Fix detection of old Solaris versions. Autotools need double square brackets.

Fix --enable-v4-mapped configuration on *BSD. PR 53824. Submitted by: olli hauer <ohauer gmx.de> Committed by: ylavic

* configure.in, acinclude.m4: Move systemd check to APACHE_CHECK_SYSTEMD and use pkg-config where available.

Fix syntax error introduced in r1612921.

Check for correct minimum PCRE version in configure, do not check in source code.

Fix PCRE_DUPNAMES check to find the header file for PCRE when PCRE isn't in the default includes path.

Check if PCRE_DUPNAMES exists at configure time

Allow user to explicitly determine whether or not BSD syntax makefiles should be created... The default is NO unless we are on a *BSD* system, in which case we actually see if 'make' is gmake or not.

For *bsd* systems, see if we are using BSD make and, if so, then use that format for include/ifdef/else/endif

FreeBSD: Disable IPv4-mapped listening sockets by default for versions 5+ instead of just for FreeBSD 5. PR: 53824

Don't link support/* against systemd libs. * Makefile.in: Link httpd using $(HTTPD_LIBS). * configure.in: Add systemd to HTTPD_LIBS rather than LIBS. Add PCRE libs to HTTPD_LIBS.

* configure.in: Simplify/fix systemd detection: move later to fix autoconf warnings; define HAVE_SYSTEMD rather than using CPPFLAGS. * server/listen.c: Use HAVE_SYSTEMD for systemd #define. * modules/arch/unix/config5.m4: Update systemd headers check.

Add support for systemd socket activation to listener sockets.

maintainer-mode should auto-enable mod_bucketeer

core: Support the SINGLE_LISTEN_UNSERIALIZED_ACCEPT optimization on Linux kernel versions 3.x and above. PR: 55121 Submitted by: Bradley Heilbrun <apache heilbrun.org> Reviewed by: trawick

Avoid valgrind warnings in mod_ssl random number generator We intentionally add uninitialized stack memory. To avoid warnings, make valgrind believe that the memory is defined. Add configure option to enable valgrind support

follow-up to r1389481: find/use getloadavg() prototype on Solaris

configure: Fix processing of --disable-FEATURE for various features.

Would be nice to have some sort of canonical definition of server loading for Apache. So create a struct that holds some useful data. The hope is that for those platforms that lack getloadavg(), people will write replacements.

consolidate hints/instructions about using APR (and maybe APR-util) with httpd

Spin off module-specific build options into separate build vars. [second try, supersedes r1385214] Add MOD_CFLAGS, MOD_LDFLAGS variables etc. to the build commands, which are available to modules for customization on a per-subdir basis (by adding definitions to modules.mk). Reduces the risk of side-effects when a module needs to add CFLAGS, LDFLAGS etc. and these would be added to the global settings (ALL_CFLAGS etc.). Adapt build settings for mod_ssl, mod_socache_dc, mod_deflate, mod_xml2enc, mod_proxy_html, and mod_lua to use the new MOD_xxx build variables. Change PICFLAGS, SHLTCFLAGS and LTCFLAGS into config vars, instead of AC_SUBSTing them in build/rules.mk.in. For support/ab, introduce ab_CFLAGS and ab_LDFLAGS, and define explicit make targets where they appear at the proper position in the build commands. Consistently use "--with-xxx=PATH" in configure help strings which are used to specify a path to the installation directory of an auxiliary package.

revert r1385214, as I inadvertently left out acinclude.m4.

Spin off module-specific build options into separate build vars. Add MOD_CFLAGS, MOD_LDFLAGS variables etc. to the build commands, which are available to modules for customization on a per-subdir basis (by adding definitions to modules.mk). Reduces the risk of side-effects when a module needs to add CFLAGS, LDFLAGS etc. and these would be added to the global settings (ALL_CFLAGS etc.). Adapt build settings for mod_ssl, mod_socache_dc, mod_deflate, mod_xml2enc, mod_proxy_html, and mod_lua to use the new MOD_xxx build variables. Change PICFLAGS, SHLTCFLAGS and LTCFLAGS into config vars, instead of AC_SUBSTing them in build/rules.mk.in. For support/ab, introduce ab_CFLAGS and ab_LDFLAGS, and define explicit make targets where they appear at the proper position in the build commands. Consistently use "--with-xxx=PATH" in configure help strings which are used to specify a path to the installation directory of an auxiliary package.

fix APREQ declarations for MinGW builds

* configure.in: Enforce 1.4.0 as minimum APR version requirement. PR: 52748

* configure.in: Correct handling of --with-suexec-logfile in r1341905, thanks to rpluem.

Make ap_regcomp() return AP_REG_ESPACE if out of memory. Make ap_pregcomp() abort if out of memory. This raises the minimum PCRE requirement to version 6.0, released in 2005.

suexec: Support use of setgid/setuid capability bits on Linux, a weaker set of privileges than the full setuid/setgid root binary. * configure.in: Add --enable-suexec-capabilites flag. * Makefile.in: If configured, use setcap instead of chmod 7555 on installed suexec binary. * modules/arch/unix/mod_unixd.c (unixd_pre_config): Drop test for setuid bit if capability bits are used. * docs/manual/: Add docs.

* configure.in: Use AC_MSG_NOTICE so passing --silent works as advertised.

suexec: Add support for logging to syslog as an alternative to a logfile. * support/suexec.c (err_output) [AP_LOG_SYSLOG]: Log to syslog. (main): Close syslog fd if open, before execv. Add -V output for AP_LOG_SYSLOG. * configure.in: Add --with-suexec-syslog argument; allow --without-suexec-logfile to omit definition of AP_LOG_EXEC.

Final step to make cross compilation working. In case of cross compilation set CC_FOR_BUILD to cc unless we got already CC_FOR_BUILD from environment.

To early, not enough drugs, sorry for the noise.

Fix out of tree build with srclib apr. Fix error message for included apr flag. Forward port of r1241897 and r1243670 from 2.4.x plus one shell syntax fix. Will backport the fix to 2.4.x.

configure: Move the --with-included-apr check so that srcdir!=builddir builds aren't broken by it.

configure: Improve the error message when --with-included-apr is specified but either APR or APR-Util is not included.

mod_policy: Add a new testing module to help server administrators enforce a configurable level of protocol compliance on their servers and application servers behind theirs.

libapreq is not a bunch of C files now, no special configure magic required

Set --enable-load-all-modules implicitely in maintainer mode unless disabled explicitely.

A stab at libapreq build integration, still disabled by default.

disable libapreq stuff until it works generically

have libapreq install honor the top level layout options

hook up srclib/libapreq to the build system

Add default proxy-html.conf file (defines links to be rewritten according to W3C HTML 4 and XHTML 1).

Load all modules in the generated default configuration when using --enable-load-all-modules with configure.

By default, only load those modules that are either required or explicitly selected by a configure --enable-foo argument. The LoadModule statements for modules enabled by --enable-mods-shared=most and friends will be commented out.

Add some CFLAGS in maintainer mode if supported by gcc: -Wformat -Wformat-security -Werror=format-security -Werror=declaration-after-statement Add new macro APACHE_ADD_GCC_CFLAGS to do the compatibility check.

Merge branch revert-ap-ldap: Revert ap_ldap integration due to veto by Graham Leggett Mailing list threads: http://mail-archives.apache.org/mod_mbox/httpd-dev/201106.mbox/%3C4192DC1D-C0B9-42BB-B614-C3A41290F18B@sharp.fm%3E http://mail-archives.apache.org/mod_mbox/httpd-dev/201107.mbox/%3C4E15E51E.4090700@rowe-clan.net%3E

Move PCRE_LIBS from HTTPD_LDFLAGS to PROGRAM_LDADD. Otherwise -lpcre comes before libmain which leads to undefined recences during linking of httpd. Followup to r1142739.

Line already contained in aclocal.m4.

Only link the httpd binary against PCRE. No other support binary needs PCRE.

Add support to ErrorLogFormat for logging the system unique thread id under Linux

Several fixes for the ap_ldap build logic. This should fix compilation without --with-ldap.

Incorporate the ap_ldap incomplete API, as there is no interest or effort at APR to make this a complete abstraction, and it was voted 'off the island' with APR 2.0. This will allow httpd 2.3 to build against either apr-2.0 or apr+util 1.x.

Silence autoconf 2.68 warnings. Add AC_LANG_SOURCE to the two occurences of AC_COMPILE_IFELSE in configure.in.

keep -DAP_HOOK_PROBES_ENABLED (--enable-hook-probes) and -DAP_DECLARE_EXPORT (MinGW builds) out of the *CPPFLAGS used for third-party module builds introduce INTERNAL_CPPFLAGS for that; apxs bypasses this

Use AP_ instead of APR_ and move into the compile CPP flags

Start of ap_hooks.h setup

move -L/path/to/pcrelib -lpcre out beyond all the LDFLAGS this fixes unresolved symbol errors on on MinGW

AP_ENABLE_V4_MAPPED with WinNT MPM mpm_winnt.c: Fail to compile if enabled, and add a hint about the issue to address before it will work. configure.in: Move the --enable-v4-mapped check to below the MPM logic, and default to --disable-v4-mapped if using the WinNT MPM

Allow apr and apr-util to be in 2 different location. Otherwise: apr_version.h is not found when apr-util is tested.

