Merge android-4.4.97 (46d256d) into msm-4.4

* refs/heads/tmp-46d256d
  Linux 4.4.97
  staging: r8712u: Fix Sparse warning in rtl871x_xmit.c
  xen: don't print error message in case of missing Xenstore entry
  bt8xx: fix memory leak
  s390/dasd: check for device error pointer within state change interrupts
  mei: return error on notification request to a disconnected client
  exynos4-is: fimc-is: Unmap region obtained by of_iomap()
  staging: lustre: ptlrpc: skip lock if export failed
  staging: lustre: hsm: stack overrun in hai_dump_data_field
  staging: lustre: llite: don't invoke direct_IO for the EOF case
  platform/x86: intel_mid_thermal: Fix module autoload
  scsi: aacraid: Process Error for response I/O
  xen/manage: correct return value check on xenbus_scanf()
  cx231xx: Fix I2C on Internal Master 3 Bus
  perf tools: Only increase index if perf_evsel__new_idx() succeeds
  drm/amdgpu: when dpm disabled, also need to stop/start vce.
  i2c: riic: correctly finish transfers
  ext4: do not use stripe_width if it is not set
  ext4: fix stripe-unaligned allocations
  staging: rtl8712u: Fix endian settings for structs describing network packets
  mfd: axp20x: Fix axp288 PEK_DBR and PEK_DBF irqs being swapped
  mfd: ab8500-sysctrl: Handle probe deferral
  ARM: pxa: Don't rely on public mmc header to include leds.h
  mmc: s3cmci: include linux/interrupt.h for tasklet_struct
  PM / wakeirq: report a wakeup_event on dedicated wekup irq
  Fix tracing sample code warning.
  tracing/samples: Fix creation and deletion of simple_thread_fn creation
  drm/msm: fix an integer overflow test
  drm/msm: Fix potential buffer overflow issue
  perf tools: Fix build failure on perl script context
  ocfs2: fstrim: Fix start offset of first cluster group during fstrim
  ARM: 8715/1: add a private asm/unaligned.h
  ARM: dts: mvebu: pl310-cache disable double-linefill
  arm64: ensure __dump_instr() checks addr_limit
  ASoC: adau17x1: Workaround for noise bug in ADC
  KEYS: fix out-of-bounds read during ASN.1 parsing
  KEYS: return full count in keyring_read() if buffer is too small
  cifs: check MaxPathNameComponentLength != 0 before using it
  ALSA: seq: Fix nested rwsem annotation for lockdep splat
  ALSA: timer: Add missing mutex lock for compat ioctls
  BACKPORT: xfrm: Clear sk_dst_cache when applying per-socket policy.
  Revert "ANDROID: sched/rt: schedtune: Add boost retention to RT"
  cpufreq: Drop schedfreq governor
  ANDROID: sched/rt: schedtune: Add boost retention to RT
  ANDROID: sched/rt: add schedtune accounting
  ANDROID: Revert "arm64: move ELF_ET_DYN_BASE to 4GB / 4MB"
  ANDROID: Revert "arm: move ELF_ET_DYN_BASE to 4MB"
  sched: EAS: Fix the calculation of group util in group_idle_state()
  sched: EAS: update trg_cpu to backup_cpu if no energy saving for target_cpu
  sched: EAS: Fix the condition to distinguish energy before/after

Conflicts:
	drivers/cpufreq/Kconfig
	drivers/gpu/drm/msm/msm_gem_submit.c
	kernel/sched/core.c
	kernel/sched/fair.c
	kernel/sched/rt.c
	kernel/sched/sched.h

Change-Id: I0d8c5287cb67fd47c8944a002c0ca71adcdef537
Signed-off-by: Srinivasarao P <spathi@codeaurora.org>

1 files changed, 13 insertions, 100 deletions

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index b3a8411bac2b..e515311aa93c 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -53,7 +53,6 @@ unsigned int sysctl_sched_latency = 6000000ULL;
 unsigned int normalized_sysctl_sched_latency = 6000000ULL;
 
 unsigned int sysctl_sched_sync_hint_enable = 1;
-unsigned int sysctl_sched_initial_task_util = 0;
 unsigned int sysctl_sched_cstate_aware = 1;
 
