1 files changed, 359 insertions, 94 deletions

diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c
index bcf1198e8991..8ccb483204fa 100644
--- a/drivers/idle/intel_idle.c
+++ b/drivers/idle/intel_idle.c
@@ -48,14 +48,20 @@
 #include <trace/events/power.h>
 #include <linux/sched.h>
 #include <linux/sched/smt.h>
+#include <linux/mutex.h>
 #include <linux/notifier.h>
 #include <linux/cpu.h>
 #include <linux/moduleparam.h>
+#include <linux/sysfs.h>
+#include <asm/cpuid/api.h>
 #include <asm/cpu_device_id.h>
 #include <asm/intel-family.h>
 #include <asm/mwait.h>
 #include <asm/spec-ctrl.h>
+#include <asm/msr.h>
+#include <asm/tsc.h>
 #include <asm/fpu/api.h>
+#include <asm/smp.h>
 
 #define INTEL_IDLE_VERSION "0.5.1"
 
@@ -88,11 +94,16 @@ struct idle_cpu {
 	 * Indicate which enable bits to clear here.
 	 */
 	unsigned long auto_demotion_disable_flags;
-	bool byt_auto_demotion_disable_flag;
 	bool disable_promotion_to_c1e;
+	bool c1_demotion_supported;
 	bool use_acpi;
 };
 
+static bool c1_demotion_supported;
+static DEFINE_MUTEX(c1_demotion_mutex);
+
+static struct device *sysfs_root __initdata;
+
 static const struct idle_cpu *icpu __initdata;
 static struct cpuidle_state *cpuidle_state_table __initdata;
 
@@ -121,6 +132,12 @@ static unsigned int mwait_substates __initdata;
 #define CPUIDLE_FLAG_INIT_XSTATE	BIT(17)
 
 /*
+ * Ignore the sub-state when matching mwait hints between the ACPI _CST and
+ * custom tables.
+ */
+#define CPUIDLE_FLAG_PARTIAL_HINT_MATCH	BIT(18)
+
+/*
  * MWAIT takes an 8-bit "hint" in EAX "suggesting"
  * the C-state (top nibble) and sub-state (bottom nibble)
  * 0x00 means "MWAIT(C1)", 0x10 means "MWAIT(C2)" etc.
@@ -221,6 +238,15 @@ static __cpuidle int intel_idle_s2idle(struct cpuidle_device *dev,
 	return 0;
 }
 
+static void intel_idle_enter_dead(struct cpuidle_device *dev, int index)
+{
+	struct cpuidle_driver *drv = cpuidle_get_cpu_driver(dev);
+	struct cpuidle_state *state = &drv->states[index];
+	unsigned long eax = flg2MWAIT(state->flags);
+
+	mwait_play_dead(eax);
+}
+
 /*
  * States are indexed by the cstate number,
  * which is also the index into the MWAIT hint array.
@@ -1022,6 +1048,88 @@ static struct cpuidle_state spr_cstates[] __initdata = {
 		.enter = NULL }
 };
 
