1 files changed, 356 insertions, 152 deletions

diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c
index 75fd2a7b0842..d55606608ac8 100644
--- a/drivers/idle/intel_idle.c
+++ b/drivers/idle/intel_idle.c
@@ -41,6 +41,7 @@
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
+#include <linux/acpi.h>
 #include <linux/kernel.h>
 #include <linux/cpuidle.h>
 #include <linux/tick.h>
@@ -62,6 +63,7 @@ static struct cpuidle_driver intel_idle_driver = {
 };
 /* intel_idle.max_cstate=0 disables driver */
 static int max_cstate = CPUIDLE_STATE_MAX - 1;
+static unsigned int disabled_states_mask;
 
 static unsigned int mwait_substates;
 
@@ -79,6 +81,7 @@ struct idle_cpu {
 	unsigned long auto_demotion_disable_flags;
 	bool byt_auto_demotion_disable_flag;
 	bool disable_promotion_to_c1e;
+	bool use_acpi;
 };
 
 static const struct idle_cpu *icpu;
@@ -90,6 +93,11 @@ static void intel_idle_s2idle(struct cpuidle_device *dev,
 static struct cpuidle_state *cpuidle_state_table;
 
 /*
+ * Enable this state by default even if the ACPI _CST does not list it.
+ */
+#define CPUIDLE_FLAG_ALWAYS_ENABLE	BIT(15)
+
+/*
  * Set this flag for states where the HW flushes the TLB for us
  * and so we don't need cross-calls to keep it consistent.
  * If this flag is set, SW flushes the TLB, so even if the
@@ -124,7 +132,7 @@ static struct cpuidle_state nehalem_cstates[] = {
 	{
 		.name = "C1E",
 		.desc = "MWAIT 0x01",
-		.flags = MWAIT2flg(0x01),
+		.flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE,
 		.exit_latency = 10,
 		.target_residency = 20,
 		.enter = &intel_idle,
@@ -161,7 +169,7 @@ static struct cpuidle_state snb_cstates[] = {
 	{
 		.name = "C1E",
 		.desc = "MWAIT 0x01",
-		.flags = MWAIT2flg(0x01),
+		.flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE,
 		.exit_latency = 10,
 		.target_residency = 20,
 		.enter = &intel_idle,
@@ -296,7 +304,7 @@ static struct cpuidle_state ivb_cstates[] = {
 	{
 		.name = "C1E",
 		.desc = "MWAIT 0x01",
-		.flags = MWAIT2flg(0x01),
+		.flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE,
 		.exit_latency = 10,
 		.target_residency = 20,
 		.enter = &intel_idle,
@@ -341,7 +349,7 @@ static struct cpuidle_state ivt_cstates[] = {
 	{
 		.name = "C1E",
 		.desc = "MWAIT 0x01",
-		.flags = MWAIT2flg(0x01),
+		.flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE,
 		.exit_latency = 10,
 		.target_residency = 80,
 		.enter = &intel_idle,
@@ -378,7 +386,7 @@ static struct cpuidle_state ivt_cstates_4s[] = {
 	{
 		.name = "C1E",
 		.desc = "MWAIT 0x01",
-		.flags = MWAIT2flg(0x01),
+		.flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE,
 		.exit_latency = 10,
 		.target_residency = 250,
 		.enter = &intel_idle,
@@ -415,7 +423,7 @@ static struct cpuidle_state ivt_cstates_8s[] = {
 	{
 		.name = "C1E",
 		.desc = "MWAIT 0x01",
-		.flags = MWAIT2flg(0x01),
+		.flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE,
 		.exit_latency = 10,
 		.target_residency = 500,
 		.enter = &intel_idle,
@@ -452,7 +460,7 @@ static struct cpuidle_state hsw_cstates[] = {
 	{
 		.name = "C1E",
 		.desc = "MWAIT 0x01",
-		.flags = MWAIT2flg(0x01),
+		.flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE,
 		.exit_latency = 10,
 		.target_residency = 20,
 		.enter = &intel_idle,
@@ -520,7 +528,7 @@ static struct cpuidle_state bdw_cstates[] = {
 	{
 		.name = "C1E",
 		.desc = "MWAIT 0x01",
-		.flags = MWAIT2flg(0x01),
+		.flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE,
 		.exit_latency = 10,
 		.target_residency = 20,
 		.enter = &intel_idle,
@@ -589,7 +597,7 @@ static struct cpuidle_state skl_cstates[] = {
 	{
 		.name = "C1E",
 		.desc = "MWAIT 0x01",
-		.flags = MWAIT2flg(0x01),
+		.flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE,
 		.exit_latency = 10,
 		.target_residency = 20,
 		.enter = &intel_idle,
@@ -658,7 +666,7 @@ static struct cpuidle_state skx_cstates[] = {
 	{
 		.name = "C1E",
 		.desc = "MWAIT 0x01",
-		.flags = MWAIT2flg(0x01),
+		.flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE,
 		.exit_latency = 10,
 		.target_residency = 20,
 		.enter = &intel_idle,
@@ -808,7 +816,7 @@ static struct cpuidle_state bxt_cstates[] = {
 	{
 		.name = "C1E",
 		.desc = "MWAIT 0x01",
-		.flags = MWAIT2flg(0x01),
+		.flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE,
 		.exit_latency = 10,
 		.target_residency = 20,
 		.enter = &intel_idle,
@@ -869,7 +877,7 @@ static struct cpuidle_state dnv_cstates[] = {
 	{
 		.name = "C1E",
 		.desc = "MWAIT 0x01",
-		.flags = MWAIT2flg(0x01),
+		.flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE,
 		.exit_latency = 10,
 		.target_residency = 20,
 		.enter = &intel_idle,
@@ -944,37 +952,19 @@ static void intel_idle_s2idle(struct cpuidle_device *dev,
 	mwait_idle_with_hints(eax, ecx);
 }
 
