Re: [PATCH v3 6/7] powerpc: kprobes: emulate instructions on kprobe handler re-entry
From: Naveen N. Rao
Date: Wed Apr 19 2017 - 12:44:09 EST
Excerpts from Masami Hiramatsu's message of April 19, 2017 20:13:
BTW, as I pointed, 5/7 and 6/7 should be merged since this actually
makes meaningful change.
Yes, sorry if I wasn't clear in my previous reply in the (!) previous
patch series.
Since this has to go through the powerpc tree, I followed this since I
felt that Michael Ellerman prefers to keep functional changes separate
from refactoring. I'm fine with either approach.
Michael?
Thanks!
- Naveen
Thank you,
On Wed, 19 Apr 2017 18:21:05 +0530
"Naveen N. Rao" <naveen.n.rao@xxxxxxxxxxxxxxxxxx> wrote:
On kprobe handler re-entry, try to emulate the instruction rather than
single stepping always.
Acked-by: Ananth N Mavinakayanahalli <ananth@xxxxxxxxxxxxxxxxxx>
Signed-off-by: Naveen N. Rao <naveen.n.rao@xxxxxxxxxxxxxxxxxx>
---
arch/powerpc/kernel/kprobes.c | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/arch/powerpc/kernel/kprobes.c b/arch/powerpc/kernel/kprobes.c
index 46e8c1e03ce4..067e9863bfdf 100644
--- a/arch/powerpc/kernel/kprobes.c
+++ b/arch/powerpc/kernel/kprobes.c
@@ -276,6 +276,14 @@ int __kprobes kprobe_handler(struct pt_regs *regs)
kprobes_inc_nmissed_count(p);
prepare_singlestep(p, regs);
kcb->kprobe_status = KPROBE_REENTER;
+ if (p->ainsn.boostable >= 0) {
+ ret = try_to_emulate(p, regs);
+
+ if (ret > 0) {
+ restore_previous_kprobe(kcb);
+ return 1;
+ }
+ }
return 1;
} else {
if (*addr != BREAKPOINT_INSTRUCTION) {
--
2.12.1
--
Masami Hiramatsu <mhiramat@xxxxxxxxxx>