+static struct cpuidle_state gnr_cstates[] __initdata = {
+	{
+		.name = "C1",
+		.desc = "MWAIT 0x00",
+		.flags = MWAIT2flg(0x00),
+		.exit_latency = 1,
+		.target_residency = 1,
+		.enter = &intel_idle,
+		.enter_s2idle = intel_idle_s2idle, },
+	{
+		.name = "C1E",
+		.desc = "MWAIT 0x01",
+		.flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE,
+		.exit_latency = 4,
+		.target_residency = 4,
+		.enter = &intel_idle,
+		.enter_s2idle = intel_idle_s2idle, },
+	{
+		.name = "C6",
+		.desc = "MWAIT 0x20",
+		.flags = MWAIT2flg(0x20) | CPUIDLE_FLAG_TLB_FLUSHED |
+					   CPUIDLE_FLAG_INIT_XSTATE |
+					   CPUIDLE_FLAG_PARTIAL_HINT_MATCH,
+		.exit_latency = 170,
+		.target_residency = 650,
+		.enter = &intel_idle,
+		.enter_s2idle = intel_idle_s2idle, },
+	{
+		.name = "C6P",
+		.desc = "MWAIT 0x21",
+		.flags = MWAIT2flg(0x21) | CPUIDLE_FLAG_TLB_FLUSHED |
+					   CPUIDLE_FLAG_INIT_XSTATE |
+					   CPUIDLE_FLAG_PARTIAL_HINT_MATCH,
+		.exit_latency = 210,
+		.target_residency = 1000,
+		.enter = &intel_idle,
+		.enter_s2idle = intel_idle_s2idle, },
+	{
+		.enter = NULL }
+};
+
+static struct cpuidle_state gnrd_cstates[] __initdata = {
+	{
+		.name = "C1",
+		.desc = "MWAIT 0x00",
+		.flags = MWAIT2flg(0x00),
+		.exit_latency = 1,
+		.target_residency = 1,
+		.enter = &intel_idle,
+		.enter_s2idle = intel_idle_s2idle, },
+	{
+		.name = "C1E",
+		.desc = "MWAIT 0x01",
+		.flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE,
+		.exit_latency = 4,
+		.target_residency = 4,
+		.enter = &intel_idle,
+		.enter_s2idle = intel_idle_s2idle, },
+	{
+		.name = "C6",
+		.desc = "MWAIT 0x20",
+		.flags = MWAIT2flg(0x20) | CPUIDLE_FLAG_TLB_FLUSHED |
+					   CPUIDLE_FLAG_INIT_XSTATE |
+					   CPUIDLE_FLAG_PARTIAL_HINT_MATCH,
+		.exit_latency = 220,
+		.target_residency = 650,
+		.enter = &intel_idle,
+		.enter_s2idle = intel_idle_s2idle, },
+	{
+		.name = "C6P",
+		.desc = "MWAIT 0x21",
+		.flags = MWAIT2flg(0x21) | CPUIDLE_FLAG_TLB_FLUSHED |
+					   CPUIDLE_FLAG_INIT_XSTATE |
+					   CPUIDLE_FLAG_PARTIAL_HINT_MATCH,
+		.exit_latency = 240,
+		.target_residency = 750,
+		.enter = &intel_idle,
+		.enter_s2idle = intel_idle_s2idle, },
+	{
+		.enter = NULL }
+};
+
 static struct cpuidle_state atom_cstates[] __initdata = {
 	{
 		.name = "C1E",
@@ -1315,7 +1423,8 @@ static struct cpuidle_state srf_cstates[] __initdata = {
 	{
 		.name = "C6S",
 		.desc = "MWAIT 0x22",
-		.flags = MWAIT2flg(0x22) | CPUIDLE_FLAG_TLB_FLUSHED,
+		.flags = MWAIT2flg(0x22) | CPUIDLE_FLAG_TLB_FLUSHED |
+					   CPUIDLE_FLAG_PARTIAL_HINT_MATCH,
 		.exit_latency = 270,
 		.target_residency = 700,
 		.enter = &intel_idle,
@@ -1323,7 +1432,8 @@ static struct cpuidle_state srf_cstates[] __initdata = {
 	{
 		.name = "C6SP",
 		.desc = "MWAIT 0x23",
-		.flags = MWAIT2flg(0x23) | CPUIDLE_FLAG_TLB_FLUSHED,
+		.flags = MWAIT2flg(0x23) | CPUIDLE_FLAG_TLB_FLUSHED |
+					   CPUIDLE_FLAG_PARTIAL_HINT_MATCH,
 		.exit_latency = 310,
 		.target_residency = 900,
 		.enter = &intel_idle,
@@ -1372,13 +1482,11 @@ static const struct idle_cpu idle_cpu_snx __initconst = {
 static const struct idle_cpu idle_cpu_byt __initconst = {
 	.state_table = byt_cstates,
 	.disable_promotion_to_c1e = true,
-	.byt_auto_demotion_disable_flag = true,
 };
 
 static const struct idle_cpu idle_cpu_cht __initconst = {
 	.state_table = cht_cstates,
 	.disable_promotion_to_c1e = true,
-	.byt_auto_demotion_disable_flag = true,
 };
 
