Skip to contentSkip to conversation

kprobes: Update ftrace_ops when clearing ftrace-based aggrprobe's post_handler

PATCH In review Li Huafei 2 participants 3 messages Nov 11, 2022 – Nov 12, 2022 kprobes

lore
Share & tools
lore

Li Huafei

author
#0 lore raw

PATCH 1/1 kprobes: Update ftrace_ops when clearing ftrace-based aggrprobe's post_handler

In __unregister_kprobe_top(), if the currently unregistered probe has post_handler but other child probes of the aggrprobe do not have post_handler, the post_handler of the aggrprobe is cleared. If this is a ftrace-based probe, there is a problem. In later calls to disarm_kprobe(), we will use kprobe_ftrace_ops because post_handler is NULL. But we're armed with kprobe_ipmodify_ops. This triggers a WARN in __disarm_kprobe_ftrace() and may even cause use-after-free:

Failed to disarm kprobe-ftrace at kernel_clone+0x0/0x3c0 (error -2)

  WARNING: CPU: 5 PID: 137 at kernel/kprobes.c:1135 __disarm_kprobe_ftrace.isra.21+0xcf/0xe0
  Modules linked in: testKprobe_007(-)
  CPU: 5 PID: 137 Comm: rmmod Not tainted 6.1.0-rc4-dirty #18
  [...]
  Call Trace:
   <TASK>
   __disable_kprobe+0xcd/0xe0
   __unregister_kprobe_top+0x12/0x150
   ? mutex_lock+0xe/0x30
   unregister_kprobes.part.23+0x31/0xa0
   unregister_kprobe+0x32/0x40
   __x64_sys_delete_module+0x15e/0x260
   ? do_user_addr_fault+0x2cd/0x6b0
   do_syscall_64+0x3a/0x90
   entry_SYSCALL_64_after_hwframe+0x63/0xcd
   [...]

For ftrace kprobe, update post_handler at the same time update ftrace_ops, moving it from kprobe_ipmodify_ops to kprobe_ftrace_ops.

Fixes 0bc11ed5ab60 ("kprobes: Allow kprobes coexist with livepatch")
---
 kernel/kprobes.c | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)
kernel/kprobes.c +11 -1
index cd9f5a66a690..f8bec48a9cf9 100644
@@ -1766,7 +1766,17 @@ static int __unregister_kprobe_top(struct kprobe *p)
 				if ((list_p != p) && (list_p->post_handler))
 					goto noclean;
 			}
-			ap->post_handler = NULL;
+			/*
+			 * For ftrace kprobe, we need to update ftrace_ops
+			 * at the same time as we update post_handler, moving
+			 * it from kprobe_ipmodify_ops to kprobe_ftrace_ops.
+			 */
+			if (unlikely(kprobe_ftrace(ap))) {
+				disarm_kprobe(ap, false);
+				ap->post_handler = NULL;
+				arm_kprobe(ap);
+			} else
+				ap->post_handler = NULL;
 		}
 noclean:
 		/*
Signature · 1 line
2.17.1
#1 lore raw
Quoted 9 lines · On Fri, 11 Nov 2022 18:10:06 +0800 Li Huafei <[email protected]> wrote:

In __unregister_kprobe_top(), if the currently unregistered probe has post_handler but other child probes of the aggrprobe do not have post_handler, the post_handler of the aggrprobe is cleared. If this is a ftrace-based probe, there is a problem. In later calls to disarm_kprobe(), we will use kprobe_ftrace_ops because post_handler is NULL. But we're armed with kprobe_ipmodify_ops. This triggers a WARN in __disarm_kprobe_ftrace() and may even cause use-after-free:

Failed to disarm kprobe-ftrace at kernel_clone+0x0/0x3c0 (error -2)

Quoted 16 lines
  WARNING: CPU: 5 PID: 137 at kernel/kprobes.c:1135 __disarm_kprobe_ftrace.isra.21+0xcf/0xe0
  Modules linked in: testKprobe_007(-)
  CPU: 5 PID: 137 Comm: rmmod Not tainted 6.1.0-rc4-dirty #18
  [...]
  Call Trace:
   <TASK>
   __disable_kprobe+0xcd/0xe0
   __unregister_kprobe_top+0x12/0x150
   ? mutex_lock+0xe/0x30
   unregister_kprobes.part.23+0x31/0xa0
   unregister_kprobe+0x32/0x40
   __x64_sys_delete_module+0x15e/0x260
   ? do_user_addr_fault+0x2cd/0x6b0
   do_syscall_64+0x3a/0x90
   entry_SYSCALL_64_after_hwframe+0x63/0xcd
   [...]

Ah, good catch! :D

Quoted 3 lines

For ftrace kprobe, update post_handler at the same time update ftrace_ops, moving it from kprobe_ipmodify_ops to kprobe_ftrace_ops.

Hmm, but I would not like this because there can be a time window when it can miss an event. What about just skipping clearing ap->post_handler in kprobe-on-ftrace case?

