summaryrefslogtreecommitdiff
path: root/Configure
Commit message (Collapse)AuthorAgeFilesLines
* Both cygwin and djgpp on one machineH.Merijn Brand2001-10-031-2/+5
| | | | | Message-Id: <20011003150222.A474.H.M.BRAND@hccnet.nl> p4raw-id: //depot/perl@12314
* Add a few glibc cpp symbols to probe for.Jarkko Hietaniemi2001-09-241-5/+10
| | | p4raw-id: //depot/perl@12167
* Remove iconv references since we are not using itJarkko Hietaniemi2001-09-201-14/+2
| | | | | | | and not going to; regen perltoc, perlmodlib. (Some references still linger in perltoc, have to figure out where they are coming from.) p4raw-id: //depot/perl@12095
* The lseek fix (#11991) was no fix.Jarkko Hietaniemi2001-09-111-2/+2
| | | p4raw-id: //depot/perl@11999
* Configure not recording lseek protoBlair Zajac2001-09-111-3/+3
| | | | | | | Message-ID: <3B9D6055.BA334A31@orcaware.com> plus regen Configure. p4raw-id: //depot/perl@11991
* Configure blooperH.Merijn Brand2001-08-221-1/+1
| | | | | Message-Id: <20010822123750.A797.H.M.BRAND@hccnet.nl> p4raw-id: //depot/perl@11724
* Allow the threads extension to be found if present.Jarkko Hietaniemi2001-08-171-1/+9
| | | p4raw-id: //depot/perl@11705
* Introduce more floating point classifying APIs;Jarkko Hietaniemi2001-08-091-1/+73
| | | | | | | | | | | | | try to define (most importantly) Perl_is{inf,nan}(x). For this first cut, assume none such APIs exist in non-UNIXoid platforms-- this is of course too unkind: Win32 rumoredly has _fpclass() and _isnan(), and VMS/VOS very probably have some of the APIs available. The only thing Perl uses at the moment is Perl_isnan() (has been in use for a while now). NOTE: the Win32 and NetWare config_h.?c need regenerating. p4raw-id: //depot/perl@11618
* Multiple modules didn't work; plus typo fixes.Jarkko Hietaniemi2001-07-311-9/+10
| | | p4raw-id: //depot/perl@11524
* Add -Wall in cflags, not in Configure.Jarkko Hietaniemi2001-07-311-9/+1
| | | | | | | (This way -Wall affects only the core and the core XS, not user modules' XS. Concern from Paul Marquess, implementation idea from Andy Dougherty.) p4raw-id: //depot/perl@11516
* Configure's detection of system manualSean Dague2001-07-311-4/+6
| | | | | Message-ID: <20010513155315.A15833@underhill.dague.net> p4raw-id: //depot/perl@11514
* extras.make, extras.test, and extras.install make targets.Jarkko Hietaniemi2001-07-301-4/+3
| | | p4raw-id: //depot/perl@11495
* Better to have the list of extras in $extras than 'define'.Jarkko Hietaniemi2001-07-301-1/+2
| | | p4raw-id: //depot/perl@11493
* Add the support of asking for extra modules or bundles of modulesJarkko Hietaniemi2001-07-301-1/+54
| | | | | | | | to be fetched from the CPAN and installed as a part of the Perl build process. None of this installation is yet really done; only the list is asked, and saved away in extras.lst for the build and installation process to worry about later. p4raw-id: //depot/perl@11490
* Darwin: -fpic not so goodWilfredo Sánchez2001-07-291-1/+2
| | | | | Message-Id: <EC1496A6-82FB-11D5-BFC6-003065D59FBA@thor.sbay.org> p4raw-id: //depot/perl@11477
* Too volatile failsafe check.Jarkko Hietaniemi2001-07-231-3/+3
| | | p4raw-id: //depot/perl@11450
* #include <stdlib.h> so that exit() is prototyped,Jarkko Hietaniemi2001-07-141-1/+3
| | | | | noticed by Hugo. p4raw-id: //depot/perl@11369
* Add HAS_PTHREAD_ATFORK.Jarkko Hietaniemi2001-07-121-1/+7
