Re: objtool warning for next-20221118

From: Paul E. McKenney
Date: Wed Nov 23 2022 - 12:50:09 EST


On Tue, Nov 22, 2022 at 05:48:12PM -0800, Josh Poimboeuf wrote:
> On Tue, Nov 22, 2022 at 04:22:58PM -0800, Paul E. McKenney wrote:
> > On Tue, Nov 22, 2022 at 09:35:17AM +0100, Peter Zijlstra wrote:
> > > On Mon, Nov 21, 2022 at 09:16:05PM -0800, Josh Poimboeuf wrote:
> > >
> > > > It's complaining about an unreachable instruction after a call to
> > > > arch_cpu_idle_dead(). In this case objtool detects the fact
> > > > arch_cpu_idle_dead() doesn't return due to its call to the
> > > > non-CONFIG_SMP version of play_dead(). But GCC has no way of detecting
> > > > that because the caller is in another translation unit.
> > > >
> > > > As far as I can tell, that function should never return. Though it
> > > > seems to have some dubious semantics (see xen_pv_play_dead() for
> > > > example, which *does* seem to return?). I'm thinking it would be an
> > > > improvement to enforce that noreturn behavior across all arches and
> > > > platforms, sprinkling __noreturn and BUG() on arch_cpu_idle_dead() and
> > > > maybe some of it callees, where needed.
> > > >
> > > > Peter, what do you think? I could attempt a patch.
> > >
> > > I'm thinking the Xen case makes all this really rather difficult :/
> > >
> > > While normally a CPU is brought up through a trampoline, Xen seems to
> > > have implemented it by simply returning from play_dead(), and afaict
> > > that is actually a valid way to go about doing it.
> > >
> > > Perhaps the best way would be to stick a REACHABLE annotation in
> > > arch_cpu_idle_dead() or something?
> >
> > When I apply this on -next, I still get the objtool complaint.
> > Is there something else I should also be doing?
>
> Silly GCC is folding the inline asm. This works (but still doesn't seem
> like the right approach):
>
> diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
> index 26e8f57c75ad..128e7d78fedf 100644
> --- a/arch/x86/kernel/process.c
> +++ b/arch/x86/kernel/process.c
> @@ -702,7 +702,7 @@ static void (*x86_idle)(void);
> #ifndef CONFIG_SMP
> static inline void play_dead(void)
> {
> - BUG();
> + _BUG_FLAGS(ASM_UD2, 0, ASM_REACHABLE);
> }
> #endif

I tried this, and still get:

vmlinux.o: warning: objtool: do_idle+0x156: unreachable instruction

Maybe my gcc is haunted?

Thanx, Paul