 static const struct idle_cpu idle_cpu_ivb __initconst = {
@@ -1450,6 +1558,21 @@ static const struct idle_cpu idle_cpu_gmt __initconst = {
 static const struct idle_cpu idle_cpu_spr __initconst = {
 	.state_table = spr_cstates,
 	.disable_promotion_to_c1e = true,
+	.c1_demotion_supported = true,
+	.use_acpi = true,
+};
+
+static const struct idle_cpu idle_cpu_gnr __initconst = {
+	.state_table = gnr_cstates,
+	.disable_promotion_to_c1e = true,
+	.c1_demotion_supported = true,
+	.use_acpi = true,
+};
+
+static const struct idle_cpu idle_cpu_gnrd __initconst = {
+	.state_table = gnrd_cstates,
+	.disable_promotion_to_c1e = true,
+	.c1_demotion_supported = true,
 	.use_acpi = true,
 };
 
@@ -1475,6 +1598,10 @@ static const struct idle_cpu idle_cpu_dnv __initconst = {
 	.use_acpi = true,
 };
 
+static const struct idle_cpu idle_cpu_tmt __initconst = {
+	.disable_promotion_to_c1e = true,
+};
+
 static const struct idle_cpu idle_cpu_snr __initconst = {
 	.state_table = snr_cstates,
 	.disable_promotion_to_c1e = true,
@@ -1484,63 +1611,70 @@ static const struct idle_cpu idle_cpu_snr __initconst = {
 static const struct idle_cpu idle_cpu_grr __initconst = {
 	.state_table = grr_cstates,
 	.disable_promotion_to_c1e = true,
+	.c1_demotion_supported = true,
 	.use_acpi = true,
 };
 
 static const struct idle_cpu idle_cpu_srf __initconst = {
 	.state_table = srf_cstates,
 	.disable_promotion_to_c1e = true,
+	.c1_demotion_supported = true,
 	.use_acpi = true,
 };
 
