On Tue, 2013-03-26 at 09:18 -0700, Greg KH wrote:
> On Tue, Mar 26, 2013 at 08:56:20AM -0700, Kees Cook wrote:
> > On Sun, Mar 24, 2013 at 12:31 PM, Jiri Slaby <jslaby@suse.cz> wrote:
> > > On 03/20/2013 11:25 PM, Sebastian Gottschall wrote:
> > >
> > >> Am 20.03.2013 21:15, schrieb Greg KH:
> > >>> I'm announcing the release of the 3.8.4 kernel.
> > > ...
> > >>>        signal: always clear sa_restorer on execve
> > >> consider that this kernel is not compileable on mips targets due the
> > >> patch in kernel/signal.c
> > >>
> > >> SA_RESTORER is defined in mips, but sa_restorer does not exist on that
> > >> platform
> > >
> > > The same for ia64...
> > >
> > > kernel/signal.c: In function 'flush_signal_handlers':
> > > kernel/signal.c:441: error: 'struct sigaction' has no member named
> > > 'sa_restorer'
> > >
> > > Do we need this in 3.0, 3.4, 3.8 too?
> > > commit 522cff142d7d2f9230839c9e1f21a4d8bcc22a4a
> > > Author: Andrew Morton <akpm@linux-foundation.org>
> > > Date:   Wed Mar 13 14:59:34 2013 -0700
> > >
> > >     kernel/signal.c: use __ARCH_HAS_SA_RESTORER instead of SA_RESTORER
> > >
> > >     __ARCH_HAS_SA_RESTORER is the preferred conditional for use in 3.9 and
> > >     later kernels, per Kees.
> > 
> > Hrm, if so, we'll need to also backport the changes that aded
> > __ARCH_HAS_SA_RESTORER...
> 
> Ok, well, can someone provide me the needed patches?  Otherwise I will
> have to revert the offending change from the stable releases.

For the third time... use the attached patch followed by Andrew's commit
522cff142d7d2f9230839c9e1f21a4d8bcc22a4a.

Ben.

-- 
Ben Hutchings
I'm not a reverse psychological virus.  Please don't copy me into your sig.