summaryrefslogtreecommitdiff
path: root/Configure
Commit message (Collapse)AuthorAgeFilesLines
* [perl #64126] ./Configure -de -Dusevendorprefix didn't defaultH.Merijn Brand2013-05-271-4/+5
|
* [perl #66604] 64bit compilation Problem on Solaris x86_64Andy Dougherty2013-05-101-1/+14
| | | | Configure changes
* Spell-checking on meta-unitsH.Merijn Brand2013-05-091-32/+32
| | | | | | Tried to be as consistent as possible. Removed all trailing whitespace of changed files. Resyncs Glossary from previously missed updates.
* maildomain to support /etc/mailnameH.Merijn Brand2013-05-091-1/+7
| | | | Now backported to upstream dist/meta
* The URL for the dist source repo changedH.Merijn Brand2013-05-071-2/+2
|
* Get Configure in sync with metaconfigH.Merijn Brand2013-03-091-2/+3
|
* Add bootstrap_charset/BOOTSTRAP_CHARSET for non-ASCII architecturesH.Merijn Brand2013-03-091-29/+36
|
* Configure: bail out when neither stdarg.h nor varargs.hAaron Crane2013-03-031-0/+10
| | | | | | | | | | | Perl uses variadic functions, so we need at least one of the relevant mechanisms available; for this reason, the CompCert C compiler is unsuitable for our purposes. If the person building Perl wants to carry on regardless, they can use the "-K" option. This partly addresses RT#112494.
* Fix [perl #116523] Configure is confused by '=' in output from uname -aAndy Dougherty2013-02-191-1/+5
| | | | | | | | | | | | | | | | | | | | | | Previously, Configure used to try to post-process the output of uname -a. This was to support Xenix, which gave output like this: sysname=XENIX nodename=whatever release=2.3.2 .. etc... Configure attempted to parse that output and build up a $myuname variable that resembled those generated on traditional Unix systems. More recently, a user reported uname -a output on a CVS checkout of NetBSD that also included information such as root=username@hostname:/dirname module=src branch=HEAD The Configure code misparsed that and ended up throwing most of it away. This patch takes the simplest approach and just skips the attempt to process the Xenix output. This is based on commit 05a827803ab3908a795df94f6ed3890e24a6df2 in the metaconfig repository.
* use_versioned_archname => versionedarchnameH.Merijn Brand2013-02-151-11/+9
| | | | (now synced with meta)
* Add use_versioned_archname to ConfigureDavid Golden2013-02-051-0/+30
| | | | | | | | | | | | | | | | | | | | | This patch adds 'use_versioned_archname' as a Configure option. When set, it includes 'api_versionstring' in 'archname'. E.g. x86_64-linux-5.13.6-thread-multi. It is unset by default. This feature was requested by Tim Bunce, who observed that INSTALL_BASE creates a library structure that does not differentiate by perl version. Instead, it places architecture specific files in "$install_base/lib/perl5/$archname". This makes it difficult to use a common INSTALL_BASE library path with multiple versions of perl. By setting -Duse_versioned_archname, the $archname will be distinct for architecture *and* API version, allowing mixed use of INSTALL_BASE. This patch demonstrates the desired effect for discussion. Actual changes would need to be made upstream in metaconfig and integrated back into blead.
* Let Configure honor the external MAILDOMAIN environment variable, if set.H.Merijn Brand2013-01-281-3/+7
|
* Remove the Rhapsody port.Nicholas Clark2013-01-161-1/+1
| | | | | | | Rhapsody was an Apple OS that later evolved into Darwin and Mac OS X. It was initially only released to developers, but later became Mac OS X Server, with releases in 1999 and 2000. It was obsoleted by Mac OS X 10.0, released in March 2001.
* Remove register keywork from randfunc tests in ConfigureH.Merijn Brand2012-12-241-4/+4
|
* Remove BeOS special-casing from Configure.Nicholas Clark2012-12-141-11/+2
|
* Fix [perl #114812] Configure not finding isblank().Andy Dougherty2012-09-101-2/+2
| | | | | | Configure would not find isblank() when run with g++ because the probe used exit() without including <stdlib.h>. The simplest fix is to have the probe use return instead.
* Fix alignment for darwin with -Dusemorebits.Andy Dougherty2012-09-071-3/+13
| | | | | | | | | | | | | | | | | By default, the darwin build assumes a "multiarchitecture" build. Configure has a hardwired default of '8' for alignbytes (and then proceeds to ignore it with another hard-wired '8' in config.h). That '8' was supposed to be a safe value, in case perl was built on one architecture but run on another with a stricter constraint. With darwin and -Dusemorebits, however, the alignment should be on 16-byte boundaries. We don't want to penalize all darwin builds for this unlikely configuration, but we do want to allow it. This patch causes Configure to compute alignbytes even for multiarch builds, but if the result is less than 8, it sets it to 8 (which preserves the previous behavior). If, however, alignbytes is 16, Configure won't decrease it. Then, this patch also fixes config_h.SH so that it uses the value determined by Configure instead of the previous hardwired value.