 static const struct x86_cpu_id intel_idle_ids[] __initconst = {
-	X86_MATCH_INTEL_FAM6_MODEL(NEHALEM_EP,		&idle_cpu_nhx),
-	X86_MATCH_INTEL_FAM6_MODEL(NEHALEM,		&idle_cpu_nehalem),
-	X86_MATCH_INTEL_FAM6_MODEL(NEHALEM_G,		&idle_cpu_nehalem),
-	X86_MATCH_INTEL_FAM6_MODEL(WESTMERE,		&idle_cpu_nehalem),
-	X86_MATCH_INTEL_FAM6_MODEL(WESTMERE_EP,		&idle_cpu_nhx),
-	X86_MATCH_INTEL_FAM6_MODEL(NEHALEM_EX,		&idle_cpu_nhx),
-	X86_MATCH_INTEL_FAM6_MODEL(ATOM_BONNELL,	&idle_cpu_atom),
-	X86_MATCH_INTEL_FAM6_MODEL(ATOM_BONNELL_MID,	&idle_cpu_lincroft),
-	X86_MATCH_INTEL_FAM6_MODEL(WESTMERE_EX,		&idle_cpu_nhx),
-	X86_MATCH_INTEL_FAM6_MODEL(SANDYBRIDGE,		&idle_cpu_snb),
-	X86_MATCH_INTEL_FAM6_MODEL(SANDYBRIDGE_X,	&idle_cpu_snx),
-	X86_MATCH_INTEL_FAM6_MODEL(ATOM_SALTWELL,	&idle_cpu_atom),
-	X86_MATCH_INTEL_FAM6_MODEL(ATOM_SILVERMONT,	&idle_cpu_byt),
-	X86_MATCH_INTEL_FAM6_MODEL(ATOM_SILVERMONT_MID,	&idle_cpu_tangier),
-	X86_MATCH_INTEL_FAM6_MODEL(ATOM_AIRMONT,	&idle_cpu_cht),
-	X86_MATCH_INTEL_FAM6_MODEL(IVYBRIDGE,		&idle_cpu_ivb),
-	X86_MATCH_INTEL_FAM6_MODEL(IVYBRIDGE_X,		&idle_cpu_ivt),
-	X86_MATCH_INTEL_FAM6_MODEL(HASWELL,		&idle_cpu_hsw),
-	X86_MATCH_INTEL_FAM6_MODEL(HASWELL_X,		&idle_cpu_hsx),
-	X86_MATCH_INTEL_FAM6_MODEL(HASWELL_L,		&idle_cpu_hsw),
-	X86_MATCH_INTEL_FAM6_MODEL(HASWELL_G,		&idle_cpu_hsw),
-	X86_MATCH_INTEL_FAM6_MODEL(ATOM_SILVERMONT_D,	&idle_cpu_avn),
-	X86_MATCH_INTEL_FAM6_MODEL(BROADWELL,		&idle_cpu_bdw),
-	X86_MATCH_INTEL_FAM6_MODEL(BROADWELL_G,		&idle_cpu_bdw),
-	X86_MATCH_INTEL_FAM6_MODEL(BROADWELL_X,		&idle_cpu_bdx),
-	X86_MATCH_INTEL_FAM6_MODEL(BROADWELL_D,		&idle_cpu_bdx),
-	X86_MATCH_INTEL_FAM6_MODEL(SKYLAKE_L,		&idle_cpu_skl),
-	X86_MATCH_INTEL_FAM6_MODEL(SKYLAKE,		&idle_cpu_skl),
-	X86_MATCH_INTEL_FAM6_MODEL(KABYLAKE_L,		&idle_cpu_skl),
-	X86_MATCH_INTEL_FAM6_MODEL(KABYLAKE,		&idle_cpu_skl),
-	X86_MATCH_INTEL_FAM6_MODEL(SKYLAKE_X,		&idle_cpu_skx),
-	X86_MATCH_INTEL_FAM6_MODEL(ICELAKE_X,		&idle_cpu_icx),
-	X86_MATCH_INTEL_FAM6_MODEL(ICELAKE_D,		&idle_cpu_icx),
-	X86_MATCH_INTEL_FAM6_MODEL(ALDERLAKE,		&idle_cpu_adl),
-	X86_MATCH_INTEL_FAM6_MODEL(ALDERLAKE_L,		&idle_cpu_adl_l),
-	X86_MATCH_INTEL_FAM6_MODEL(METEORLAKE_L,	&idle_cpu_mtl_l),
-	X86_MATCH_INTEL_FAM6_MODEL(ATOM_GRACEMONT,	&idle_cpu_gmt),
-	X86_MATCH_INTEL_FAM6_MODEL(SAPPHIRERAPIDS_X,	&idle_cpu_spr),
-	X86_MATCH_INTEL_FAM6_MODEL(EMERALDRAPIDS_X,	&idle_cpu_spr),
-	X86_MATCH_INTEL_FAM6_MODEL(XEON_PHI_KNL,	&idle_cpu_knl),
-	X86_MATCH_INTEL_FAM6_MODEL(XEON_PHI_KNM,	&idle_cpu_knl),
-	X86_MATCH_INTEL_FAM6_MODEL(ATOM_GOLDMONT,	&idle_cpu_bxt),
-	X86_MATCH_INTEL_FAM6_MODEL(ATOM_GOLDMONT_PLUS,	&idle_cpu_bxt),
-	X86_MATCH_INTEL_FAM6_MODEL(ATOM_GOLDMONT_D,	&idle_cpu_dnv),
-	X86_MATCH_INTEL_FAM6_MODEL(ATOM_TREMONT_D,	&idle_cpu_snr),
-	X86_MATCH_INTEL_FAM6_MODEL(ATOM_CRESTMONT,	&idle_cpu_grr),
