Hi Oups sorry yes of course I'm wrong
Sorry for the bug Best regards 2011/12/7 Stephan Bergmann <sberg...@redhat.com> > -#if defined(LINUX) >> -/* The linux kernel thread implemention, always return the pid of the >> - thread subprocess and not of the main process. So we save the main >> - pid at startup >> +#if defined(LINUX) && defined (__GLIBC__) && __GLIBC__ == 2 || >> __GLIBC_MINOR__ < 4 >> > > The above looks broken, given that && binds tighter than ||. From the > comments, I gather that what actually should be checked is that > __GLIBC__.__GLIBC_MINOR__ < 2.4, right? So, since we would not care about > glibc 1, anyway, the || should be replaced with a && I think. > > Stephan > > +/* The linux kernel 2.4 getpid implemention always return the pid of the >> + thread subprocess and not of the main process, the NPTL implementation >> + with a Linux kernel 2.6 kernel return the pid. So when possibly >> + their is the wrong implementation of getpid, we save the pid at >> startup. >> FIXME: when our Linux base-line is above: >> - + Linux kernel version 2.6.18 or higher; -> clone() for NTPL >> - + glibc2 version 2.5 or higher; -> NPTL >> + + Linux kernel version 2.6 or higher; -> clone() for NTPL >> + + glibc2 version 2.4 or higher; -> No longer LinuxThreads, only >> NPTL >> Then we get a working getpid() and can remove this hack. >> + FIXME: getppid is also wrong in this situation >> */ >> > > > ______________________________**_________________ > LibreOffice mailing list > LibreOffice@lists.freedesktop.**org <LibreOffice@lists.freedesktop.org> > http://lists.freedesktop.org/**mailman/listinfo/libreoffice<http://lists.freedesktop.org/mailman/listinfo/libreoffice> > -- Arnaud Versini
_______________________________________________ LibreOffice mailing list LibreOffice@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/libreoffice