Fix link errors for crypt() with --no-add-needed/--no-copy-dt-needed-entries, which is the default for the gold linker.

revert r1066944, r1067177 for now breaks MacOS X and wherever this code will go, server/main.c does not seem to be the right place. See http://mail-archives.apache.org/mod_mbox/httpd-dev/201102.mbox/%3C01930193-8A3B-4635-9EB0-5EE31349E49C@jaguNET.com%3E

Reload resolv.conf on graceful restarts PR: 50619 Submitted by: Matt Miller <m miller f5 com>, Stefan Fritsch

A new mode that is maintainer + O0 (for better gdb)

Move some "nice to be able to change" balancer stuff to shm

Mkae gcc no longer silent when doing arith on void*...

Use -Wdeclaration-after-statement in maintainer mode if we use a recent enough gcc.

Support large log files in suexec PR: 45856

prevent an error message if an old autoconf without AC_PROG_CC_C99 is used but print a meaningful warning instead

Building DSOs by default should set MOD_SO_ENABLED. Without this patch apxs doesn't recognize we have DSO support.

- Introduce log levels trace1/.../trace8 - Add macro wrappers for ap_log_*error. On C99, this will save argument preparation and function call overhead when a message is not logged because of the configured loglevel. - Introduce per-module loglevel configuration.

Fix lookup of apu-config for VPATH builds For included APU, apu-config is not in the srcdir but instead in the builddir. That only matters for VPATH builds, ie. when you build outside of the source tree. Similar to r951769 (apr-config).

For included APR, apr-config is not in the srcdir but instead in the builddir. That only matters for VPATH builds, ie. when you build outside of the source tree.

more configure-time support for dynamically loadable MPMs: . axe --with-mpm="shared" hack, replace with --enable-mpms-shared={all|list} . replace singular MPM_NAME with access to the list of enabled MPMs . replace singular MPM_SUBDIR with list MPM_SUBDIRS . enable OS/2 MPM in same manner as others with configure support instead of hard-coding in configure.in Current state: MPMs are built as static archives (but not linked to httpd) with --enable-mpms-shared, so they still have to be built with apxs to load dynamically.

Make sure that --enable-maintainer-mode increase warnings/errors level.

Always make mod_*.h files available where they exist, modpath by modpath

Bring back OS/2 support. Reverses r758929 with a little bit of conflict resolution.

Here's the simpler solution to the two groups of mappers/ modules... The mod_so and mod_watchdog are truly not mappers/ at all. Very open to better names than modules/core/ but these are clearly API's which are to be consumed by the httpd core, or generically, any module.

Move logic to decide if an MPM is supported, and whether the MPM is threaded, down to the MPM itself. (server/mpm/FOO/config.m4, which runs before the actual MPM selection) server/mpm/config.m4 makes some general platform checks that can be used for MPM decisions, and contains some functions related to MPMs. XXX The check here for whether APR_POLLSET_THREADSAFE is available is a rough approximation and needs to be replaced by a run-time check. Replace the limited per-platform hard-coded MPM selection and the current defaulting to event (whether or not it works) with a selection based on which MPMs work on the platform, as reported by the MPMs themselves. (config2.m4, which runs after the MPMs record whether they are supported) Order of preference: WinNT (mingw32 only) then Event then Worker then Prefork

stop exporting APACHE_MPM_DIR to modules -- unclear why this was ever necessary, and it makes no sense with loadable MPMs axe the unnecessary MPM_DIR build variable move the stray APACHE_SUBST(MPM_LIB) from configure.in to server/mpm/config.m4, where the other MPM_foos are handled

The Event MPM is no longer experimental; move it out of that ghetto. Yank the few bits that implement the concept of an experimental MPM.

remove OS/2 platform support

remove BeOS OS support

finish removing BeOS MPM

* configure.in: Allow configure to be built without find_apu.m4, at least temporarily to allow buildconf against apr trunk.

fix shell syntax error in check for apr major version on OS X: acceptable_majors=""1 2""

Introduce a new set of APIs to allow MPMs to be proper modules instead of integral parts which share global variables, functions, and macros with the rest of httpd. Converted now: prefork, worker, event, simple, WinNT* *WinNT hasn't been built or tested, and relies on a hack to include the WinNT mpm.h to disable Unixy MPM support routines in mpm_common.c

* configure.in: If APR version is 2, APR-util does not exist.

improve acceptance of APR/APR-Util trunk/2.0-dev

Allow --with-included-apr to pick up APR trunk/2.x

pcre-config --cflags actually emits include paths not cflags, so add it to correct variable. Also, reuse APR_INCLUDES/APU_INCLUDES variables so as to save an extra invocation of config files. * configure.in: Add PCRE_CONFIG --cflags to INCLUDES; reuse AP{RU}_INCLUDES variables.

Fix correct linking against PCRE. r730882 broke builds on MacOSX. Now we put the PCRE libs into a new variable PCRE_LIBS and add it to HTTPD_LDFLAGS. That way the detected libs for PCRE could be reused in other parts of the build system if necessary. Side effect: I had to change the handling of HTTPD_FLAGS on AIX and OS390 from hard setting of the variable to using APR_ADDTO(). I hope that macro works on those platforms.

Only link libhttpd against pcre.

We need to link only ab against libm (because of sqrt()).

Import the core parts of the DTrace patch. This is a modified and updated version of what was submited by Theo before: <http://mail-archives.apache.org/mod_mbox/httpd-dev/200805.mbox/%3C6AFBCFE8-4CCA-4A02-8A43-F9170689695D@omniti.com%3E> Note, this does not hook it up into the build system at this time, because the original patch was a little too creative there. Submitted By: Theo Schlossnagle <jesus omniti.com>

* Enable the use of autoconf >= 2.62 without causing APR / APR-UTIL options passed to the configure script issue warnings about unknown options.

* buildconf, configure.in: Don't attempt to configure the no-longer-bundled pcre.

Stop bundling PCRE in trunk.

Bump trunk to require apr-util >=1.3.x, since we already require apr >=1.3.x.

Adopt conditional handling for the new AC_USE_SYSTEM_EXTENSIONS as of 2.60. Submitted by: Arfrever Frehtes Taifersar Arahesis <arfrever.fta gmail.com>

AC_USE_SYSTEM_EXTENSIONS encompases _MINIX and _AIX and silences autoconf 2.62

For *trunk/* development, presume at least apr-1.3 (now released) for seperation of ldap link options.

Determine registered authn/z providers directly in ap_setup_auth_internal(), which allows optional functions that just wrapped ap_list_provider_names() to be removed from authn/z modules. This change requires modules/aaa/mod_auth.h to be included into server/request.c, which necessitates a minor change to configure.in for Unix platforms. I'm unable to tell whether a similar change is necessary for Windows and NetWare builds or not. Could developers with access to those platforms please test and make any needed configuration or build alterations? Thanks!

Revert unintentional changes in r657354

configure: Don't reject libtool 2.x PR 44817

mod_session: Add a generic session interface to unify the different attempts at saving persistent sessions across requests.

Enabled to configure the https port as we do with http port too.

As APR 1.2.x cannot solve it's issues with apr_proc_create(), mandate the working version of APR (1.3.0-dev or later).

Fix VPATH builds of httpd with the apr_memcache backend for ssl sessions, when APR and APR-Util are also in a VPATH. This is caused by APR_INCLUDEDIR actually having multiple paths in a VPATH builds, so we need to use _INCLUDES instead.

Add alternative fixes for CVE-2007-3304: * configure.in: Check for getpgid. * include/mpm_common.h (ap_mpm_safe_kill): New prototype. * server/mpm_common.c (reclaim_one_pid): Ensure pid validity before calling apr_proc_wait(). (ap_mpm_safe_kill): New function. * server/mpm/prefork/prefork.c, server/mpm/worker/worker.c, server/mpm/experimental/event/event.c: Use ap_mpm_safe_kill() on pids from the scoreboard, throughout. * include/ap_mmn.h: Minor bump.

* configure.in: Substitute MOD_SO_ENABLED variable. * support/apxs.in: Use it to eliminate run-time check for whether httpd has mod_so built-in. Submitted by: David M. Lee <dmlee crossroads.com> PR: 40653

* configure.in: Bump autoconf version requirement (matching APR).

Disable the getpwnam and getgrnam functions, since they aren't available on win32.

include/ap_config.h: Include the autoconf generated header that defines what is available. configure.in: The times function is not really available on mingw32. Disable em.

