02:00
Compile with jemalloc to reduce memory allocator contention The jemalloc allocator is scalable high performance allocator, this is the first in the series of commits that will add jemalloc as a memory allocator for BIND 9. This commit adds configure.ac check and Makefile modifications to use jemalloc as BIND 9 allocator.
2021-05-11 12:29:57 +02:00
#
Don't enforce jemalloc on NetBSD The NetBSD system allocator is in fact based on the jemalloc, but it doesn't export the extended interface, so we can't use that. Remove the jemalloc enforcement for the NetBSD.
2022-10-31 14:46:30 +00:00
# Compile with jemalloc (either provided as package or wired in the system on FreeBSD)
Compile with jemalloc to reduce memory allocator contention The jemalloc allocator is scalable high performance allocator, this is the first in the series of commits that will add jemalloc as a memory allocator for BIND 9. This commit adds configure.ac check and Makefile modifications to use jemalloc as BIND 9 allocator.
2021-05-11 12:29:57 +02:00
#
Use system allocator when jemalloc is unavailable This commit adds support for systems where the jemalloc library is not available as a package, here's the quick summary: * On Linux - the jemalloc is usually available as a package, if configured --without-jemalloc, the shim would be used around malloc(), free(), realloc() and malloc_usable_size() * On macOS - the jemalloc is available from homebrew or macports, if configured --without-jemalloc, the shim would be used around malloc(), free(), realloc() and malloc_size() * On FreeBSD - the jemalloc is *the* system allocator, we just need to check for <malloc_np.h> header to get access to non-standard API * On NetBSD - the jemalloc is *the* system allocator, we just need to check for <jemalloc/jemalloc.h> header to get access to non-standard API * On a system hostile to users and developers (read OpenBSD) - the jemalloc API is emulated by using ((size_t *)ptr)[-1] field to hold the size information. The OpenBSD developers care only for themselves, so why should we care about speed on OpenBSD?
2021-05-25 12:46:00 +02:00
# [pairwise: --with-jemalloc=detect, --with-jemalloc=yes, --without-jemalloc]
AC_ARG_WITH([jemalloc],
[AS_HELP_STRING([--with-jemalloc=detect],[enable jemalloc memory allocator (default is detect)])],
[],[with_jemalloc=detect])
AS_CASE([$with_jemalloc],
[no],[],
[yes],[AX_CHECK_JEMALLOC(
[AC_DEFINE([HAVE_JEMALLOC], [1], [Define to 1 if jemalloc is available])],
[AC_MSG_ERROR([jemalloc not found])])],
[AX_CHECK_JEMALLOC(
[AC_DEFINE([HAVE_JEMALLOC], [1], [Define to 1 if jemalloc is available])
with_jemalloc=yes],
[AC_MSG_WARN([jemalloc not found; performance will be reduced])
with_jemalloc=no])])
AS_IF([test "$with_jemalloc" = "no"],
A more helpful error when --without-jemalloc is impossible When jemalloc is the system allocator (on FreeBSD and NetBSD), trying to build --without-jemalloc caused an obscure compiler error. Instead, complain at configure time that --without-jemalloc cannot work. (It needs to remain an error because it is vexing when configure quietly ignores an explicit direction.)
2022-09-21 17:44:48 +01:00
[AS_CASE([$host],
Don't enforce jemalloc on NetBSD The NetBSD system allocator is in fact based on the jemalloc, but it doesn't export the extended interface, so we can't use that. Remove the jemalloc enforcement for the NetBSD.
2022-10-31 14:46:30 +00:00
[*-freebsd*],[AC_MSG_ERROR([You cannot compile without jemalloc; jemalloc is the system allocator on FreeBSD])])
A more helpful error when --without-jemalloc is impossible When jemalloc is the system allocator (on FreeBSD and NetBSD), trying to build --without-jemalloc caused an obscure compiler error. Instead, complain at configure time that --without-jemalloc cannot work. (It needs to remain an error because it is vexing when configure quietly ignores an explicit direction.)
2022-09-21 17:44:48 +01:00
AC_CHECK_FUNCS([malloc_size malloc_usable_size])])
Compile with jemalloc to reduce memory allocator contention The jemalloc allocator is scalable high performance allocator, this is the first in the series of commits that will add jemalloc as a memory allocator for BIND 9. This commit adds configure.ac check and Makefile modifications to use jemalloc as BIND 9 allocator.
2021-05-11 12:29:57 +02:00
Properly handle JEMALLOC_* Autoconf variables The AX_CHECK_JEMALLOC() m4 macro sets the JEMALLOC_CFLAGS variable, not JEMALLOC_CPPFLAGS. Furthermore, the JEMALLOC_CFLAGS and JEMALLOC_LIBS variables should only be included in the build flags if jemalloc was successfully configured. Tweak lib/isc/Makefile.am accordingly.
2021-10-12 10:44:30 +02:00
AM_CONDITIONAL([HAVE_JEMALLOC], [test "$with_jemalloc" = "yes"])
Add autoconf option to enable memory leak detection in libraries There's a known memory leak in the engine_pkcs11 at the time of writing this and it interferes with the named ability to check for memory leaks in the OpenSSL memory context by default. Add an autoconf option to explicitly enable the memory leak detection, and use it in the CI except for pkcs11 enabled builds. When this gets fixed in the engine_pkc11, the option can be enabled by default.
2022-09-27 13:35:26 +02:00
#
# Check memory leaks in external libraries
#
# [pairwise: --enable-leak-detection, --disable-leak-detection]
AC_ARG_ENABLE([leak-detection],
[AS_HELP_STRING([--enable-leak-detection],[enable the memory leak detection in external libraries (libxml2, libuv, OpenSSL) (disabled by default)])],
[],[enable_leak_detection=no])
AS_CASE([$enable_leak_detection],
[yes],[AC_DEFINE([ENABLE_LEAK_DETECTION], [1], [Define to enable memory leak detection in external libraries])])
Add a --enable-querytrace configure switch for very verbose query tracelogging (#37520)
2015-02-26 16:49:27 +05:30
#
add a quick-and-dirty method of debugging a single query when built with "configure --enable-singletrace", named will produce detailed query logging at the highest debug level for any query with query ID zero. this enables monitoring of the progress of a single query by specifying the QID using "dig +qid=0". the "client" logging category should be set to a low severity level to suppress logging of other queries. (the chance of another query using QID=0 at the same time is only 1 in 2^16.) "--enable-singletrace" turns on "--enable-querytrace" as well, so if the logging severity is not lowered, all other queries will be logged verbosely as well. compiling with either of these options will impair query performance; they should only be turned on when testing or troubleshooting.
