1 files changed, 147 insertions, 32 deletions

diff --git a/kernel/signal.c b/kernel/signal.c
index 80789a59b4db..d3efafd8109a 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -25,6 +25,7 @@
 #include <linux/posix-timers.h>
 #include <linux/signal.h>
 #include <linux/audit.h>
+#include <linux/capability.h>
 #include <asm/param.h>
 #include <asm/uaccess.h>
 #include <asm/unistd.h>
@@ -329,13 +330,20 @@ void __exit_sighand(struct task_struct *tsk)
         /* Ok, we're done with the signal handlers */
         tsk->sighand = NULL;
         if (atomic_dec_and_test(&sighand->count))
-                kmem_cache_free(sighand_cachep, sighand);
+                sighand_free(sighand);
 }
 
 void exit_sighand(struct task_struct *tsk)
 {
         write_lock_irq(&tasklist_lock);
-        __exit_sighand(tsk);
+        rcu_read_lock();
+        if (tsk->sighand != NULL) {
+                struct sighand_struct *sighand = rcu_dereference(tsk->sighand);
+                spin_lock(&sighand->siglock);
+                __exit_sighand(tsk);
+                spin_unlock(&sighand->siglock);
+        }
+        rcu_read_unlock();
         write_unlock_irq(&tasklist_lock);
 }
 
@@ -345,19 +353,20 @@ void exit_sighand(struct task_struct *tsk)
 void __exit_signal(struct task_struct *tsk)
 {
         struct signal_struct * sig = tsk->signal;
-        struct sighand_struct * sighand = tsk->sighand;
+        struct sighand_struct * sighand;
 
         if (!sig)
                 BUG();
         if (!atomic_read(&sig->count))
                 BUG();
+        rcu_read_lock();
+        sighand = rcu_dereference(tsk->sighand);
         spin_lock(&sighand->siglock);
         posix_cpu_timers_exit(tsk);
         if (atomic_dec_and_test(&sig->count)) {
                 posix_cpu_timers_exit_group(tsk);
-                if (tsk == sig->curr_target)
-                        sig->curr_target = next_thread(tsk);
                 tsk->signal = NULL;
+                __exit_sighand(tsk);
                 spin_unlock(&sighand->siglock);
                 flush_sigqueue(&sig->shared_pending);
         } else {
@@ -389,9 +398,11 @@ void __exit_signal(struct task_struct *tsk)
                 sig->nvcsw += tsk->nvcsw;
                 sig->nivcsw += tsk->nivcsw;
                 sig->sched_time += tsk->sched_time;
+                __exit_sighand(tsk);
                 spin_unlock(&sighand->siglock);
                 sig = NULL;     /* Marker for below.  */
         }
+        rcu_read_unlock();
         clear_tsk_thread_flag(tsk,TIF_SIGPENDING);
         flush_sigqueue(&tsk->pending);
         if (sig) {
@@ -465,7 +476,7 @@ unblock_all_signals(void)
         spin_unlock_irqrestore(&current->sighand->siglock, flags);
 }
 
-static inline int collect_signal(int sig, struct sigpending *list, siginfo_t *info)
+static int collect_signal(int sig, struct sigpending *list, siginfo_t *info)
 {
         struct sigqueue *q, *first = NULL;
         int still_pending = 0;
@@ -513,16 +524,7 @@ static int __dequeue_signal(struct sigpending *pending, sigset_t *mask,
 {
         int sig = 0;
 
-        /* SIGKILL must have priority, otherwise it is quite easy
-         * to create an unkillable process, sending sig < SIGKILL
-         * to self */
-        if (unlikely(sigismember(&pending->signal, SIGKILL))) {
-                if (!sigismember(mask, SIGKILL))
-                        sig = SIGKILL;
-        }
-
-        if (likely(!sig))
-                sig = next_signal(pending, mask);
+        sig = next_signal(pending, mask);
         if (sig) {
                 if (current->notifier) {
                         if (sigismember(current->notifier_mask, sig)) {
@@ -622,6 +624,33 @@ void signal_wake_up(struct task_struct *t, int resume)
  * Returns 1 if any signals were found.
  *
  * All callers must be holding the siglock.
+ *
+ * This version takes a sigset mask and looks at all signals,
+ * not just those in the first mask word.
+ */
+static int rm_from_queue_full(sigset_t *mask, struct sigpending *s)
+{
+        struct sigqueue *q, *n;
+        sigset_t m;
+
+        sigandsets(&m, mask, &s->signal);
+        if (sigisemptyset(&m))
+                return 0;
+
+        signandsets(&s->signal, &s->signal, mask);
+        list_for_each_entry_safe(q, n, &s->list, list) {
+                if (sigismember(mask, q->info.si_signo)) {
+                        list_del_init(&q->list);
+                        __sigqueue_free(q);
+                }
+        }
+        return 1;
+}
+/*
+ * Remove signals in mask from the pending set and queue.
+ * Returns 1 if any signals were found.
+ *
+ * All callers must be holding the siglock.
  */
 static int rm_from_queue(unsigned long mask, struct sigpending *s)
 {
@@ -1089,18 +1118,29 @@ void zap_other_threads(struct task_struct *p)
 }
 
