diff mbox series

[v3,1/2] perf: Fix required permissions if sigtrap is requested

Message ID 20210705084453.2151729-1-elver@google.com (mailing list archive)
State New, archived
Headers show
Series [v3,1/2] perf: Fix required permissions if sigtrap is requested | expand

Commit Message

Marco Elver July 5, 2021, 8:44 a.m. UTC
If perf_event_open() is called with another task as target and
perf_event_attr::sigtrap is set, and the target task's user does not
match the calling user, also require the CAP_KILL capability or
PTRACE_MODE_ATTACH permissions.

Otherwise, with the CAP_PERFMON capability alone it would be possible
for a user to send SIGTRAP signals via perf events to another user's
tasks. This could potentially result in those tasks being terminated if
they cannot handle SIGTRAP signals.

Note: The check complements the existing capability check, but is not
supposed to supersede the ptrace_may_access() check. At a high level we
now have:

	capable of CAP_PERFMON and (CAP_KILL if sigtrap)
		OR
	ptrace_may_access(...) // also checks for same thread-group and uid

Fixes: 97ba62b27867 ("perf: Add support for SIGTRAP on perf events")
Cc: <stable@vger.kernel.org> # 5.13+
Reported-by: Dmitry Vyukov <dvyukov@google.com>
Signed-off-by: Marco Elver <elver@google.com>
---
v3:
* Upgrade ptrace mode check to ATTACH if attr.sigtrap, otherwise it's
  possible to change the target task (send signal) even if only read
  ptrace permissions were granted (reported by Eric W. Biederman).

v2: https://lkml.kernel.org/r/20210701083842.580466-1-elver@google.com
* Drop kill_capable() and just check CAP_KILL (reported by Ondrej Mosnacek).
* Use ns_capable(__task_cred(task)->user_ns, CAP_KILL) to check for
  capability in target task's ns (reported by Ondrej Mosnacek).

v1: https://lkml.kernel.org/r/20210630093709.3612997-1-elver@google.com
---
 kernel/events/core.c | 25 ++++++++++++++++++++++++-
 1 file changed, 24 insertions(+), 1 deletion(-)

Comments

Dmitry Vyukov July 6, 2021, 6:16 a.m. UTC | #1
On Mon, Jul 5, 2021 at 10:45 AM Marco Elver <elver@google.com> wrote:
>
> If perf_event_open() is called with another task as target and
> perf_event_attr::sigtrap is set, and the target task's user does not
> match the calling user, also require the CAP_KILL capability or
> PTRACE_MODE_ATTACH permissions.
>
> Otherwise, with the CAP_PERFMON capability alone it would be possible
> for a user to send SIGTRAP signals via perf events to another user's
> tasks. This could potentially result in those tasks being terminated if
> they cannot handle SIGTRAP signals.
>
> Note: The check complements the existing capability check, but is not
> supposed to supersede the ptrace_may_access() check. At a high level we
> now have:
>
>         capable of CAP_PERFMON and (CAP_KILL if sigtrap)
>                 OR
>         ptrace_may_access(...) // also checks for same thread-group and uid
>
> Fixes: 97ba62b27867 ("perf: Add support for SIGTRAP on perf events")
> Cc: <stable@vger.kernel.org> # 5.13+
> Reported-by: Dmitry Vyukov <dvyukov@google.com>
> Signed-off-by: Marco Elver <elver@google.com>