2020-05-15 16:37:44 -07:00
# was --enable-querytrace or --enable-singletrace specified?
Add a --enable-querytrace configure switch for very verbose query tracelogging (#37520)
2015-02-26 16:49:27 +05:30
#
Add pairwise testing Pairwise testing is a test case generation technique based on the observation that most faults are caused by interactions of at most two factors. For BIND, its configure options can be thought of as such factors. Process BIND configure options into a model that is subsequently processed by the PICT tool in order to find an effective test vector. That test vector is then used for configuring and building BIND using various combinations of configure options.
2020-07-01 10:29:36 +02:00
# [pairwise: --enable-singletrace --enable-querytrace, --disable-singletrace --enable-querytrace, --disable-singletrace --disable-querytrace]
add a quick-and-dirty method of debugging a single query when built with "configure --enable-singletrace", named will produce detailed query logging at the highest debug level for any query with query ID zero. this enables monitoring of the progress of a single query by specifying the QID using "dig +qid=0". the "client" logging category should be set to a low severity level to suppress logging of other queries. (the chance of another query using QID=0 at the same time is only 1 in 2^16.) "--enable-singletrace" turns on "--enable-querytrace" as well, so if the logging severity is not lowered, all other queries will be logged verbosely as well. compiling with either of these options will impair query performance; they should only be turned on when testing or troubleshooting.
2020-05-15 16:37:44 -07:00
AC_ARG_ENABLE([singletrace],
AS_HELP_STRING([--enable-singletrace],
[enable single-query trace logging
[default=no]]),
enable_singletrace="$enableval", enable_singletrace="no")
AC_MSG_CHECKING([whether to enable single-query trace logging])
case "$enable_singletrace" in
yes)
enable_querytrace=yes
AC_MSG_RESULT(yes)
AC_DEFINE(WANT_SINGLETRACE, 1, [Define to enable single-query tracing.])
;;
no)
AC_MSG_RESULT(no)
;;
*)
AC_MSG_ERROR("--enable-singletrace requires yes or no")
;;
esac
Add pairwise testing Pairwise testing is a test case generation technique based on the observation that most faults are caused by interactions of at most two factors. For BIND, its configure options can be thought of as such factors. Process BIND configure options into a model that is subsequently processed by the PICT tool in order to find an effective test vector. That test vector is then used for configuring and building BIND using various combinations of configure options.
2020-07-01 10:29:36 +02:00
# [pairwise: skip]
Add a --enable-querytrace configure switch for very verbose query tracelogging (#37520)
2015-02-26 16:49:27 +05:30
AC_ARG_ENABLE(querytrace,
4814. [cleanup] Use AS_HELP_STRING for consistent help text. [RT #46521]
2017-11-13 11:36:45 +11:00
AS_HELP_STRING([--enable-querytrace],
[enable very verbose query trace logging
[default=no]]),
add a quick-and-dirty method of debugging a single query when built with "configure --enable-singletrace", named will produce detailed query logging at the highest debug level for any query with query ID zero. this enables monitoring of the progress of a single query by specifying the QID using "dig +qid=0". the "client" logging category should be set to a low severity level to suppress logging of other queries. (the chance of another query using QID=0 at the same time is only 1 in 2^16.) "--enable-singletrace" turns on "--enable-querytrace" as well, so if the logging severity is not lowered, all other queries will be logged verbosely as well. compiling with either of these options will impair query performance; they should only be turned on when testing or troubleshooting.
2020-05-15 16:37:44 -07:00
enable_querytrace="$enableval",
enable_querytrace="$enable_singletrace")
Add a --enable-querytrace configure switch for very verbose query tracelogging (#37520)
2015-02-26 16:49:27 +05:30
Code cleanups (#41656)
2016-03-04 11:12:23 +05:30
AC_MSG_CHECKING([whether to enable query trace logging])
fixed incorrect output from the configuration summary
2020-04-23 14:29:07 -07:00
case "$enable_querytrace" in
Add a --enable-querytrace configure switch for very verbose query tracelogging (#37520)
2015-02-26 16:49:27 +05:30
yes)
AC_MSG_RESULT(yes)
AC_DEFINE(WANT_QUERYTRACE, 1, [Define to enable very verbose query trace logging.])
;;
no)
add a quick-and-dirty method of debugging a single query when built with "configure --enable-singletrace", named will produce detailed query logging at the highest debug level for any query with query ID zero. this enables monitoring of the progress of a single query by specifying the QID using "dig +qid=0". the "client" logging category should be set to a low severity level to suppress logging of other queries. (the chance of another query using QID=0 at the same time is only 1 in 2^16.) "--enable-singletrace" turns on "--enable-querytrace" as well, so if the logging severity is not lowered, all other queries will be logged verbosely as well. compiling with either of these options will impair query performance; they should only be turned on when testing or troubleshooting.
2020-05-15 16:37:44 -07:00
AS_IF([test "$enable_singletrace" = "yes"],
[AC_MSG_ERROR([--enable-singletrace requires --enable-querytrace])])
Add a --enable-querytrace configure switch for very verbose query tracelogging (#37520)
2015-02-26 16:49:27 +05:30
AC_MSG_RESULT(no)
;;
*)
import libdummyrpz test library for DNSRPS libdummyrpz is a limited version of the fastrpz library for use in testing the dnsrps API.
2023-03-13 12:47:00 -07:00
AC_MSG_ERROR(["--enable-querytrace requires yes or no (not $enable_querytrace)"])
Add a --enable-querytrace configure switch for very verbose query tracelogging (#37520)
2015-02-26 16:49:27 +05:30
;;
esac
Set "dnssec-validation auto" by default - the default setting for dnssec-validation is now "auto", which activates DNSSEC validation using the IANA root key. The old behavior can be restored by explicitly setting "dnssec-validation yes", which "yes", which activates DNSSEC validation only if keys are explicitly configured in named.conf. - the ARM has been updated to describe the new behavior
2018-05-16 11:58:15 -07:00
#
# Was --disable-auto-validation specified?
#
Add pairwise testing Pairwise testing is a test case generation technique based on the observation that most faults are caused by interactions of at most two factors. For BIND, its configure options can be thought of as such factors. Process BIND configure options into a model that is subsequently processed by the PICT tool in order to find an effective test vector. That test vector is then used for configuring and building BIND using various combinations of configure options.
2020-07-01 10:29:36 +02:00
Add VALIDATION_DEFAULT to Windows Configure script
2018-06-06 12:02:40 +02:00
validation_default=auto
Add pairwise testing Pairwise testing is a test case generation technique based on the observation that most faults are caused by interactions of at most two factors. For BIND, its configure options can be thought of as such factors. Process BIND configure options into a model that is subsequently processed by the PICT tool in order to find an effective test vector. That test vector is then used for configuring and building BIND using various combinations of configure options.