 /*
@@ -746,9 +745,7 @@ void init_entity_runnable_average(struct sched_entity *se)
 	sa->load_sum = sa->load_avg * LOAD_AVG_MAX;
 	/*
 	 * In previous Android versions, we used to have:
-	 * 	sa->util_avg =  sched_freq() ?
-	 * 		sysctl_sched_initial_task_util :
-	 *		scale_load_down(SCHED_LOAD_SCALE);
+	 * 	sa->util_avg = scale_load_down(SCHED_LOAD_SCALE);
 	 * 	sa->util_sum = sa->util_avg * LOAD_AVG_MAX;
 	 * However, that functionality has been moved to enqueue.
 	 * It is unclear if we should restore this in enqueue.
@@ -5759,23 +5756,6 @@ unsigned long boosted_cpu_util(int cpu);
 #define boosted_cpu_util(cpu) cpu_util_freq(cpu)
 #endif
 
-#if defined(CONFIG_SMP) && defined(CONFIG_CPU_FREQ_GOV_SCHED)
-static void update_capacity_of(int cpu)
-{
-	unsigned long req_cap;
-
-	if (!sched_freq())
-		return;
-
-	/* Normalize scale-invariant capacity to cpu. */
-	req_cap = boosted_cpu_util(cpu);
-	req_cap = req_cap * SCHED_CAPACITY_SCALE / capacity_orig_of(cpu);
-	set_cfs_cpu_capacity(cpu, true, req_cap);
-}
-#else
-#define update_capacity_of(X) do {} while(0)
-#endif /* SMP and CPU_FREQ_GOV_SCHED */
-
 /*
  * The enqueue_task method is called before nr_running is
  * increased. Here we update the fair scheduling stats and
@@ -5788,7 +5768,6 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags)
 	struct sched_entity *se = &p->se;
 #ifdef CONFIG_SMP
 	int task_new = flags & ENQUEUE_WAKEUP_NEW;
-	int task_wakeup = flags & ENQUEUE_WAKEUP;
 #endif
 
 	/*
@@ -5863,19 +5842,6 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags)
 			rq->rd->overutilized = true;
 			trace_sched_overutilized(true);
 		}
-
-	}
-
-	if (!se) {
-		/*
-		 * We want to potentially trigger a freq switch
-		 * request only for tasks that are waking up; this is
-		 * because we get here also during load balancing, but
-		 * in these cases it seems wise to trigger as single
-		 * request after load balancing is done.
-		 */
-		if (task_new || task_wakeup)
-			update_capacity_of(cpu_of(rq));
 	}
 
 #endif /* CONFIG_SMP */
@@ -5953,23 +5919,6 @@ static void dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags)
 	 */
 	schedtune_dequeue_task(p, cpu_of(rq));
 
-	if (!se) {
-		/*
-		 * We want to potentially trigger a freq switch
-		 * request only for tasks that are going to sleep;
-		 * this is because we get here also during load
-		 * balancing, but in these cases it seems wise to
-		 * trigger as single request after load balancing is
-		 * done.
-		 */
-		if (task_sleep) {
-			if (rq->cfs.nr_running)
-				update_capacity_of(cpu_of(rq));
-			else if (sched_freq())
-				set_cfs_cpu_capacity(cpu_of(rq), false, 0); /* no normalization required for 0 */
-		}
-	}
-
 #endif /* CONFIG_SMP */
 
 	hrtick_update(rq);
@@ -6521,13 +6470,6 @@ static int group_idle_state(struct energy_env *eenv, struct sched_group *sg)
 	/* Take non-cpuidle idling into account (active idle/arch_cpu_idle()) */
 	state++;
 
-	/*
-	 * Try to estimate if a deeper idle state is
-	 * achievable when we move the task.
-	 */
-	for_each_cpu(i, sched_group_cpus(sg))
-		grp_util += cpu_util(i);
-
 	src_in_grp = cpumask_test_cpu(eenv->src_cpu, sched_group_cpus(sg));
 	dst_in_grp = cpumask_test_cpu(eenv->dst_cpu, sched_group_cpus(sg));
 	if (src_in_grp == dst_in_grp) {
@@ -6536,10 +6478,16 @@ static int group_idle_state(struct energy_env *eenv, struct sched_group *sg)
 		 */
 		goto end;
 	}
-	/* add or remove util as appropriate to indicate what group util
-	 * will be (worst case - no concurrent execution) after moving the task
+
+	/*
+	 * Try to estimate if a deeper idle state is
+	 * achievable when we move the task.
 	 */
-	grp_util += src_in_grp ? -eenv->util_delta : eenv->util_delta;
+	for_each_cpu(i, sched_group_cpus(sg)) {
+		grp_util += cpu_util_wake(i, eenv->task);
+		if (unlikely(i == eenv->trg_cpu))
+			grp_util += eenv->util_delta;
+	}
 
 	if (grp_util <=
 		((long)sg->sgc->max_capacity * (int)sg->group_weight)) {
@@ -6626,13 +6574,13 @@ static int sched_group_energy(struct energy_env *eenv)
 