Acked-by: Dmitry Vyukov <dvyukov@google.com>

> ---
> v3:
> * Upgrade ptrace mode check to ATTACH if attr.sigtrap, otherwise it's
>   possible to change the target task (send signal) even if only read
>   ptrace permissions were granted (reported by Eric W. Biederman).
>
> v2: https://lkml.kernel.org/r/20210701083842.580466-1-elver@google.com
> * Drop kill_capable() and just check CAP_KILL (reported by Ondrej Mosnacek).
> * Use ns_capable(__task_cred(task)->user_ns, CAP_KILL) to check for
>   capability in target task's ns (reported by Ondrej Mosnacek).
>
> v1: https://lkml.kernel.org/r/20210630093709.3612997-1-elver@google.com
> ---
>  kernel/events/core.c | 25 ++++++++++++++++++++++++-
>  1 file changed, 24 insertions(+), 1 deletion(-)
>
> diff --git a/kernel/events/core.c b/kernel/events/core.c
> index fe88d6eea3c2..f79ee82e644a 100644
> --- a/kernel/events/core.c
> +++ b/kernel/events/core.c
> @@ -12152,10 +12152,33 @@ SYSCALL_DEFINE5(perf_event_open,
>         }
>
>         if (task) {
> +               unsigned int ptrace_mode = PTRACE_MODE_READ_REALCREDS;
> +               bool is_capable;
> +
>                 err = down_read_interruptible(&task->signal->exec_update_lock);
>                 if (err)
>                         goto err_file;
>
> +               is_capable = perfmon_capable();
> +               if (attr.sigtrap) {
> +                       /*
> +                        * perf_event_attr::sigtrap sends signals to the other
> +                        * task. Require the current task to also have
> +                        * CAP_KILL.
> +                        */
> +                       rcu_read_lock();
> +                       is_capable &= ns_capable(__task_cred(task)->user_ns, CAP_KILL);
> +                       rcu_read_unlock();
> +
> +                       /*
> +                        * If the required capabilities aren't available, checks
> +                        * for ptrace permissions: upgrade to ATTACH, since
> +                        * sending signals can effectively change the target
> +                        * task.
> +                        */
> +                       ptrace_mode = PTRACE_MODE_ATTACH_REALCREDS;
> +               }
> +
>                 /*
>                  * Preserve ptrace permission check for backwards compatibility.
>                  *
> @@ -12165,7 +12188,7 @@ SYSCALL_DEFINE5(perf_event_open,
>                  * perf_event_exit_task() that could imply).
>                  */
>                 err = -EACCES;
> -               if (!perfmon_capable() && !ptrace_may_access(task, PTRACE_MODE_READ_REALCREDS))
> +               if (!is_capable && !ptrace_may_access(task, ptrace_mode))
>                         goto err_cred;
>         }
>
> --
> 2.32.0.93.g670b81a890-goog
>
Marco Elver July 12, 2021, 10:32 a.m. UTC | #2
It'd be good to get this sorted -- please take another look.

