summaryrefslogtreecommitdiff
path: root/kernel/kprobes.c
diff options
context:
space:
mode:
authorAnanth N Mavinakayanahalli <ananth@in.ibm.com>2006-02-03 14:03:43 +0300
committerLinus Torvalds <torvalds@g5.osdl.org>2006-02-03 19:32:00 +0300
commit278ff9537030bbb292b33504f5e1f6e0126793eb (patch)
tree0087828bdf4d9a4fb57c9bcafd068ff058b39797 /kernel/kprobes.c
parente65cefe87beda627c0bfba39b387ee4bffedc93c (diff)
downloadlinux-278ff9537030bbb292b33504f5e1f6e0126793eb.tar.xz
[PATCH] Kprobes: Fix deadlock in function-return probes
When two function-return probes are inserted on kfree()[1] and the second on say, sys_link()[2], and later [2] is unregistered, we have a deadlock as kfree is called with the kretprobe_lock held and the function-return probe on kfree will also try to grab the same lock. However, we can move the kfree() during unregistration to outside the spinlock as we are sure that no instances from the free list will be used after synchronized_sched() returns during the unregistration process. Thanks to Masami Hiramatsu for spotting this. Signed-off-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Diffstat (limited to 'kernel/kprobes.c')
-rw-r--r--kernel/kprobes.c2
1 files changed, 1 insertions, 1 deletions
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index 95ad7f8db3d6..fef1af8a73ce 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -631,12 +631,12 @@ void __kprobes unregister_kretprobe(struct kretprobe *rp)
unregister_kprobe(&rp->kp);
/* No race here */
spin_lock_irqsave(&kretprobe_lock, flags);
- free_rp_inst(rp);
while ((ri = get_used_rp_inst(rp)) != NULL) {
ri->rp = NULL;
hlist_del(&ri->uflist);
}
spin_unlock_irqrestore(&kretprobe_lock, flags);
+ free_rp_inst(rp);
}
static int __init init_kprobes(void)