AC_SYS_LARGEFILE and old OSX framework

2013-10-30 Thread Yann Diorcet
I notice an issue with autoconf when i'm using an old osx framework (or old version) : -mmacosx-version-min < 10.6 When we use AC_SYS_LARGEFILE the autoconf add # define _DARWIN_USE_64_BIT_INODE 1 in the config.h due to AH_VERBATIM([_DARWIN_USE_64_BIT_INODE], [/* Enable large inode numbers on

Re: AC_SYS_LARGEFILE and old OSX framework

2013-10-30 Thread Paul Eggert
On 10/30/2013 05:38 AM, Yann Diorcet wrote: > This may cause segmentation fault on generated file (certainly linked to a > different struct stat) This suggests that (1) parts of your software package includes config.h, which defines _DARWIN_USE_64_BIT_INODE and gets one struct stat, (2) other part

Re: AC_SYS_LARGEFILE and old OSX framework

2013-10-30 Thread Diorcet Yann
Le 30/10/2013 20:41, Paul Eggert a écrit : > On 10/30/2013 05:38 AM, Yann Diorcet wrote: >> This may cause segmentation fault on generated file (certainly linked to a >> different struct stat) > This suggests that (1) parts of your software package includes config.h, > which defines _DARWIN_USE_64_

Re: AC_SYS_LARGEFILE and old OSX framework

2013-10-30 Thread Paul Eggert
On 10/30/2013 12:46 PM, Diorcet Yann wrote: > Could be an explanation. but how do you explain the issue on that > https://gitorious.org/glib/android/source/003dbec6c84798cb5c6a8759258f2b7180e5fa10:glib/gkeyfile.c#L354-1871 > S_ISREG is done just after fstat ... and there is #include "config.h". I

Re: AC_SYS_LARGEFILE and old OSX framework

2013-10-30 Thread Diorcet Yann
Le 30/10/2013 20:56, Paul Eggert a écrit : > On 10/30/2013 12:46 PM, Diorcet Yann wrote: >> Could be an explanation. but how do you explain the issue on that >> https://gitorious.org/glib/android/source/003dbec6c84798cb5c6a8759258f2b7180e5fa10:glib/gkeyfile.c#L354-1871 >> S_ISREG is done just after

Re: AC_SYS_LARGEFILE and old OSX framework

2013-10-30 Thread Paul Eggert
Diorcet Yann wrote: > I tried your idea ... same result. It sounds like it's not an Autoconf-related problem, then. But you'll have to debug it, I'm afraid. Possibly your program is trying to store an ino_t value into an int variable; that could explain the problem.