2020-07-01 10:29:36 +02:00
# [pairwise: --enable-auto-validation, --disable-auto-validation]
Set "dnssec-validation auto" by default - the default setting for dnssec-validation is now "auto", which activates DNSSEC validation using the IANA root key. The old behavior can be restored by explicitly setting "dnssec-validation yes", which "yes", which activates DNSSEC validation only if keys are explicitly configured in named.conf. - the ARM has been updated to describe the new behavior
2018-05-16 11:58:15 -07:00
AC_ARG_ENABLE(auto-validation,
AS_HELP_STRING([--enable-auto-validation],
[turn on DNSSEC validation by default, using the IANA root key [default=yes]]),
Add VALIDATION_DEFAULT to Windows Configure script
2018-06-06 12:02:40 +02:00
[:],[enable_auto_validation=yes])
AS_IF([test "$enable_auto_validation" = "no"],[validation_default=yes])
AC_DEFINE_UNQUOTED([VALIDATION_DEFAULT], ["$validation_default"], [the default value of dnssec-validation option])
Set "dnssec-validation auto" by default - the default setting for dnssec-validation is now "auto", which activates DNSSEC validation using the IANA root key. The old behavior can be restored by explicitly setting "dnssec-validation yes", which "yes", which activates DNSSEC validation only if keys are explicitly configured in named.conf. - the ARM has been updated to describe the new behavior
2018-05-16 11:58:15 -07:00
Remove check for isc_port_t; Windows never have it, Unix always have it
2018-08-21 14:35:57 +02:00
#
Make dnsrps play well with dlopen
2018-08-25 13:15:06 +02:00
#
# Response policy rewriting using DNS Response Policy Service (DNSRPS)
# interface.
#
# DNSRPS can be compiled into BIND everywhere with a reasonably
# modern C compiler. It is enabled on systems with dlopen() and librpz.so.
#
dnsrps_avail=yes
AC_MSG_CHECKING([for librpz __attribute__s])
AC_COMPILE_IFELSE(
[AC_LANG_PROGRAM(
[[]],
[[
extern void f(char *p __attribute__((unused)), ...)
__attribute__((format(printf,1,2))) __attribute__((__noreturn__));
]])],
[
librpz_have_attr=yes
AC_DEFINE([LIBRPZ_HAVE_ATTR], [1], [have __attribute__s used in librpz.h])
AC_MSG_RESULT([yes])
],[
librpz_have_attr=no
AC_MSG_RESULT([no])
])
Add DNSRPS builds to pairwise testing The --enable-dnsrps-dl switch for ./configure enables preparing a DNSRPS-enabled build of BIND 9 that is not directly linked against a DNSRPS provider library (dlopen() at runtime is used instead). Employ this switch to test DNSRPS-enabled builds in the pairwise testing job in GitLab CI.
2023-02-01 11:22:59 +01:00
# [pairwise: --enable-dnsrps --enable-dnsrps-dl, --disable-dnsrps]
Make dnsrps play well with dlopen
2018-08-25 13:15:06 +02:00
AC_ARG_ENABLE([dnsrps-dl],
[AS_HELP_STRING([--enable-dnsrps-dl],
[DNS Response Policy Service delayed link
allow configuration of dnsrps library path for testing purposes, we need to be able to specify a library path from which to load the dnsrps implementation. this can now be done with the "dnsrps-library" option. DNSRPS can now be enabled in configure regardless of whether librpz.so is currently installed on the system.
2023-03-13 14:13:39 -07:00
[default=yes]])],
Replace enable_dnsprs_dl with enable_dnsrps_dl "enable_dnsprs_dl" is a typo where "enable_dnsrps_dl" was intended.
2023-03-29 13:12:50 +02:00
[enable_dnsrps_dl="$enableval"], [enable_dnsrps_dl="yes"])
Make dnsrps play well with dlopen
2018-08-25 13:15:06 +02:00
Replace enable_dnsprs_dl with enable_dnsrps_dl "enable_dnsprs_dl" is a typo where "enable_dnsrps_dl" was intended.
2023-03-29 13:12:50 +02:00
AS_IF([test "$enable_dnsrps_dl" = "yes" -a "$with_dlopen" = "no"],
Make dnsrps play well with dlopen
2018-08-25 13:15:06 +02:00
[AC_MSG_ERROR([DNS Response Policy Service delayed link requires dlopen to be enabled])])
Add pairwise testing Pairwise testing is a test case generation technique based on the observation that most faults are caused by interactions of at most two factors. For BIND, its configure options can be thought of as such factors. Process BIND configure options into a model that is subsequently processed by the PICT tool in order to find an effective test vector. That test vector is then used for configuring and building BIND using various combinations of configure options.
2020-07-01 10:29:36 +02:00
# [pairwise: skip]
Make dnsrps play well with dlopen
2018-08-25 13:15:06 +02:00
AC_ARG_WITH([dnsrps-libname],
[AS_HELP_STRING([--with-dnsrps-libname],
[DNSRPS provider library name (librpz.so)])],
[librpz_name="$withval"], [librpz_name="librpz.so"])
Add pairwise testing Pairwise testing is a test case generation technique based on the observation that most faults are caused by interactions of at most two factors. For BIND, its configure options can be thought of as such factors. Process BIND configure options into a model that is subsequently processed by the PICT tool in order to find an effective test vector. That test vector is then used for configuring and building BIND using various combinations of configure options.
2020-07-01 10:29:36 +02:00
# [pairwise: skip]
Make dnsrps play well with dlopen
2018-08-25 13:15:06 +02:00
AC_ARG_WITH([dnsrps-dir],
[AS_HELP_STRING([--with-dnsrps-dir],
[path to DNSRPS provider library])],
[librpz_path="$withval/$librpz_name"], [librpz_path="$librpz_name"])
AC_DEFINE_UNQUOTED([DNSRPS_LIBRPZ_PATH], ["$librpz_path"],
[dnsrps $librpz_name])
allow configuration of dnsrps library path for testing purposes, we need to be able to specify a library path from which to load the dnsrps implementation. this can now be done with the "dnsrps-library" option. DNSRPS can now be enabled in configure regardless of whether librpz.so is currently installed on the system.
2023-03-13 14:13:39 -07:00
AS_IF([test "$enable_dnsrps_dl" = "yes"],
Make dnsrps play well with dlopen
2018-08-25 13:15:06 +02:00
[
dnsrps_lib_open=2
],[
dnsrps_lib_open=1
# Add librpz.so to linked libraries if we are not using dlopen()
AC_SEARCH_LIBS([librpz_client_create], [rpz], [],
[dnsrps_lib_open=0
dnsrps_avail=no])
])
AC_DEFINE_UNQUOTED([DNSRPS_LIB_OPEN], [$dnsrps_lib_open],
[0=no DNSRPS 1=static link 2=dlopen()])
Add pairwise testing Pairwise testing is a test case generation technique based on the observation that most faults are caused by interactions of at most two factors. For BIND, its configure options can be thought of as such factors. Process BIND configure options into a model that is subsequently processed by the PICT tool in order to find an effective test vector. That test vector is then used for configuring and building BIND using various combinations of configure options.