Bring in basic support for building the winnt MPM under mingw32, using the autoconf build system. Currently this will get you to the point of linking errors on gen_test_char. configure.in: Default to winnt MPM on mingw32 hosts. configure.in: Add AP_DECLARE_EXPORT for mingw32 hsots. os/config.m4: Use win32 dir under mingw32. os/win32/Makefile.in: Add in a new makefile to generate libos on win32. server/mpm/config.m4: Add winnt MPM to the help string. server/mpm/config.m4: winnt MPM is a threaded MPM. server/mpm/winnt/child.c: SO_UPDATE_ACCEPT_CONTEXT is defined under a different header in mingw32. server/mpm/winnt/{Makefile.in,config.m4}: Base build for winnt MPM. os/win32,server/mpm/winnt: Update svn:ignore for .deps/.libs

* configure.in: Use common variables for the required APR/APR-util version.

The syntax ">&/dev/null" is csh-style, not bourne-style shell. Change it so that all bourne-style shells understand it. I am going to make the same fix in 2.2.x

* configure.in: Add --with-included-apr flag to force use of the bundled copies of APR and APR-util.

* include/ap_config.h: Define AP_HAVE_DESIGNATED_INITIALIZER for GCC (but not G++) or any compiler which claims C99 support. * configure.in: Don't define AP_HAVE_DESIGNATED_INITIALIZER here.

* configure.in: Move apr/apr-util version checks later to fix in-tree apr/apr-util build failures due to $EGREP not getting set early enough.

* configure.in: Require apr >= 1.2.0 and apr-util >= 1.2.0.

make mod_dbd.h available for all who need it

Move the AC_TRY_RUN expansion much further down, after we have had a chance to initialise the compile-environment for all platforms.

One way or another, whether we leave the casts in or out, right now we are using "void *" containers to store long integers; Add a configure-time check to ensure that it is valid to assume that a "void *" container is large enough to store a long int. See <200510031150.j93Boh700945@devsys.jaguNET.com> and ensuing discussion on httpd-dev.

Use rsync to install manual and htdocs to speed up installation times. * Makefile.in: Update install-man and install-htdocs to use rsync if available, otherwise fail back to current cp -rp and rm -rf .svn/ strategy. * configure.in: Look for rsync. * acinclude.m4: Include RSYNC in exported config_vars.mk (Incorporates suggestions on rsync flags from colm.) Reviewed by: colm, trawick (concept), niq (concept)

Remove support for the threaded-mpm's on Linux 2.0, and de-autoconf the graceful restart signal. Frees up SIGWINCH for implementing graceful stop. See <20050727114058.GA3390@stdlib.net> and subsequent thread on httpd-dev.

* configure.in: Define MKINSTALLDIRS and INSTALL in config_vars.mk. Don't use AC_PROG_INSTALL since any system "install" command is never used anyway. * build/rules.mk.in: Remove fixed reference to install.sh and mkdir.sh in abs_srcdir. * build/config_vars.sh: Munge definitions of MKINSTALLDIRS and INSTALL to use installed scripts.

Sanitize the installed config_vars.mk to remove references to the build and source directories, and to remove variables which are internal to the httpd build system: * build/config_vars.sh.in: New file. * Makefile.in (build/config_vars.out): New target. (install-build): Install build/config_vars.out as config_vars.mk. * configure.in: Add include path for srclib/pcre to INCLUDES rather than CPPFLAGS, and use a topdir-relative not absolute path name. Generate build/config_vars.sh. Reviewed by: Paul Querna PR: 20874, 28145

* configure.in: Disallow use of bundled APR with external APR-util, or bundled APR-util with external APR, another cause of the dreaded PR21718.

Merged the simple-conf branch changes r159781:160695 into the trunk.

Support use of an external copy of the PCRE library: * configure.in: Set abs_{builddir,srcdir} higher. Add --with-pcre flag; build against external PCRE library if used. * Makefile.in (install-include): Don't install pcre headers any more. * srclib/Makefile.in (SUBDIRS): Remove. PR: 27550 (part two) Submitted by: Andres Salomon <dilinger voxel.net>, Joe Orton

Move the POSIX reg* implementations into the ap_* namespace; internalise the ap_reg*<->PCRE wrapper: * configure.in: Add srclib/pcre to the include path. * include/ap_regex.h: Renamed from include/pcreposix.h. Prefix all constants with AP_; prefix all functions and types with ap_. Define AP_DECLARE to nothing if necessary. Remove regcomp error codes. * include/httpd.h: Include ap_regex.h not pcreposix.h. (ap_pregcomp, ap_regexec, ap_regfree): s/regex_t/ap_regex_t/. (ap_regexec, ap_regerror): Prototypes moved to ap_regex.h. * server/util.c (regex_cleanup, ap_pregcomp, ap_pregsub, ap_pregfree): Adjust for ap_ prefixed types. (ap_regexec, ap_regerror): Removed. * server/Makefile.in: Build util_pcre.c. * server/util_pcre.c: Copied from srclib/pcre/pcreposix.c; remove use of PCRE-internals to do error mapping; rename types to add AP_/ap_ prefixes as above. Use APR includes. (ap_regerror): Use apr_snprintf. * srclib/pcre/Makefile.in: Don't build pcreposix.c into libpcre.la. * modules/*: Update to use new type and constant names. PR: 27750 (part one) Submitted by: Andres Salomon <dilinger voxel.net>, Joe Orton

* configure.in: Add --enable-pie flag; define PICFLAGS and PILDFLAGS. * build/rules.mk.in: Factor out BASE_CC and BASE_CXX from COMPILE. Substitute PICFLAGS into COMPILE and CXX_COMPILE, but not into {SH,LT}{,_CXX}_COMPILE. Substitute PILDFLAGS into LINK but not {SH,MOD}_LINK. Reviewed by: jerenkrantz

Fix VPATH errors introduced in r124104. * configure.in: Use correct source path to get-version.sh and include/ dir.

Add a build script to create a solaris package.

The current config.guess in APR outputs i386-pc-os2-emx on OS/2 rather than i386-pc-os2_emx (dash instead of underscore), adjust the tests to match.

* configure.in: Only export CC and CPP if a cache file is not specified.

* configure.in: Ensure that $CC and $CPP are correctly passed through to the pcre configure script if config caching is disabled (the autoconf 2.5x default).

Echo a warning from configure if you select an experimental MPM. Reviewed by: Jeff Trawick

* configure.in: Only build against APR 1.x.

* configure.in: Ensure that apr and apr-util exported include directories are later than local source directories in INCLUDES. PR: 18959

* configure.in: Define AP_HAVE_DESIGNATED_INITIALIZER via ap_config.h so third-party modules don't have to pick up NOTEST_CPPFLAGS to avoid gcc warnings from http_config.h.

* configure.in, acinclude.m4: Substitute AP[RU]_CONFIG with location of installed ap[ru]-config scripts so third-party modules can use `apxs -q APR_CONFIG`. * support/apxs.in: Use new AP[RU]_CONFIG variables; use apr-config --apr-libtool.

Teach apxs how to find ap{ru}-$AP{RU}_MAJOR_VERSION-config (httpd-2.1+ will be APR 1.0 or higher which uses this convention.)

Workaround RHEL 3 'cuteness' for OpenSSL by calling pkg-config openssl. For more info, see: http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=82369

Nobody objected: <IfModule> now takes the module identifier in addition to the filename. CAUTION: this commit breaks the build on netware. It still needs to be adjusted. (but I don't have any netware knowledge...) Also, developers need to re-run buildconf on unices. PR: 29003 Submitted by: Edward Rudd <eddie omegaware.com>, Andr� Malo

*) SECURITY: CAN-2004-0174 (cve.mitre.org) Fix starvation issue on listening sockets where a short-lived connection on a rarely-accessed listening socket will cause a child to hold the accept mutex and block out new connections until another connection arrives on that rarely-accessed listening socket. With Apache 2.x there is no performance concern about enabling the logic for platforms which don't need it, so it is enabled everywhere except for Win32. [Jeff Trawick] (already in 2.0.49, propagating to mirrors now)

* configure.in: Remove long since unnecessary endianness checks.

* configure.in: Remove unnecessary $builddir/server/mpm/$mpm and $srcdir from INCLUDES. Remove unused USE_VPATH variable.

* configure.in: Check for sqrt in libm (needed by ab).

Add fatal exception hook for use by debug modules. The hook is only available if the --enable-exception-hook configure parm is used. Sample users at http://httpd.apache.org/~trawick/exception_hook.html

"Listen 80" just works now, using v4-only, v6-only, mapped address or non-mapped address. Submitted by: Colm MacCarthaigh <colm@stdlib.net>

Add optional hook to extend mod_status output. * configure.in: Add modules/generators to include path. * Makefile.in: Install mod_status.h. * modules/generators/mod_status.h: New file. * modules/generators/mod_status.c: Implement ap_run_status_hook. (status_handler): Run status hooks.

Use three-argument form of AC_DEFINE everywhere; remove acconfig.h and the warning message from autoconf 2.5x.

Play nice with libtool 1.5.

Fix a VPATH build issue - our command lines can sometimes hit the wall against the shell's max command line length. We had a number of needless paths included in VPATH builds; this change will spare us from -I'ncluding the module paths with no .h files.

