Skip to contentSkip to conversation

kernel/fork: validate exit_signal in kernel_clone()

PATCH Superseded Deepanshu Kartikey 2 participants 2 messages Mar 16

Subsystems (4)

v2 of 3 Superseded Latest is v3 →

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 kernel/fork: validate exit_signal in kernel_clone()

When a child process exits, it sends exit_signal to its parent via do_notify_parent(). The clone() syscall constructs exit_signal as:

(lower_32_bits(clone_flags) & CSIGNAL)

CSIGNAL is 0xff, so values in the range 65-255 are possible. However, valid_signal() only accepts signals up to _NSIG (64 on x86_64), causing a WARN_ON in do_notify_parent() when the process exits:

  WARNING: kernel/signal.c:2174 do_notify_parent+0xc7e/0xd70

The syzkaller reproducer triggers this by calling clone() with flags=0x80, resulting in exit_signal = (0x80 & CSIGNAL) = 128, which exceeds _NSIG and is not a valid signal.

The v1 of this patch added the check only in the clone() syscall handler, which is incomplete. kernel_clone() has other callers such as sys_ia32_clone() which would remain unprotected. Move the check to kernel_clone() to cover all callers.

clone3() already validates exit_signal in copy_clone_args_from_user(). The comment above kernel_clone() states that callers are expected to validate exit_signal, but several callers never did. Adding the check to kernel_clone() enforces this for all callers centrally.

Note that this is a user-visible change: previously, passing an invalid exit_signal to clone() was silently accepted. The man page for clone() does not document any defined behavior for invalid exit_signal values, so rejecting them with -EINVAL is the correct behavior. It is unlikely that any sane application relies on passing an invalid exit_signal.

---
 kernel/fork.c | 2 ++
 1 file changed, 2 insertions(+)
kernel/fork.c +2 -0
index 947a8dbce06a..89d7eb67baf5 100644
@@ -2687,6 +2687,8 @@ pid_t kernel_clone(struct kernel_clone_args *args)
 	    (args->pidfd == args->parent_tid))
 		return -EINVAL;
 
+	if (!valid_signal(args->exit_signal))
+		return -EINVAL;
 	/*
 	 * Determine whether and which event to report to ptracer.  When
 	 * called from kernel_thread or CLONE_UNTRACED is explicitly
Signature · 1 line
2.43.0
#1 lore raw

Deepanshu,

Let me repeat, the changelog should be updated.

Quoted 6 lines · Deepanshu Kartikey

CSIGNAL is 0xff, so values in the range 65-255 are possible. However, valid_signal() only accepts signals up to _NSIG (64 on x86_64), causing a WARN_ON in do_notify_parent() when the process exits:

  WARNING: kernel/signal.c:2174 do_notify_parent+0xc7e/0xd70

Again, do_notify_parent-sanitize-the-valid_signal-checks.patch was dropped. do_notify_parent() won't WARN() in this case.

Quoted 5 lines

Note that this is a user-visible change: previously, passing an invalid exit_signal to clone() was silently accepted. The man page for clone() does not document any defined behavior for invalid exit_signal values, so rejecting them with -EINVAL is the correct behavior. It is unlikely that any sane application relies on passing an invalid exit_signal.

Yes, it only documents that if exit_signal == 0 then the parent process is not signaled when the child terminates. But in fact a non-zero non-valid signal acts the same way.

kernel/fork.c pid_t kernel_clone(struct kernel_clone_args *args) · line 2691

Quoted 8 lines
kernel/fork.c +2 -0
@@ -2687,6 +2687,8 @@ pid_t kernel_clone(struct kernel_clone_args *args)
 	    (args->pidfd == args->parent_tid))
 		return -EINVAL;
 
+	if (!valid_signal(args->exit_signal))
+		return -EINVAL;

OK, but then it also makes sense to remove the same valid_signal() check in copy_clone_args_from_user() ?

Oleg.