On Wed, Sep 03, 2008 at 07:49:10AM +1000, Benjamin Herrenschmidt wrote: > On Tue, 2008-09-02 at 16:30 -0500, Olof Johansson wrote: > > On Wed, Sep 03, 2008 at 07:06:15AM +1000, Benjamin Herrenschmidt wrote: > > > On Tue, 2008-09-02 at 13:06 -0500, Olof Johansson wrote: > > > > On Tue, Sep 02, 2008 at 03:48:45PM +1000, Benjamin Herrenschmidt wrote: > > > > > Some stuff I'm working on that isn't released yet needs > > > > > a syscall number but I can't say what it's for until I > > > > > get permission to release it :-) > > > > > > > > > > In the meantime, it would be nice if a number could be set > > > > > aside so the tools being developped to use that stuff don't > > > > > all need to be updated when things are eventually released. > > > > > > > > > > Signed-off-by: Benjamin Herrenschmidt <[EMAIL PROTECTED]> > > > > > --- > > > > > > > > > > I wonder if I'll get hanged for that one :-) > > > > > > > > Yes. Merge your code or live with the pain of things changing > > > > underneath. Period. > > > > > > Pain isn't for me :-) It's for those writing userspace stuff based on > > > my kernel changing all the time :-) > > > > Lame excuse. Merge the code. > > man, you know as well as I do I can't do that before the lawyers let > me :-)
Then don't try to reserve a system call to be used only by an out-of-tree, proprietary kernel extension. Which your kernel change in essence is until it's released. Would we accept a system call number reservation that was submitted by, say, nVidia? No? Well, there you go. It's really that simple. -Olof _______________________________________________ Linuxppc-dev mailing list Linuxppc-dev@ozlabs.org https://ozlabs.org/mailman/listinfo/linuxppc-dev
