On Mon, Mar 04, 2013 at 08:35:08PM +0100, Miloslav Trmač wrote:
> On Mon, Mar 4, 2013 at 7:50 PM, Josh Boyer <jwbo...@gmail.com> wrote:
> >> 1: Long-term ABI for applications that we don't want to break without
> >> significant discussion.
> >> For now, this will include the stable kernel and libc ABIs
> >
> > Please define what you mean by "stable kernel ABI". Do you mean the
> > kernel <-> userspace syscall ABI? If you mean anything other than that
> > I really don't think it's going to work.
>
> This means "whatever the Linux kernel project says is their stable
> ABI". It was not at all intended to expand the ABI boundary.
The only really guaranteed stable ABI the kernel exports is the syscall/ioctl
interface, and to a lesser extent the proc/sysfs ABI.
Kernels in rawhide do differ from what we end up shipping in releases,
because they are continually rebased during the merge window/rc phase,
wherein it's entirely possible that a new interface is refined.
We've had situations for example where a new syscall added during the merge
window
has had additional parameters added/existing ones changed during -rc phase.
This hasn't been a problem because typically, nothing relies upon an interface
in
unreleased kernels, but we need to make it clear here that nothing in new
interfaces
is frozen until a .0 release, in case people start thinking "you shipped it in
rawhide,
so it must be stable".
Dave
--
devel mailing list
devel@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/devel