* Add Configure probe for ip_mreq_sourceH.Merijn Brand2012-08-261-1/+41
| | | | Needed to upgrade Socket from CPAN
* Fix RT [perl #72156] Re: Perl 5.12.0 RC 0 - Pager detectionAndy Dougherty2012-08-021-1/+1
| | | | | | | | Instruct ./getfile to trust the default pager value. It might not begin with a slash, or it might include some options, such as "/usr/bin/less -R". This is based on commit 53ddb3ba3e23ab80444c30b4deefa64114408438 in the metaconfig repository.
* Add IP probe for ip_mreqH.Merijn Brand2012-07-111-1/+21
| | | | Backport 2f1eb816b5cba6977b1a8159
* Configure: Avoid Cppsym warnings for extra tokens [perl #113024]Andy Dougherty2012-06-061-2/+3
| | | | | | | | | | The cppsymbols can include macros such as __INT16_C(c), which can't be tested with a simple #ifdef. This patch strips off the opening parenthesis and everything following it. These macros were generated by cpp -dM. Also ensure Cppsym.true list is sorted for later input to comm. (I noticed this while testing this change on Solaris.)
* Replace $compile_ok by $compile for two probes that can fail.Andy Dougherty2012-06-061-2/+2
| | | | | | The probes for _NSGetExecutablePath and KERN_PROC_PATHNAME will fail on systems where they are not available. This is normal, so use $compile to throw away the error messages.
* Determine if ELF even if ld is not ''H.Merijn Brand2012-04-011-7/+12
|
* Add ld_can_script probe to ConfigureH.Merijn Brand2012-03-311-1/+31
| | | | | | | With this new probe, requested by rafl, we could start building perl as it would be built on OSs like AIX and Windows and actually test visibility of the API and thus early detect if tests would fail on these restrictive OSs
* Add new probes for IPv6 (LeoNerd)H.Merijn Brand2012-02-161-1/+41
|
* Revert "Add strptime probe"Ævar Arnfjörð Bjarmason2012-02-151-7/+1
| | | | This reverts commit 8852e312c3c616ab731ccbe7da54fb04eb8c3d30.
* Add strptime probeH.Merijn Brand2012-02-121-1/+7
|
* Add blank lines around Configure's duplicate extension diagnostic.Nicholas Clark2012-01-311-1/+3
| | | | | | | The diagnostic message containing the duplicate extension did not stand out, particularly given that the immediately previous output is a very long line of found extensions. So add blank lines on either to make it easier to spot.
* Configure now aborts with a diagnostic if it detects a duplicate extension.Nicholas Clark2012-01-311-0/+17
| | | | | | | | | | | | | | | | | | During the development process, if an extension module changes dual-life status it can move between ext/, dist/ and cpan/. If one uses git pull in a built tree across such a change, one can end up with the build tree in a messy state from which the 'distclean' target can't recover. Configure will spot both the old and new directories, and may wrongly categorise the extension as non-XS, resulting in obscure build failures after miniperl is build. In the general case, the "old" directory might contain files the user wants to keep, so it's not safe to automatically delete it. The only safe option is to ask the user to resolve the problem, hence with this change, if Configure detects such a situation it aborts with a diagnostic suggesting what to do. This isn't a problem that affects released tarballs.
* mymalloc isn't thread safe (rurban)H.Merijn Brand2012-01-271-1/+2
|
* 'A' is not blankH.Merijn Brand2012-01-091-2/+2
|
* Add probe for isblank() (requested by khw)H.Merijn Brand2012-01-091-1/+27
|
* Regen Configure and friendsH.Merijn Brand2011-10-011-217/+217
| | | | After backporting Nicholas' work and slimming down metaconfig.h
* Teach Configure about "procselfexe" on Solaris and NetBSDNicholas Clark2011-09-281-1/+3
| | | | | | | | | | | | Configure would already find /proc/self/exe on NetBSD, where /proc/self is a symlink to /proc/curproc. However, the revised probe avoids the extra symlink traversal. Configure did not previously probe for the relevant path on Solaris, /proc/self/path/a.out Rename the description of /proc/curproc/file from BSD to FreeBSD, as it seems that of the "big 3" BSDs, only FreeBSD uses this path. Based on a patch from Johann 'Myrkraverk' Oskarsson.
* In Configure, refactor the test for procselfexe into a loop.Nicholas Clark2011-09-271-15/+14
| | | | | This removes code duplication, and makes it easy to add more variants. Based on a patch from Johann 'Myrkraverk' Oskarsson.
