Commit f3fbd7ec authored by Masami Hiramatsu's avatar Masami Hiramatsu Committed by Jon Medhurst

arm: kprobes: Allow to handle reentered kprobe on single-stepping

This is arm port of commit 6a5022a5 ("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: default avatarMasami Hiramatsu <mhiramat@kernel.org>
Signed-off-by: default avatarJon Medhurst <tixy@linaro.org>
parent 4495c08e
...@@ -271,6 +271,7 @@ void __kprobes kprobe_handler(struct pt_regs *regs) ...@@ -271,6 +271,7 @@ void __kprobes kprobe_handler(struct pt_regs *regs)
switch (kcb->kprobe_status) { switch (kcb->kprobe_status) {
case KPROBE_HIT_ACTIVE: case KPROBE_HIT_ACTIVE:
case KPROBE_HIT_SSDONE: case KPROBE_HIT_SSDONE:
case KPROBE_HIT_SS:
/* A pre- or post-handler probe got us here. */ /* A pre- or post-handler probe got us here. */
kprobes_inc_nmissed_count(p); kprobes_inc_nmissed_count(p);
save_previous_kprobe(kcb); save_previous_kprobe(kcb);
...@@ -279,6 +280,11 @@ void __kprobes kprobe_handler(struct pt_regs *regs) ...@@ -279,6 +280,11 @@ void __kprobes kprobe_handler(struct pt_regs *regs)
singlestep(p, regs, kcb); singlestep(p, regs, kcb);
restore_previous_kprobe(kcb); restore_previous_kprobe(kcb);
break; 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: default:
/* impossible cases */ /* impossible cases */
BUG(); BUG();
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment