[PATCH 4.9.y] arm: kprobes: Allow to handle reentered kprobe on single-stepping

[Greg KH]

On Wed, Mar 03, 2021 at 09:18:27PM +0900, Masami Hiramatsu wrote:
> Hi ShaoBo,
> 
> Thanks for backporting and real bug report!
> 
> On Wed, 3 Mar 2021 15:10:52 +0800
> ShaoBo Huang <huangshaobo6 at huawei.com> wrote:
> 
> > From: Masami Hiramatsu <mhiramat at kernel.org>
> > 
> > commit f3fbd7ec62dec1528fb8044034e2885f2b257941 upstream
> > 
> > This is arm port of commit 6a5022a56ac3 ("kprobes/x86: Allow to
> > handle reentered kprobe on single-stepping")
> > 
> > Since the FIQ handlers can interrupt in the single stepping
> > (or preparing the single stepping, do_debug etc.), we should
> > consider a kprobe is hit in the NMI handler. Even in that
> > case, the kprobe is allowed to be reentered as same as the
> > kprobes hit in kprobe handlers
> > (KPROBE_HIT_ACTIVE or KPROBE_HIT_SSDONE).
> > 
> > The real issue will happen when a kprobe hit while another
> > reentered kprobe is processing (KPROBE_REENTER), because
> > we already consumed a saved-area for the previous kprobe.
> > 
> > Signed-off-by: Masami Hiramatsu <mhiramat at kernel.org>
> > Signed-off-by: Jon Medhurst <tixy at linaro.org>
> > Fixes: 24ba613c9d6c ("ARM kprobes: core code")
> > Cc: stable at vger.kernel.org #v2.6.25~v4.11
> > Signed-off-by: huangshaobo <huangshaobo6 at huawei.com>
> > ---
> >  arch/arm/probes/kprobes/core.c | 6 ++++++
> >  1 file changed, 6 insertions(+)
> > 
> > diff --git a/arch/arm/probes/kprobes/core.c b/arch/arm/probes/kprobes/core.c
> > index 3eb018fa1a1f..c3362ddd6c4c 100644
> > --- a/arch/arm/probes/kprobes/core.c
> > +++ b/arch/arm/probes/kprobes/core.c
> > @@ -270,6 +270,7 @@ void __kprobes kprobe_handler(struct pt_regs *regs)
> >  			switch (kcb->kprobe_status) {
> >  			case KPROBE_HIT_ACTIVE:
> >  			case KPROBE_HIT_SSDONE:
> > +			case KPROBE_HIT_SS:
> >  				/* A pre- or post-handler probe got us here. */
> >  				kprobes_inc_nmissed_count(p);
> >  				save_previous_kprobe(kcb);
> > @@ -278,6 +279,11 @@ void __kprobes kprobe_handler(struct pt_regs *regs)
> >  				singlestep(p, regs, kcb);
> >  				restore_previous_kprobe(kcb);
> >  				break;
> > +			case KPROBE_REENTER:
> > +				/* A nested probe was hit in FIQ, it is a BUG */
> > +				pr_warn("Unrecoverable kprobe detected at %p.\n",
> > +					p->addr);
> > +				/* fall through */
> >  			default:
> >  				/* impossible cases */
> >  				BUG();
> > -- 
> > 2.12.3
> > 

Also queued up to 4.4.y as well.

thanks,

greg k-h