2020-07-01 10:29:36 +02:00
# [pairwise: skip]
Make dnsrps play well with dlopen
2018-08-25 13:15:06 +02:00
AC_ARG_ENABLE([dnsrps],
AS_HELP_STRING([--enable-dnsrps],
[enable DNS Response Policy Service API]),
[enable_dnsrps=$enableval], [enable_dnsrps=no])
AS_IF([test "$enable_dnsrps" != "no"],[
AS_IF([test "$dnsrps_lib_open" = "0"],
[AC_MSG_ERROR([dlopen and librpz.so needed for DNSRPS])])
AC_DEFINE([USE_DNSRPS], [1], [Enable DNS Response Policy Service API])
])
import libdummyrpz test library for DNSRPS libdummyrpz is a limited version of the fastrpz library for use in testing the dnsrps API.
2023-03-13 12:47:00 -07:00
AM_CONDITIONAL([DNSRPS], [test "$enable_dnsrps" != "no"])
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AC_CHECK_HEADERS([glob.h])
1913. [func] Integrate contibuted DLZ code into named. [RT #11382]
2005-09-05 00:12:29 +00:00
Use library constructor/destructor to initialize OpenSSL Instead of calling isc_tls_initialize()/isc_tls_destroy() explicitly use gcc/clang attributes on POSIX and DLLMain on Windows to initialize and shutdown OpenSSL library. This resolves the issue when isc_nm_create() / isc_nm_destroy() was called multiple times and it would call OpenSSL library destructors from isc_nm_destroy(). At the same time, since we now have introduced the ctor/dtor for libisc, this commit moves the isc_mem API initialization (the list of the contexts) and changes the isc_mem_checkdestroyed() to schedule the checking of memory context on library unload instead of executing the code immediately.
2021-02-09 17:44:40 +01:00
#
# Support for constructor and destructor attributes
#
AX_GCC_FUNC_ATTRIBUTE([constructor])
AX_GCC_FUNC_ATTRIBUTE([destructor])
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
#
# Files to configure. These are listed here because we used to
# specify them as arguments to AC_OUTPUT.
#
1913. [func] Integrate contibuted DLZ code into named. [RT #11382]
2005-09-05 00:12:29 +00:00
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
# Top
Added a isc_glob() function that wraps glob() calls for POSIX systems and implement a custom glob() function on Windows systems.
2019-10-23 16:25:06 -03:00
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AC_CONFIG_FILES([Makefile])
1956. [bug] Improve cross compile support, 'gen' is now built by native compiler. See README for additional cross compile support information. [RT #15148]
2006-01-05 00:30:34 +00:00
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
# Binaries
2105. [func] GSS-TSIG support (RFC 3645).
2006-12-04 01:54:53 +00:00
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AC_CONFIG_FILES([bin/Makefile
bin/named/Makefile
bin/rndc/Makefile
bin/dig/Makefile
bin/delv/Makefile
bin/dnssec/Makefile
bin/tools/Makefile
bin/nsupdate/Makefile
bin/check/Makefile
bin/confgen/Makefile
bin/plugins/Makefile])
1913. [func] Integrate contibuted DLZ code into named. [RT #11382]
2005-09-05 00:12:29 +00:00
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
# Libraries
1856. [doc] Switch Docbook toolchain from DSSSL to XSL.
2005-05-11 05:55:41 +00:00
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AC_CONFIG_FILES([lib/Makefile
lib/isc/Makefile
lib/dns/Makefile
lib/ns/Makefile
lib/isccfg/Makefile
Remove libbind9 It is now empty.
2022-12-16 10:51:36 +00:00
lib/isccc/Makefile])
1856. [doc] Switch Docbook toolchain from DSSSL to XSL.
2005-05-11 05:55:41 +00:00
Convert the documentation to Sphinx documentation format The ARM and the manpages have been converted into Sphinx documentation format. Sphinx uses reStructuredText as its markup language, and many of its strengths come from the power and straightforwardness of reStructuredText and its parsing and translating suite, the Docutils.
2019-03-31 12:49:47 +02:00
# Documentation
AC_CONFIG_FILES([doc/Makefile
doc/arm/Makefile
doc/man/Makefile
doc/misc/Makefile])
Tests and benchmark for isc_ascii The test is to verify basic functionality. The benchmark compares a number of alternative tolower() implementations on large and small strings.
2022-06-25 15:29:54 +01:00
# Unit tests and benchmarks
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
Move all the unit tests to /tests/<libname>/ The unit tests are now using a common base, which means that lib/dns/tests/ code now has to include lib/isc/include/isc/test.h and link with lib/isc/test.c and lib/ns/tests has to include both libisc and libdns parts. Instead of cross-linking code between the directories, move the /lib/<foo>/test.c to /tests/<foo>.c and /lib/<foo>/include/<foo>test.h to /tests/include/tests/<foo>.h and create a single libtest.la convenience library in /tests/. At the same time, move the /lib/<foo>/tests/ to /tests/<foo>/ (but keep it symlinked to the old location) and adjust paths accordingly. In few places, we are now using absolute paths instead of relative paths, because the directory level has changed. By moving the directories under the /tests/ directory, the test-related code is kept in a single place and we can avoid referencing files between libns->libdns->libisc which is unhealthy because they live in a separate Makefile-space. In the future, the /bin/tests/ should be merged to /tests/ and symlink kept, and the /fuzz/ directory moved to /tests/fuzz/.
2022-05-03 11:37:31 +02:00
AC_CONFIG_FILES([tests/Makefile
Tests and benchmark for isc_ascii The test is to verify basic functionality. The benchmark compares a number of alternative tolower() implementations on large and small strings.
2022-06-25 15:29:54 +01:00
tests/bench/Makefile
Move all the unit tests to /tests/<libname>/ The unit tests are now using a common base, which means that lib/dns/tests/ code now has to include lib/isc/include/isc/test.h and link with lib/isc/test.c and lib/ns/tests has to include both libisc and libdns parts. Instead of cross-linking code between the directories, move the /lib/<foo>/test.c to /tests/<foo>.c and /lib/<foo>/include/<foo>test.h to /tests/include/tests/<foo>.h and create a single libtest.la convenience library in /tests/. At the same time, move the /lib/<foo>/tests/ to /tests/<foo>/ (but keep it symlinked to the old location) and adjust paths accordingly. In few places, we are now using absolute paths instead of relative paths, because the directory level has changed. By moving the directories under the /tests/ directory, the test-related code is kept in a single place and we can avoid referencing files between libns->libdns->libisc which is unhealthy because they live in a separate Makefile-space. In the future, the /bin/tests/ should be merged to /tests/ and symlink kept, and the /fuzz/ directory moved to /tests/fuzz/.