 /*
- * Must be called with the tasklist_lock held for reading!
+ * Must be called under rcu_read_lock() or with tasklist_lock read-held.
  */
 int group_send_sig_info(int sig, struct siginfo *info, struct task_struct *p)
 {
         unsigned long flags;
+        struct sighand_struct *sp;
         int ret;
 
+retry:
         ret = check_kill_permission(sig, info, p);
-        if (!ret && sig && p->sighand) {
-                spin_lock_irqsave(&p->sighand->siglock, flags);
+        if (!ret && sig && (sp = rcu_dereference(p->sighand))) {
+                spin_lock_irqsave(&sp->siglock, flags);
+                if (p->sighand != sp) {
+                        spin_unlock_irqrestore(&sp->siglock, flags);
+                        goto retry;
+                }
+                if ((atomic_read(&sp->count) == 0) ||
+                                (atomic_read(&p->usage) == 0)) {
+                        spin_unlock_irqrestore(&sp->siglock, flags);
+                        return -ESRCH;
+                }
                 ret = __group_send_sig_info(sig, info, p);
-                spin_unlock_irqrestore(&p->sighand->siglock, flags);
+                spin_unlock_irqrestore(&sp->siglock, flags);
         }
 
         return ret;
@@ -1145,14 +1185,21 @@ int
 kill_proc_info(int sig, struct siginfo *info, pid_t pid)
 {
         int error;
+        int acquired_tasklist_lock = 0;
         struct task_struct *p;
 
-        read_lock(&tasklist_lock);
+        rcu_read_lock();
+        if (unlikely(sig_kernel_stop(sig) || sig == SIGCONT)) {
+                read_lock(&tasklist_lock);
+                acquired_tasklist_lock = 1;
+        }
         p = find_task_by_pid(pid);
         error = -ESRCH;
         if (p)
                 error = group_send_sig_info(sig, info, p);
-        read_unlock(&tasklist_lock);
+        if (unlikely(acquired_tasklist_lock))
+                read_unlock(&tasklist_lock);
+        rcu_read_unlock();
         return error;
 }
 
@@ -1172,8 +1219,7 @@ int kill_proc_info_as_uid(int sig, struct siginfo *info, pid_t pid,
                 ret = -ESRCH;
                 goto out_unlock;
         }
-        if ((!info || ((unsigned long)info != 1 &&
-                        (unsigned long)info != 2 && SI_FROMUSER(info)))
+        if ((info == SEND_SIG_NOINFO || (!is_si_special(info) && SI_FROMUSER(info)))
             && (euid != p->suid) && (euid != p->uid)
             && (uid != p->suid) && (uid != p->uid)) {
                 ret = -EPERM;
@@ -1364,16 +1410,54 @@ send_sigqueue(int sig, struct sigqueue *q, struct task_struct *p)
 {
         unsigned long flags;
         int ret = 0;
+        struct sighand_struct *sh;
 
         BUG_ON(!(q->flags & SIGQUEUE_PREALLOC));
-        read_lock(&tasklist_lock);
+
+        /*
+         * The rcu based delayed sighand destroy makes it possible to
+         * run this without tasklist lock held. The task struct itself
+         * cannot go away as create_timer did get_task_struct().
+         *
+         * We return -1, when the task is marked exiting, so
+         * posix_timer_event can redirect it to the group leader
+         */
+        rcu_read_lock();
 
         if (unlikely(p->flags & PF_EXITING)) {
                 ret = -1;
                 goto out_err;
         }
 
