Merge branch 'dynamic_sg' into rdma.git for-next

From Maor Gottlieb says:

====================
This series extends __sg_alloc_table_from_pages to allow chaining of new
pages to an already initialized SG table.

This allows for drivers to utilize the optimization of merging contiguous
pages without a need to pre allocate all the pages and hold them in a very
large temporary buffer prior to the call to SG table initialization.

The last patch changes the Infiniband core to use the new API. It removes
duplicate functionality from the code and benefits from the optimization
of allocating dynamic SG table from pages.

In huge pages system of 2MB page size, without this change, the SG table
would contain x512 SG entries.
====================

* branch 'dynamic_sg':
  RDMA/umem: Move to allocate SG table from pages
  lib/scatterlist: Add support in dynamic allocation of SG table from pages
  tools/testing/scatterlist: Show errors in human readable form
  tools/testing/scatterlist: Rejuvenate bit-rotten test

1 files changed, 61 insertions, 48 deletions

diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c
index 71a30b0d0f05..f66236cff69b 100644
--- a/drivers/acpi/processor_idle.c
+++ b/drivers/acpi/processor_idle.c
@@ -161,18 +161,10 @@ static void lapic_timer_propagate_broadcast(struct acpi_processor *pr)
 }
 
 /* Power(C) State timer broadcast control */
-static void lapic_timer_state_broadcast(struct acpi_processor *pr,
-				       struct acpi_processor_cx *cx,
-				       int broadcast)
+static bool lapic_timer_needs_broadcast(struct acpi_processor *pr,
+					struct acpi_processor_cx *cx)
 {
-	int state = cx - pr->power.states;
-
-	if (state >= pr->power.timer_broadcast_on_state) {
-		if (broadcast)
-			tick_broadcast_enter();
-		else
-			tick_broadcast_exit();
-	}
+	return cx - pr->power.states >= pr->power.timer_broadcast_on_state;
 }
 
 #else
@@ -180,10 +172,11 @@ static void lapic_timer_state_broadcast(struct acpi_processor *pr,
 static void lapic_timer_check_state(int state, struct acpi_processor *pr,
 				   struct acpi_processor_cx *cstate) { }
 static void lapic_timer_propagate_broadcast(struct acpi_processor *pr) { }
-static void lapic_timer_state_broadcast(struct acpi_processor *pr,
-				       struct acpi_processor_cx *cx,
-				       int broadcast)
+
+static bool lapic_timer_needs_broadcast(struct acpi_processor *pr,
+					struct acpi_processor_cx *cx)
 {
+	return false;
 }
 
 #endif
@@ -566,32 +559,43 @@ static DEFINE_RAW_SPINLOCK(c3_lock);
 
 /**
  * acpi_idle_enter_bm - enters C3 with proper BM handling
+ * @drv: cpuidle driver
  * @pr: Target processor
  * @cx: Target state context
- * @timer_bc: Whether or not to change timer mode to broadcast
+ * @index: index of target state
  */
-static void acpi_idle_enter_bm(struct acpi_processor *pr,
-			       struct acpi_processor_cx *cx, bool timer_bc)
+static int acpi_idle_enter_bm(struct cpuidle_driver *drv,
+			       struct acpi_processor *pr,
+			       struct acpi_processor_cx *cx,
+			       int index)
 {
-	acpi_unlazy_tlb(smp_processor_id());
-
-	/*
-	 * Must be done before busmaster disable as we might need to
-	 * access HPET !
-	 */
-	if (timer_bc)
-		lapic_timer_state_broadcast(pr, cx, 1);
+	static struct acpi_processor_cx safe_cx = {
+		.entry_method = ACPI_CSTATE_HALT,
+	};
 
 	/*
 	 * disable bus master
 	 * bm_check implies we need ARB_DIS
 	 * bm_control implies whether we can do ARB_DIS
 	 *
-	 * That leaves a case where bm_check is set and bm_control is
-	 * not set. In that case we cannot do much, we enter C3
-	 * without doing anything.
+	 * That leaves a case where bm_check is set and bm_control is not set.
+	 * In that case we cannot do much, we enter C3 without doing anything.
 	 */
-	if (pr->flags.bm_control) {
+	bool dis_bm = pr->flags.bm_control;
+
+	/* If we can skip BM, demote to a safe state. */
+	if (!cx->bm_sts_skip && acpi_idle_bm_check()) {
+		dis_bm = false;
+		index = drv->safe_state_index;
+		if (index >= 0) {
+			cx = this_cpu_read(acpi_cstate[index]);
+		} else {
+			cx = &safe_cx;
+			index = -EBUSY;
+		}
+	}
+
+	if (dis_bm) {
 		raw_spin_lock(&c3_lock);
 		c3_cpu_count++;
 		/* Disable bus master arbitration when all CPUs are in C3 */
@@ -600,18 +604,21 @@ static void acpi_idle_enter_bm(struct acpi_processor *pr,
 		raw_spin_unlock(&c3_lock);
 	}
 