2022-05-03 11:37:31 +02:00
tests/isc/Makefile
tests/dns/Makefile
tests/ns/Makefile
tests/irs/Makefile
Stop the unit tests from running twice Move the libtest code into a 'libtest' subdirectory and make it one of the SUBDIRS in the tests Makefile. having it at the top level required having "." as one of the subdirs, and that caused the unit tests to be executed twice.
2022-05-03 19:43:23 -07:00
tests/isccfg/Makefile
tests/libtest/Makefile])
Move all the unit tests to /tests/<libname>/ The unit tests are now using a common base, which means that lib/dns/tests/ code now has to include lib/isc/include/isc/test.h and link with lib/isc/test.c and lib/ns/tests has to include both libisc and libdns parts. Instead of cross-linking code between the directories, move the /lib/<foo>/test.c to /tests/<foo>.c and /lib/<foo>/include/<foo>test.h to /tests/include/tests/<foo>.h and create a single libtest.la convenience library in /tests/. At the same time, move the /lib/<foo>/tests/ to /tests/<foo>/ (but keep it symlinked to the old location) and adjust paths accordingly. In few places, we are now using absolute paths instead of relative paths, because the directory level has changed. By moving the directories under the /tests/ directory, the test-related code is kept in a single place and we can avoid referencing files between libns->libdns->libisc which is unhealthy because they live in a separate Makefile-space. In the future, the /bin/tests/ should be merged to /tests/ and symlink kept, and the /fuzz/ directory moved to /tests/fuzz/.
2022-05-03 11:37:31 +02:00
AC_CONFIG_FILES([tests/unit-test-driver.sh],
[chmod +x tests/unit-test-driver.sh])
Provide unit test driver This adds a unit test driver for BIND with Automake. It runs the unit test program provided as its sole command line argument and then looks for a core dump generated by that test program. If one is found, the driver prints the backtrace into the test log.
2020-04-03 11:26:02 +02:00
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
# System Tests
AC_CONFIG_FILES([bin/tests/Makefile
bin/tests/system/Makefile
bin/tests/system/conf.sh
bin/tests/system/dyndb/driver/Makefile
create system test with asynchronous plugin the test-async plugin uses ns_query_hookasync() at the NS_QUERY_DONE_SEND hook point to call an asynchronous function. the only effect is to change the query response code to "NOTIMP", so we can confirm that the hook ran and resumed correctly.
2020-11-19 17:58:45 -08:00
bin/tests/system/dlzexternal/driver/Makefile
import libdummyrpz test library for DNSRPS libdummyrpz is a limited version of the fastrpz library for use in testing the dnsrps API.
2023-03-13 12:47:00 -07:00
bin/tests/system/hooks/driver/Makefile
bin/tests/system/rpz/testlib/Makefile
])
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AC_CONFIG_FILES([bin/tests/system/ifconfig.sh],
[chmod +x bin/tests/system/ifconfig.sh])
Invoke pytest runner from run.sh Utilize developers' muscle memory to incentivize using the pytest runner instead of the legacy one. The script also serves as basic examples of how to run the pyest command to achieve the same results as the legacy runner. Invoking pytest directly should be the end goal, since it offers many potentially useful options (refer to pytest --help).
2023-01-13 16:32:39 +01:00
AC_CONFIG_FILES([bin/tests/system/legacy.run.sh],
[chmod +x bin/tests/system/legacy.run.sh])
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
AC_CONFIG_FILES([bin/tests/system/start.sh],
[chmod +x bin/tests/system/start.sh])
AC_CONFIG_FILES([bin/tests/system/stop.sh],
[chmod +x bin/tests/system/stop.sh])
Re-enable the fuzzing tests The fuzzing tests were temporarily disabled when the build system has been converted to automake. This commit restores the functionality to run the fuzzing tests as part of the `make check`. When the afl or libfuzzer is enabled via ./configure, it uses a custom LOG_DRIVER (fuzz/<fuzzer.sh>). Currently only libfuzzer.sh has been implemented that runs each fuzz test for 5 seconds each.
2020-07-31 15:20:56 +02:00
# Fuzz Tests
AC_CONFIG_FILES([fuzz/Makefile])
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
# Misc
AC_CONFIG_FILES([util/check-make-install])
1913. [func] Integrate contibuted DLZ code into named. [RT #11382]
2005-09-05 00:12:29 +00:00
#
# Do it
#
AC_OUTPUT
Have Emacs automatically use shell mode to edit configure.in.
2000-07-04 01:20:53 +00:00
Add pairwise testing Pairwise testing is a test case generation technique based on the observation that most faults are caused by interactions of at most two factors. For BIND, its configure options can be thought of as such factors. Process BIND configure options into a model that is subsequently processed by the PICT tool in order to find an effective test vector. That test vector is then used for configuring and building BIND using various combinations of configure options.
2020-07-01 10:29:36 +02:00
# [pairwise: --enable-full-report, --disable-full-report]
[master] add configure summary 3615. [cleanup] "configure" now finishes by printing a summary of optional BIND features and whether they are active or inactive. ("configure --enable-full-report" increases the verbosity of the summary.) [RT #31777]
2013-07-10 20:54:08 -07:00
AC_ARG_ENABLE(full-report,
4814. [cleanup] Use AS_HELP_STRING for consistent help text. [RT #46521]
2017-11-13 11:36:45 +11:00
AS_HELP_STRING([--enable-full-report],
[report values of all configure options]))
[master] add configure summary 3615. [cleanup] "configure" now finishes by printing a summary of optional BIND features and whether they are active or inactive. ("configure --enable-full-report" increases the verbosity of the summary.) [RT #31777]
2013-07-10 20:54:08 -07:00
[master] some output was not silenced with configure -q 4589. [cleanup] "configure -q" is now silent. [RT #44829]
2017-04-20 15:09:28 -07:00
report() {
echo "==============================================================================="
echo "Configuration summary:"
echo "-------------------------------------------------------------------------------"
echo "Optional features enabled:"
Use system allocator when jemalloc is unavailable This commit adds support for systems where the jemalloc library is not available as a package, here's the quick summary: * On Linux - the jemalloc is usually available as a package, if configured --without-jemalloc, the shim would be used around malloc(), free(), realloc() and malloc_usable_size() * On macOS - the jemalloc is available from homebrew or macports, if configured --without-jemalloc, the shim would be used around malloc(), free(), realloc() and malloc_size() * On FreeBSD - the jemalloc is *the* system allocator, we just need to check for <malloc_np.h> header to get access to non-standard API * On NetBSD - the jemalloc is *the* system allocator, we just need to check for <jemalloc/jemalloc.h> header to get access to non-standard API * On a system hostile to users and developers (read OpenBSD) - the jemalloc API is emulated by using ((size_t *)ptr)[-1] field to hold the size information. The OpenBSD developers care only for themselves, so why should we care about speed on OpenBSD?
