Merge tag 'omap-for-v4.6/fixes-rc2-v2-signed' of git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap into fixes

Merge "omap fixes against v4.6-rc2" from Tony Lindgren

Fixes for omaps against v4.6-rc2, mostly to fix suspend for beagle-x15
that broke when we added runtime based SoC revision detection earlier.

It seems suspend worked earlier as things were only partially initialized,
while now we initialize things properly for dra7.

Note that the "ARM: OMAP: Catch callers of revision information prior
to it being populated" had to be reverted as it caused bogus warnings
for other SoCs because omap initcalls bail out based on revision being
set to 0 for other SoCs. These initcalls will mostly just disappear
when we drop support for omap3 legacy booting.

Also included is a fix for dra7 sys_32k_ck clock source that is not
enabled on boot making system fall back to using emulated clock.

* tag 'omap-for-v4.6/fixes-rc2-v2-signed' of git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap: (198 commits)
  Revert "ARM: OMAP: Catch callers of revision information prior to it being populated"
  ARM: OMAP: Catch callers of revision information prior to it being populated
  ARM: dts: dra7: Correct clock tree for sys_32k_ck
  ARM: OMAP: DRA7: Provide proper class to omap2_set_globals_tap
  ARM: OMAP: DRA7: wakeupgen: Skip SAR save for wakeupgen
  Linux 4.6-rc2
  v4l2-mc: avoid warning about unused variable
  Convert straggling drivers to new six-argument get_user_pages()
  .mailmap: add Christophe Ricard
  Make CONFIG_FHANDLE default y
  mm/page_isolation.c: fix the function comments
  oom, oom_reaper: do not enqueue task if it is on the oom_reaper_list head
  mm/page_isolation: fix tracepoint to mirror check function behavior
  mm/rmap: batched invalidations should use existing api
  x86/mm: TLB_REMOTE_SEND_IPI should count pages
  mm: fix invalid node in alloc_migrate_target()
  include/linux/huge_mm.h: return NULL instead of false for pmd_trans_huge_lock()
  mm, kasan: fix compilation for CONFIG_SLAB
  MAINTAINERS: orangefs mailing list is subscribers-only
  net: mvneta: fix changing MTU when using per-cpu processing
  ...

6 files changed, 143 insertions, 38 deletions

diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 2a2efe1bc76c..adc5e4bd74f8 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -137,11 +137,13 @@ static void bpf_map_show_fdinfo(struct seq_file *m, struct file *filp)
 		   "map_type:\t%u\n"
 		   "key_size:\t%u\n"
 		   "value_size:\t%u\n"
-		   "max_entries:\t%u\n",
+		   "max_entries:\t%u\n"
+		   "map_flags:\t%#x\n",
 		   map->map_type,
 		   map->key_size,
 		   map->value_size,
-		   map->max_entries);
+		   map->max_entries,
+		   map->map_flags);
 }
 #endif
 
diff --git a/kernel/events/core.c b/kernel/events/core.c
index de24fbce5277..52bedc5a5aaa 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -2417,14 +2417,24 @@ static void ctx_sched_out(struct perf_event_context *ctx,
 			cpuctx->task_ctx = NULL;
 	}
 
-	is_active ^= ctx->is_active; /* changed bits */
-
+	/*
+	 * Always update time if it was set; not only when it changes.
+	 * Otherwise we can 'forget' to update time for any but the last
+	 * context we sched out. For example:
+	 *
+	 *   ctx_sched_out(.event_type = EVENT_FLEXIBLE)
+	 *   ctx_sched_out(.event_type = EVENT_PINNED)
+	 *
+	 * would only update time for the pinned events.
+	 */
 	if (is_active & EVENT_TIME) {
 		/* update (and stop) ctx time */
 		update_context_time(ctx);
 		update_cgrp_time_from_cpuctx(cpuctx);
 	}
 