Ensure that ssl-std.conf is generated at configure time, and switch to using the expanded config variables to work the same as httpd-std.conf PR: 19611

By default, use the same CC and CPP with which APR was built. The user can override with CC and CPP environment variables. This applies to out-of-tree APR builds. Prior to this change, if you had multiple compilers on the system and you told APR to use a non-default one, you'd have to remember to tell Apache too. Otherwise, Apache would use the default compiler but use CFLAGS that went with the compiler that APR was built with.

Move the dropped '--enable-layout' subdir logic into APR_SUBDIR_CONFIG macro as a third parameter rather than manually computing it ourselves. Add the ' prefix'd --enable-layout so that this can work with an autoconf-2.54+ generated configure. (Previously, the option would not be dropped with 2.54 or 2.57 since it unconditionally adds single-quotes which our case statement would not catch.) Try to shorten long invocations of APR_SUBDIR_CONFIG where possible.

Fix a build problem with passing unsupported --enable-layout args to apr and apr-util. This broke binbuild.sh as well as user-specified layout parameters. PR: 18649 Original concept code submitted by: Justin Erenkrantz Mangled by: Jeff Trawick

use APR_CHECK_APR_DEFINE() where possible

Linux 2.4+: enable coredumps when Apache is started as root if CoreDumpDirectory is explicitly coded

Use APR_LAYOUT instead of APACHE_LAYOUT in configure. (We need a bug fix in this code, so let's merge the macro first.) PR: 15679

Remove files on 'make distclean' that should be. PR: 15592

Use correct quoting mechanism for APR_SUBDIR_CONFIG.

Allow apachectl to perform status with links and elinks as well if lynx is not available.

update a test compile to find apr.h when apr isn't in srclib (the only one of several such places in Apache which didn't get fixed to work with separate APR... no fun when v4 sockets won't talk to Apache anymore)

Make buildconf work with out-of-tree apr and apr-util by adding --with-apr and --with-apr-util flags. Submitted by: Thom May <thom@planetarytramp.net> Reviewed by: Justin Erenkrantz (with minor stylistic nit)

Allow dependencies to be generated by something other than GCC. This allows Sun's cpp to generate valid .deps dependencies. (This is a similar patch as applied to APR.)

Take apr include dir from beginning, not end of list given by apu-config. Submitted by: Brian Havard

Find apr, apr-util automatically if no --with-apr{-util} flag, same as other libs.

If apr and apr-util are not in-tree, we need to be able to find the include directory for each in order to generate the server/exports.c and server/export_vars.h files. configure.in: - Provide APR_INCLUDEDIR, APU_INCLUDEDIR. server/Makefile.in: - Use $APR_INCLUDEDIR $APU_INCLUDEDIR vars. - Add export_files target to generate a list of headers with symbols to export, use that list to generate exports.c and export_vars.h.

Changes to reduce dependency on in-tree apr and apr-util. Change AC_CONFIG_AUX_DIR in configure to build from srclib/apr/build. Have buildconf copy the needed files (config.guess, config.sub, PrintPath) from srclib/apr/build to build. Change binbuild.sh to stop using srclib/apr. This gets to to where ./configure --with-apr --with-apr-util actually works without an in-tree apr and apr-util, though we still need them in-tree to run buildconf.

Don't include apr_network.m4 and apr_threads.m4; we don't use them.

Add --[enable|disable]-v4-mapped configure option to control whether or not Apache expects to handle IPv4 connections on IPv6 listening sockets. Either setting will work on systems with the IPV6_V6ONLY socket option. --enable-v4-mapped must be used on systems that always allow IPv4 connections on IPv6 listening sockets. Note: As the ssl config file is not automatically generated and it is expected to require editing anyway to work, the only change there was to suggest the required Listen statements in a comment. PR: PR 14037 (Bugzilla), PR 7492 (Gnats), various dups of these PRs

Rework the apxs -cp command to work when the source dir isn't available by querying apr-config and apu-config at link-time instead. (Note that the $apr_config value in configure.in isn't sufficient as we need to know where apr-config will be when we execute apxs later on - hence the --bindir option to apr-config and apu-config.)

OS/390: re-instate support for single listen unserialized accept Reported by: Marcel_Schmidt at swissre dot com

Use APR_PARSE_ARGUMENTS instead of APACHE_PARSE_ARGUMENTS Submitted by: Thom May <thom@planetarytramp.net> Reviewed by: Justin Erenkrantz

AIX: Apache is slightly faster with this malloc setting

set some environment variables on AIX that improve Apache performance on SMP machines... no difference was seen on single CPU machines

Make the mime types configuration file path relative to {sysconfdir}, just like the other config files.

Fix the build on BSDi machines. The bsd_makefile trick didn't completely work as autoconf may run config.status and remove the fixes we've just applied, so this allows us to workaround that problem. Initailly suggested by Cliff, this is a modified version.

the TCP guy in Austin (AIX-land) swears that 4.3.2 GA had the fix for thundering herd, so tweak the prior change just a bit

AIX 4.3.3 and above: Define SINGLE_LISTEN_UNSERIALIZED_ACCEPT. These levels of AIX don't have a thundering herd problem with accept().

Fix breakage introduced in mod_cgi commit since we need to now include mod_include's headers all the time. (Also fix mod_proxy header not being listed in builddir for VPATH builds)

get APR to tell us the name of the shared library path variable

Remove APACHE_MKDIR_P_CHECK in favor of new APR_MKDIR_P_CHECK.

Instead of using the hardcoded libtool path of srclib/apr, use the output of apr-config --apr-libtool.

Instead of hardcoding srclib/apr{-util} in the INCLUDES, get the information from ap{ru}-config --includes which knows how to deal with VPATH builds itself.

First pass at trying to standardize httpd-2.0's usage of APR and APR-util so that they can be separated - use an installed APR and APR-util with httpd-2.0. This is not expected to achieve full separation, but is merely the first step in the process. The tree should still be buildable in the source-only case. This commit does the following things: - Use the standard find_apr.m4 and find_apu.m4 m4 macro files that flood and Subversion already use for locating APR and APR-util. This adds the --with-apr and --with-apr-util configure options. - Stop using APRVARS and export_vars.sh. - Use apr-config and apu-config to determine dependencies. - Add AP_BUILD_SRCLIB_DIRS and AP_CLEAN_SRCLIB_DIRS values (names stink) to dynamically figure out what subdirs in srclib we will build. (What we really want to do is set one value - go L->R on build and R->L on clean. I can't think of a way to do this.) - If we are using APR and APR-util that are already installed, don't specify them in srclib. (This still needs some work I bet.)

Allow VPATH builds to succeed when configured from an empty directory. (Justin removed some unneeded changes in Makefile.in.) Submitted by: Thom May <thom@planetarytramp.net> Reviewed by: Justin Erenkrantz

Added support for MPMs that live in subdirectories under server/mpm like server/mpm/experimental/*

Pass the --includedir flag to the apr and apr-util configure calls. Submitted by: Pier Fumagalli <pier@betaversion.org> Reviewed by: Aaron Bannert

AIX: Fix the syntax for setting the LDR_CNTRL and AIXTHREAD_SCOPE environment variables in the envvars file.

Remove hardcoded paths from MPMs and move them to a more central place. Expand paths in httpd-std.conf in a similar fashion as we do in our headers. Submitted by: Pier Fumagalli <pier@betaversion.org>

Added exp_foo and rel_foo variables to config_vars.mk for all Apache and Autoconf path variables (like --sysconfdir, --sbindir, etc). exp_foo is the "expanded" version, which means that all internal variable references have been interpolated. rel_foo is the same as $exp_foo, only relative to $prefix if they share a common path. This will be used in apxs.in and apachectl.in to allow Apache installs to be relocated as long as the same relative directory structure remains. Currently this is only possible if you use the Apache layout, but not with other layouts or any overridden --foodir paths.

Solaris 8 doesn't have a thundering herd problem according to the sun experts

Stop copying APR's libtool. And pass enough parameters to apr's subconfig so that it can figure out where we would like libtool to be installed.

Fix a bug that would cause SERVER_CONFIG_FILE to be an absolute path, which would prevent the relocation of an httpd binary without having to specify the -f flag. Now it will be relative to the $prefix if $sysconfdir shares a common prefix, or it will remain relative to $sysconfdir. Submitted by: Thom May <thom@planetarytramp.net>, Aaron Bannert

64-bit build on AIX: fix major bogosity in the logic to avoid setting LDR_CNTRL=MAXDATA=xxx for a 64-bit build... the name of the environment variable was wrong and even the test usage was bogus

Fix a problem in the new --enable-layout functionality where it wouldn't allow overrides from variables like --prefix, --bindir, etc. Obtained from: Partly from an autoconf-generated configure script. Submitted by: Thom May <thom@planetarytramp.net> Reviewed by: Aaron Bannert, Justin Erenkrantz

For some reason AC_PREFIX_DEFAULT must not be called before the apr m4 files are included. This fixes builds on autoconf 2.52 again. Obtained from: Sander Striker Submitted by: Thom May <thom@planetarytramp.net> Reviewed by: Aaron Bannert

No longer assume SERVER_CONFIG_FILE is in the "conf" directory. Instead look for it in the $sysconfdir. This required the use of a new m4 function from APR that retrieves the fully "expanded", or recursively interpolated, value of the $sysconfdir variable.