2021-05-25 12:46:00 +02:00
if test "yes" = "$with_jemalloc"; then
echo " Memory allocator: jemalloc"
fi
minor cosmetic fixes - the configuration summary reported zlib compression was not supported even when it was. - when bind.keys.h was regenerated it violated clang-format style.
2020-02-19 00:17:03 -08:00
if test "yes" = "$enable_full_report" -o "standard" = "$with_locktype"; then
echo " Mutex lock type: $with_locktype"
[master] some output was not silenced with configure -q 4589. [cleanup] "configure -q" is now silent. [RT #44829]
2017-04-20 15:09:28 -07:00
fi
fixed incorrect output from the configuration summary
2020-04-23 14:29:07 -07:00
test "no" = "$enable_dnstap" || \
4814. [cleanup] Use AS_HELP_STRING for consistent help text. [RT #46521]
2017-11-13 11:36:45 +11:00
echo " Allow 'dnstap' packet logging (--enable-dnstap)"
use configure --enable-geoip to turn on GeoIP2 - instead of --with-geoip2, use --enable-geoip to turn on the GeoIP feature - --with-maxminddb can be used to set the prefix for maxminddb
2019-06-27 21:19:30 -07:00
test -z "$MAXMINDDB_LIBS" || echo " GeoIP2 access control (--enable-geoip)"
fixed incorrect output from the configuration summary
2020-04-23 14:29:07 -07:00
test -z "$GSSAPI_LIBS" || echo " GSS-API (--with-gssapi)"
[master] address several issues with native pkcs11
2014-01-18 11:51:07 -08:00
Remove check for isc_port_t; Windows never have it, Unix always have it
2018-08-21 14:35:57 +02:00
# these lines are only printed if run with --enable-full-report
[master] some output was not silenced with configure -q 4589. [cleanup] "configure -q" is now silent. [RT #44829]
2017-04-20 15:09:28 -07:00
if test "yes" = "$enable_full_report"; then
fixed incorrect output from the configuration summary
2020-04-23 14:29:07 -07:00
test -z "$LIBXML2_LIBS" || echo " XML statistics (--with-libxml2)"
test -z "$JSON_C_LIBS" || echo " JSON statistics (--with-json-c): $JSON_C_CFLAGS $JSON_C_LIBS"
test -z "$ZLIB_LIBS" || echo " HTTP zlib compression (--with-zlib)"
test -z "$LMDB_LIBS" || echo " LMDB database to store configuration for 'addzone' zones (--with-lmdb)"
test -z "$LIBIDN2_LIBS" || echo " IDN support (--with-libidn2)"
[master] some output was not silenced with configure -q 4589. [cleanup] "configure -q" is now silent. [RT #44829]
2017-04-20 15:09:28 -07:00
fi
[master] address several issues with native pkcs11
2014-01-18 11:51:07 -08:00
[master] DNS Response Policy Service API 4713. [func] Added support for the DNS Response Policy Service (DNSRPS) API, which allows named to use an external response policy daemon when built with "configure --enable-dnsrps". Thanks to Vernon Schryver and Farsight Security. [RT #43376]
2017-09-11 11:53:42 -07:00
test "yes" = "$enable_dnsrps" && \
4814. [cleanup] Use AS_HELP_STRING for consistent help text. [RT #46521]
2017-11-13 11:36:45 +11:00
echo " DNS Response Policy Service interface (--enable-dnsrps)"
fixed incorrect output from the configuration summary
2020-04-23 14:29:07 -07:00
test "yes" = "$enable_fixed_rrset" && \
4814. [cleanup] Use AS_HELP_STRING for consistent help text. [RT #46521]
2017-11-13 11:36:45 +11:00
echo " Allow 'fixed' rrset-order (--enable-fixed-rrset)"
add --enable-fips-mode to configure summary Report when FIPS mode is enabled or disabled in the configure summary output.
2021-10-13 16:50:14 -07:00
test "yes" = "$enable_fips_mode" && \
echo " FIPS mode in OpenSSL (--enable-fips-mode)"
fixed incorrect output from the configuration summary
2020-04-23 14:29:07 -07:00
test "yes" = "$enable_querytrace" && \
4814. [cleanup] Use AS_HELP_STRING for consistent help text. [RT #46521]
2017-11-13 11:36:45 +11:00
echo " Very verbose query trace logging (--enable-querytrace)"
add a quick-and-dirty method of debugging a single query when built with "configure --enable-singletrace", named will produce detailed query logging at the highest debug level for any query with query ID zero. this enables monitoring of the progress of a single query by specifying the QID using "dig +qid=0". the "client" logging category should be set to a low severity level to suppress logging of other queries. (the chance of another query using QID=0 at the same time is only 1 in 2^16.) "--enable-singletrace" turns on "--enable-querytrace" as well, so if the logging severity is not lowered, all other queries will be logged verbosely as well. compiling with either of these options will impair query performance; they should only be turned on when testing or troubleshooting.
2020-05-15 16:37:44 -07:00
test "yes" = "$enable_singletrace" && \
echo " Single-query trace logging (--enable-singletrace)"
fixed incorrect output from the configuration summary
2020-04-23 14:29:07 -07:00
test -z "$HAVE_CMOCKA" || echo " CMocka Unit Testing Framework (--with-cmocka)"
[master] some output was not silenced with configure -q 4589. [cleanup] "configure -q" is now silent. [RT #44829]
2017-04-20 15:09:28 -07:00
configure summary failed to report --disable-auto-validation correctly
2019-04-04 10:43:33 -07:00
test "auto" = "$validation_default" && echo " DNSSEC validation active by default (--enable-auto-validation)"
Cleanup the configure feature report
2018-06-22 13:47:45 +02:00
[master] add configured prefixes to summary 4746. [cleanup] Add configured prefixes to configure summary output. [RT #46153]
2017-09-30 10:01:01 -07:00
echo "-------------------------------------------------------------------------------"
[master] some output was not silenced with configure -q 4589. [cleanup] "configure -q" is now silent. [RT #44829]
2017-04-20 15:09:28 -07:00
echo "Features disabled or unavailable on this platform:"
Use system allocator when jemalloc is unavailable This commit adds support for systems where the jemalloc library is not available as a package, here's the quick summary: * On Linux - the jemalloc is usually available as a package, if configured --without-jemalloc, the shim would be used around malloc(), free(), realloc() and malloc_usable_size() * On macOS - the jemalloc is available from homebrew or macports, if configured --without-jemalloc, the shim would be used around malloc(), free(), realloc() and malloc_size() * On FreeBSD - the jemalloc is *the* system allocator, we just need to check for <malloc_np.h> header to get access to non-standard API * On NetBSD - the jemalloc is *the* system allocator, we just need to check for <jemalloc/jemalloc.h> header to get access to non-standard API * On a system hostile to users and developers (read OpenBSD) - the jemalloc API is emulated by using ((size_t *)ptr)[-1] field to hold the size information. The OpenBSD developers care only for themselves, so why should we care about speed on OpenBSD?