-        spin_lock_irqsave(&p->sighand->siglock, flags);
+retry:
+        sh = rcu_dereference(p->sighand);
+
+        spin_lock_irqsave(&sh->siglock, flags);
+        if (p->sighand != sh) {
+                /* We raced with exec() in a multithreaded process... */
+                spin_unlock_irqrestore(&sh->siglock, flags);
+                goto retry;
+        }
+
+        /*
+         * We do the check here again to handle the following scenario:
+         *
+         * CPU 0                CPU 1
+         * send_sigqueue
+         * check PF_EXITING
+         * interrupt            exit code running
+         *                      __exit_signal
+         *                      lock sighand->siglock
+         *                      unlock sighand->siglock
+         * lock sh->siglock
+         * add(tsk->pending)    flush_sigqueue(tsk->pending)
+         *
+         */
+
+        if (unlikely(p->flags & PF_EXITING)) {
+                ret = -1;
+                goto out;
+        }
 
         if (unlikely(!list_empty(&q->list))) {
                 /*
@@ -1397,9 +1481,9 @@ send_sigqueue(int sig, struct sigqueue *q, struct task_struct *p)
                 signal_wake_up(p, sig == SIGKILL);
 
 out:
-        spin_unlock_irqrestore(&p->sighand->siglock, flags);
+        spin_unlock_irqrestore(&sh->siglock, flags);
 out_err:
-        read_unlock(&tasklist_lock);
+        rcu_read_unlock();
 
         return ret;
 }
@@ -1411,7 +1495,9 @@ send_group_sigqueue(int sig, struct sigqueue *q, struct task_struct *p)
         int ret = 0;
 
         BUG_ON(!(q->flags & SIGQUEUE_PREALLOC));
+
         read_lock(&tasklist_lock);
+        /* Since it_lock is held, p->sighand cannot be NULL. */
         spin_lock_irqsave(&p->sighand->siglock, flags);
         handle_stop_signal(sig, p);
 
@@ -1445,7 +1531,7 @@ send_group_sigqueue(int sig, struct sigqueue *q, struct task_struct *p)
 out:
         spin_unlock_irqrestore(&p->sighand->siglock, flags);
         read_unlock(&tasklist_lock);
-        return(ret);
+        return ret;
 }
 
 /*
@@ -1795,7 +1881,7 @@ do_signal_stop(int signr)
  * We return zero if we still hold the siglock and should look
  * for another signal without checking group_stop_count again.
  */
-static inline int handle_group_stop(void)
+static int handle_group_stop(void)
 {
         int stop_count;
 
@@ -2347,6 +2433,7 @@ int
 do_sigaction(int sig, const struct k_sigaction *act, struct k_sigaction *oact)
 {
         struct k_sigaction *k;
+        sigset_t mask;
 
         if (!valid_signal(sig) || sig < 1 || (act && sig_kernel_only(sig)))
                 return -EINVAL;
@@ -2394,9 +2481,11 @@ do_sigaction(int sig, const struct k_sigaction *act, struct k_sigaction *oact)
                         *k = *act;
                         sigdelsetmask(&k->sa.sa_mask,
                                       sigmask(SIGKILL) | sigmask(SIGSTOP));
-                        rm_from_queue(sigmask(sig), &t->signal->shared_pending);
+                        sigemptyset(&mask);
+                        sigaddset(&mask, sig);
+                        rm_from_queue_full(&mask, &t->signal->shared_pending);
                         do {
-                                rm_from_queue(sigmask(sig), &t->pending);
+                                rm_from_queue_full(&mask, &t->pending);
                                 recalc_sigpending_tsk(t);
                                 t = next_thread(t);
                         } while (t != current);
@@ -2632,6 +2721,32 @@ sys_pause(void)
 
 #endif
 
+#ifdef __ARCH_WANT_SYS_RT_SIGSUSPEND
+asmlinkage long sys_rt_sigsuspend(sigset_t __user *unewset, size_t sigsetsize)
+{
+        sigset_t newset;
+
+        /* XXX: Don't preclude handling different sized sigset_t's.  */
+        if (sigsetsize != sizeof(sigset_t))
+                return -EINVAL;
+
+        if (copy_from_user(&newset, unewset, sizeof(newset)))
+                return -EFAULT;
+        sigdelsetmask(&newset, sigmask(SIGKILL)|sigmask(SIGSTOP));
+
+        spin_lock_irq(&current->sighand->siglock);
+        current->saved_sigmask = current->blocked;
+        current->blocked = newset;
+        recalc_sigpending();
+        spin_unlock_irq(&current->sighand->siglock);
+
+        current->state = TASK_INTERRUPTIBLE;
+        schedule();
+        set_thread_flag(TIF_RESTORE_SIGMASK);
+        return -ERESTARTNOHAND;
+}
+#endif /* __ARCH_WANT_SYS_RT_SIGSUSPEND */
+
 void __init signals_init(void)
 {
         sigqueue_cachep =