-static void __setup_broadcast_timer(bool on)
-{
-	if (on)
-		tick_broadcast_enable();
-	else
-		tick_broadcast_disable();
-}
-
-static void auto_demotion_disable(void)
-{
-	unsigned long long msr_bits;
-
-	rdmsrl(MSR_PKG_CST_CONFIG_CONTROL, msr_bits);
-	msr_bits &= ~(icpu->auto_demotion_disable_flags);
-	wrmsrl(MSR_PKG_CST_CONFIG_CONTROL, msr_bits);
-}
-static void c1e_promotion_disable(void)
-{
-	unsigned long long msr_bits;
-
-	rdmsrl(MSR_IA32_POWER_CTL, msr_bits);
-	msr_bits &= ~0x2;
-	wrmsrl(MSR_IA32_POWER_CTL, msr_bits);
-}
-
 static const struct idle_cpu idle_cpu_nehalem = {
 	.state_table = nehalem_cstates,
 	.auto_demotion_disable_flags = NHM_C1_AUTO_DEMOTE | NHM_C3_AUTO_DEMOTE,
 	.disable_promotion_to_c1e = true,
 };
 
+static const struct idle_cpu idle_cpu_nhx = {
+	.state_table = nehalem_cstates,
+	.auto_demotion_disable_flags = NHM_C1_AUTO_DEMOTE | NHM_C3_AUTO_DEMOTE,
+	.disable_promotion_to_c1e = true,
+	.use_acpi = true,
+};
+
 static const struct idle_cpu idle_cpu_atom = {
 	.state_table = atom_cstates,
 };
@@ -993,6 +983,12 @@ static const struct idle_cpu idle_cpu_snb = {
 	.disable_promotion_to_c1e = true,
 };
 
+static const struct idle_cpu idle_cpu_snx = {
+	.state_table = snb_cstates,
+	.disable_promotion_to_c1e = true,
+	.use_acpi = true,
+};
+
 static const struct idle_cpu idle_cpu_byt = {
 	.state_table = byt_cstates,
 	.disable_promotion_to_c1e = true,
@@ -1013,6 +1009,7 @@ static const struct idle_cpu idle_cpu_ivb = {
 static const struct idle_cpu idle_cpu_ivt = {
 	.state_table = ivt_cstates,
 	.disable_promotion_to_c1e = true,
+	.use_acpi = true,
 };
 
 static const struct idle_cpu idle_cpu_hsw = {
@@ -1020,11 +1017,23 @@ static const struct idle_cpu idle_cpu_hsw = {
 	.disable_promotion_to_c1e = true,
 };
 
+static const struct idle_cpu idle_cpu_hsx = {
+	.state_table = hsw_cstates,
+	.disable_promotion_to_c1e = true,
+	.use_acpi = true,
+};
+
 static const struct idle_cpu idle_cpu_bdw = {
 	.state_table = bdw_cstates,
 	.disable_promotion_to_c1e = true,
 };
 