2021-05-25 12:46:00 +02:00
if test "no" = "$with_jemalloc"; then
echo " Memory allocator: system"
Tweak jemalloc warning text in configure.ac
2021-08-10 12:20:52 +02:00
echo
echo " +------------------------------------------+"
echo " | ==== WARNING ==== |"
echo " | |"
echo " | This is NOT a recommended configuration. |"
echo " | Using the system memory allocator causes |"
echo " | reduced performance and increased memory |"
echo " | fragmentation. Installing the jemalloc |"
echo " | memory allocator (version >= 4.0.0) is |"
echo " | strongly recommended. |"
echo " +------------------------------------------+"
echo
Use system allocator when jemalloc is unavailable This commit adds support for systems where the jemalloc library is not available as a package, here's the quick summary: * On Linux - the jemalloc is usually available as a package, if configured --without-jemalloc, the shim would be used around malloc(), free(), realloc() and malloc_usable_size() * On macOS - the jemalloc is available from homebrew or macports, if configured --without-jemalloc, the shim would be used around malloc(), free(), realloc() and malloc_size() * On FreeBSD - the jemalloc is *the* system allocator, we just need to check for <malloc_np.h> header to get access to non-standard API * On NetBSD - the jemalloc is *the* system allocator, we just need to check for <jemalloc/jemalloc.h> header to get access to non-standard API * On a system hostile to users and developers (read OpenBSD) - the jemalloc API is emulated by using ((size_t *)ptr)[-1] field to hold the size information. The OpenBSD developers care only for themselves, so why should we care about speed on OpenBSD?
2021-05-25 12:46:00 +02:00
fi
[master] some output was not silenced with configure -q 4589. [cleanup] "configure -q" is now silent. [RT #44829]
2017-04-20 15:09:28 -07:00
fixed incorrect output from the configuration summary
2020-04-23 14:29:07 -07:00
test "no" = "$enable_dnstap" && \
4814. [cleanup] Use AS_HELP_STRING for consistent help text. [RT #46521]
2017-11-13 11:36:45 +11:00
echo " Allow 'dnstap' packet logging (--enable-dnstap)"
use configure --enable-geoip to turn on GeoIP2 - instead of --with-geoip2, use --enable-geoip to turn on the GeoIP feature - --with-maxminddb can be used to set the prefix for maxminddb
2019-06-27 21:19:30 -07:00
test -z "$MAXMINDDB_LIBS" && echo " GeoIP2 access control (--enable-geoip)"
fixed incorrect output from the configuration summary
2020-04-23 14:29:07 -07:00
test -z "$GSSAPI_LIBS" && echo " GSS-API (--with-gssapi)"
[master] some output was not silenced with configure -q 4589. [cleanup] "configure -q" is now silent. [RT #44829]
2017-04-20 15:09:28 -07:00
[master] DNS Response Policy Service API 4713. [func] Added support for the DNS Response Policy Service (DNSRPS) API, which allows named to use an external response policy daemon when built with "configure --enable-dnsrps". Thanks to Vernon Schryver and Farsight Security. [RT #43376]
2017-09-11 11:53:42 -07:00
test "no" = "$enable_dnsrps" && \
4814. [cleanup] Use AS_HELP_STRING for consistent help text. [RT #46521]
2017-11-13 11:36:45 +11:00
echo " DNS Response Policy Service interface (--enable-dnsrps)"
[master] DNS Response Policy Service API 4713. [func] Added support for the DNS Response Policy Service (DNSRPS) API, which allows named to use an external response policy daemon when built with "configure --enable-dnsrps". Thanks to Vernon Schryver and Farsight Security. [RT #43376]
2017-09-11 11:53:42 -07:00
fixed incorrect output from the configuration summary
2020-04-23 14:29:07 -07:00
test "yes" = "$enable_fixed_rrset" || \
4814. [cleanup] Use AS_HELP_STRING for consistent help text. [RT #46521]
2017-11-13 11:36:45 +11:00
echo " Allow 'fixed' rrset-order (--enable-fixed-rrset)"
[master] some output was not silenced with configure -q 4589. [cleanup] "configure -q" is now silent. [RT #44829]
2017-04-20 15:09:28 -07:00
fixed incorrect output from the configuration summary
2020-04-23 14:29:07 -07:00
test "yes" = "$validation_default" && echo " DNSSEC validation requires configuration (--enable-auto-validation)"
Set "dnssec-validation auto" by default - the default setting for dnssec-validation is now "auto", which activates DNSSEC validation using the IANA root key. The old behavior can be restored by explicitly setting "dnssec-validation yes", which "yes", which activates DNSSEC validation only if keys are explicitly configured in named.conf. - the ARM has been updated to describe the new behavior
2018-05-16 11:58:15 -07:00
add --enable-fips-mode to configure summary Report when FIPS mode is enabled or disabled in the configure summary output.
2021-10-13 16:50:14 -07:00
test "yes" = "$enable_fips_mode" || \
echo " FIPS mode in OpenSSL (--enable-fips-mode)"
fixed incorrect output from the configuration summary
2020-04-23 14:29:07 -07:00
test "yes" = "$enable_querytrace" || \
4814. [cleanup] Use AS_HELP_STRING for consistent help text. [RT #46521]
2017-11-13 11:36:45 +11:00
echo " Very verbose query trace logging (--enable-querytrace)"
add a quick-and-dirty method of debugging a single query when built with "configure --enable-singletrace", named will produce detailed query logging at the highest debug level for any query with query ID zero. this enables monitoring of the progress of a single query by specifying the QID using "dig +qid=0". the "client" logging category should be set to a low severity level to suppress logging of other queries. (the chance of another query using QID=0 at the same time is only 1 in 2^16.) "--enable-singletrace" turns on "--enable-querytrace" as well, so if the logging severity is not lowered, all other queries will be logged verbosely as well. compiling with either of these options will impair query performance; they should only be turned on when testing or troubleshooting.