 				if (sg->group_weight == 1) {
 					/* Remove capacity of src CPU (before task move) */
-					if (eenv->util_delta == 0 &&
+					if (eenv->trg_cpu == eenv->src_cpu &&
 					    cpumask_test_cpu(eenv->src_cpu, sched_group_cpus(sg))) {
 						eenv->cap.before = sg->sge->cap_states[cap_idx].cap;
 						eenv->cap.delta -= eenv->cap.before;
 					}
 					/* Add capacity of dst CPU  (after task move) */
-					if (eenv->util_delta != 0 &&
+					if (eenv->trg_cpu == eenv->dst_cpu &&
 					    cpumask_test_cpu(eenv->dst_cpu, sched_group_cpus(sg))) {
 						eenv->cap.after = sg->sge->cap_states[cap_idx].cap;
 						eenv->cap.delta += eenv->cap.after;
@@ -7813,6 +7761,7 @@ static int select_energy_cpu_brute(struct task_struct *p, int prev_cpu, int sync
 			/* No energy saving for target_cpu, try backup */
 			target_cpu = tmp_backup;
 			eenv.dst_cpu = target_cpu;
+			eenv.trg_cpu = target_cpu;
 			if (tmp_backup < 0 ||
 			    tmp_backup == prev_cpu ||
 			    energy_diff(&eenv) >= 0) {
@@ -8865,10 +8814,6 @@ static void attach_one_task(struct rq *rq, struct task_struct *p)
 {
 	raw_spin_lock(&rq->lock);
 	attach_task(rq, p);
-	/*
-	 * We want to potentially raise target_cpu's OPP.
-	 */
-	update_capacity_of(cpu_of(rq));
 	raw_spin_unlock(&rq->lock);
 }
 
@@ -8890,11 +8835,6 @@ static void attach_tasks(struct lb_env *env)
 		attach_task(env->dst_rq, p);
 	}
 
-	/*
-	 * We want to potentially raise env.dst_cpu's OPP.
-	 */
-	update_capacity_of(env->dst_cpu);
-
 	raw_spin_unlock(&env->dst_rq->lock);
 }
 
@@ -10454,11 +10394,6 @@ more_balance:
 		 * ld_moved     - cumulative load moved across iterations
 		 */
 		cur_ld_moved = detach_tasks(&env);
-		/*
-		 * We want to potentially lower env.src_cpu's OPP.
-		 */
-		if (cur_ld_moved)
-			update_capacity_of(env.src_cpu);
 
 		/*
 		 * We've detached some tasks from busiest_rq. Every
@@ -10708,7 +10643,6 @@ static int idle_balance(struct rq *this_rq)
 	struct sched_domain *sd;
 	int pulled_task = 0;
 	u64 curr_cost = 0;
-	long removed_util=0;
 
 	if (cpu_isolated(this_cpu))
 		return 0;
@@ -10735,17 +10669,6 @@ static int idle_balance(struct rq *this_rq)
 
 	raw_spin_unlock(&this_rq->lock);
 
-	/*
-	 * If removed_util_avg is !0 we most probably migrated some task away
-	 * from this_cpu. In this case we might be willing to trigger an OPP
-	 * update, but we want to do so if we don't find anybody else to pull
-	 * here (we will trigger an OPP update with the pulled task's enqueue
-	 * anyway).
-	 *
-	 * Record removed_util before calling update_blocked_averages, and use
-	 * it below (before returning) to see if an OPP update is required.
-	 */
-	removed_util = atomic_long_read(&(this_rq->cfs).removed_util_avg);
 	update_blocked_averages(this_cpu);
 	rcu_read_lock();
 	for_each_domain(this_cpu, sd) {
@@ -10813,12 +10736,6 @@ out:
 	if (pulled_task) {
 		idle_exit_fair(this_rq);
 		this_rq->idle_stamp = 0;
-	} else if (removed_util) {
-		/*
-		 * No task pulled and someone has been migrated away.
-		 * Good case to trigger an OPP update.
-		 */
-		update_capacity_of(this_cpu);
 	}
 
 	return pulled_task;
@@ -10903,10 +10820,6 @@ static int active_load_balance_cpu_stop(void *data)
 		p = detach_one_task(&env);
 		if (p) {
 			schedstat_inc(sd, alb_pushed);
-			/*
-			 * We want to potentially lower env.src_cpu's OPP.
-			 */
-			update_capacity_of(env.src_cpu);
 			moved = true;
 		} else {
 			schedstat_inc(sd, alb_failed);