+static const struct idle_cpu idle_cpu_bdx = {
+	.state_table = bdw_cstates,
+	.disable_promotion_to_c1e = true,
+	.use_acpi = true,
+};
+
 static const struct idle_cpu idle_cpu_skl = {
 	.state_table = skl_cstates,
 	.disable_promotion_to_c1e = true,
@@ -1033,15 +1042,18 @@ static const struct idle_cpu idle_cpu_skl = {
 static const struct idle_cpu idle_cpu_skx = {
 	.state_table = skx_cstates,
 	.disable_promotion_to_c1e = true,
+	.use_acpi = true,
 };
 
 static const struct idle_cpu idle_cpu_avn = {
 	.state_table = avn_cstates,
 	.disable_promotion_to_c1e = true,
+	.use_acpi = true,
 };
 
 static const struct idle_cpu idle_cpu_knl = {
 	.state_table = knl_cstates,
+	.use_acpi = true,
 };
 
 static const struct idle_cpu idle_cpu_bxt = {
@@ -1052,20 +1064,21 @@ static const struct idle_cpu idle_cpu_bxt = {
 static const struct idle_cpu idle_cpu_dnv = {
 	.state_table = dnv_cstates,
 	.disable_promotion_to_c1e = true,
+	.use_acpi = true,
 };
 
 static const struct x86_cpu_id intel_idle_ids[] __initconst = {
-	INTEL_CPU_FAM6(NEHALEM_EP,		idle_cpu_nehalem),
+	INTEL_CPU_FAM6(NEHALEM_EP,		idle_cpu_nhx),
 	INTEL_CPU_FAM6(NEHALEM,			idle_cpu_nehalem),
 	INTEL_CPU_FAM6(NEHALEM_G,		idle_cpu_nehalem),
 	INTEL_CPU_FAM6(WESTMERE,		idle_cpu_nehalem),
-	INTEL_CPU_FAM6(WESTMERE_EP,		idle_cpu_nehalem),
-	INTEL_CPU_FAM6(NEHALEM_EX,		idle_cpu_nehalem),
+	INTEL_CPU_FAM6(WESTMERE_EP,		idle_cpu_nhx),
+	INTEL_CPU_FAM6(NEHALEM_EX,		idle_cpu_nhx),
 	INTEL_CPU_FAM6(ATOM_BONNELL,		idle_cpu_atom),
 	INTEL_CPU_FAM6(ATOM_BONNELL_MID,	idle_cpu_lincroft),
-	INTEL_CPU_FAM6(WESTMERE_EX,		idle_cpu_nehalem),
+	INTEL_CPU_FAM6(WESTMERE_EX,		idle_cpu_nhx),
 	INTEL_CPU_FAM6(SANDYBRIDGE,		idle_cpu_snb),
-	INTEL_CPU_FAM6(SANDYBRIDGE_X,		idle_cpu_snb),
+	INTEL_CPU_FAM6(SANDYBRIDGE_X,		idle_cpu_snx),
 	INTEL_CPU_FAM6(ATOM_SALTWELL,		idle_cpu_atom),
 	INTEL_CPU_FAM6(ATOM_SILVERMONT,		idle_cpu_byt),
 	INTEL_CPU_FAM6(ATOM_SILVERMONT_MID,	idle_cpu_tangier),
@@ -1073,14 +1086,14 @@ static const struct x86_cpu_id intel_idle_ids[] __initconst = {
 	INTEL_CPU_FAM6(IVYBRIDGE,		idle_cpu_ivb),
 	INTEL_CPU_FAM6(IVYBRIDGE_X,		idle_cpu_ivt),
 	INTEL_CPU_FAM6(HASWELL,			idle_cpu_hsw),
-	INTEL_CPU_FAM6(HASWELL_X,		idle_cpu_hsw),
+	INTEL_CPU_FAM6(HASWELL_X,		idle_cpu_hsx),
 	INTEL_CPU_FAM6(HASWELL_L,		idle_cpu_hsw),
 	INTEL_CPU_FAM6(HASWELL_G,		idle_cpu_hsw),
 	INTEL_CPU_FAM6(ATOM_SILVERMONT_D,	idle_cpu_avn),
 	INTEL_CPU_FAM6(BROADWELL,		idle_cpu_bdw),
 	INTEL_CPU_FAM6(BROADWELL_G,		idle_cpu_bdw),
-	INTEL_CPU_FAM6(BROADWELL_X,		idle_cpu_bdw),
-	INTEL_CPU_FAM6(BROADWELL_D,		idle_cpu_bdw),
+	INTEL_CPU_FAM6(BROADWELL_X,		idle_cpu_bdx),
+	INTEL_CPU_FAM6(BROADWELL_D,		idle_cpu_bdx),
 	INTEL_CPU_FAM6(SKYLAKE_L,		idle_cpu_skl),
 	INTEL_CPU_FAM6(SKYLAKE,			idle_cpu_skl),
 	INTEL_CPU_FAM6(KABYLAKE_L,		idle_cpu_skl),
@@ -1095,76 +1108,178 @@ static const struct x86_cpu_id intel_idle_ids[] __initconst = {
 	{}
 };
 
