On Mon, May 28, 2018 at 03:12:19PM +0100, Al Viro wrote:
> On Mon, May 28, 2018 at 03:49:11PM +0200, Christian Brauner wrote:
> > may_ptrace_stop() already behaves like a boolean function. Let's actually
> > declare it as such too.
> >
> > Signed-off-by: Christian Brauner <christ...@brauner.io>
> > ---
> > kernel/signal.c | 8 ++++----
> > 1 file changed, 4 insertions(+), 4 deletions(-)
> >
> > diff --git a/kernel/signal.c b/kernel/signal.c
> > index 81be01d193f4..6c2e7b45cba1 100644
> > --- a/kernel/signal.c
> > +++ b/kernel/signal.c
> > @@ -1889,10 +1889,10 @@ static void do_notify_parent_cldstop(struct
> > task_struct *tsk,
> > spin_unlock_irqrestore(&sighand->siglock, flags);
> > }
> >
> > -static inline int may_ptrace_stop(void)
> > +static inline bool may_ptrace_stop(void)
> > {
> > if (!likely(current->ptrace))
> > - return 0;
> > + return false;
> > /*
> > * Are we in the middle of do_coredump?
> > * If so and our tracer is also part of the coredump stopping
> > @@ -1908,9 +1908,9 @@ static inline int may_ptrace_stop(void)
> > */
> > if (unlikely(current->mm->core_state) &&
> > unlikely(current->mm == current->parent->mm))
> > - return 0;
> > + return false;
> >
> > - return 1;
> > + return true;
>
> return !current->mm->core_state || current->mm != current->parent->mm;
>
> or, if it gives any measurably better code generation,
>
> return likely(!current->mm->core_state ||
> current->mm != current->parent->mm);
Makes sense. I put this in v1.
Thanks!
Christian