Many thanks,
-- Marco

On Mon, 5 Jul 2021 at 10:45, Marco Elver <elver@google.com> wrote:
> If perf_event_open() is called with another task as target and
> perf_event_attr::sigtrap is set, and the target task's user does not
> match the calling user, also require the CAP_KILL capability or
> PTRACE_MODE_ATTACH permissions.
>
> Otherwise, with the CAP_PERFMON capability alone it would be possible
> for a user to send SIGTRAP signals via perf events to another user's
> tasks. This could potentially result in those tasks being terminated if
> they cannot handle SIGTRAP signals.
>
> Note: The check complements the existing capability check, but is not
> supposed to supersede the ptrace_may_access() check. At a high level we
> now have:
>
>         capable of CAP_PERFMON and (CAP_KILL if sigtrap)
>                 OR
>         ptrace_may_access(...) // also checks for same thread-group and uid
>
> Fixes: 97ba62b27867 ("perf: Add support for SIGTRAP on perf events")
> Cc: <stable@vger.kernel.org> # 5.13+
> Reported-by: Dmitry Vyukov <dvyukov@google.com>
> Signed-off-by: Marco Elver <elver@google.com>
> ---
> v3:
> * Upgrade ptrace mode check to ATTACH if attr.sigtrap, otherwise it's
>   possible to change the target task (send signal) even if only read
>   ptrace permissions were granted (reported by Eric W. Biederman).
>
> v2: https://lkml.kernel.org/r/20210701083842.580466-1-elver@google.com
> * Drop kill_capable() and just check CAP_KILL (reported by Ondrej Mosnacek).
> * Use ns_capable(__task_cred(task)->user_ns, CAP_KILL) to check for
>   capability in target task's ns (reported by Ondrej Mosnacek).
>
> v1: https://lkml.kernel.org/r/20210630093709.3612997-1-elver@google.com
> ---
>  kernel/events/core.c | 25 ++++++++++++++++++++++++-
>  1 file changed, 24 insertions(+), 1 deletion(-)
>
> diff --git a/kernel/events/core.c b/kernel/events/core.c
> index fe88d6eea3c2..f79ee82e644a 100644
> --- a/kernel/events/core.c
> +++ b/kernel/events/core.c
> @@ -12152,10 +12152,33 @@ SYSCALL_DEFINE5(perf_event_open,
>         }
>
>         if (task) {
> +               unsigned int ptrace_mode = PTRACE_MODE_READ_REALCREDS;
> +               bool is_capable;
> +
>                 err = down_read_interruptible(&task->signal->exec_update_lock);
>                 if (err)
>                         goto err_file;
>
> +               is_capable = perfmon_capable();
> +               if (attr.sigtrap) {
> +                       /*
> +                        * perf_event_attr::sigtrap sends signals to the other
> +                        * task. Require the current task to also have
> +                        * CAP_KILL.
> +                        */
> +                       rcu_read_lock();
> +                       is_capable &= ns_capable(__task_cred(task)->user_ns, CAP_KILL);
> +                       rcu_read_unlock();
> +
> +                       /*
> +                        * If the required capabilities aren't available, checks
> +                        * for ptrace permissions: upgrade to ATTACH, since
> +                        * sending signals can effectively change the target
> +                        * task.
> +                        */
> +                       ptrace_mode = PTRACE_MODE_ATTACH_REALCREDS;
> +               }
> +
>                 /*
>                  * Preserve ptrace permission check for backwards compatibility.
>                  *
> @@ -12165,7 +12188,7 @@ SYSCALL_DEFINE5(perf_event_open,
>                  * perf_event_exit_task() that could imply).
>                  */
>                 err = -EACCES;
> -               if (!perfmon_capable() && !ptrace_may_access(task, PTRACE_MODE_READ_REALCREDS))
> +               if (!is_capable && !ptrace_may_access(task, ptrace_mode))
>                         goto err_cred;
>         }
>
> --
> 2.32.0.93.g670b81a890-goog
>
Peter Zijlstra July 13, 2021, 9:48 a.m. UTC | #3
On Mon, Jul 12, 2021 at 12:32:33PM +0200, Marco Elver wrote:
> It'd be good to get this sorted -- please take another look.

Thanks!

I'll queue them into perf/urgent.
diff mbox series

Patch

diff --git a/kernel/events/core.c b/kernel/events/core.c
index fe88d6eea3c2..f79ee82e644a 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -12152,10 +12152,33 @@  SYSCALL_DEFINE5(perf_event_open,
 	}
 
 	if (task) {
+		unsigned int ptrace_mode = PTRACE_MODE_READ_REALCREDS;
+		bool is_capable;
+
 		err = down_read_interruptible(&task->signal->exec_update_lock);
 		if (err)
 			goto err_file;
 
+		is_capable = perfmon_capable();
+		if (attr.sigtrap) {
+			/*
+			 * perf_event_attr::sigtrap sends signals to the other
+			 * task. Require the current task to also have
+			 * CAP_KILL.
+			 */
+			rcu_read_lock();
+			is_capable &= ns_capable(__task_cred(task)->user_ns, CAP_KILL);
+			rcu_read_unlock();
+
+			/*
+			 * If the required capabilities aren't available, checks
+			 * for ptrace permissions: upgrade to ATTACH, since
+			 * sending signals can effectively change the target
+			 * task.
+			 */
+			ptrace_mode = PTRACE_MODE_ATTACH_REALCREDS;
+		}
+
 		/*
 		 * Preserve ptrace permission check for backwards compatibility.
 		 *
@@ -12165,7 +12188,7 @@  SYSCALL_DEFINE5(perf_event_open,
 		 * perf_event_exit_task() that could imply).
 		 */
 		err = -EACCES;
-		if (!perfmon_capable() && !ptrace_may_access(task, PTRACE_MODE_READ_REALCREDS))
+		if (!is_capable && !ptrace_may_access(task, ptrace_mode))
 			goto err_cred;
 	}