-	X86_MATCH_INTEL_FAM6_MODEL(ATOM_CRESTMONT_X,	&idle_cpu_srf),
+	X86_MATCH_VFM(INTEL_NEHALEM_EP,		&idle_cpu_nhx),
+	X86_MATCH_VFM(INTEL_NEHALEM,		&idle_cpu_nehalem),
+	X86_MATCH_VFM(INTEL_NEHALEM_G,		&idle_cpu_nehalem),
+	X86_MATCH_VFM(INTEL_WESTMERE,		&idle_cpu_nehalem),
+	X86_MATCH_VFM(INTEL_WESTMERE_EP,	&idle_cpu_nhx),
+	X86_MATCH_VFM(INTEL_NEHALEM_EX,		&idle_cpu_nhx),
+	X86_MATCH_VFM(INTEL_ATOM_BONNELL,	&idle_cpu_atom),
+	X86_MATCH_VFM(INTEL_ATOM_BONNELL_MID,	&idle_cpu_lincroft),
+	X86_MATCH_VFM(INTEL_WESTMERE_EX,	&idle_cpu_nhx),
+	X86_MATCH_VFM(INTEL_SANDYBRIDGE,	&idle_cpu_snb),
+	X86_MATCH_VFM(INTEL_SANDYBRIDGE_X,	&idle_cpu_snx),
+	X86_MATCH_VFM(INTEL_ATOM_SALTWELL,	&idle_cpu_atom),
+	X86_MATCH_VFM(INTEL_ATOM_SILVERMONT,	&idle_cpu_byt),
+	X86_MATCH_VFM(INTEL_ATOM_SILVERMONT_MID, &idle_cpu_tangier),
+	X86_MATCH_VFM(INTEL_ATOM_AIRMONT,	&idle_cpu_cht),
+	X86_MATCH_VFM(INTEL_IVYBRIDGE,		&idle_cpu_ivb),
+	X86_MATCH_VFM(INTEL_IVYBRIDGE_X,	&idle_cpu_ivt),
+	X86_MATCH_VFM(INTEL_HASWELL,		&idle_cpu_hsw),
+	X86_MATCH_VFM(INTEL_HASWELL_X,		&idle_cpu_hsx),
+	X86_MATCH_VFM(INTEL_HASWELL_L,		&idle_cpu_hsw),
+	X86_MATCH_VFM(INTEL_HASWELL_G,		&idle_cpu_hsw),
+	X86_MATCH_VFM(INTEL_ATOM_SILVERMONT_D,	&idle_cpu_avn),
+	X86_MATCH_VFM(INTEL_BROADWELL,		&idle_cpu_bdw),
+	X86_MATCH_VFM(INTEL_BROADWELL_G,	&idle_cpu_bdw),
+	X86_MATCH_VFM(INTEL_BROADWELL_X,	&idle_cpu_bdx),
+	X86_MATCH_VFM(INTEL_BROADWELL_D,	&idle_cpu_bdx),
+	X86_MATCH_VFM(INTEL_SKYLAKE_L,		&idle_cpu_skl),
+	X86_MATCH_VFM(INTEL_SKYLAKE,		&idle_cpu_skl),
+	X86_MATCH_VFM(INTEL_KABYLAKE_L,		&idle_cpu_skl),
+	X86_MATCH_VFM(INTEL_KABYLAKE,		&idle_cpu_skl),
+	X86_MATCH_VFM(INTEL_SKYLAKE_X,		&idle_cpu_skx),
+	X86_MATCH_VFM(INTEL_ICELAKE_X,		&idle_cpu_icx),
+	X86_MATCH_VFM(INTEL_ICELAKE_D,		&idle_cpu_icx),
+	X86_MATCH_VFM(INTEL_ALDERLAKE,		&idle_cpu_adl),
+	X86_MATCH_VFM(INTEL_ALDERLAKE_L,	&idle_cpu_adl_l),
+	X86_MATCH_VFM(INTEL_METEORLAKE_L,	&idle_cpu_mtl_l),
+	X86_MATCH_VFM(INTEL_ATOM_GRACEMONT,	&idle_cpu_gmt),
+	X86_MATCH_VFM(INTEL_SAPPHIRERAPIDS_X,	&idle_cpu_spr),
+	X86_MATCH_VFM(INTEL_EMERALDRAPIDS_X,	&idle_cpu_spr),
+	X86_MATCH_VFM(INTEL_GRANITERAPIDS_X,	&idle_cpu_gnr),
+	X86_MATCH_VFM(INTEL_GRANITERAPIDS_D,	&idle_cpu_gnrd),
+	X86_MATCH_VFM(INTEL_XEON_PHI_KNL,	&idle_cpu_knl),
+	X86_MATCH_VFM(INTEL_XEON_PHI_KNM,	&idle_cpu_knl),
+	X86_MATCH_VFM(INTEL_ATOM_GOLDMONT,	&idle_cpu_bxt),
+	X86_MATCH_VFM(INTEL_ATOM_GOLDMONT_PLUS,	&idle_cpu_bxt),
+	X86_MATCH_VFM(INTEL_ATOM_GOLDMONT_D,	&idle_cpu_dnv),
+	X86_MATCH_VFM(INTEL_ATOM_TREMONT,       &idle_cpu_tmt),
+	X86_MATCH_VFM(INTEL_ATOM_TREMONT_L,     &idle_cpu_tmt),
+	X86_MATCH_VFM(INTEL_ATOM_TREMONT_D,	&idle_cpu_snr),
+	X86_MATCH_VFM(INTEL_ATOM_CRESTMONT,	&idle_cpu_grr),
+	X86_MATCH_VFM(INTEL_ATOM_CRESTMONT_X,	&idle_cpu_srf),
+	X86_MATCH_VFM(INTEL_ATOM_DARKMONT_X,	&idle_cpu_srf),
 	{}
 };
 
