Re: [PATCH v1] perf/core: take over CAP_SYS_PTRACE creds to CAP_PERFMON cap

From: Arnaldo Carvalho de Melo
Date: Thu Aug 06 2020 - 13:42:08 EST


Em Wed, Aug 05, 2020 at 10:56:56AM +0300, Alexey Budankov escreveu:
>
> Open access to per-process monitoring for CAP_PERFMON only
> privileged processes [1]. Extend ptrace_may_access() check
> in perf_events subsystem with perfmon_capable() to simplify
> user experience and make monitoring more secure by reducing
> attack surface.


Acked-by: Arnaldo Carvalho de Melo <acme@xxxxxxxxxx>

Peter, can you pick up this one? You said it was ok last time we
discussed this.

- Arnaldo

> [1] https://lore.kernel.org/lkml/7776fa40-6c65-2aa6-1322-eb3a01201000@xxxxxxxxxxxxxxx/
>
> Signed-off-by: Alexey Budankov <alexey.budankov@xxxxxxxxxxxxxxx>
> ---
> kernel/events/core.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/kernel/events/core.c b/kernel/events/core.c
> index 856d98c36f56..f50d528af444 100644
> --- a/kernel/events/core.c
> +++ b/kernel/events/core.c
> @@ -11587,7 +11587,7 @@ SYSCALL_DEFINE5(perf_event_open,
> goto err_task;
>
> /*
> - * Reuse ptrace permission checks for now.
> + * Preserve ptrace permission check for backwards compatibility.
> *
> * We must hold exec_update_mutex across this and any potential
> * perf_install_in_context() call for this new event to
> @@ -11595,7 +11595,7 @@ SYSCALL_DEFINE5(perf_event_open,
> * perf_event_exit_task() that could imply).
> */
> err = -EACCES;
> - if (!ptrace_may_access(task, PTRACE_MODE_READ_REALCREDS))
> + if (!perfmon_capable() && !ptrace_may_access(task, PTRACE_MODE_READ_REALCREDS))
> goto err_cred;
> }
>
> --
> 2.24.1
>

--

- Arnaldo