On Mon, Feb 11, 2008 at 01:41:03PM -0500, Andres Mejia wrote: > > do not pass the --enable-inotify parameter to the configure script; > > presence of inotify will be detected automatically, if not available it > > will be disabled on the fly. However, when the --enable-inotify parameter > > is passed by the user, configure will abort with the error that you are > > seeing if inotify checks fail. > > This is done on purpose so that no package on a different architecture/kernel > builds packages sucessfully with a different set of options enabled, else I'm > sure there would be a different set of bugs sooner or later. The inotify > option is not enabled for any non-linux kernel using machine. > > The parameters passed to the configure script can be manually overridden by a > user, but they won't be for the buildd machines.
Well, if you are sure that the target machines *do have inotify support* the only thing that could help is the config.log output of the failed configure process. > > What I still need to do is, to adapt the inotify check for cross compiling, > > right now it will always fail when cross compiled. > > Actually, no cross compiling is done. The powerpc buildd machine is a powerpc > machine. The same goes for the other buildd machines. Then it should work; we do check the headers and then we run some small test code which is calling inotify_init(). If the inotify_init() function returns -1 (error) we will compile without inotify support; however - if the --enable-inotify parameter was specified configure will abort because it can not do what the user wanted. So the question: is inotify really working on those systems? Kind regards, Jin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
