Skip to content

Commit

Permalink
kprobes: Remove NMI context check
Browse files Browse the repository at this point in the history
The in_nmi() check in pre_handler_kretprobe() is meant to avoid
recursion, and blindly assumes that anything NMI is recursive.

However, since commit:

  9b38cc7 ("kretprobe: Prevent triggering kretprobe from within kprobe_flush_task")

there is a better way to detect and avoid actual recursion.

By setting a dummy kprobe, any actual exceptions will terminate early
(by trying to handle the dummy kprobe), and recursion will not happen.

Employ this to avoid the kretprobe_table_lock() recursion, replacing
the over-eager in_nmi() check.

Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lkml.kernel.org/r/159870615628.1229682.6087311596892125907.stgit@devnote2
  • Loading branch information
Masami Hiramatsu authored and Ingo Molnar committed Sep 8, 2020
1 parent 5e96ce8 commit e03b4a0
Showing 1 changed file with 4 additions and 12 deletions.
16 changes: 4 additions & 12 deletions kernel/kprobes.c
Original file line number Diff line number Diff line change
Expand Up @@ -1359,7 +1359,8 @@ static void cleanup_rp_inst(struct kretprobe *rp)
struct hlist_node *next;
struct hlist_head *head;

/* No race here */
/* To avoid recursive kretprobe by NMI, set kprobe busy here */
kprobe_busy_begin();
for (hash = 0; hash < KPROBE_TABLE_SIZE; hash++) {
kretprobe_table_lock(hash, &flags);
head = &kretprobe_inst_table[hash];
Expand All @@ -1369,6 +1370,8 @@ static void cleanup_rp_inst(struct kretprobe *rp)
}
kretprobe_table_unlock(hash, &flags);
}
kprobe_busy_end();

free_rp_inst(rp);
}
NOKPROBE_SYMBOL(cleanup_rp_inst);
Expand Down Expand Up @@ -2035,17 +2038,6 @@ static int pre_handler_kretprobe(struct kprobe *p, struct pt_regs *regs)
unsigned long hash, flags = 0;
struct kretprobe_instance *ri;

/*
* To avoid deadlocks, prohibit return probing in NMI contexts,
* just skip the probe and increase the (inexact) 'nmissed'
* statistical counter, so that the user is informed that
* something happened:
*/
if (unlikely(in_nmi())) {
rp->nmissed++;
return 0;
}

/* TODO: consider to only swap the RA after the last pre_handler fired */
hash = hash_ptr(current, KPROBE_HASH_BITS);
raw_spin_lock_irqsave(&rp->lock, flags);
Expand Down

0 comments on commit e03b4a0

Please sign in to comment.