+	is_active ^= ctx->is_active; /* changed bits */
+
 	if (!ctx->nr_active || !(is_active & EVENT_ALL))
 		return;
 
@@ -8532,6 +8542,7 @@ SYSCALL_DEFINE5(perf_event_open,
 					f_flags);
 	if (IS_ERR(event_file)) {
 		err = PTR_ERR(event_file);
+		event_file = NULL;
 		goto err_context;
 	}
 
diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c
index 53ab2f85d77e..2324ba5310db 100644
--- a/kernel/locking/lockdep.c
+++ b/kernel/locking/lockdep.c
@@ -2000,6 +2000,77 @@ static inline int get_first_held_lock(struct task_struct *curr,
 }
 
 /*
+ * Returns the next chain_key iteration
+ */
+static u64 print_chain_key_iteration(int class_idx, u64 chain_key)
+{
+	u64 new_chain_key = iterate_chain_key(chain_key, class_idx);
+
+	printk(" class_idx:%d -> chain_key:%016Lx",
+		class_idx,
+		(unsigned long long)new_chain_key);
+	return new_chain_key;
+}
+
+static void
+print_chain_keys_held_locks(struct task_struct *curr, struct held_lock *hlock_next)
+{
+	struct held_lock *hlock;
+	u64 chain_key = 0;
+	int depth = curr->lockdep_depth;
+	int i;
+
+	printk("depth: %u\n", depth + 1);
+	for (i = get_first_held_lock(curr, hlock_next); i < depth; i++) {
+		hlock = curr->held_locks + i;
+		chain_key = print_chain_key_iteration(hlock->class_idx, chain_key);
+
+		print_lock(hlock);
+	}
+
+	print_chain_key_iteration(hlock_next->class_idx, chain_key);
+	print_lock(hlock_next);
+}
+
+static void print_chain_keys_chain(struct lock_chain *chain)
+{
+	int i;
+	u64 chain_key = 0;
+	int class_id;
+
+	printk("depth: %u\n", chain->depth);
+	for (i = 0; i < chain->depth; i++) {
+		class_id = chain_hlocks[chain->base + i];
+		chain_key = print_chain_key_iteration(class_id + 1, chain_key);
+
+		print_lock_name(lock_classes + class_id);
+		printk("\n");
+	}
+}
+
+static void print_collision(struct task_struct *curr,
+			struct held_lock *hlock_next,
+			struct lock_chain *chain)
+{
+	printk("\n");
+	printk("======================\n");
+	printk("[chain_key collision ]\n");
+	print_kernel_ident();
+	printk("----------------------\n");
+	printk("%s/%d: ", current->comm, task_pid_nr(current));
+	printk("Hash chain already cached but the contents don't match!\n");
+
+	printk("Held locks:");
+	print_chain_keys_held_locks(curr, hlock_next);
+
+	printk("Locks in cached chain:");
+	print_chain_keys_chain(chain);
+
+	printk("\nstack backtrace:\n");
+	dump_stack();
+}
+
+/*
  * Checks whether the chain and the current held locks are consistent
  * in depth and also in content. If they are not it most likely means
  * that there was a collision during the calculation of the chain_key.
@@ -2014,14 +2085,18 @@ static int check_no_collision(struct task_struct *curr,
 
 	i = get_first_held_lock(curr, hlock);
 
-	if (DEBUG_LOCKS_WARN_ON(chain->depth != curr->lockdep_depth - (i - 1)))
+	if (DEBUG_LOCKS_WARN_ON(chain->depth != curr->lockdep_depth - (i - 1))) {
+		print_collision(curr, hlock, chain);
 		return 0;
+	}
 
 	for (j = 0; j < chain->depth - 1; j++, i++) {
 		id = curr->held_locks[i].class_idx - 1;
 