Fix a typo in the default cgidir variable. Set a couple more defaults if they haven't already been set, just as a precaution.

Set defaults for non-autoconf path variables. Also export a couple variables that for whatever reason weren't being placed into config_vars.mk. Warning: This commit alters the "default" layout (ie. when no --enable-layout is specified). The most notable change will be htdocs in $prefix/share/htdocs, which works with the autoconf defaults, but may not be what we want. This will require more work.

Fix --enable-layout to work again. Fix bug where --sysconfdir and --localstatedir were being ignored by APR. Caution: When specifying --enable-layout, common arguments like --prefix, --exec-prefix, etc. will be ignored and the settings from the layout will be used instead. PR: 9124, 9888, 9873, 9885 Submitted by: Thom May <thom@planetarytramp.net>, Aaron Bannert Reviewed by: Justin Erenkrantz

Fix the AC_OUTPUT() macro for the BSD users. Submitted by: Zack Weinberg <zack@codesourcery.com>

Check for timegm() and for tm_gmtoff in struct tm. These are used for the USE_MDTM case in proxy_ftp.

new var. APRUTIL_LDFLAGS

Per Ian's change, add the APRUTIL_EXPORT_INCLUDES vars to our environment.

OS/2: Supply the configured libraries to the linker when building loadable modules. This allows things like mod_deflate to build (needs -lz).

This old variable does not appear to be in use any longer.

On AIX, DSOs which reference external libraries need those libraries available to ld when the DSO is linked. This gets mod_deflate, which needs zlib, to work as a DSO on AIX.

fix the value of the _CEE_RUNOPTS envvar on OS/390

AIX: don't set LDR_CNTRL environment variable for 64-bit build

Move any load library path environment variables out of apachectl and into a separate environment variable file which can be more easily tailored by the admin. The environment variable file as built by Apache may have additional system- specific settings. For example, on OS/390 we tailor the heap settings to allow lots of threads.

change Apache/apr/apr-util to use run-time linking on AIX currently, a kludge (-uXML_Parse) is needed to get a reference to expat in the Apache executable programs; I think this is related to the fact that expat libtool is getting generated a little differently than apr libtool and is choosing to build a different flavor of shared library

separate logic specific to building Apache DSOs from logic needed when anybody's DSOs should be built (e.g., via apxs) so that on a few platforms with special issues third-party modules can be loaded as long as mod_so is built into httpd

get binbuild working when apr, apr-util, and expat are built shared in directories that are not well known to the loader. How it works: * libtool figures out the name of this system's library path environment variable (i.e. LD_LIBRARY_PATH, LIBPATH, DYLD_LIBRARY_PATH) at apr configure time (existing code) * save this variable name in SHLIBPATH_VAR in APRVARS * substitute this name when apachectl is created * uncomment the lines containing this variable during binbuild * install_bindist.sh updates the path to lib/ when the binary is installed (existing code) not done yet, but a good idea: * create a file or directory containing all environment variable settings for apachectl. Others env vars need to be tweaked, at least on AIX and OS/390, so keeping them separate should reduce headaches long term.

Get shared builds of libapr and libaprutil, as well as Apache DSOs, working on AIX. Submitted by: a cast of many, reverse engineering libtool to figure out how to work around its oddities/limitations/ bugs on AIX (Jeff Trawick coded these changes but it took a lot of help)

Allow SUEXEC_BIN to be defined by the configure script.

Cleanup the proxy code that creates a request to the origin server. This change adds an optional hook, which allows modules to gain control while the request is created if the proxy module is loaded. The purpose of this hook is to allow modules to add input and/or output filters to the request to the origin. While I was at it, I made the core use this hook, so that proxy request creation uses some of the code from the core. This can still be greatly improved, but this is a good start.

Fix a typo in the configure script. This get AIX building cleanly. Submitted by: Jens-Uwe Mager <jum@helios.de> Reviewed by: Aaron Bannert <aaron@clove.org> and Ryan Bloom

Kill two autoconf-related birds with one commit. - substr used in APACHE_MODULE for the help string did not parse correctly with autoconf 2.50+ so we had to punt there anyway and use AC_HELP_STRING. - Add APACHE_HELP_STRING define that will call AC_HELP_STRING on 2.50+ (actually not 2.13 - look at the regex call) or do our custom variation of it. This function can't have any extra spaces or it will be returned in the help string. So noted. If anyone can figure out how to insert a line break like 2.50+ does when we go over the 26th column, I'd appreciate it. I tried and I'm way too tired to figure it out now. Adding this would greatly simplify two or three HELP_STRING uses. - Switch all of those annoying WITH and ENABLE functions to use the APACHE_HELP_STRING. This makes everything consistent now. I've always had to go through and keep aligning everything every few months or so because I'm the only one who cares. No more. I refuse to do it any more! Use APACHE_HELP_STRING or be crucified. Looks decent with autoconf-2.13 and autoconf-2.52.

httpd.exp is a generated file, and therefore, it is generated into the build directory, not the source directory. When we go to use it, we have to look in the build dir for it.

kill -l requires the short version of the signal (without SIG prefix). (apachectl graceful was broken.) The best way to do this is to change configure to internally represent the signal without the SIG prefix and prepend SIG before substitution (and export a version without the SIG prefix for apachectl). This highlights why we want to move apachectl's core functionality into httpd.

Switch back to SIGUSR1 for graceful restarts on all platforms that support it. This defines a symbol called AP_SIG_GRACEFUL in ap_config_auto.h which will have the appropriate signal value. All direct references to SIGWINCH have been replaced with AP_SIG_GRACEFUL. On Linux 2.0, use SIGWINCH instead since SIGUSR1 is used by glibc 2.0's user-space threading library to control threads. All later versions of Linux/glibc don't have this problem. (Not to mention the security holes in older Linux versions which make it unsuitable for use as a web server.) If your platform doesn't have SIGUSR1, use the appropriate mojo in configure to define what your graceful restart signal should be. In theory, a configure switch could be added to allow the admin to specify the appropriate signal that should be used. This is left as an exercise to the reader for now. The docs need to be updated. Since the signal is now configurable, just saying SIGUSR1 for graceful restart isn't completely true. Also, the apachectl functionality needs to be moved into httpd - this is what Win32 does and it makes us consistent across platforms. Roy issued a veto against use of SIGWINCH by default, so this should resolve that veto.

Fix test for ltconfig, m4 processing removes the square brackets. (I just happened to notice "-f: not found" while running configure)

Remove an error during configure. Libtool 1.4 doesn't have ltconfig.

on AIX we need to pass in --disable-shared to apr and apr-util configurations; otherwise we get goofy executable files obviously libtool isn't doing the right thing, but I don't know what that is

Fix the interaction between --prefix and --with-layout. We were over-writing the specified prefix during --with-layout processing. By keeping a copy of the specified prefix, and resetting it after the layout processing, this is fixed. PR: 7755

* configure.in: added Cygwin specific APR_SETVARs. * os/config.m4: added Cygwin specific values for $OS and $OS_DIR. $OS_DIR will change to "cygwin" as soon as I get the DSO/DLL things clean as with 1.3.20. * support/logresolve.c: added OS specific #ifdef statement to define h_errno as extern __declspec(dllimport); it's imported from the cygwin1.dll itself. Submitted by: Stipe Tolj <tolj@wapme-systems.de>

Remove some unneeded autoconf checks. We now have no duplicate libraries in the generated EXTRA_LIBS in config_vars.mk (at least on Linux 2.4). APR already determines if we need nsl, socket - checking for it in httpd is merely redundant. None of the code in httpd-2.0 seems to deal with HAVE_GMTOFF as all of that code was moved to APR a long time ago.

Add rules.mk back to the substitute list. I forgot to add the rules.mk.in files last night, but they are there now.

We don't have rules.mk.in, so autoconf/m4/sed will overwrite the good rules.mk with a blank file. This gets us building again. From what I see of rbb's libtool 1.4 commit, I didn't see any change to rules.mk that would require variable substitution. Ryan?

Allow Apache to use libtool 1.4. In order to do this, we check the libtool version of the machine, and we add the correct flags to the build. We also had to remove the --disable-shared in order to make this work correctly.

Make mpmt_os2 the default MPM on OS/2.

The Unix MPMs other than perchild now allow child server processes to use the accept mutex when starting as root and using SysV sems for the accept mutex. Previously, this combination would lead to fatal errors in the child server processes. perchild can't use SysV sems because of security issues. translation: steal apache 1.3 code to call semop(IPC_SET) on the semaphore to set permissions and uid/gid

allow non-Unix MPMs to use mpm_common.c by not automatically building in Unix-specific code

OS/2: Fix 'make install' handling of EXEs.