@@ -1583,6 +1717,10 @@ 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 bool no_native __read_mostly; /* No effect if no_acpi is set. */
+module_param_named(no_native, no_native, bool, 0444);
+MODULE_PARM_DESC(no_native, "Ignore cpu specific (native) idle states in lieu of ACPI idle states");
+
 static struct acpi_processor_power acpi_state_table __initdata;
 
 /**
@@ -1687,12 +1825,16 @@ static void __init intel_idle_init_cstates_acpi(struct cpuidle_driver *drv)
 		if (intel_idle_state_needs_timer_stop(state))
 			state->flags |= CPUIDLE_FLAG_TIMER_STOP;
 
+		if (cx->type > ACPI_STATE_C1 && !boot_cpu_has(X86_FEATURE_NONSTOP_TSC))
+			mark_tsc_unstable("TSC halts in idle");
+
 		state->enter = intel_idle;
+		state->enter_dead = intel_idle_enter_dead;
 		state->enter_s2idle = intel_idle_s2idle;
 	}
 }
 
-static bool __init intel_idle_off_by_default(u32 mwait_hint)
+static bool __init intel_idle_off_by_default(unsigned int flags, u32 mwait_hint)
 {
 	int cstate, limit;
 
@@ -1709,17 +1851,34 @@ static bool __init intel_idle_off_by_default(u32 mwait_hint)
 	 * the interesting states are ACPI_CSTATE_FFH.
 	 */
 	for (cstate = 1; cstate < limit; cstate++) {
-		if (acpi_state_table.states[cstate].address == mwait_hint)
+		u32 acpi_hint = acpi_state_table.states[cstate].address;
+		u32 table_hint = mwait_hint;
+
+		if (flags & CPUIDLE_FLAG_PARTIAL_HINT_MATCH) {
+			acpi_hint &= ~MWAIT_SUBSTATE_MASK;
+			table_hint &= ~MWAIT_SUBSTATE_MASK;
+		}
+
+		if (acpi_hint == table_hint)
 			return false;
 	}
 	return true;
 }
+
+static inline bool ignore_native(void)
+{
+	return no_native && !no_acpi;
+}
 #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; }
+static inline bool intel_idle_off_by_default(unsigned int flags, u32 mwait_hint)
+{
+	return false;
+}
+static inline bool ignore_native(void) { return false; }
 #endif /* !CONFIG_ACPI_PROCESSOR_CSTATE */
 
 /**
@@ -1783,35 +1942,35 @@ static void __init bxt_idle_state_table_update(void)
 	unsigned long long msr;
 	unsigned int usec;
 
-	rdmsrl(MSR_PKGC6_IRTL, msr);
+	rdmsrq(MSR_PKGC6_IRTL, msr);
 	usec = irtl_2_usec(msr);
 	if (usec) {
 		bxt_cstates[2].exit_latency = usec;
 		bxt_cstates[2].target_residency = usec;
 	}
 
-	rdmsrl(MSR_PKGC7_IRTL, msr);
+	rdmsrq(MSR_PKGC7_IRTL, msr);
 	usec = irtl_2_usec(msr);
 	if (usec) {
 		bxt_cstates[3].exit_latency = usec;
 		bxt_cstates[3].target_residency = usec;
 	}
 