| | | p4raw-id: //depot/perl@11334
* Re: [PATCH Configure] perl -V brokenAndy Dougherty2001-07-121-5/+10
| | | | | Message-ID: <Pine.SOL.4.10.10107111329060.2021-100000@maxwell.phys.lafayette.edu> p4raw-id: //depot/perl@11286
* Detypo, problem noticed by Gerrit P. Haase.Jarkko Hietaniemi2001-07-081-2/+2
| | | p4raw-id: //depot/perl@11217
* Add I18N::Langinfo, which is basically a wrapper aroundJarkko Hietaniemi2001-07-081-1/+6
| | | | | | nl_langinfo(), which is an additional way to query locale specific information. p4raw-id: //depot/perl@11207
* Somehow the #ifdefs of the added code (in #11093) madeJarkko Hietaniemi2001-07-041-7/+1
| | | | | | | HP-UX to fail the fcntl locking test, without the ifdefs the test seems to be working again. Reason unknown: HP-UX cc doesn't complain either way. p4raw-id: //depot/perl@11146
* Whitespace allowed at the ends of /lib/syscalls.exp linesJarkko Hietaniemi2001-07-021-2/+2
| | | | | | (from Richard Hatch) (this was the cause of pipes() and times() mysteriously not being found) p4raw-id: //depot/perl@11115
* The fcntl locking test may hang if NFS locking messed up;Jarkko Hietaniemi2001-07-021-2/+24
| | | | | break out with alarm(10). p4raw-id: //depot/perl@11093
* I thought this Configure glitch for AIX was just recently fixed?Jarkko Hietaniemi2001-06-291-2/+2
| | | p4raw-id: //depot/perl@11027
* Add Configure probes for nl_langinfo() and <langinfo.h>.Jarkko Hietaniemi2001-06-281-1/+13
| | | | | | Expected not to exist in non-UNIX excepting in VMS, where according to a quick web survey they just might. p4raw-id: //depot/perl@11003
* Move use64bitint and use64bitall before uselargefiles.Jarkko Hietaniemi2001-06-281-394/+394
| | | p4raw-id: //depot/perl@11000
* Re: [PATCH 5.6.1] OS2 ConfigureIlya Zakharevich2001-06-241-1/+13
| | | | | Message-ID: <20010624052440.A27698@math.ohio-state.edu> p4raw-id: //depot/perl@10892
* In Cygwin do not build NDBM_File and ODBM_File unlessJarkko Hietaniemi2001-06-231-3/+3
| | | | | -lndbm and -ldbm are available. p4raw-id: //depot/perl@10863
* Cleanup of the cross-compilation unit.Jarkko Hietaniemi2001-06-221-14/+17
| | | p4raw-id: //depot/perl@10837
* Further tiny bcopy cleanupAndy Dougherty2001-06-221-43/+41
| | | | | Message-ID: <Pine.SOL.4.10.10106221417030.28044-100000@maxwell.phys.lafayette.edu> p4raw-id: //depot/perl@10831
* Tell that you are removing the unneeded -lsfio.Jarkko Hietaniemi2001-06-221-5/+10
| | | p4raw-id: //depot/perl@10807
* More d'ohs.Jarkko Hietaniemi2001-06-211-2/+2
| | | p4raw-id: //depot/perl@10802
* Add fchdir() probe (in non-UNIX systems assume it doesn't exist).Jarkko Hietaniemi2001-06-211-1/+7
| | | p4raw-id: //depot/perl@10797
* Cross-configuration changes: make $from to rm the filesJarkko Hietaniemi2001-06-211-9/+55
| | | | | | | | | | before copying them over (this means that he rm can be removed from the fflushall test), create $targetmkdir (not documented in install since it is supposed to internal use only), make $run to copy over the executable only if a stamp file .xok doesn't already exist (to avoid unnecessary copying of the executable) p4raw-id: //depot/perl@10788
* Do not remove the ccsym* Cppsym* files sinceJarkko Hietaniemi2001-06-201-2/+1
| | | | | | they are useful for metaconfig units coming after us. (This is the Configure nit mentioned in #10752.) p4raw-id: //depot/perl@10760