Use the autoconf variable "host" consistently for platform checks.

Allow all parts of Apache 2.0 to build when --srcdir is used. This required exposing a build directory and a source directory to all parts of Apache's build system. It also required a small hack in APR-util, if we are using the bundled Expat, and we are using VPATH support, then we have hard-coded the xml/expat location. I couldn't figure out how to allow the configure script to determine the correct location. I added a comment, but if somebody else figures it out, we should fix that at some point. PR: 7630

We need to pass the prefix to APR, APR-util, and PCRE. If we don't pass that information, then all three libraries will install themselves into their default locations. PR: 7750

get threaded.c to compile on non-AIX systems which have sys/processor.h if we want to call bindprocessor(), we should check for bindprocessor(), not some header file (prefork and perchild to be fixed shortly)

SINGLE_LISTEN_UNSERIALIZED_ACCEPT applies to every kernel after 2.2, not just 2.2.

Make the processor unbinding/re-binding stuff on AIX work again.

Generate httpd.exp on the fly. This should allow DSOs to work on AIX, without the headache of maintaining the httpd.exp file. This is adapted from OS/2's generation of ApacheCoreOS2.def. There exist a few bugs still: 1) mod_dav and mod_proxy may not yet work, due to certain namespace issues. 2) Some symbols may need to be added, a la core_header.def Once these have been fixed, the old httpd.exp file will be deleted.

Get the compile flag types right and in order.

Change the make targets and rules to be consistent in all of the Apache-owned source trees. Sanity is a good thing.

clean up DSO builds on OS/390 a bit by ignoring some generated files and by passing libtool the OS/390-specific options needed to split the main httpd executable into a small executable with main() and a dll with everything else (previously this information was hard-coded in the OS/390 libtool program)

Separate enablement of AP_DEBUG code from enablement of code which depends on the compiler supporting designated initializers.

Implement Jeff's idea for an autoconf-defined symbol to enable the special command function prototype stuff. AP_DEBUG is always set in maintainer mode and AP_DEBUG_HAVE_GCC is set when the compiler is GCC. Submitted by: Jeff Trawick

get rid of an errant semicolon which broke gcc builds (sorry!)

Only turn on AP_DEBUG if we're using gcc. The AP_DEBUG flavor of the cmd_func declarations is not portable. Last night Roy removed the check for gcc before turning on AP_DEBUG. Interestingly (or not :) ) the AP_DEBUG flavor of cmd_func compiles cleanly on gcc but the non-AP_DEBUG flavor doesn't. The non-AP_DEBUG flavor compiles cleanly with Tru64 cc and AIX xlc but the AP_DEBUG flavor doesn't. (On Tru64 we get a bunch of warnings with the AP_DEBUG flavor but at least it compiles; the same cannot be said of AIX xlc.) Perhaps we should use a symbol other than AP_DEBUG to distinguish between the cmd_func flavors. Then AP_DEBUG can be turned on regardless of compiler. The new flag can be turned on for gcc+ --enable-maintainer-mode, or perhaps for gcc always since the non- AP_DEBUG flavor won't compile cleanly with gcc. (what a mess; this probably wouldn't work right for different levels of gcc anyway)

Fix httpd's definition of LTFLAGS to be consistent with that of apr and apr-util, allow it to be overridden by the configure command-line (default="--silent") and introduce LT_LDFLAGS to replace what we were formally abusing as LTFLAGS.

It is easier to re-run configure's test and pick up the cached values than it is to replicate them in our own macros. This isn't too expensive because CC and CPP have already been set by APR. Too bad it generates a lot of output. Also, we don't need to repeat command-line options that are already being handled by APR now that we use APR's generated flags. The switches are still usable on the command-line.

When checking for gcc, use variable ac_cv_prog_gcc instead of GCC. This gets AP_DEBUG set properly for --enable-maintainer-mode builds. (As libtool configuration is removed from other places -- e.g., pcre, expat -- the same change will be needed of those configuration scripts use GCC.)

Simplify the configuration process by moving all libtool stuff to APR and moving the last bits of hints.m4 inline. Now we only run every test four times instead of five. One down, three to go.

No need to sinclude macros that we no longer use.

We don't need no stinkin libtool when APR already has one for us to use.

Pass in the full path for the AP_LIBS so we can always find them when not installed.

Completely revamp configure so that it preserves the standard make variables CPPFLAGS, CFLAGS, CXXFLAGS, LDFLAGS and LIBS by moving the configure additions to EXTRA_* variables. Also, allow the user to specify NOTEST_* values for all of the above, which eliminates the need for THREAD_CPPFLAGS, THREAD_CFLAGS, and OPTIM. Fix the setting of INCLUDES and EXTRA_INCLUDES. Check flags as they are added to avoid pointless duplications. Fix the order in which flags are given on the compile and link lines.

Remove unused variable.

This changes the build process as outlined in my email earlier this week. Essentially we now build the static objects, then allow a platform to run a given command, build the shared modules, then run another command before exiting. This gets us building dso's on beos and may provide a better framework for some of the other "strange" platforms out there without stopping the normal ones working.

Temporary (hopefully) reversion to the previously used hack to allow --prefix to work for now. This means that part of the Apache layout is duplicated in configure.in again. There are a _LOT_ of inconsistencies in different places about what items in config.layout are actually used and how they can be set. Also, change the couple of other things that are necessary to make /usr/local/apache2 the default prefix when using this way of doing things. PR: Obtained from: Submitted by: Reviewed by:

Oh bother. OPTIM isn't treated the same in apr and httpd-2.0. While being fixed, allow to build

Minor sync-up of how httpd-2.0 and apr handles these. -g is more an Optimization flag, and the others are more compiler flags

Make Apache check APR's configuration when determining the default MPM. This way, if APR is built without threads, then Apache defaults to the prefork MPM.

Use APR's macro to generate config.nice

Changes to get AIX DSOs working, sans APR changes. Basically, this allows linker flags to be passed for the httpd executable and DSOs only.

Switch to using aplibtool for building under OS/2 & add support for building loadable modules as OS/2 DLLs. PR: Obtained from: Submitted by: Reviewed by:

if the perl interpreter is not found in PATH, use "/replace/with/path/to/perl/interpreter" for the path when substituting in split-logfiles et al so that an unlucky user will have a better clue about why these programs won't work

switch to --enable-* rather than --with-* for the debug options

Untangled the buildconf script and eliminated the need for build's aclocal.m4, generated_lists, build.mk, build2.mk, and a host of other libtool muck that is now under srclib/apr/build. More to follow.

Remove some unused bits and improve comments.

Cleanup the build process a bit more. The Apache configure script no longer creates its own helper scripts, it just uses APR's. Submitted by: jean-frederic clere <jfrederic.clere@fujitsu-siemens.com>

Bring the other Perl scripts into the autoconf-edited find-the-interpreter fold, like apxs and apachectl.

Bring forward the suexec umask setting option from 1.3, and also fix some configure.in bits that would define macros that needed to be numbers as strings instead. Obtained from: Apache HTTP Server 1.3

Get us closer to building on Unix. Not perfect yet, but we no longer should look for expat in srclib.

Rely on APR to determine whether or not we have netinet/tcp.h. This test requires a special work-around for IRIX 6.5, which will be in APR soon. There is no need for it in Apache. The include of netinet/tcp.h was removed from perchild.c and threaded.c. They no longer seem to be needed there. PR: 6459

Shift XML handling code from Apache down into APRUTIL - teach APRUTIL how to find Expat and respond to --with-expat - Apache's configure points APRUTIL's configure at its srclib/expat-lite (this will go away; aprutil can work against installed expats or fallback to an expat bundled within aprutil) - shift some of the timing of INCLUDES and top_builddir processing in the APRUTIL configure.in script - expose the new apr_xml functions in apr_xml.h, apr_xml.c (by building it), and apu_compat.h - rewrite util_xml.[ch] in terms of apr_xml

Work on a BSD/OS system where GNU make is used instead of the native make. Roy suggests using the variable set by AC_PROG_MAKE; we don't use that yet, though I guess we could. Greg suggests leaving this to Roy since he will revamp this stuff soon anyway. Inspired by: Joe Orton's patch to APR to accomplish the same thing

Update APR macro usage to new name-protected names. Replace REENTRANCY_FLAGS with direct use of THREAD_CPPFLAGS. This is a temporary band-aid, since the way we currently set the standard make vars (CFLAGS, CPPFLAGS, LDFLAGS, etc.) is bogus and frequently redundant [to be fixed later].

Intermediate change of hardcoded "helpers" reference to "build" until I get a chance to revamp this to use APR's variables directly.

implement ability to do --enable-modules and --enable-mods-shared for "all" and "most". still need to tweak various config.m4 files to sort them out.

Fix the configure scripts, so that we don't try to substitute into test/Makefile unless we actually have a test directory. The problem is that we don't distribute test directories with our tarballs, so our configure script outputs error messages. This fixes that problem.

