Commit Graph

116 Commits

Author SHA1 Message Date
Guido van Rossum 02a1c40051 Pat Knight:
Solaris 2 has stub implementations of the POSIX thread functions such as
pthread_detach in libc. This means that configure tries to use them without
-lpthread, then the test of pthread_create fails and the configuration
falls back to the Solaris thread library. This patch moves the test for
pthread_create in -lpthread ahead of the test for pthread_detach in libc.
The patch also ensures that -lpthread is at the start of the library list
when linking, to pick up POSIX thread semantics for fork (see below).

Justification.
Use of POSIX threads on Solaris ensures that the fork() call only runs the
thread that called fork() in the child. This is desirable to prevent (for
example) parent server or database threads running in the child. Sun's
-lthread library uses a traditional fork() which replicates all the
parent's threads in the child. I find this undesirable.

Digression.
The configure.in seems to always test for -lthread even if a POSIX library
is found. I'm not enough of a configure.in wizard to decide whether this is
desirable or how to fix it. It is also irrelevant to this patch - I just
spotted it while testing.
End of Digression.
2000-02-25 19:26:31 +00:00
Guido van Rossum 9c6ba9b598 According to Harri Pasanen, the Solaris+gcc test for LINKFORSHARED
needed an adaptation for the latest gcc -- we need to grep stderr as
well as stdout for the magic string BFD.  (Does anybody know what BFD
means?)
2000-02-03 13:42:50 +00:00
Guido van Rossum e97ee18115 Changes to auto-detect the correct dynload_<platform>.c file.
NOTE: Windows, Mac and OS/2 build procedures must be adapted manually!

This is part of a set of patches by Greg Stein.
1999-12-20 21:27:22 +00:00
Guido van Rossum 9501219303 Patch by Geoff Furnish to make compiling with C++ more gentle.
(The configure script is regenerated, not from his patch.)
1999-12-16 17:50:52 +00:00
Fred Drake f581db15a3 Remove test for getlogin_r(); the interface is not clearly defined, at
least on Solaris (sometimes it's Unix98, sometimes it conforms to an
early draft).

Properly generate config.h.in using autoheader instead of editing it
manually; thanks, Guido!
1999-12-14 21:11:47 +00:00
Fred Drake 3bb3e5773b Added detection of getlogin() and getlogin_r(). 1999-12-14 20:48:17 +00:00
Fred Drake 35a092fece Added detection for getgroups(), fpathconf(), pathconf(),
confstr(), and sysconf().
1999-12-13 16:23:35 +00:00
Fred Drake 89143df520 Added support for ctermid, tempnam, tmpfile, tmpnam, and tmpnam_r. 1999-12-09 22:03:25 +00:00
Guido van Rossum 0eefa3fbbf Patch by Vladimir Marangozov, inspired by a bug report from Gary
Duzan, for AIX, to support C++ objects with static initializers, when
using the genuine IBM C++ compiler (namely xlC/xlC_r).

See accompanying patches to acconfig.h and importdl.c.
1999-11-16 15:57:37 +00:00
Guido van Rossum f5957eaf0a Dynamic linking support for BSD/OS 4.x as suggested by Vivek Khera 1999-10-05 21:59:33 +00:00
Guido van Rossum 458e7fab9b Erik Lindvall contributed a BSD/OS 4 patch for LDSHARED. 1999-09-17 15:40:40 +00:00
Guido van Rossum a8acf72afa Need to double the macro brackets in FreeBSD check.
(My fault, not Dom's -- he submitted a correct patch to configure.)
1999-04-27 22:06:08 +00:00
Guido van Rossum eeb63fd6c2 FreeBSD/[34]* change, Dom Mitchell. 1999-04-21 17:03:13 +00:00
Guido van Rossum 8bc1dfd2fb Quote a bunch of shell variables used in test, related to long-long. 1999-04-10 16:01:48 +00:00
Guido van Rossum 8137680704 For BeOS PowerPC. Chris Herborth. 1999-04-06 15:50:51 +00:00
Guido van Rossum 6b9da45969 Chris Herborth: the new compiler in R4.1 needs some new options to work... 1999-03-24 17:48:12 +00:00
Guido van Rossum 8db7d8bf6b Ack! It never defined HAVE_GETHOSTBYNAME_R so that code was never tested! 1999-03-23 16:40:33 +00:00
Guido van Rossum a96f0ba7cc Add code to test for all sorts of gethostbyname_r variants,
donated by David Arnold.
1999-03-22 21:49:51 +00:00
Guido van Rossum caa69fdf4c Should test for gethost*by*name_r, not for gethostname_r (which
doesn't exist and doesn't make sense).
1999-03-15 21:40:59 +00:00
Guido van Rossum 24f8579ee4 Get rid of AC_CHECK_LIB(m, pow) since this is taken care of later with
LIBM (from --with-libm=...); this actually broke the customizability
offered by the latter option.  Thanks go to Clay Spence for reporting
this.
1999-02-23 18:00:56 +00:00
Guido van Rossum 54ecc3d24f Patches by William Lewis for Nextstep descendants. 1999-01-27 17:53:11 +00:00
Guido van Rossum 7a5f420e67 Chris Herborth's first pass at making config and install
for BeOS better.
1999-01-12 20:30:23 +00:00
Guido van Rossum df69365f5f Ty Sarna writes:
The following patches (relative to 1.5.2b1) enable Python dynamic
loading to work on NetBSD platforms that use ELF (presnetly mips and
alpha systems).  They automaticly determine wether the system is ELF or
a.out rather than using astatic list of platforms so that when other
NetBSD platforms move to ELF, python will continue to work without
change.
1999-01-07 21:50:41 +00:00
Guido van Rossum 00f0f6ef03 Changes for long file support by Steve Clift.
(Slightly modified to isolate the test for various funcs more.)
1999-01-06 18:52:29 +00:00
Guido van Rossum 2666a15576 Several tiny changes:
- remove PowerPC specific compiler switch for BeOS (Chris Herborth)
- FreeBSD/3 support
- check for pthread_detach in libc instead of _create
1998-12-17 18:07:35 +00:00
Guido van Rossum dd997f79b4 Add test for BSDI -- by Nigel Head and otto@mail.olympus.net. 1998-10-07 19:58:26 +00:00
Guido van Rossum fb4533c124 Added tweak for SET_CCC (which I don't even know what it is!) for Linux.
Added AC_CHECK_SIZEOF(void *).
Added yet another test for pthreads (pthread_create may be in libc).
Added pthread_init to list of functions whose existence is checked.
1998-10-02 14:39:42 +00:00
Guido van Rossum ec95c7bd2b Changes for BeOS, QNX and long long, by Chris Herborth. 1998-08-04 17:59:56 +00:00
Guido van Rossum f867812288 Fix two unrelated problems in the same area:
(1) reorder the tests for -Olimit 1500 and -OPT:Olimit=0 so that the
latter test is performed first, and if it works, the former test is
skipped.  This should get rid of the problem that the new SGI
compilers accept both but emit a warning about -Olimit 1500.