* Where available, use _NSGetExecutablePath() to make $^X absolute.Nicholas Clark2011-09-271-0/+102
| | | | | | | | | In Configure, check whether _NSGetExecutablePath() can be used to find the absolute pathname of the executable. If so, set usensgetexecutablepath in config.sh and USE_NSGETEXECUTABLEPATH in config.h. If this is set, then use this approach in S_set_caret_X() to canonicalise $^X as an absolute path. This approach works on OS X, and possible on other platforms that use dyld.
* Where available, use sysctl() with KERN_PROC_PATHNAME to make $^X absolute.Nicholas Clark2011-09-271-0/+117
| | | | | | | | | | In Configure, check whether sysctl() and KERN_PROC_PATHNAME can be used to find the absolute pathname of the executable. If so, set usekernprocpathname in config.sh and USE_KERN_PROC_PATHNAME in config.h. If this is set, then use this approach in S_set_caret_X() to canonicalise $^X as an absolute path. This approach works on (at least) FreeBSD, and doesn't rely on the /proc filesystem existing, or /proc/curproc/file being present.
* Use a real compile to test for stdbool.hH.Merijn Brand2011-09-221-3/+28
| | | | | | | | | On some Solaris systems, <stdbool.h> is present, but will give an error if used by a compiler that is not sufficiently c99-compliant. Check for this by including <stdbool.h> and trying to compile a short program that uses bool. Signed-off-by: H.Merijn Brand <h.m.brand@xs4all.nl>
* Regenerate. Sorry for the massive re-orderH.Merijn Brand2011-09-161-6/+6
|
* Probe for <stdbool.h>, and if found use it in handy.hNicholas Clark2011-09-161-0/+6
| | | | | | | | | This means that the core uses the compiler's bool type if one exists. This avoids potential problems of clashes between perl's own implementation of bool and the compiler's bool type, which otherwise occur when one attempts to include headers which in turn include <stdbool.h>. Signed-off-by: H.Merijn Brand <h.m.brand@xs4all.nl>
* Revert "Fix Configure's csym test for gcc's link time optimisation"H.Merijn Brand2011-08-261-8/+8
| | | | | | There are too many related problems arising from this change This reverts commit e820c6d6a6d0a8828aa68a6895696b659c471f2f.
* Fix Configure's csym test for gcc's link time optimisationH.Merijn Brand2011-08-221-8/+8
| | | | | This introduces a volatile into the test program so gcc cannot optimise out the symbol itself as being unused.
* Add st_ino size/sign probesH.Merijn Brand2011-07-311-1/+73
| | | | Work initiated by Tony Cook
* Chainsaw out the possibility to disable useperlioH.Merijn Brand2011-07-311-115/+32
| | | | This work was actually committed to metaconfig on 16 Feb 2011
* Subject: optimize matching -g still enables -DDEBUGINGH.Merijn Brand2011-02-141-2/+2
| | | | | | | | | | | | | | | | | | | | | | | From: Nicholas Clark <nick@ccl4.org> Date: Mon, 14 Feb 2011 16:01:49 +0000 Message-ID: <20110214160149.GU24189@plum.flirble.org> INSTALL says this: =item -DEBUGGING=-g =item -Doptimize=-g Adds -g to optimize, but does not set -DDEBUGGING. (Note: Your system may actually require something like cc -g2. Check your man pages for cc(1) and also any hint file for your system.) However, we *aren't* compliant with our documentation, as currently -Doptimize=-g (or anything containing -g) still causes -DDEBUGGING to be added to the C flags. Signed-off-by: H.Merijn Brand <h.m.brand@xs4all.nl>
* From: Benny Siegert <bsiegert@gmail.com>H.Merijn Brand2011-01-231-2/+2
| | | | | | | | | | | | Date: Thu, 20 Jan 2011 19:27:04 +0100 Subject: [PATCH] MirBSD: fix rpath for shared libperl If you build perl with a shared libperl.so on MirBSD (the default config), it will work up to the installation; however, once installed, it will be unable to find libperl. Treat rpath handling like in the other BSD dialects. Signed-off-by: H.Merijn Brand <h.m.brand@xs4all.nl>
* Re-generated Configure after the spell-check fixes from Peter J. AcklamH.Merijn Brand2011-01-071-6/+6
| | | | | | | 1. re-generate Configure and config_h.SH 2. update Porting/config_H, which was very outdated Signed-off-by: H.Merijn Brand <h.m.brand@xs4all.nl>
* No reference to $Id: in generated filesH.Merijn Brand2010-12-211-3/+1
| | | | | | They might have served a purpose in the original files, but Nicholas and Zefram expressed their concern that in the generated files, these tags are misleading and unneeded.
* Restore a reference to $perlpath. Installation needs itH.Merijn Brand2010-12-211-70/+103
| | | | cbeaa1895 removed the reference to $perlpath
* Add sin6_scope_id probe (LeoNerd)H.Merijn Brand2010-12-201-103/+90
|
View Lorry Controller Status