Merge branch 'perf/urgent' into perf/core

Merge reason: We are going to queue up a dependent patch.

Signed-off-by: Ingo Molnar <mingo@elte.hu>

5 files changed, 97 insertions, 36 deletions

diff --git a/kernel/fork.c b/kernel/fork.c
index e2cd3e2a5ae8..26a7a6707fa7 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -668,6 +668,38 @@ struct mm_struct *mm_access(struct task_struct *task, unsigned int mode)
         return mm;
 }
 
+static void complete_vfork_done(struct task_struct *tsk)
+{
+        struct completion *vfork;
+
+        task_lock(tsk);
+        vfork = tsk->vfork_done;
+        if (likely(vfork)) {
+                tsk->vfork_done = NULL;
+                complete(vfork);
+        }
+        task_unlock(tsk);
+}
+
+static int wait_for_vfork_done(struct task_struct *child,
+                                struct completion *vfork)
+{
+        int killed;
+
+        freezer_do_not_count();
+        killed = wait_for_completion_killable(vfork);
+        freezer_count();
+
+        if (killed) {
+                task_lock(child);
+                child->vfork_done = NULL;
+                task_unlock(child);
+        }
+
+        put_task_struct(child);
+        return killed;
+}
+
 /* Please note the differences between mmput and mm_release.
  * mmput is called whenever we stop holding onto a mm_struct,
  * error success whatever.
@@ -683,8 +715,6 @@ struct mm_struct *mm_access(struct task_struct *task, unsigned int mode)
  */
 void mm_release(struct task_struct *tsk, struct mm_struct *mm)
 {
-        struct completion *vfork_done = tsk->vfork_done;
-
         /* Get rid of any futexes when releasing the mm */
 #ifdef CONFIG_FUTEX
         if (unlikely(tsk->robust_list)) {
@@ -704,17 +734,15 @@ void mm_release(struct task_struct *tsk, struct mm_struct *mm)
         /* Get rid of any cached register state */
         deactivate_mm(tsk, mm);
 
-        /* notify parent sleeping on vfork() */
-        if (vfork_done) {
-                tsk->vfork_done = NULL;
-                complete(vfork_done);
-        }
+        if (tsk->vfork_done)
+                complete_vfork_done(tsk);
 
         /*
          * If we're exiting normally, clear a user-space tid field if
          * requested.  We leave this alone when dying by signal, to leave
          * the value intact in a core dump, and to save the unnecessary
-         * trouble otherwise.  Userland only wants this done for a sys_exit.
+         * trouble, say, a killed vfork parent shouldn't touch this mm.
+         * Userland only wants this done for a sys_exit.
          */
         if (tsk->clear_child_tid) {
                 if (!(tsk->flags & PF_SIGNALED) &&
@@ -1018,7 +1046,6 @@ static void copy_flags(unsigned long clone_flags, struct task_struct *p)
 
         new_flags &= ~(PF_SUPERPRIV | PF_WQ_WORKER);
         new_flags |= PF_FORKNOEXEC;
-        new_flags |= PF_STARTING;
         p->flags = new_flags;
 }
 
@@ -1548,16 +1575,9 @@ long do_fork(unsigned long clone_flags,
                 if (clone_flags & CLONE_VFORK) {
                         p->vfork_done = &vfork;
                         init_completion(&vfork);
+                        get_task_struct(p);
                 }
 
-                /*
-                 * We set PF_STARTING at creation in case tracing wants to
-                 * use this to distinguish a fully live task from one that
-                 * hasn't finished SIGSTOP raising yet.  Now we clear it
-                 * and set the child going.
-                 */
-                p->flags &= ~PF_STARTING;
-
                 wake_up_new_task(p);
 
                 /* forking complete and child started to run, tell ptracer */
@@ -1565,10 +1585,8 @@ long do_fork(unsigned long clone_flags,
                         ptrace_event(trace, nr);
 
                 if (clone_flags & CLONE_VFORK) {
-                        freezer_do_not_count();
-                        wait_for_completion(&vfork);
-                        freezer_count();
-                        ptrace_event(PTRACE_EVENT_VFORK_DONE, nr);
+                        if (!wait_for_vfork_done(p, &vfork))
+                                ptrace_event(PTRACE_EVENT_VFORK_DONE, nr);
                 }
         } else {
                 nr = PTR_ERR(p);
diff --git a/kernel/hung_task.c b/kernel/hung_task.c
index 2e48ec0c2e91..c21449f85a2a 100644
--- a/kernel/hung_task.c
+++ b/kernel/hung_task.c
@@ -119,15 +119,20 @@ static void check_hung_task(struct task_struct *t, unsigned long timeout)
  * For preemptible RCU it is sufficient to call rcu_read_unlock in order
  * to exit the grace period. For classic RCU, a reschedule is required.
  */
-static void rcu_lock_break(struct task_struct *g, struct task_struct *t)
+static bool rcu_lock_break(struct task_struct *g, struct task_struct *t)
 {
+        bool can_cont;
+
         get_task_struct(g);
         get_task_struct(t);
         rcu_read_unlock();
         cond_resched();
         rcu_read_lock();
+        can_cont = pid_alive(g) && pid_alive(t);
         put_task_struct(t);
         put_task_struct(g);
+
+        return can_cont;
 }
 