(2) The DGUX hack was somehow split in two by the Olimit tests,
probably as the result of a non-context diff.  Moved this back
together again, after the Olimit tests.
1998-07-07 21:05:09 +00:00
Guido van Rossum 08d6be4871 Add timegm to list of functions tested, for Marc-Andre L. 1998-06-11 15:19:05 +00:00
Guido van Rossum b93a86297d support hpux pthreads 1998-05-07 13:27:32 +00:00
Guido van Rossum 2cb5660da1 Get rid of test for -lsun (very old IRIX); add test for -lm (needed on
enough platforms).
1998-05-07 13:25:56 +00:00
Guido van Rossum ee21f418b0 Remove bogus CCSHARED for cc on IRIX 6.x. 1998-04-20 18:51:54 +00:00
Guido van Rossum 1f74016735 Support for OpenBSD :-( 1998-04-13 15:27:21 +00:00
Guido van Rossum 70c7f48b9b Get rid of the check for -linet -- recent Sequent Dynix systems don't
need this any more and apparently it screws up their configuration.
1998-03-26 18:44:10 +00:00
Guido van Rossum 1809975a29 Added mktime to checked functions 1998-03-06 17:14:34 +00:00
Guido van Rossum 56896dd5bb Setting LINKFORSHARED for IRIX was a mistake. 1998-03-05 16:59:36 +00:00
Guido van Rossum 2b5ca00305 Some changes because gcc on SGI doesn't support '-all' 1998-03-05 15:41:09 +00:00
Guido van Rossum 69f65801ab Misc changes for AIX: install the python.exp file, and properly use $(srcdir). 1998-03-03 21:47:00 +00:00
Guido van Rossum 357428567c Different way to suppores the DGUX rule 1998-02-19 21:32:40 +00:00
Guido van Rossum 8f006272fe Some shifting around of the DGUX hack so that it won't break on DEC Unix
where Make doesn't know that '#' as a command is a comment.
1998-01-13 19:03:09 +00:00
Guido van Rossum 3c4bb802d0 Patch by Brian Gallew for DG/UX. I'm not quite sure what it does but
it seems harmless for other platforms.  It plays tricks with the name
of the library used to link with.  Apparently DG/UX really wants a
shared library to link with if it wants shared modules to use symbols
from the library.  I'm not sure why this wasn't an issue with 1.4;
DG/UX seems to be the only platform where moving to a single library
made things harder!

BTW This adds a target to create libpython$(VERSION).so; however this
target is *only* for DG/UX.
1997-12-18 23:55:32 +00:00
Guido van Rossum 8f4ceb168c Two unrelated changes:
- add test for strptime (not used by the core but needed by Marc Lemburg's
Date object).

- Test for GNU ld on Solaris; need to add an extra linker option to
export symbols in that case.
1997-12-18 23:42:19 +00:00
Guido van Rossum 009f78754d When --with-thread(s) is specified, the thread module is now automatically
enabled.  This is done through a substitution in Modules/Setup.thread(.in).
Bill Janssen will be happy.  The original idea was by Lele Gaifax (though
I changed the implementation to use a separate file).
1997-12-04 00:51:42 +00:00
Guido van Rossum 4954595036 Changes for FreeBSD and NetBSD: NetBSD does shared libs just like FreeBSD
(big surprise), and FreeBSD has pthread support in -lc_r -- maybe NetBSD
has too.
1997-12-02 19:28:29 +00:00
Guido van Rossum 2d9feeddf3 Change the option used on Linux to export all symbols
from the main program to shared libraries.  On mklinux, the old
'-rdynamic' doesn't work; the new '-Xlinker -export-dynamic' works
both there and on Intel Linux platforms.
1997-11-24 17:36:21 +00:00
Guido van Rossum 2efa34ba9b Daniel Larsson discovered another cut-n-paste bug in the Olimit story... 1997-10-23 17:43:11 +00:00
Guido van Rossum 304dd2defb Switch around OPT assignments in Don Beaudry's patch for Olimit (his
code had the assignments switched around).

Also rename sharedmodules to lib-dynload.
1997-10-20 23:10:56 +00:00
Guido van Rossum 91922677ea Don Beaudry's changes to support SGI_ABI on Irix 6.x. 1997-10-09 20:24:13 +00:00
Guido van Rossum 9601315140 Add checks for setlocale() and <locale.h> 1997-10-08 22:47:52 +00:00