-	rdmsrl(MSR_PKGC8_IRTL, msr);
+	rdmsrq(MSR_PKGC8_IRTL, msr);
 	usec = irtl_2_usec(msr);
 	if (usec) {
 		bxt_cstates[4].exit_latency = usec;
 		bxt_cstates[4].target_residency = usec;
 	}
 
-	rdmsrl(MSR_PKGC9_IRTL, msr);
+	rdmsrq(MSR_PKGC9_IRTL, msr);
 	usec = irtl_2_usec(msr);
 	if (usec) {
 		bxt_cstates[5].exit_latency = usec;
 		bxt_cstates[5].target_residency = usec;
 	}
 
-	rdmsrl(MSR_PKGC10_IRTL, msr);
+	rdmsrq(MSR_PKGC10_IRTL, msr);
 	usec = irtl_2_usec(msr);
 	if (usec) {
 		bxt_cstates[6].exit_latency = usec;
@@ -1839,7 +1998,7 @@ static void __init sklh_idle_state_table_update(void)
 	if ((mwait_substates & (0xF << 28)) == 0)
 		return;
 
-	rdmsrl(MSR_PKG_CST_CONFIG_CONTROL, msr);
+	rdmsrq(MSR_PKG_CST_CONFIG_CONTROL, msr);
 
 	/* PC10 is not enabled in PKG C-state limit */
 	if ((msr & 0xF) != 8)
@@ -1851,7 +2010,7 @@ static void __init sklh_idle_state_table_update(void)
 	/* if SGX is present */
 	if (ebx & (1 << 2)) {
 
-		rdmsrl(MSR_IA32_FEAT_CTL, msr);
+		rdmsrq(MSR_IA32_FEAT_CTL, msr);
 
 		/* if SGX is enabled */
 		if (msr & (1 << 18))
@@ -1870,7 +2029,7 @@ static void __init skx_idle_state_table_update(void)
 {
 	unsigned long long msr;
 
-	rdmsrl(MSR_PKG_CST_CONFIG_CONTROL, msr);
+	rdmsrq(MSR_PKG_CST_CONFIG_CONTROL, msr);
 
 	/*
 	 * 000b: C0/C1 (no package C-state support)
@@ -1923,7 +2082,7 @@ static void __init spr_idle_state_table_update(void)
 	 * C6. However, if PC6 is disabled, we update the numbers to match
 	 * core C6.
 	 */
-	rdmsrl(MSR_PKG_CST_CONFIG_CONTROL, msr);
+	rdmsrq(MSR_PKG_CST_CONFIG_CONTROL, msr);
 
 	/* Limit value 2 and above allow for PC6. */
 	if ((msr & 0x7) < 2) {
@@ -1932,9 +2091,19 @@ static void __init spr_idle_state_table_update(void)
 	}
 }
 
+/**
+ * byt_cht_auto_demotion_disable - Disable Bay/Cherry Trail auto-demotion.
+ */
+static void __init byt_cht_auto_demotion_disable(void)
+{
+	wrmsrq(MSR_CC6_DEMOTION_POLICY_CONFIG, 0);
+	wrmsrq(MSR_MC6_DEMOTION_POLICY_CONFIG, 0);
+}
+
 static bool __init intel_idle_verify_cstate(unsigned int mwait_hint)
 {
-	unsigned int mwait_cstate = MWAIT_HINT2CSTATE(mwait_hint) + 1;
+	unsigned int mwait_cstate = (MWAIT_HINT2CSTATE(mwait_hint) + 1) &
+					MWAIT_CSTATE_MASK;
 	unsigned int num_substates = (mwait_substates >> mwait_cstate * 4) &
 					MWAIT_SUBSTATE_MASK;
 