 /*
@@ -154,9 +159,7 @@ static void check_hung_uninterruptible_tasks(unsigned long timeout)
                         goto unlock;
                 if (!--batch_count) {
                         batch_count = HUNG_TASK_BATCHING;
-                        rcu_lock_break(g, t);
-                        /* Exit if t or g was unhashed during refresh. */
-                        if (t->state == TASK_DEAD || g->state == TASK_DEAD)
+                        if (!rcu_lock_break(g, t))
                                 goto unlock;
                 }
                 /* use "==" to skip the TASK_KILLABLE tasks waiting on NFS */
diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c
index 32313c084442..0f0d4704ddd8 100644
--- a/kernel/irq/manage.c
+++ b/kernel/irq/manage.c
@@ -985,6 +985,11 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new)
 
                 /* add new interrupt at end of irq queue */
                 do {
+                        /*
+                         * Or all existing action->thread_mask bits,
+                         * so we can find the next zero bit for this
+                         * new action.
+                         */
                         thread_mask |= old->thread_mask;
                         old_ptr = &old->next;
                         old = *old_ptr;
@@ -993,14 +998,41 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new)
         }
 
         /*
-         * Setup the thread mask for this irqaction. Unlikely to have
-         * 32 resp 64 irqs sharing one line, but who knows.
+         * Setup the thread mask for this irqaction for ONESHOT. For
+         * !ONESHOT irqs the thread mask is 0 so we can avoid a
+         * conditional in irq_wake_thread().
          */
-        if (new->flags & IRQF_ONESHOT && thread_mask == ~0UL) {
-                ret = -EBUSY;
-                goto out_mask;
+        if (new->flags & IRQF_ONESHOT) {
+                /*
+                 * Unlikely to have 32 resp 64 irqs sharing one line,
+                 * but who knows.
+                 */
+                if (thread_mask == ~0UL) {
+                        ret = -EBUSY;
+                        goto out_mask;
+                }
+                /*
+                 * The thread_mask for the action is or'ed to
+                 * desc->thread_active to indicate that the
+                 * IRQF_ONESHOT thread handler has been woken, but not
+                 * yet finished. The bit is cleared when a thread
+                 * completes. When all threads of a shared interrupt
+                 * line have completed desc->threads_active becomes
+                 * zero and the interrupt line is unmasked. See
+                 * handle.c:irq_wake_thread() for further information.
+                 *
+                 * If no thread is woken by primary (hard irq context)
+                 * interrupt handlers, then desc->threads_active is
+                 * also checked for zero to unmask the irq line in the
+                 * affected hard irq flow handlers
+                 * (handle_[fasteoi|level]_irq).
+                 *
+                 * The new action gets the first zero bit of
+                 * thread_mask assigned. See the loop above which or's
+                 * all existing action->thread_mask bits.
+                 */
+                new->thread_mask = 1 << ffz(thread_mask);
         }
-        new->thread_mask = 1 << ffz(thread_mask);
 
         if (!shared) {
                 init_waitqueue_head(&desc->wait_for_threads);
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index 9788c0ec6f43..c62b8546cc90 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -1334,8 +1334,10 @@ int __kprobes register_kprobe(struct kprobe *p)
         if (!kernel_text_address((unsigned long) p->addr) ||
             in_kprobes_functions((unsigned long) p->addr) ||
             ftrace_text_reserved(p->addr, p->addr) ||
-            jump_label_text_reserved(p->addr, p->addr))
-                goto fail_with_jump_label;
+            jump_label_text_reserved(p->addr, p->addr)) {
+                ret = -EINVAL;
+                goto cannot_probe;
+        }
 
         /* User can pass only KPROBE_FLAG_DISABLED to register_kprobe */
         p->flags &= KPROBE_FLAG_DISABLED;
@@ -1352,7 +1354,7 @@ int __kprobes register_kprobe(struct kprobe *p)
                  * its code to prohibit unexpected unloading.
                  */
                 if (unlikely(!try_module_get(probed_mod)))
-                        goto fail_with_jump_label;
+                        goto cannot_probe;
 
                 /*
                  * If the module freed .init.text, we couldn't insert
@@ -1361,7 +1363,7 @@ int __kprobes register_kprobe(struct kprobe *p)
                 if (within_module_init((unsigned long)p->addr, probed_mod) &&
                     probed_mod->state != MODULE_STATE_COMING) {
                         module_put(probed_mod);
-                        goto fail_with_jump_label;
+                        goto cannot_probe;
                 }
                 /* ret will be updated by following code */
         }
@@ -1409,7 +1411,7 @@ out:
 
         return ret;
 
-fail_with_jump_label:
+cannot_probe:
         preempt_enable();
         jump_label_unlock();
         return ret;
diff --git a/kernel/printk.c b/kernel/printk.c
index cb8a6bd697c6..0b3ea2cbd5fb 100644
--- a/kernel/printk.c
+++ b/kernel/printk.c
@@ -707,6 +707,9 @@ static bool printk_time = 0;
 #endif
 module_param_named(time, printk_time, bool, S_IRUGO | S_IWUSR);
 
+static bool always_kmsg_dump;
+module_param_named(always_kmsg_dump, always_kmsg_dump, bool, S_IRUGO | S_IWUSR);
+
 /* Check if we have any console registered that can be called early in boot. */
 static int have_callable_console(void)
 {
@@ -1737,6 +1740,9 @@ void kmsg_dump(enum kmsg_dump_reason reason)
         unsigned long l1, l2;
         unsigned long flags;
 
+        if ((reason > KMSG_DUMP_OOPS) && !always_kmsg_dump)
+                return;
+
         /* Theoretically, the log could move on after we do this, but
            there's not a lot we can do about that. The new messages
            will overwrite the start of what we dump. */