[ missed in last commit ] clean out some headers and libs from the configure process (not needed)

*) clean out some old crap from ap_config.h; other simplifications; don't include stuff randomly (let users include them explicitly) *) toss ap_config_path.h -- we can put those values into ap_config_auto.h

Clean up the help text for some of the options. (cosmetic)

Replace configure --with-optim option by using and saving the environment variable OPTIM instead. This is needed because configure options do not support multiple flags separated by spaces.

Force --with-optim and --with-port to have argument values, since they result in later compile errors otherwise. I would have removed --with-optim in favor of just setting it with an environment variable, but then it wouldn't be saved in config.nice. A better solution is needed because these arguments can only accept a single value (quotes and spaces are stripped). Perhaps setting OPTIM within config.nice? First things first.

Libtool takes care of the lib prefix so get rid of our own workaround.

Fix a compile break with the new APR libtool setup. Basically, we are using a libtool library from MM, but it isn't getting linked into the Apache executable, so we fail the link stage. This makes APR publicize the libtool libraries that it uses, so that Apache can link them in.

a couple tweaks to get Apache building with the new libtool-ized APR

Set MKDIR for APR's RUN_SUBDIR_CONFIG_NOW macro.

seems nobody is using sethostent()

We cannot set the SUEXEC_BIN directory from autoconf, because all of the instructions say that in order to modify this variable, you have to modify httpd.h

Get suexec compiling again.

Allow SuExec to be configured from the ./configure command line

Back out the .libs thing from the aprutil linking. Add --disable-shared (for now) to the subdir config to ensure that we only get a static library for aprutil (which allows for linking all of its functions into the executable). [ APR will need something similar if/when it gets libtool-ized ] Mark a note in STATUS indicating we should eventually toss the --disable-shared flag.

Actually link apr-util's .a file, instead of relying on libtool. We may want to change this later, but right now, we want a single binary that has every function it requires.

Switch to the APR-provided APR_CHARSET_EBCDIC feature test macro.

Find the RM program during configuration.

pick up exported build variables from APRUTIL

remove APACHE_CHECK_DBM and the usage of srclib/sdbm/. this stuff is part of APRUTIL now.

Build apr-util correctly when building the rest of the server.

Add apr-util to the build process. This should now build once everything has been checked out to the correct locations.

Apache builds again, although the test directory doesn't. I will work on the test directory soon. The server does NOT currently run, but it does build cleanly.

Get the support programs building properly. Httpd itself is next.

Get base Apache to build properly. This doesn't solve the problem of not having apr-util, but we should be getting to the link stage now.

Get Apache to configure completely again. This still doesn't build, but now we are getting very close.

Start pointing the configuration scripts to the correct directories

Get buildconf working in the new repository. Nothing builds, but we can at least configure everything.

reorder the libs for proper inter-lib dependencies

Split the hints file into two files, one in APR and one in Apache. The APR hints file just sets build variables, the Apache hints file just sets Apache variables. This is meant to clean up parts of APR, so that they don't include Apache information.

add --with-dbm switch to configure, allowing the user to set the type of DBM that apu_dbm will use (defaults to builtin sdbm). first pass at gdbm config support (seems to work fine in my simple test). - added src/lib/aputil/apu_private.h.in to hold aputil config info - apu_dbm.c now uses #if for the config values (rather than #ifdef) - cleared out temp hack in configure.in to force sdbm usage

Always include lib/sdbm in INCLUDES (until somebody that can spell lbi/aputil teaches the configure to be a little smarter). This lets apu_dbm compile, which in turns lets us build httpd again.

*) Put lib/aputil/ into the INCLUDES path *) add a Makefile.in to test/ to help with building test programs (these are not part of the build; just manual compiles right now) *) add test/dbu.c as a quick test of apu_dbm [ dbu.c originally came from the SDBM package ]

Start the oft-discussed, but oft-punted lib/aputil library. Drop in a first cut at an "any" DBM set of cover functions.

Get the hints file working on Linux. This also gets SINGLE_LISTEN.... working on linux 2.2.

Get SINGLE_LISTEN_UNSERIALIZED_ACCEPT working again. This defines that macro on those platforms that have it defined in 1.3. The only platform that doesn't work yet is Linux 2.2.

Not all platforms have INADDR_NONE defined by default. Apache used to check for this and the define it if needed. Since APR also needs this check it makes more sense for APR to just check and export a symobl that is always available. Submitted by: Branko �ibej <brane@xbc.nu> Reviewed by: Ryan Bloom

Create a feature macro that determines if Autoconf was used to configure the server. The ap_ugly_hack variable is only valid if Autoconf was used for the configuration.

Simplify the configuration mechanisms for threading in APR. Revise how Apache configs APR and allow MPMs (or other config scripts) to pass cmd line params down to APR. For the MPMs that specifically require or disallow threads, have them force APR threads on/off. Submitted by: Joe Orton <joe@light.plus.com> Reviewed by: Greg Stein

Create modules.c in the build tree. This file depends on your configuration and thus cannot be shared by multiple build trees.

Use autoconf to substitute the proper values into apachectl, and install it when we install everything else.

Multiple build and configuration fixes Build process: -add datadir and localstatedir substitutions -fix layout name -fix logfilename misspelling -fix evaluation of installation dir variables and -replace $foobar by $(foobar) to be usefull in the makefile Cross compile: -add rules for cross-compiling in rules.mk. Okay, rule to check for $CC_FOR_BUILD is still missing -use CHECK_TOOL instead of CHECK_PROG for ranlib -add missing "AR=@AR@" to severaly Makefile.in's -cache result for "struct rlimit" -compile all helper programs with native and cross compiler and use the native version to generate header file PR: 6384 Submitted by: "R�diger" Kuhlmann <Tadu@gmx.de> Reviewed by: Ryan Bloom

Prepare our autoconf setup for autoconf 2.14a and for cross-compiling PR: 6379 Submitted by: "R�diger" Kuhlmann <Tadu@gmx.de> Reviewed by: Ryan Bloom

Set BSD_MAKEFILE in configure.in. The macro APACHE_GEN_MAKEFILES is executed in the context of config.status with variable substitution already performed, so the check in the macro itself did not work.

Add Convert-to-BSD-Makefiles-On-The-Fly system Note 1: Only needed for BSD/OS. Note 2: Invent better heuristic (timestamp checks for build/*.mk). Note 3: Conversion is currently one-way Note 4: Don't package on BSD/OS.

Finish fixing the pcre build. PCRE's configure script should be called, and we can't clean config.h and then touch the file, we actually need some of the data that config.h contains.

Tweak apache/apr builds to support dsos on OS/390. The OS/390 compile options are hard-coded because we need the special DLL,EXPORTALL on *all* object files that are part of the core so that dsos can reference symbols in the core. (Of course, we use the option on more object files than that, but it doesn't seem to hurt anything.) We hard-code the enablement of DSOs on OS/390 because the library functions on OS/390 to manage explicit loading are not portable, so there seems to be no use in probing for them. Submitted by: Greg Ames

Add recognition of the --with-port=n (as was added to 1.3), and raise an error if --enable-layout is specified but we can't find the definition file.

Fix VPATH builds

config magic for enabling DAV, Expat, and SDBM. *) AP_LIBS contains the target .a files from src/lib/ *) AP_LIB_DIRS contains add'l directories (besides pcre and apr) that we will build (and generate Makefiles for) *) apache_need_expat and apache_need_sdbm are two new configure feature macros; set by the DAV stuff as needed *) hackery to deal with dav/fs/ since it isn't really a module *) do endian testing, so we can feed it into Expat to optimize that code

Fix the problem with ./buildconf not working right out of CVS. The general problem, is that we were using AC_CONFIG_AUX_DIR_DEFAULT, which the autoconf source says shouldn't be called directly. It should be called using AC_REQUIRE, but I couldn't make that work. So, since we know exactly where the files that we are looking for are located, we'll just point autoconf directly to them, and ignore all of the hoops. :-) I am not convinced the AC_CONFIG_AUX_DIR(.) call that I am using is strictly necessary, but it doesn't cause any problems and it logically makes sense.

Finish the Configuration cleanup. This basically merges all of the autoconf scripts that APR provides for other programs into apr_common.m4. Other programs can then just include that file to get the APR checks that are required. Submitted by: Sascha Schumann <sascha@schumann.cx>

Include strings.h for strcasecmp(), strncasecmp(), and bzero(). Include time.h for time(). This removes a bunch of compiler warnings with gcc -Wall on AIX. Submitted by: Jeff Trawick, Victor Orlikowski

Fix the configure process. This does a couple of things to allow APR and Apache to share information. 1) Move the calling of APR after Apache decides which MPM is run, but before Apache generates the Makefiles and related files. This allows Apache to setup the threading cache values, while still allowing APR to generate APRVARS in time for Apache to use it. 2) Setup the cache files correctly for RUN_NOW configured subdirectories.

Move the checks for bcopy and memmove down to APR. APR now defines memmove and bzero on platforms that do not natively support them.