-		if (DEBUG_LOCKS_WARN_ON(chain_hlocks[chain->base + j] != id))
+		if (DEBUG_LOCKS_WARN_ON(chain_hlocks[chain->base + j] != id)) {
+			print_collision(curr, hlock, chain);
 			return 0;
+		}
 	}
 #endif
 	return 1;
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index d8465eeab8b3..8b489fcac37b 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -321,6 +321,24 @@ static inline void init_hrtick(void)
 }
 #endif	/* CONFIG_SCHED_HRTICK */
 
+/*
+ * cmpxchg based fetch_or, macro so it works for different integer types
+ */
+#define fetch_or(ptr, mask)						\
+	({								\
+		typeof(ptr) _ptr = (ptr);				\
+		typeof(mask) _mask = (mask);				\
+		typeof(*_ptr) _old, _val = *_ptr;			\
+									\
+		for (;;) {						\
+			_old = cmpxchg(_ptr, _val, _val | _mask);	\
+			if (_old == _val)				\
+				break;					\
+			_val = _old;					\
+		}							\
+	_old;								\
+})
+
 #if defined(CONFIG_SMP) && defined(TIF_POLLING_NRFLAG)
 /*
  * Atomically set TIF_NEED_RESCHED and test for TIF_POLLING_NRFLAG,
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index 084b79f5917e..58e3310c9b21 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -157,52 +157,50 @@ static void tick_sched_handle(struct tick_sched *ts, struct pt_regs *regs)
 cpumask_var_t tick_nohz_full_mask;
 cpumask_var_t housekeeping_mask;
 bool tick_nohz_full_running;
-static unsigned long tick_dep_mask;
+static atomic_t tick_dep_mask;
 
-static void trace_tick_dependency(unsigned long dep)
+static bool check_tick_dependency(atomic_t *dep)
 {
-	if (dep & TICK_DEP_MASK_POSIX_TIMER) {
+	int val = atomic_read(dep);
+
+	if (val & TICK_DEP_MASK_POSIX_TIMER) {
 		trace_tick_stop(0, TICK_DEP_MASK_POSIX_TIMER);
-		return;
+		return true;
 	}
 
-	if (dep & TICK_DEP_MASK_PERF_EVENTS) {
+	if (val & TICK_DEP_MASK_PERF_EVENTS) {
 		trace_tick_stop(0, TICK_DEP_MASK_PERF_EVENTS);
-		return;
+		return true;
 	}
 
-	if (dep & TICK_DEP_MASK_SCHED) {
+	if (val & TICK_DEP_MASK_SCHED) {
 		trace_tick_stop(0, TICK_DEP_MASK_SCHED);
-		return;
+		return true;
 	}
 
-	if (dep & TICK_DEP_MASK_CLOCK_UNSTABLE)
+	if (val & TICK_DEP_MASK_CLOCK_UNSTABLE) {
 		trace_tick_stop(0, TICK_DEP_MASK_CLOCK_UNSTABLE);
+		return true;
+	}
+
+	return false;
 }
 
 static bool can_stop_full_tick(struct tick_sched *ts)
 {
 	WARN_ON_ONCE(!irqs_disabled());
 
-	if (tick_dep_mask) {
-		trace_tick_dependency(tick_dep_mask);
+	if (check_tick_dependency(&tick_dep_mask))
 		return false;
-	}
 
-	if (ts->tick_dep_mask) {
-		trace_tick_dependency(ts->tick_dep_mask);
+	if (check_tick_dependency(&ts->tick_dep_mask))
 		return false;
-	}
 
-	if (current->tick_dep_mask) {
-		trace_tick_dependency(current->tick_dep_mask);
+	if (check_tick_dependency(&current->tick_dep_mask))
 		return false;
-	}
 
-	if (current->signal->tick_dep_mask) {
-		trace_tick_dependency(current->signal->tick_dep_mask);
+	if (check_tick_dependency(&current->signal->tick_dep_mask))
 		return false;
-	}
 
 	return true;
 }
@@ -259,12 +257,12 @@ static void tick_nohz_full_kick_all(void)
 	preempt_enable();
 }
 
