On Apr 21 16:28, Christopher Faylor wrote: > On Thu, Apr 21, 2005 at 08:15:33PM +0000, Eric Blake wrote: > I > >originally wrote this program to discover that inode reporting in > >readdir() is broken (to which you replied that fixing it would cause > >too much of a slowdown), > > Yes, both Corinna and I had potential "fixes" (neither being a complete > fix) for this in our sandboxes but they didn't make it into 1.5.15. > Corinna got this working for >=XP which means that it wouldn't be right > only on NTFS file systems on NT3.5, NT4, and NT2K. I just was going to > get rid of the inode field entirely because, as you'd previously said, > any well-designed software would adapt to the lack of that field.
I'm not sure this presumption is correct. The d_ino field is not marked as optional in SUSv3, it's marked as an XSI extension. The crux with XSI extensions is that (quote SuSv3) "Application writers may confidently make use of an extension on all systems supporting the X/Open System Interfaces Extension." This covers practically every serious system in the POSIX world right now. If we drop d_ino, I'd expect another round of suddenly broken applications. Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Project Co-Leader mailto:cygwin@cygwin.com Red Hat, Inc. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