-/*
- * intel_idle_probe()
+#define INTEL_CPU_FAM6_MWAIT \
+	{ X86_VENDOR_INTEL, 6, X86_MODEL_ANY, X86_FEATURE_MWAIT, 0 }
+
+static const struct x86_cpu_id intel_mwait_ids[] __initconst = {
+	INTEL_CPU_FAM6_MWAIT,
+	{}
+};
+
+static bool __init intel_idle_max_cstate_reached(int cstate)
+{
+	if (cstate + 1 > max_cstate) {
+		pr_info("max_cstate %d reached\n", max_cstate);
+		return true;
+	}
+	return false;
+}
+
+#ifdef CONFIG_ACPI_PROCESSOR_CSTATE
+#include <acpi/processor.h>
+
+static bool no_acpi __read_mostly;
+module_param(no_acpi, bool, 0444);
+MODULE_PARM_DESC(no_acpi, "Do not use ACPI _CST for building the idle states list");
+
+static bool force_use_acpi __read_mostly; /* No effect if no_acpi is set. */
+module_param_named(use_acpi, force_use_acpi, bool, 0444);
+MODULE_PARM_DESC(use_acpi, "Use ACPI _CST for building the idle states list");
+
+static struct acpi_processor_power acpi_state_table __initdata;
+
+/**
+ * intel_idle_cst_usable - Check if the _CST information can be used.
+ *
+ * Check if all of the C-states listed by _CST in the max_cstate range are
+ * ACPI_CSTATE_FFH, which means that they should be entered via MWAIT.
  */
-static int __init intel_idle_probe(void)
+static bool __init intel_idle_cst_usable(void)
 {
-	unsigned int eax, ebx, ecx;
-	const struct x86_cpu_id *id;
+	int cstate, limit;
 
-	if (max_cstate == 0) {
-		pr_debug("disabled\n");
-		return -EPERM;
-	}
+	limit = min_t(int, min_t(int, CPUIDLE_STATE_MAX, max_cstate + 1),
+		      acpi_state_table.count);
 
-	id = x86_match_cpu(intel_idle_ids);
-	if (!id) {
-		if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL &&
-		    boot_cpu_data.x86 == 6)
-			pr_debug("does not run on family %d model %d\n",
-				 boot_cpu_data.x86, boot_cpu_data.x86_model);
-		return -ENODEV;
+	for (cstate = 1; cstate < limit; cstate++) {
+		struct acpi_processor_cx *cx = &acpi_state_table.states[cstate];
+
+		if (cx->entry_method != ACPI_CSTATE_FFH)
+			return false;
 	}
 
-	if (!boot_cpu_has(X86_FEATURE_MWAIT)) {
-		pr_debug("Please enable MWAIT in BIOS SETUP\n");
-		return -ENODEV;
+	return true;
+}
+
+static bool __init intel_idle_acpi_cst_extract(void)
+{
+	unsigned int cpu;
+
+	if (no_acpi) {
+		pr_debug("Not allowed to use ACPI _CST\n");
+		return false;
 	}
 
-	if (boot_cpu_data.cpuid_level < CPUID_MWAIT_LEAF)
-		return -ENODEV;
+	for_each_possible_cpu(cpu) {
+		struct acpi_processor *pr = per_cpu(processors, cpu);
 
-	cpuid(CPUID_MWAIT_LEAF, &eax, &ebx, &ecx, &mwait_substates);
+		if (!pr)
+			continue;
 
-	if (!(ecx & CPUID5_ECX_EXTENSIONS_SUPPORTED) ||
-	    !(ecx & CPUID5_ECX_INTERRUPT_BREAK) ||
-	    !mwait_substates)
-			return -ENODEV;
+		if (acpi_processor_evaluate_cst(pr->handle, cpu, &acpi_state_table))
+			continue;
 
