On Friday, November 30, 2012 12:44:06 PM Greg KH wrote: > On Fri, Nov 30, 2012 at 12:09:40PM -0800, Dmitry Torokhov wrote: > > On Friday, November 30, 2012 10:57:55 AM Greg KH wrote: > > > On Fri, Nov 30, 2012 at 10:45:44AM -0800, Dmitry Torokhov wrote: > > > > However you snipped the rest of my reply: do we really need to > > > > renumber > > > > ioctls? There is no benefit for the driver as its ioctl handler does > > > > not parse the numbers into components. > > > > > > I don't know if you need to renumber, I really don't understand what you > > > were trying to do with this code, and as it was acting differently from > > > all other kernel ioctl declarations, I asked for some clarity. > > > > > > If you can rewrite it to look sane, and keep the same numbers, that's > > > fine with me. > > > > OK, it looks like we can redo them as: > > > > #define IOCTL_VMCI_VERSION _IO(7, 0x9f) /* 1951 */ > > #define IOCTL_VMCI_INIT_CONTEXT _IO(7, 0xa0) /* 1952 */ > > > > Is this acceptable? > > Sure, that's better. You also got lucky, '7' happens to be unused right > now. Excellent. You said you want the next drop after -rc1, right? -- Dmitry _______________________________________________ Virtualization mailing list Virtualization@xxxxxxxxxxxxxxxxxxxxxxxxxx https://lists.linuxfoundation.org/mailman/listinfo/virtualization