* After some reading I don't think we can blindlyJarkko Hietaniemi2001-06-201-7/+1
| | | | | | | | | use systems' realpath(). Too many security problems, too many buggy implementations. TODO: the realpath() emulation code in util.c doesn't seem fully operational? (readlink(), for example?) p4raw-id: //depot/perl@10744
* [ID 20010619.012] Not OK: perl v5.7.1 +DEVEL10721 +devel-10722 on ↵Spider Boardman2001-06-191-1/+8
| | | | | | | | | | | alpha-dec_osf-per lio 4.0f (UNINSTALLED) Message-Id: <200106192315.TAA18531@Orb.Nashua.NH.US> The bigintpm.t #1183 failing in Tru64 (and also Cygwin) because "1e+129" being expected but "1.e+129" being returned. The culprit (at least in Tru64) is the use of gcvt(). p4raw-id: //depot/perl@10738
* [ID 20010619.011] Not OK: perl v5.7.1 +DEVEL10721 +devel-10722 on ↵Spider Boardman2001-06-191-4/+7
| | | | | | | | | alpha-dec_osf-per lio 4.0f (UNINSTALLED) Message-Id: <200106192048.QAA20849@Orb.Nashua.NH.US> Further fixes to fflush(NULL) detection. p4raw-id: //depot/perl@10733
* AIX test -s hangs with non-existent files?Jarkko Hietaniemi2001-06-191-4/+5
| | | | | (A genuine logic bug in Configure, but AIX also broken?) p4raw-id: //depot/perl@10732
* More cross-compilation defaults gleaned from -DccJarkko Hietaniemi2001-06-181-31/+66
| | | | | if it is CPU-OS-gcc. p4raw-id: //depot/perl@10690
* The extraction changed directories.Jarkko Hietaniemi2001-06-161-1/+3
| | | p4raw-id: //depot/perl@10629
* Move the initialization of libnames to the beginning soJarkko Hietaniemi2001-06-161-2/+2
| | | | | that can one 'seed' it (as OS/2 does.) p4raw-id: //depot/perl@10624
* The first steps towards cross-compilation.Jarkko Hietaniemi2001-06-151-158/+337
| | | | | | | | | | | | | | | | | | | | | | Abstract execution of compiled test executables with $run, and abstract transfer of files with $to and $from. Under cross-compilation the $run, $to, and $from will point to appropriate wrapper scripts, by default ssh and scp, but also rsh, rcp, and cp are supported. If not cross-compiling, they will be set to '', ':', and ':', respectively. With these patches I was able to get Configure for iPAQ ARM Linux on an Intel Linux about 95% right (only a few tests failed to execute or they produced incorrect results), and I was able to compile a functional miniperl. The symbol crosscompile renamed to be usecrosscompile, the corresponding C symbol from CROSSCOMPILE to USE_CROSS_COMPILE. p4raw-id: //depot/perl@10592
* It would be also be good if I could type.Jarkko Hietaniemi2001-06-131-2/+2
| | | p4raw-id: //depot/perl@10573
* A non-empty default is a good thing.Jarkko Hietaniemi2001-06-131-1/+2
| | | p4raw-id: //depot/perl@10571
* Add Configure directive -Dusereentrant for Artur's ithreads work,Jarkko Hietaniemi2001-06-131-1/+3
| | | | | extremely experimental, not even prompted for. p4raw-id: //depot/perl@10569
* Buglet found by Mark Bixby: "./try >&2 >/dev/null" shouldJarkko Hietaniemi2001-06-111-3/+3
| | | | | be "./try 2>&1 >/dev/null". p4raw-id: //depot/perl@10522
* Don't build Thread.pm if USEITHREADSArtur Bergman2001-06-101-3/+6
| | | | | Message-ID: <B7483FB6.1398%artur@contiller.se> p4raw-id: //depot/perl@10492
* Be inspired more by Hugo-- introduce HAS_MODFL_POW32_BUG.Jarkko Hietaniemi2001-06-081-2/+6
| | | p4raw-id: //depot/perl@10479
View Lorry Controller Status