2020-05-15 16:37:44 -07:00
test "yes" = "$enable_singletrace" || \
echo " Single-query trace logging (--enable-singletrace)"
[master] some output was not silenced with configure -q 4589. [cleanup] "configure -q" is now silent. [RT #44829]
2017-04-20 15:09:28 -07:00
Fix typo
2019-02-06 12:08:34 +01:00
test "no" = "$with_cmocka" && echo " CMocka Unit Testing Framework (--with-cmocka)"
[master] some output was not silenced with configure -q 4589. [cleanup] "configure -q" is now silent. [RT #44829]
2017-04-20 15:09:28 -07:00
fixed incorrect output from the configuration summary
2020-04-23 14:29:07 -07:00
test -z "$LIBXML2_LIBS" && echo " XML statistics (--with-libxml2)"
test -z "$JSON_C_LIBS" && echo " JSON statistics (--with-json-c)"
test -z "$ZLIB_LIBS" && echo " HTTP zlib compression (--with-zlib)"
test -z "$LMDB_LIBS" && echo " LMDB database to store configuration for 'addzone' zones (--with-lmdb)"
test -z "$LIBIDN2_LIBS" && echo " IDN support (--with-libidn2)"
[master] some output was not silenced with configure -q 4589. [cleanup] "configure -q" is now silent. [RT #44829]
2017-04-20 15:09:28 -07:00
[master] add configured prefixes to summary 4746. [cleanup] Add configured prefixes to configure summary output. [RT #46153]
2017-09-30 10:01:01 -07:00
echo "-------------------------------------------------------------------------------"
echo "Configured paths:"
echo " prefix: $prefix"
echo " sysconfdir: $sysconfdir"
echo " localstatedir: $localstatedir"
report "$CC --version" output as part of the configure summary
2018-09-07 10:59:04 -07:00
echo "-------------------------------------------------------------------------------"
echo "Compiler: $CC"
Use correct compiler version flag in the autoconf script The autoconf script prints used compiler version at the end of the configure script. Solaris native compiler doesn't support --version, and -V has to be used which in turn isn't supported by Gcc/Clang. Detect which version flag has to be used and call $CC with it.
2021-10-11 21:18:49 +02:00
AS_IF([test "$GCC" = "yes"],
[$CC --version 2>&1 | sed 's/^/ /'],
[AS_CASE([$host],
[*-solaris*],[$CC -V 2>&1 | sed 's/^/ /'],
[$CC --version 2>&1 | sed 's/^/ /'])])
Complete rewrite the BIND 9 build system The rewrite of BIND 9 build system is a large work and cannot be reasonable split into separate merge requests. Addition of the automake has a positive effect on the readability and maintainability of the build system as it is more declarative, it allows conditional and we are able to drop all of the custom make code that BIND 9 developed over the years to overcome the deficiencies of autoconf + custom Makefile.in files. This squashed commit contains following changes: - conversion (or rather fresh rewrite) of all Makefile.in files to Makefile.am by using automake - the libtool is now properly integrated with automake (the way we used it was rather hackish as the only official way how to use libtool is via automake - the dynamic module loading was rewritten from a custom patchwork to libtool's libltdl (which includes the patchwork to support module loading on different systems internally) - conversion of the unit test executor from kyua to automake parallel driver - conversion of the system test executor from custom make/shell to automake parallel driver - The GSSAPI has been refactored, the custom SPNEGO on the basis that all major KRB5/GSSAPI (mit-krb5, heimdal and Windows) implementations support SPNEGO mechanism. - The various defunct tests from bin/tests have been removed: bin/tests/optional and bin/tests/pkcs11 - The text files generated from the MD files have been removed, the MarkDown has been designed to be readable by both humans and computers - The xsl header is now generated by a simple sed command instead of perl helper - The <irs/platform.h> header has been removed - cleanups of configure.ac script to make it more simpler, addition of multiple macros (there's still work to be done though) - the tarball can now be prepared with `make dist` - the system tests are partially able to run in oot build Here's a list of unfinished work that needs to be completed in subsequent merge requests: - `make distcheck` doesn't yet work (because of system tests oot run is not yet finished) - documentation is not yet built, there's a different merge request with docbook to sphinx-build rst conversion that needs to be rebased and adapted on top of the automake - msvc build is non functional yet and we need to decide whether we will just cross-compile bind9 using mingw-w64 or fix the msvc build - contributed dlz modules are not included neither in the autoconf nor automake
2018-08-07 16:46:53 +02:00
echo "CFLAGS: $STD_CFLAGS $CFLAGS"
echo "CPPFLAGS: $STD_CPPFLAGS $CPPFLAGS"
echo "LDFLAGS: $LDFLAGS"
[master] add configured prefixes to summary 4746. [cleanup] Add configured prefixes to configure summary output. [RT #46153]
2017-09-30 10:01:01 -07:00
[master] some output was not silenced with configure -q 4589. [cleanup] "configure -q" is now silent. [RT #44829]
2017-04-20 15:09:28 -07:00
if test "X$ac_unrecognized_opts" != "X"; then
add a quick-and-dirty method of debugging a single query when built with "configure --enable-singletrace", named will produce detailed query logging at the highest debug level for any query with query ID zero. this enables monitoring of the progress of a single query by specifying the QID using "dig +qid=0". the "client" logging category should be set to a low severity level to suppress logging of other queries. (the chance of another query using QID=0 at the same time is only 1 in 2^16.) "--enable-singletrace" turns on "--enable-querytrace" as well, so if the logging severity is not lowered, all other queries will be logged verbosely as well. compiling with either of these options will impair query performance; they should only be turned on when testing or troubleshooting.
2020-05-15 16:37:44 -07:00
echo "-------------------------------------------------------------------------------"
4814. [cleanup] Use AS_HELP_STRING for consistent help text. [RT #46521]
2017-11-13 11:36:45 +11:00
echo "Unrecognized options:"
echo " $ac_unrecognized_opts"
[master] some output was not silenced with configure -q 4589. [cleanup] "configure -q" is now silent. [RT #44829]
2017-04-20 15:09:28 -07:00
fi
[master] add configured prefixes to summary 4746. [cleanup] Add configured prefixes to configure summary output. [RT #46153]
2017-09-30 10:01:01 -07:00
use 'test "constant" <condition> "$variable"' [RT #45486]
2017-07-19 15:53:01 +10:00
if test "yes" != "$enable_full_report"; then
4814. [cleanup] Use AS_HELP_STRING for consistent help text. [RT #46521]
2017-11-13 11:36:45 +11:00
echo "-------------------------------------------------------------------------------"
echo "For more detail, use --enable-full-report."
[master] some output was not silenced with configure -q 4589. [cleanup] "configure -q" is now silent. [RT #44829]
2017-04-20 15:09:28 -07:00
fi
echo "==============================================================================="
}
use 'test "constant" <condition> "$variable"' [RT #45486]
2017-07-19 15:53:01 +10:00
if test "yes" != "$silent"; then
4814. [cleanup] Use AS_HELP_STRING for consistent help text. [RT #46521]
2017-11-13 11:36:45 +11:00
report
[master] mention --enable-full-report in config summary output
2015-12-02 14:02:53 -08:00
fi
[master] add configure summary 3615. [cleanup] "configure" now finishes by printing a summary of optional BIND features and whether they are active or inactive. ("configure --enable-full-report" increases the verbosity of the summary.) [RT #31777]
2013-07-10 20:54:08 -07:00
Have Emacs automatically use shell mode to edit configure.in.
2000-07-04 01:20:53 +00:00
# Tell Emacs to edit this file in shell mode.
# Local Variables:
# mode: sh
# End:
Reference in New Issue Copy Permalink