Issue690317
This issue tracker has been migrated to GitHub,
and is currently read-only.
For more information,
see the GitHub FAQs in the Python's Developer Guide.
Created on 2003-02-20 22:06 by jeffbauer, last changed 2022-04-10 16:06 by admin. This issue is now closed.
Messages (9) | |||
---|---|---|---|
msg14732 - (view) | Author: Jeff Bauer (jeffbauer) | Date: 2003-02-20 22:06 | |
gcc 2.95.2 SunOS 5.6 sparc gcc -DNDEBUG -g -O3 -Wall -Wstrict-prototypes -I. -I./Include -DPy_BUILD_CORE -c ./Modules/posixmodule.c -o Modules/posixmodule.o In file included from Include/stringobject.h:10, from Include/Python.h:83, from ./Modules/posixmodule.c:16: /usr/local/lib/gcc-lib/sparc-sun-solaris2.6/2.95.2/include/stdarg.h:170: warning : redefinition of `va_list' /usr/include/stdio.h:90: warning: `va_list' previously declared here ./Modules/posixmodule.c: In function `posix_access': ./Modules/posixmodule.c:1261: warning: implicit declaration of function `access' ./Modules/posixmodule.c: In function `posix_ttyname': ./Modules/posixmodule.c:1302: warning: implicit declaration of function `ttyname ' ---- <snip> <snip> <snip> --- f' ./Modules/posixmodule.c: In function `posix_confstr': ./Modules/posixmodule.c:6443: warning: implicit declaration of function `confstr ' ./Modules/posixmodule.c: In function `posix_sysconf': ./Modules/posixmodule.c:6985: warning: implicit declaration of function `sysconf ' ./Modules/posixmodule.c: In function `setup_confname_table': ./Modules/posixmodule.c:7024: warning: implicit declaration of function `qsort' make: *** [Modules/posixmodule.o] Error 1 |
|||
msg14733 - (view) | Author: Neal Norwitz (nnorwitz) * | Date: 2003-02-20 23:08 | |
Logged In: YES user_id=33168 Jeff, I don't know if anyone has access to Solaris 5.6. Can you provide a patch to get posixmodule to compile? From the bug report, I only see warnings, not an actual error which is breaking the build. |
|||
msg14734 - (view) | Author: Martin v. Löwis (loewis) * | Date: 2003-03-05 15:40 | |
Logged In: YES user_id=21627 I find several things confusing in this report: - why is va_list redefined? To my knowledge, gcc 2.95.2 supports Solaris 2.6 correctly, so it shouldn't provide a conflicting va_list definition. The typical cause is that the gcc installation doesn't match the operating system version (i.e. fixincludes was run for different headers). Since the gcc installation appears to have been built for 2.6, I infer that the system being used might not be Solaris 2.6. - All the problems reported are warnings, which don't cause compilation to fail. What is the actual error (use -w to suppress all warnings). |
|||
msg14735 - (view) | Author: Stuart Bishop (zenzen) | Date: 2003-03-27 20:54 | |
Logged In: YES user_id=46639 Adding a #include <unistd.h> Fixes the following: ./Modules/posixmodule.c: In function `posix_fchdir': ./Modules/posixmodule.c:1200: `fchdir' undeclared (first use in this function) ./Modules/posixmodule.c:1200: (Each undeclared identifier is reported only once ./Modules/posixmodule.c:1200: for each function it appears in.) ./Modules/posixmodule.c: In function `posix_chroot': ./Modules/posixmodule.c:1237: `chroot' undeclared (first use in this function) ./Modules/posixmodule.c: In function `posix_fsync': ./Modules/posixmodule.c:1249: `fsync' undeclared (first use in this function) ./Modules/posixmodule.c: In function `posix_fdatasync': ./Modules/posixmodule.c:1267: `fdatasync' undeclared (first use in this function Adding #include <stdlib.h> fixes the following: ./Modules/posixmodule.c: In function `posix_system': ./Modules/posixmodule.c:1868: warning: implicit declaration of function `system' ./Modules/posixmodule.c: In function `posix_execv': ./Modules/posixmodule.c:2110: warning: implicit declaration of function `malloc' ./Modules/posixmodule.c: In function `posix_openpty': ./Modules/posixmodule.c:2613: warning: function declaration isn't a prototype ./Modules/posixmodule.c:2632: warning: function declaration isn't a prototype ./Modules/posixmodule.c:2634: warning: implicit