On Mon, Nov 12, 2018 at 3:08 PM, Alexander Popov <alex.po...@linux.com> wrote:
> The stackleak_erase() function is called on the trampoline stack at the end
> of syscall. This stack is not big enough for ftrace and kprobes operations,
> e.g. it can be exhausted if we use kprobe_events for stackleak_erase().
>
> So let's disable function tracing and kprobes for stackleak_erase().
>
> Reported-by: kernel test robot <l...@intel.com>
> Signed-off-by: Alexander Popov <alex.po...@linux.com>
Thanks! I'll get this into my tree.
-Kees
> ---
> kernel/stackleak.c | 4 +++-
> 1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/kernel/stackleak.c b/kernel/stackleak.c
> index e428929..08cb57e 100644
> --- a/kernel/stackleak.c
> +++ b/kernel/stackleak.c
> @@ -11,6 +11,7 @@
> */
>
> #include <linux/stackleak.h>
> +#include <linux/kprobes.h>
>
> #ifdef CONFIG_STACKLEAK_RUNTIME_DISABLE
> #include <linux/jump_label.h>
> @@ -47,7 +48,7 @@ int stack_erasing_sysctl(struct ctl_table *table, int write,
> #define skip_erasing() false
> #endif /* CONFIG_STACKLEAK_RUNTIME_DISABLE */
>
> -asmlinkage void stackleak_erase(void)
> +asmlinkage void notrace stackleak_erase(void)
> {
> /* It would be nice not to have 'kstack_ptr' and 'boundary' on stack
> */
> unsigned long kstack_ptr = current->lowest_stack;
> @@ -101,6 +102,7 @@ asmlinkage void stackleak_erase(void)
> /* Reset the 'lowest_stack' value for the next syscall */
> current->lowest_stack = current_top_of_stack() - THREAD_SIZE/64;
> }
> +NOKPROBE_SYMBOL(stackleak_erase);
>
> void __used stackleak_track_stack(void)
> {
> --
> 2.7.4
>
--
Kees Cook