-	pr_debug("MWAIT substates: 0x%x\n", mwait_substates);
+		acpi_state_table.count++;
 
-	icpu = (const struct idle_cpu *)id->driver_data;
-	cpuidle_state_table = icpu->state_table;
+		if (!intel_idle_cst_usable())
+			continue;
 
-	pr_debug("v" INTEL_IDLE_VERSION " model 0x%X\n",
-		 boot_cpu_data.x86_model);
+		if (!acpi_processor_claim_cst_control()) {
+			acpi_state_table.count = 0;
+			return false;
+		}
 
-	return 0;
+		return true;
+	}
+
+	pr_debug("ACPI _CST not found or not usable\n");
+	return false;
 }
 
-/*
- * intel_idle_cpuidle_devices_uninit()
- * Unregisters the cpuidle devices.
- */
-static void intel_idle_cpuidle_devices_uninit(void)
+static void __init intel_idle_init_cstates_acpi(struct cpuidle_driver *drv)
 {
-	int i;
-	struct cpuidle_device *dev;
+	int cstate, limit = min_t(int, CPUIDLE_STATE_MAX, acpi_state_table.count);
+
+	/*
+	 * If limit > 0, intel_idle_cst_usable() has returned 'true', so all of
+	 * the interesting states are ACPI_CSTATE_FFH.
+	 */
+	for (cstate = 1; cstate < limit; cstate++) {
+		struct acpi_processor_cx *cx;
+		struct cpuidle_state *state;
+
+		if (intel_idle_max_cstate_reached(cstate))
+			break;
 
-	for_each_online_cpu(i) {
-		dev = per_cpu_ptr(intel_idle_cpuidle_devices, i);
-		cpuidle_unregister_device(dev);
+		cx = &acpi_state_table.states[cstate];
+
+		state = &drv->states[drv->state_count++];
+
+		snprintf(state->name, CPUIDLE_NAME_LEN, "C%d_ACPI", cstate);
+		strlcpy(state->desc, cx->desc, CPUIDLE_DESC_LEN);
+		state->exit_latency = cx->latency;
+		/*
+		 * For C1-type C-states use the same number for both the exit
+		 * latency and target residency, because that is the case for
+		 * C1 in the majority of the static C-states tables above.
+		 * For the other types of C-states, however, set the target
+		 * residency to 3 times the exit latency which should lead to
+		 * a reasonable balance between energy-efficiency and
+		 * performance in the majority of interesting cases.
+		 */
+		state->target_residency = cx->latency;
+		if (cx->type > ACPI_STATE_C1)
+			state->target_residency *= 3;
+
+		state->flags = MWAIT2flg(cx->address);
+		if (cx->type > ACPI_STATE_C2)
+			state->flags |= CPUIDLE_FLAG_TLB_FLUSHED;
+
+		if (disabled_states_mask & BIT(cstate))
+			state->flags |= CPUIDLE_FLAG_OFF;
+
+		state->enter = intel_idle;
+		state->enter_s2idle = intel_idle_s2idle;
 	}
 }
 
+static bool __init intel_idle_off_by_default(u32 mwait_hint)
+{
+	int cstate, limit;
+
+	/*
+	 * If there are no _CST C-states, do not disable any C-states by
+	 * default.
+	 */
+	if (!acpi_state_table.count)
+		return false;
+
+	limit = min_t(int, CPUIDLE_STATE_MAX, acpi_state_table.count);
+	/*
+	 * If limit > 0, intel_idle_cst_usable() has returned 'true', so all of
+	 * the interesting states are ACPI_CSTATE_FFH.
+	 */
+	for (cstate = 1; cstate < limit; cstate++) {
+		if (acpi_state_table.states[cstate].address == mwait_hint)
+			return false;
+	}
+	return true;
+}
+#else /* !CONFIG_ACPI_PROCESSOR_CSTATE */
+#define force_use_acpi	(false)
+
+static inline bool intel_idle_acpi_cst_extract(void) { return false; }
+static inline void intel_idle_init_cstates_acpi(struct cpuidle_driver *drv) { }
+static inline bool intel_idle_off_by_default(u32 mwait_hint) { return false; }
+#endif /* !CONFIG_ACPI_PROCESSOR_CSTATE */
+
 /*
  * ivt_idle_state_table_update(void)
  *
  * Tune IVT multi-socket targets
  * Assumption: num_sockets == (max_package_num + 1)
  */
-static void ivt_idle_state_table_update(void)
+static void __init ivt_idle_state_table_update(void)
 {
 	/* IVT uses a different table for 1-2, 3-4, and > 4 sockets */
 	int cpu, package_num, num_sockets = 1;
@@ -1187,15 +1302,17 @@ static void ivt_idle_state_table_update(void)
 	/* else, 1 and 2 socket systems use default ivt_cstates */
 }
 