declaration of function `grantpt' ./Modules/posixmodule.c:2639: warning: implicit declaration of function `unlockpt' ./Modules/posixmodule.c: In function `posix_putenv': ./Modules/posixmodule.c:5486: warning: implicit declaration of function `putenv' ./Modules/posixmodule.c: In function `posix_tempnam': ./Modules/posixmodule.c:5917: warning: implicit declaration of function `free' ./Modules/posixmodule.c: In function `setup_confname_table': ./Modules/posixmodule.c:6903: warning: implicit declaration of function `qsort' Changing line 187 from: extern char *getcwd(char *, int); to: extern char *getcwd(char *, size_t); And everything builds happily. |
|||
msg14736 - (view) | Author: Martin v. Löwis (loewis) * | Date: 2003-03-27 23:55 | |
Logged In: YES user_id=21627 This gets more and more confusing. Include/Python.h does include both <unistd.h> and <stdlib.h>. Can you find out why it isn't included on your system? |
|||
msg14737 - (view) | Author: Stuart Bishop (zenzen) | Date: 2003-03-28 00:42 | |
Logged In: YES user_id=46639 Lots (possibly all) of the HAVE_*_H tests in configure are failing: In file included from /usr/include/sys/wait.h:25, from /usr/include/stdlib.h:22, from configure:2876: /usr/include/sys/siginfo.h:204: field `__value' has incomplete type I think the offending bits of code in sys/siginfo.h are: #if (!defined(_POSIX_C_SOURCE) && !defined(_XPG4_2)) || \ (_POSIX_C_SOURCE > 2) || defined(__EXTENSIONS__) union sigval { int sival_int; /* integer value */ void *sival_ptr; /* pointer value */ }; #else /* needed in siginfo_t structure */ union __sigval { int __sival_int; /* integer value */ void *__sival_ptr; /* pointer value */ }; #endif [...] #if !defined(_XPG4_2) || defined(__EXTENSIONS__) union sigval __value; #else union __sigval __value; #endif Would _POSIX_C_SOURCE be getting set to a value > 2, which would break siginfo.h? PS. This Solaris 2.6 machine isn't mine and I can't give access to it. |
|||
msg14738 - (view) | Author: Martin v. Löwis (loewis) * | Date: 2003-03-28 01:03 | |
Logged In: YES user_id=21627 Yes, _POSIX_C_SOURCE is defined as 200112L; together with _XOPEN_SOURCE (600) and _XOPEN_SOURCE_EXTENDED (1). Can you please locate the block that sets define_xopen_source in configure and see whether setting this to no helps? It looks to me that Solaris 2.6 is quite broken in this respect, but I would accept adding it to the list of known-broken systems, along with a comment that explains how precisely it breaks. |
|||
msg14739 - (view) | Author: Stuart Bishop (zenzen) | Date: 2003-03-28 01:49 | |
Logged In: YES user_id=46639 That works. Hacking configure so it looks like the following allows python to build: case $ac_sys_system/$ac_sys_release in # On OpenBSD, select(2) is not available if _XOPEN_SOURCE is defined, # even though select is a POSIX function. Reported by J. Ribbens. SunOS/5.6 | OpenBSD/2.* | OpenBSD/3.[012]) define_xopen_source=no;; esac Test summary: 4 tests failed: test_iconv_codecs test_locale test_tarfile test_tempfile 33 tests skipped: test_aepack test_al test_bsddb test_bsddb3 test_bz2 test_cd test_cl test_crypt test_curses test_email_codecs test_gdbm test_gl test_gzip test_imgfile test_linuxaudiodev test_macfs test_macostools test_mpz test_nis test_normalization test_ossaudiodev test_pep277 test_plistlib test_socket_ssl test_socketserver test_sunaudiodev test_timeout test_unicode_file test_winreg test_winsound test_zipfile test_zipimport test_zlib 6 skips unexpected on sunos5: test_sunaudiodev test_bz2 test_unicode_file test_nis test_crypt test_zipimport |
|||
msg14740 - (view) | Author: Martin v. Löwis (loewis) * | Date: 2003-03-28 18:59 | |
Logged In: YES user_id=21627 I have committed a minimally different change as configure 1.386 configure.in 1.397 I'll declare this issue closed; if further problems remain, please open a separate report, or (better yet) submit a patch. |
History | |||
---|---|---|---|
Date | User | Action | Args |
2022-04-10 16:06:59 | admin | set | github: 38020 |
2003-02-20 22:06:14 | jeffbauer | create |