Re: [PATCH 07/20] x86/uaccess: Always inline force_valid_ss()

[Andy Lutomirski]

On Thu, Mar 7, 2019 at 10:59 AM Peter Zijlstra <peterz@xxxxxxxxxxxxx> wrote:
>
> On Thu, Mar 07, 2019 at 10:05:43AM -0800, Andy Lutomirski wrote:
> > On Thu, Mar 7, 2019 at 3:52 AM Peter Zijlstra <peterz@xxxxxxxxxxxxx> wrote:
> > >
> > > arch/x86/kernel/signal.o: warning: objtool: restore_sigcontext()+0x3cc: call to force_valid_ss.isra.5() with UACCESS enabled
> > >
> > > XXX: move the callsite out of te AC=1 region instead?
> >
> > Let's move it instead.  I looked at the code and it should be fine.
>
> Something like so; or did I miss something subtle?

Reviewed-by: Andy Lutomirski <luto@xxxxxxxxxx>

Also, this stuff is pretty well covered by the x86 selftests, mostly
because getting it right in the first place was way too subtle for
comfort.

--Andy