-/*
- * Translate IRTL (Interrupt Response Time Limit) MSR to usec
+/**
+ * irtl_2_usec - IRTL to microseconds conversion.
+ * @irtl: IRTL MSR value.
+ *
+ * Translate the IRTL (Interrupt Response Time Limit) MSR value to microseconds.
  */
-
-static unsigned int irtl_ns_units[] = {
-	1, 32, 1024, 32768, 1048576, 33554432, 0, 0 };
-
-static unsigned long long irtl_2_usec(unsigned long long irtl)
+static unsigned long long __init irtl_2_usec(unsigned long long irtl)
 {
+	static const unsigned int irtl_ns_units[] __initconst = {
+		1, 32, 1024, 32768, 1048576, 33554432, 0, 0
+	};
 	unsigned long long ns;
 
 	if (!irtl)
@@ -1203,15 +1320,16 @@ static unsigned long long irtl_2_usec(unsigned long long irtl)
 
 	ns = irtl_ns_units[(irtl >> 10) & 0x7];
 
-	return div64_u64((irtl & 0x3FF) * ns, 1000);
+	return div_u64((irtl & 0x3FF) * ns, NSEC_PER_USEC);
 }
+
 /*
  * bxt_idle_state_table_update(void)
  *
  * On BXT, we trust the IRTL to show the definitive maximum latency
  * We use the same value for target_residency.
  */
-static void bxt_idle_state_table_update(void)
+static void __init bxt_idle_state_table_update(void)
 {
 	unsigned long long msr;
 	unsigned int usec;
@@ -1258,7 +1376,7 @@ static void bxt_idle_state_table_update(void)
  * On SKL-H (model 0x5e) disable C8 and C9 if:
  * C10 is enabled and SGX disabled
  */
-static void sklh_idle_state_table_update(void)
+static void __init sklh_idle_state_table_update(void)
 {
 	unsigned long long msr;
 	unsigned int eax, ebx, ecx, edx;
@@ -1284,7 +1402,7 @@ static void sklh_idle_state_table_update(void)
 	/* if SGX is present */
 	if (ebx & (1 << 2)) {
 
-		rdmsrl(MSR_IA32_FEATURE_CONTROL, msr);
+		rdmsrl(MSR_IA32_FEAT_CTL, msr);
 
 		/* if SGX is enabled */
 		if (msr & (1 << 18))
@@ -1294,16 +1412,28 @@ static void sklh_idle_state_table_update(void)
 	skl_cstates[5].flags |= CPUIDLE_FLAG_UNUSABLE;	/* C8-SKL */
 	skl_cstates[6].flags |= CPUIDLE_FLAG_UNUSABLE;	/* C9-SKL */
 }
-/*
- * intel_idle_state_table_update()
- *
- * Update the default state_table for this CPU-id
- */
 
-static void intel_idle_state_table_update(void)
+static bool __init intel_idle_verify_cstate(unsigned int mwait_hint)
 {
-	switch (boot_cpu_data.x86_model) {
+	unsigned int mwait_cstate = MWAIT_HINT2CSTATE(mwait_hint) + 1;
+	unsigned int num_substates = (mwait_substates >> mwait_cstate * 4) &
+					MWAIT_SUBSTATE_MASK;
+
+	/* Ignore the C-state if there are NO sub-states in CPUID for it. */
+	if (num_substates == 0)
+		return false;
+
+	if (mwait_cstate > 2 && !boot_cpu_has(X86_FEATURE_NONSTOP_TSC))
+		mark_tsc_unstable("TSC halts in idle states deeper than C2");
+
+	return true;
+}
+
+static void __init intel_idle_init_cstates_icpu(struct cpuidle_driver *drv)
+{
+	int cstate;
 
