On Mon, Jun 07, 2021 at 03:54:27PM +0200, Bernd Edlinger wrote: > unshare_sighand should only access oldsighand->action > while holding oldsighand->siglock, to make sure that > newsighand->action is in a consistent state. > > Signed-off-by: Bernd Edlinger <bernd.edlinger@xxxxxxxxxx> > --- > fs/exec.c | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > > diff --git a/fs/exec.c b/fs/exec.c > index d8af85f..8344fba 100644 > --- a/fs/exec.c > +++ b/fs/exec.c > @@ -1193,11 +1193,11 @@ static int unshare_sighand(struct task_struct *me) > return -ENOMEM; > > refcount_set(&newsighand->count, 1); > - memcpy(newsighand->action, oldsighand->action, > - sizeof(newsighand->action)); > > write_lock_irq(&tasklist_lock); > spin_lock(&oldsighand->siglock); > + memcpy(newsighand->action, oldsighand->action, > + sizeof(newsighand->action)); > rcu_assign_pointer(me->sighand, newsighand); > spin_unlock(&oldsighand->siglock); > write_unlock_irq(&tasklist_lock); Oh, yeah, that's a nice catch. Reviewed-by: Kees Cook <keescook@xxxxxxxxxxxx> -- Kees Cook