Quoted 17 lines
Fixes 0bc11ed5ab60 ("kprobes: Allow kprobes coexist with livepatch")
---
 kernel/kprobes.c | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)
kernel/kprobes.c +0 -1
index cd9f5a66a690..f8bec48a9cf9 100644
@@ -1766,7 +1766,17 @@ static int __unregister_kprobe_top(struct kprobe *p)
 				if ((list_p != p) && (list_p->post_handler))
 					goto noclean;
 			}
-			ap->post_handler = NULL;

kernel/kprobes.c static int __unregister_kprobe_top(struct kprobe *p) · line 1778

Quoted 10 lines
kernel/kprobes.c +10 -0
+			/*
+			 * For ftrace kprobe, we need to update ftrace_ops
+			 * at the same time as we update post_handler, moving
+			 * it from kprobe_ipmodify_ops to kprobe_ftrace_ops.
+			 */
+			if (unlikely(kprobe_ftrace(ap))) {
+				disarm_kprobe(ap, false);
+				ap->post_handler = NULL;
+				arm_kprobe(ap);
+			} else

So here, just add; (also, don't use unlikely/likely for this case, this depends on where the user probes, not a systematically rare case.)

			if (!kprobe_ftrace(ap))

kernel/kprobes.c static int __unregister_kprobe_top(struct kprobe *p) · line 1779

Quoted 1 line
kernel/kprobes.c +1 -0
+				ap->post_handler = NULL;

Thank you!

Quoted 6 lines
kernel/kprobes.c +0 -0
 		}
 noclean:
 		/*
Signature · 1 line
2.17.1
Signature · 1 line
Masami Hiramatsu (Google) <[email protected]>
author
#2 lore raw
Quoted 32 lines · Masami Hiramatsu (Google)
Quoted 9 lines · On Fri, 11 Nov 2022 18:10:06 +0800 Li Huafei <[email protected]> wrote:

In __unregister_kprobe_top(), if the currently unregistered probe has post_handler but other child probes of the aggrprobe do not have post_handler, the post_handler of the aggrprobe is cleared. If this is a ftrace-based probe, there is a problem. In later calls to disarm_kprobe(), we will use kprobe_ftrace_ops because post_handler is NULL. But we're armed with kprobe_ipmodify_ops. This triggers a WARN in __disarm_kprobe_ftrace() and may even cause use-after-free:

Failed to disarm kprobe-ftrace at kernel_clone+0x0/0x3c0 (error -2)

Quoted 16 lines
  WARNING: CPU: 5 PID: 137 at kernel/kprobes.c:1135 __disarm_kprobe_ftrace.isra.21+0xcf/0xe0
  Modules linked in: testKprobe_007(-)
  CPU: 5 PID: 137 Comm: rmmod Not tainted 6.1.0-rc4-dirty #18
  [...]
  Call Trace:
   <TASK>
   __disable_kprobe+0xcd/0xe0
   __unregister_kprobe_top+0x12/0x150
   ? mutex_lock+0xe/0x30
   unregister_kprobes.part.23+0x31/0xa0
   unregister_kprobe+0x32/0x40
   __x64_sys_delete_module+0x15e/0x260
   ? do_user_addr_fault+0x2cd/0x6b0
   do_syscall_64+0x3a/0x90
   entry_SYSCALL_64_after_hwframe+0x63/0xcd
   [...]

Ah, good catch! :D

Quoted 1 line
Quoted 7 lines
Quoted 2 lines

For ftrace kprobe, update post_handler at the same time update ftrace_ops, moving it from kprobe_ipmodify_ops to kprobe_ftrace_ops.

Hmm, but I would not like this because there can be a time window when it can miss an event. What about just skipping clearing ap->post_handler in kprobe-on-ftrace case?

Agree. I hadn't considered this time window. The effects I see if I keep ap->handler are 1) kprobe_ftrace_handler() still needs to call aggr_post_handler() and 2) other ftrace_ops still can't set IPMODIFY on the probe function. This doesn't seem to be a problem.

Thanks for the suggestion.

Quoted 22 lines
Quoted 22 lines
Fixes 0bc11ed5ab60 ("kprobes: Allow kprobes coexist with livepatch")
---
 kernel/kprobes.c | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)
kernel/kprobes.c +5 -1
index cd9f5a66a690..f8bec48a9cf9 100644
@@ -1766,7 +1766,17 @@ static int __unregister_kprobe_top(struct kprobe *p)
 				if ((list_p != p) && (list_p->post_handler))
 					goto noclean;
 			}
-			ap->post_handler = NULL;
+			/*
+			 * For ftrace kprobe, we need to update ftrace_ops
+			 * at the same time as we update post_handler, moving
+			 * it from kprobe_ipmodify_ops to kprobe_ftrace_ops.
+			 */
Quoted 9 lines

kernel/kprobes.c static int __unregister_kprobe_top(struct kprobe *p) · line 1778

Quoted 5 lines
kernel/kprobes.c +5 -0
+			if (unlikely(kprobe_ftrace(ap))) {
+				disarm_kprobe(ap, false);
+				ap->post_handler = NULL;
+				arm_kprobe(ap);
+			} else

So here, just add; (also, don't use unlikely/likely for this case, this depends on where the user probes, not a systematically rare case.)

Okay. Will fix it in the next version. Thanks!

Quoted 14 lines
			if (!kprobe_ftrace(ap))

kernel/kprobes.c static int __unregister_kprobe_top(struct kprobe *p) · line 1779

Quoted 1 line
kernel/kprobes.c +1 -0
+				ap->post_handler = NULL;

Thank you!

Quoted 6 lines
kernel/kprobes.c +0 -0
 		}
 noclean:
 		/*
Signature · 1 line
2.17.1