+	switch (boot_cpu_data.x86_model) {
 	case INTEL_FAM6_IVYBRIDGE_X:
 		ivt_idle_state_table_update();
 		break;
@@ -1315,62 +1445,38 @@ static void intel_idle_state_table_update(void)
 		sklh_idle_state_table_update();
 		break;
 	}
-}
-
-/*
- * intel_idle_cpuidle_driver_init()
- * allocate, initialize cpuidle_states
- */
-static void __init intel_idle_cpuidle_driver_init(void)
-{
-	int cstate;
-	struct cpuidle_driver *drv = &intel_idle_driver;
-
-	intel_idle_state_table_update();
-
-	cpuidle_poll_state_init(drv);
-	drv->state_count = 1;
 
 	for (cstate = 0; cstate < CPUIDLE_STATE_MAX; ++cstate) {
-		int num_substates, mwait_hint, mwait_cstate;
+		unsigned int mwait_hint;
 
-		if ((cpuidle_state_table[cstate].enter == NULL) &&
-		    (cpuidle_state_table[cstate].enter_s2idle == NULL))
+		if (intel_idle_max_cstate_reached(cstate))
 			break;
 
-		if (cstate + 1 > max_cstate) {
-			pr_info("max_cstate %d reached\n", max_cstate);
+		if (!cpuidle_state_table[cstate].enter &&
+		    !cpuidle_state_table[cstate].enter_s2idle)
 			break;
-		}
-
-		mwait_hint = flg2MWAIT(cpuidle_state_table[cstate].flags);
-		mwait_cstate = MWAIT_HINT2CSTATE(mwait_hint);
-
-		/* number of sub-states for this state in CPUID.MWAIT */
-		num_substates = (mwait_substates >> ((mwait_cstate + 1) * 4))
-					& MWAIT_SUBSTATE_MASK;
 
-		/* if NO sub-states for this state in CPUID, skip it */
-		if (num_substates == 0)
-			continue;
-
-		/* if state marked as disabled, skip it */
+		/* If marked as unusable, skip this state. */
 		if (cpuidle_state_table[cstate].flags & CPUIDLE_FLAG_UNUSABLE) {
 			pr_debug("state %s is disabled\n",
 				 cpuidle_state_table[cstate].name);
 			continue;
 		}
 
+		mwait_hint = flg2MWAIT(cpuidle_state_table[cstate].flags);
+		if (!intel_idle_verify_cstate(mwait_hint))
+			continue;
 
-		if (((mwait_cstate + 1) > 2) &&
-			!boot_cpu_has(X86_FEATURE_NONSTOP_TSC))
-			mark_tsc_unstable("TSC halts in idle"
-					" states deeper than C2");
+		/* Structure copy. */
+		drv->states[drv->state_count] = cpuidle_state_table[cstate];
 
-		drv->states[drv->state_count] =	/* structure copy */
-			cpuidle_state_table[cstate];
+		if ((disabled_states_mask & BIT(drv->state_count)) ||
+		    ((icpu->use_acpi || force_use_acpi) &&
+		     intel_idle_off_by_default(mwait_hint) &&
+		     !(cpuidle_state_table[cstate].flags & CPUIDLE_FLAG_ALWAYS_ENABLE)))
+			drv->states[drv->state_count].flags |= CPUIDLE_FLAG_OFF;
 
-		drv->state_count += 1;
+		drv->state_count++;
 	}
 
 	if (icpu->byt_auto_demotion_disable_flag) {
@@ -1379,6 +1485,42 @@ static void __init intel_idle_cpuidle_driver_init(void)
 	}
 }
 
+/*
+ * intel_idle_cpuidle_driver_init()
+ * allocate, initialize cpuidle_states
+ */
+static void __init intel_idle_cpuidle_driver_init(struct cpuidle_driver *drv)
+{
+	cpuidle_poll_state_init(drv);
+
+	if (disabled_states_mask & BIT(0))
+		drv->states[0].flags |= CPUIDLE_FLAG_OFF;
+
+	drv->state_count = 1;
+
+	if (icpu)
+		intel_idle_init_cstates_icpu(drv);
+	else
+		intel_idle_init_cstates_acpi(drv);
+}
+
+static void auto_demotion_disable(void)
+{
+	unsigned long long msr_bits;
+
+	rdmsrl(MSR_PKG_CST_CONFIG_CONTROL, msr_bits);
+	msr_bits &= ~(icpu->auto_demotion_disable_flags);
+	wrmsrl(MSR_PKG_CST_CONFIG_CONTROL, msr_bits);
+}
+
+static void c1e_promotion_disable(void)
+{
+	unsigned long long msr_bits;
+
+	rdmsrl(MSR_IA32_POWER_CTL, msr_bits);
+	msr_bits &= ~0x2;
+	wrmsrl(MSR_IA32_POWER_CTL, msr_bits);
+}
 