One bug and one fluff: need to use the '.' shell ``command'' to source the file; and remove error message first time APR's configure is run, and APRVAR doesn't exist. PR: Obtained from: Submitted by: Reviewed by:

Make sure to get the required vars for the config.guess/et.al. family first. Right now, this is semi-duplicated, since APR does the same. PR: Obtained from: Submitted by: Reviewed by:

We need to preload first, before running any other configures. PR: Obtained from: Submitted by: Reviewed by:

Detect libraries based on which platform configure is being run on. Basically, Apache now runs APR's configure script first. APR's configure script has been setup to create a new file, APRVARS. APRVARS is basically all of the environment variables that APR wants to export to the program that is using it. This allows the calling program to "source" APRVARS and get those environment variables. Removed hack to make platforms use -ldl.

Remove select checks from Apache config. These checks were used to define ap_select in ap_config.h. The only file that used ap_select() was prefork.c. I have modified prefork.c to use select instead of ap_select. In the future, this should be modified to use ap_poll like the rest of the Unix MPMs. There are other MPMs that are continuing to use select, but this too should change in time.

Fix error messages issued from MPMs which explain where to change compiled-in limits (e.g., ThreadsPerChild, MaxClients, StartTreads). missing: the same minor changes for the 2nd BeOS MPM Submitted by: Greg Ames Reviewed by: Jeff Trawick

Command handler revamp. Note that this makes the code produce a LOT of warnings!

Remove the check for EAGAIN. With APR handling the error codes, this is no longer an issue that Apache needs to deal with.

Apache doesn't use struct tm anymore, Remove a check that defines a macro that we never use.

Build only needed parts of libpcre. This is more light-wheight than having a complete build-subsystem for PCRE.

Nothing in Apache should be calling stat anymore, so we can remove the check for sys/stat.h and all references to the resulting macro.

Apache doesn't need to check for inline or pid's, because APR provides both ap_inline and ap_proc_t.

Split generating Makefiles and creating config_vars.mk. The Makefiles are now rebuilt, when you run config.status.

Fix path for helpers/PrintPath which is in $srcdir.

Remove waitpid from the config checks and all calls to waitpid from the server.

Remove a bunch of string functions from Apache. These are basically standard string functions like strstr, strcasecmp, etc that Apache used to define for platforms that don't have them. These functions and the feature tests have moved down to APR where they really belong. In doing this, I am also able to remove a bunch of tests from the Apache configure process.

Remove all occurances of strerror from Apache. ap_strerror works just fine with standard errno values, and it is more portable. This also allows me to remove the check for strerror from Apache's configure script.

Fix the check for rlim_t correctly. I wish I had known that we had created an rlim_t check in acinclude.m4. :-) Submitted by: Jeff Trawick

Cleanup APR header files a bit. Basically, this splits ap_table_t related functions into their own header file, away from apr_pools.h. At the same time as I was doing this, I removed a couple of definitions that aren't used anywhere anymore, like ap_child_info_t.

Remove all occurances of gettimeofday. Replace it with ap_now which provides the same function but works cross-paltform.

Make Apache and APR use the same flags to specify maintainer mode. I chose to use --with-maintainer-mode, because that flag can be passed down using AC_CONFIG_SUBDIRS. I couldn't find any easy way to pass environment variables to configure scripts called from within Apache's configure.

use helpers/PrintPath instead of which to find the location of the perl interpreter

Replace hsregex with PCRE. The PCRE library is always used in 2.0, regardless of what system we are on. Currently, we are using the POSIX wrappers that PCRE provides.

Make apxs mostly work for 2.0. It doesn't add the module to the config file yet, but that should be trivial to add back in.

Configure creates config.nice now containing your configure options. Syntax: ./config.nice [--more-options] Obtained from: PHP

Blurf. Move the preload down a bit to ensure that the location of config.guess is set :)

And preload from the top. Next on the list: using hints.m4 to set whether to use HSregex and to select the "default" mpm for each OS type. PR: Obtained from: Submitted by: Reviewed by:

Add VPATH support to Apache: os/$(OS_DIR) is now equivalent to the old $(OS_DIR) Necessary include paths are added only in VPATH mode. Dependencies are stored in the build directory now. They contain paths which depend on the current build, and therefore they are not general. Fixed two dependencies in main/Makefile.in.

Overall UNIX build system improvements: * Makefile header is now completely dynamic * Absolute paths everywhere (fixes Tru64 support) * Get rid of LTLIBRARY_SHARED_NAME rule in library.mk (fixes Irix support, untested) * VPATH does not contain variables anymore (fixes UnixWare support) * Remove inclusion of program.mk in support/Makefile.in (PROGRAM_NAME is empty => breaks Irix and others) * Call REENTRANCY_FLAGS earlier, so that flags are passed to header checks (might fix OpenBSD pthread.h-detection, untested)

Add the support directory back into the build process.

Fix the progname stuff to work the same way the rest of the configure scheme works. Submitted by: Sascha Schumann Reviewed by: Ryan Bloom

Define HAVE_STRING_H, HAVE_STDLIB_H, HAVE_CTYPE_H, HAVE_LIMITS_H at configure time. Some of the modules use these.

If a developer uses --with-program-name=foo, the config files are renamed to match the name of the executable. In this example, httpd.conf will be renamed to foo.conf.

Allow the Apache-2.0 executable name to be specified with the --with-program-name command line argument to configure. This does not change the name of the config files or anything else, yet.

Fix minor configuration & platform dependencies

In my libnsl, there's gethostname but not gethostbyname

Recognize an EBCDIC platform during configuration and turn on CHARSET_EBCDIC.

In mpmt_pthread.c, include <netinet/tcp.h> only if the system actually has it. (This was intended to be committed with the mpmt_pthread.c change.)

Under OS/2 (well actually EMX gcc) libraries don't have a 'lib' prefix. In main code libtool handles it. This deals with APR.

Eliminate src/apaci, and replace with a header file containing paths: include/ap_config_path.h.

Fix the generation of the apaci script, so it has valid permissions after running config.status Submitted by: Brian Martin <bmartin@penguincomputing.com> Reviewed by: Ryan Bloom

Strict prototypes are no use until someone rewrites the config API.

Oops, EXTRA_CFLAGS wasn't getting set.

A little CFLAGS reformation. Additions to Apache's CFLAGS should be using EXTRA_CFLAGS instead, so we don't affect packages that Apache configures.

Fix prefix handling. Before this patch, when using --prefix on the configure line, the binary ignored it when setting HTTPD_ROOT. This patch fixes that problem.

Remove vestiges of automake.

Move threads/reentrancy m4 macros into threads.m4. All results are now cached and can be made easily available to Apache and APR.

Rewrite pthreads check to be more tool-independent, and propagate flags to APR.

These changes are committed together, because they depend on each other. - shared modules can be built in the tree - added support for --with-layout, uses APACI's config.layout - working 'make install' - working 'make depend' - working Pthreads checks - buildconf replaced

Submitted by: Sascha Schumann <sascha@schumann.cx> Reviewed by: Rasmus Lerdorf - INADDR_NONE is not defined. Supply a replacement - APACHE_MODULE() used a non-portable shell construct - inet_addr() needs -lnsl -lsocket. Neccessary checks added - For POSIX conformance, _POSIX_PTHREAD_SEMANTICS is required. This fixes i.e. sigwait() problems.

Finish the MPM decision logic. This moves the threads checking stuff into the mpm directory where it belongs. It also adds logic so that if threads aren't detected, we automatically choose to use the prefork mpm. Lastly, I also cleaned up a variable name to make it more consistent with the other variables used in our configure script.

First stab at logic to determine which threading library to use. This also gets rid of the hack of always putting -pthread in the CFLAGS variable.

Remove a bunch of functions that are being replaced by functions in APR. Also finished porting Apache to use APR in most cases.

Speed up building, by making --disable-shared the default until we put proper DSO support into the autoconf configuration scripts.

Make building a little easier by not requiring a separate make of APR.

Eliminate the .h.stub files; AC_DEFINE can generate the necessary #defines in the config header for us.

Make Apache's configure script run APR's.

autoconf: Start adding support for selecting modules on the ./configure line. This commit only supports mod_env and mod_log_config.

autoconf: Generate modules.c based on configuration. This is the first step to supporting selection of modules, and it seems to make --with-mpm actually work now.

Fix a couple of comments

Place ap_config_auto.h in src/include. It didn't seem to work earlier, but it does now. *shrug*

Eliminate ap_config.h's checks when using autoconf. Now, ap_ac_config.h is used instead, and autoconf's configure script will check for everything else.

Deal with times() properly in autoconf. This also changes NO_TIMES to !HAVE_TIMES.

Some extra autoconf macros that might help us compile on "a POSIXized ISC UNIX" or MINIX if someone ever tries to do it.

Fix a thinko. Apache's configure.in shouldn't attempt to create APR's Makefile.

The second part of the big autoconf patch, which actually adds autoconf building to the tree.