+	rcu_idle_enter();
+
 	acpi_idle_do_entry(cx);
 
+	rcu_idle_exit();
+
 	/* Re-enable bus master arbitration */
-	if (pr->flags.bm_control) {
+	if (dis_bm) {
 		raw_spin_lock(&c3_lock);
 		acpi_write_bit_register(ACPI_BITREG_ARB_DISABLE, 0);
 		c3_cpu_count--;
 		raw_spin_unlock(&c3_lock);
 	}
 
-	if (timer_bc)
-		lapic_timer_state_broadcast(pr, cx, 0);
+	return index;
 }
 
 static int acpi_idle_enter(struct cpuidle_device *dev,
@@ -625,32 +632,21 @@ static int acpi_idle_enter(struct cpuidle_device *dev,
 		return -EINVAL;
 
 	if (cx->type != ACPI_STATE_C1) {
+		if (cx->type == ACPI_STATE_C3 && pr->flags.bm_check)
+			return acpi_idle_enter_bm(drv, pr, cx, index);
+
+		/* C2 to C1 demotion. */
 		if (acpi_idle_fallback_to_c1(pr) && num_online_cpus() > 1) {
 			index = ACPI_IDLE_STATE_START;
 			cx = per_cpu(acpi_cstate[index], dev->cpu);
-		} else if (cx->type == ACPI_STATE_C3 && pr->flags.bm_check) {
-			if (cx->bm_sts_skip || !acpi_idle_bm_check()) {
-				acpi_idle_enter_bm(pr, cx, true);
-				return index;
-			} else if (drv->safe_state_index >= 0) {
-				index = drv->safe_state_index;
-				cx = per_cpu(acpi_cstate[index], dev->cpu);
-			} else {
-				acpi_safe_halt();
-				return -EBUSY;
-			}
 		}
 	}
 
-	lapic_timer_state_broadcast(pr, cx, 1);
-
 	if (cx->type == ACPI_STATE_C3)
 		ACPI_FLUSH_CPU_CACHE();
 
 	acpi_idle_do_entry(cx);
 
-	lapic_timer_state_broadcast(pr, cx, 0);
-
 	return index;
 }
 
@@ -666,7 +662,13 @@ static int acpi_idle_enter_s2idle(struct cpuidle_device *dev,
 			return 0;
 
 		if (pr->flags.bm_check) {
-			acpi_idle_enter_bm(pr, cx, false);
+			u8 bm_sts_skip = cx->bm_sts_skip;
+
+			/* Don't check BM_STS, do an unconditional ARB_DIS for S2IDLE */
+			cx->bm_sts_skip = 1;
+			acpi_idle_enter_bm(drv, pr, cx, index);
+			cx->bm_sts_skip = bm_sts_skip;
+
 			return 0;
 		} else {
 			ACPI_FLUSH_CPU_CACHE();
@@ -682,11 +684,13 @@ static int acpi_processor_setup_cpuidle_cx(struct acpi_processor *pr,
 {
 	int i, count = ACPI_IDLE_STATE_START;
 	struct acpi_processor_cx *cx;
+	struct cpuidle_state *state;
 
 	if (max_cstate == 0)
 		max_cstate = 1;
 
 	for (i = 1; i < ACPI_PROCESSOR_MAX_POWER && i <= max_cstate; i++) {
+		state = &acpi_idle_driver.states[count];
 		cx = &pr->power.states[i];
 
 		if (!cx->valid)
@@ -694,6 +698,15 @@ static int acpi_processor_setup_cpuidle_cx(struct acpi_processor *pr,
 
 		per_cpu(acpi_cstate[count], dev->cpu) = cx;
 
+		if (lapic_timer_needs_broadcast(pr, cx))
+			state->flags |= CPUIDLE_FLAG_TIMER_STOP;
+
+		if (cx->type == ACPI_STATE_C3) {
+			state->flags |= CPUIDLE_FLAG_TLB_FLUSHED;
+			if (pr->flags.bm_check)
+				state->flags |= CPUIDLE_FLAG_RCU_IDLE;
+		}
+
 		count++;
 		if (count == CPUIDLE_STATE_MAX)
 			break;