 /*
  * intel_idle_cpu_init()
@@ -1397,6 +1539,9 @@ static int intel_idle_cpu_init(unsigned int cpu)
 		return -EIO;
 	}
 
+	if (!icpu)
+		return 0;
+
 	if (icpu->auto_demotion_disable_flags)
 		auto_demotion_disable();
 
@@ -1411,7 +1556,7 @@ static int intel_idle_cpu_online(unsigned int cpu)
 	struct cpuidle_device *dev;
 
 	if (lapic_timer_reliable_states != LAPIC_TIMER_ALWAYS_RELIABLE)
-		__setup_broadcast_timer(true);
+		tick_broadcast_enable();
 
 	/*
 	 * Some systems can hotplug a cpu at runtime after
@@ -1425,23 +1570,74 @@ static int intel_idle_cpu_online(unsigned int cpu)
 	return 0;
 }
 
+/**
+ * intel_idle_cpuidle_devices_uninit - Unregister all cpuidle devices.
+ */
+static void __init intel_idle_cpuidle_devices_uninit(void)
+{
+	int i;
+
+	for_each_online_cpu(i)
+		cpuidle_unregister_device(per_cpu_ptr(intel_idle_cpuidle_devices, i));
+}
+
 static int __init intel_idle_init(void)
 {
+	const struct x86_cpu_id *id;
+	unsigned int eax, ebx, ecx;
 	int retval;
 
 	/* Do not load intel_idle at all for now if idle= is passed */
 	if (boot_option_idle_override != IDLE_NO_OVERRIDE)
 		return -ENODEV;
 
-	retval = intel_idle_probe();
-	if (retval)
-		return retval;
+	if (max_cstate == 0) {
+		pr_debug("disabled\n");
+		return -EPERM;
+	}
+
+	id = x86_match_cpu(intel_idle_ids);
+	if (id) {
+		if (!boot_cpu_has(X86_FEATURE_MWAIT)) {
+			pr_debug("Please enable MWAIT in BIOS SETUP\n");
+			return -ENODEV;
+		}
+	} else {
+		id = x86_match_cpu(intel_mwait_ids);
+		if (!id)
+			return -ENODEV;
+	}
+
+	if (boot_cpu_data.cpuid_level < CPUID_MWAIT_LEAF)
+		return -ENODEV;
+
+	cpuid(CPUID_MWAIT_LEAF, &eax, &ebx, &ecx, &mwait_substates);
+
+	if (!(ecx & CPUID5_ECX_EXTENSIONS_SUPPORTED) ||
+	    !(ecx & CPUID5_ECX_INTERRUPT_BREAK) ||
+	    !mwait_substates)
+			return -ENODEV;
+
+	pr_debug("MWAIT substates: 0x%x\n", mwait_substates);
+
+	icpu = (const struct idle_cpu *)id->driver_data;
+	if (icpu) {
+		cpuidle_state_table = icpu->state_table;
+		if (icpu->use_acpi || force_use_acpi)
+			intel_idle_acpi_cst_extract();
+	} else if (!intel_idle_acpi_cst_extract()) {
+		return -ENODEV;
+	}
+
+	pr_debug("v" INTEL_IDLE_VERSION " model 0x%X\n",
+		 boot_cpu_data.x86_model);
 
 	intel_idle_cpuidle_devices = alloc_percpu(struct cpuidle_device);
-	if (intel_idle_cpuidle_devices == NULL)
+	if (!intel_idle_cpuidle_devices)
 		return -ENOMEM;
 
-	intel_idle_cpuidle_driver_init();
+	intel_idle_cpuidle_driver_init(&intel_idle_driver);
+
 	retval = cpuidle_register_driver(&intel_idle_driver);
 	if (retval) {
 		struct cpuidle_driver *drv = cpuidle_get_driver();
@@ -1480,3 +1676,11 @@ device_initcall(intel_idle_init);
  * is the easiest way (currently) to continue doing that.
  */
 module_param(max_cstate, int, 0444);
+/*
+ * The positions of the bits that are set in this number are the indices of the
+ * idle states to be disabled by default (as reflected by the names of the
+ * corresponding idle state directories in sysfs, "state0", "state1" ...
+ * "state<i>" ..., where <i> is the index of the given state).
+ */
+module_param_named(states_off, disabled_states_mask, uint, 0444);
+MODULE_PARM_DESC(states_off, "Mask of disabled idle states");