-static void tick_nohz_dep_set_all(unsigned long *dep,
+static void tick_nohz_dep_set_all(atomic_t *dep,
 				  enum tick_dep_bits bit)
 {
-	unsigned long prev;
+	int prev;
 
-	prev = fetch_or(dep, BIT_MASK(bit));
+	prev = atomic_fetch_or(dep, BIT(bit));
 	if (!prev)
 		tick_nohz_full_kick_all();
 }
@@ -280,7 +278,7 @@ void tick_nohz_dep_set(enum tick_dep_bits bit)
 
 void tick_nohz_dep_clear(enum tick_dep_bits bit)
 {
-	clear_bit(bit, &tick_dep_mask);
+	atomic_andnot(BIT(bit), &tick_dep_mask);
 }
 
 /*
@@ -289,12 +287,12 @@ void tick_nohz_dep_clear(enum tick_dep_bits bit)
  */
 void tick_nohz_dep_set_cpu(int cpu, enum tick_dep_bits bit)
 {
-	unsigned long prev;
+	int prev;
 	struct tick_sched *ts;
 
 	ts = per_cpu_ptr(&tick_cpu_sched, cpu);
 
-	prev = fetch_or(&ts->tick_dep_mask, BIT_MASK(bit));
+	prev = atomic_fetch_or(&ts->tick_dep_mask, BIT(bit));
 	if (!prev) {
 		preempt_disable();
 		/* Perf needs local kick that is NMI safe */
@@ -313,7 +311,7 @@ void tick_nohz_dep_clear_cpu(int cpu, enum tick_dep_bits bit)
 {
 	struct tick_sched *ts = per_cpu_ptr(&tick_cpu_sched, cpu);
 
-	clear_bit(bit, &ts->tick_dep_mask);
+	atomic_andnot(BIT(bit), &ts->tick_dep_mask);
 }
 
 /*
@@ -331,7 +329,7 @@ void tick_nohz_dep_set_task(struct task_struct *tsk, enum tick_dep_bits bit)
 
 void tick_nohz_dep_clear_task(struct task_struct *tsk, enum tick_dep_bits bit)
 {
-	clear_bit(bit, &tsk->tick_dep_mask);
+	atomic_andnot(BIT(bit), &tsk->tick_dep_mask);
 }
 
 /*
@@ -345,7 +343,7 @@ void tick_nohz_dep_set_signal(struct signal_struct *sig, enum tick_dep_bits bit)
 
 void tick_nohz_dep_clear_signal(struct signal_struct *sig, enum tick_dep_bits bit)
 {
-	clear_bit(bit, &sig->tick_dep_mask);
+	atomic_andnot(BIT(bit), &sig->tick_dep_mask);
 }
 
 /*
@@ -366,7 +364,8 @@ void __tick_nohz_task_switch(void)
 	ts = this_cpu_ptr(&tick_cpu_sched);
 
 	if (ts->tick_stopped) {
-		if (current->tick_dep_mask || current->signal->tick_dep_mask)
+		if (atomic_read(&current->tick_dep_mask) ||
+		    atomic_read(&current->signal->tick_dep_mask))
 			tick_nohz_full_kick();
 	}
 out:
diff --git a/kernel/time/tick-sched.h b/kernel/time/tick-sched.h
index eb4e32566a83..bf38226e5c17 100644
--- a/kernel/time/tick-sched.h
+++ b/kernel/time/tick-sched.h
@@ -60,7 +60,7 @@ struct tick_sched {
 	u64				next_timer;
 	ktime_t				idle_expires;
 	int				do_timer_last;
-	unsigned long			tick_dep_mask;
+	atomic_t			tick_dep_mask;
 };
 
 extern struct tick_sched *tick_get_tick_sched(int cpu);