@@ -1989,29 +2158,33 @@ 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:
+	switch (boot_cpu_data.x86_vfm) {
+	case INTEL_IVYBRIDGE_X:
 		ivt_idle_state_table_update();
 		break;
-	case INTEL_FAM6_ATOM_GOLDMONT:
-	case INTEL_FAM6_ATOM_GOLDMONT_PLUS:
+	case INTEL_ATOM_GOLDMONT:
+	case INTEL_ATOM_GOLDMONT_PLUS:
 		bxt_idle_state_table_update();
 		break;
-	case INTEL_FAM6_SKYLAKE:
+	case INTEL_SKYLAKE:
 		sklh_idle_state_table_update();
 		break;
-	case INTEL_FAM6_SKYLAKE_X:
+	case INTEL_SKYLAKE_X:
 		skx_idle_state_table_update();
 		break;
-	case INTEL_FAM6_SAPPHIRERAPIDS_X:
-	case INTEL_FAM6_EMERALDRAPIDS_X:
+	case INTEL_SAPPHIRERAPIDS_X:
+	case INTEL_EMERALDRAPIDS_X:
 		spr_idle_state_table_update();
 		break;
-	case INTEL_FAM6_ALDERLAKE:
-	case INTEL_FAM6_ALDERLAKE_L:
-	case INTEL_FAM6_ATOM_GRACEMONT:
+	case INTEL_ALDERLAKE:
+	case INTEL_ALDERLAKE_L:
+	case INTEL_ATOM_GRACEMONT:
 		adl_idle_state_table_update();
 		break;
+	case INTEL_ATOM_SILVERMONT:
+	case INTEL_ATOM_AIRMONT:
+		byt_cht_auto_demotion_disable();
+		break;
 	}
 
 	for (cstate = 0; cstate < CPUIDLE_STATE_MAX; ++cstate) {
@@ -2025,6 +2198,9 @@ static void __init intel_idle_init_cstates_icpu(struct cpuidle_driver *drv)
 		    !cpuidle_state_table[cstate].enter_s2idle)
 			break;
 
+		if (!cpuidle_state_table[cstate].enter_dead)
+			cpuidle_state_table[cstate].enter_dead = intel_idle_enter_dead;
+
 		/* If marked as unusable, skip this state. */
 		if (cpuidle_state_table[cstate].flags & CPUIDLE_FLAG_UNUSABLE) {
 			pr_debug("state %s is disabled\n",
@@ -2045,7 +2221,7 @@ static void __init intel_idle_init_cstates_icpu(struct cpuidle_driver *drv)
 
 		if ((disabled_states_mask & BIT(drv->state_count)) ||
 		    ((icpu->use_acpi || force_use_acpi) &&
-		     intel_idle_off_by_default(mwait_hint) &&
+		     intel_idle_off_by_default(state->flags, mwait_hint) &&
 		     !(state->flags & CPUIDLE_FLAG_ALWAYS_ENABLE)))
 			state->flags |= CPUIDLE_FLAG_OFF;
 
@@ -2054,11 +2230,6 @@ static void __init intel_idle_init_cstates_icpu(struct cpuidle_driver *drv)
 
 		drv->state_count++;
 	}
-
-	if (icpu->byt_auto_demotion_disable_flag) {
-		wrmsrl(MSR_CC6_DEMOTION_POLICY_CONFIG, 0);
-		wrmsrl(MSR_MC6_DEMOTION_POLICY_CONFIG, 0);
-	}
 }
 
 /**
@@ -2074,7 +2245,7 @@ static void __init intel_idle_cpuidle_driver_init(struct cpuidle_driver *drv)
 
 	drv->state_count = 1;
 
-	if (icpu)
+	if (icpu && icpu->state_table)
 		intel_idle_init_cstates_icpu(drv);
 	else
 		intel_idle_init_cstates_acpi(drv);
@@ -2084,27 +2255,27 @@ static void auto_demotion_disable(void)
 {
 	unsigned long long msr_bits;
 
-	rdmsrl(MSR_PKG_CST_CONFIG_CONTROL, msr_bits);
+	rdmsrq(MSR_PKG_CST_CONFIG_CONTROL, msr_bits);
 	msr_bits &= ~auto_demotion_disable_flags;
-	wrmsrl(MSR_PKG_CST_CONFIG_CONTROL, msr_bits);
+	wrmsrq(MSR_PKG_CST_CONFIG_CONTROL, msr_bits);
 }
 
 static void c1e_promotion_enable(void)
 {
 	unsigned long long msr_bits;
 
-	rdmsrl(MSR_IA32_POWER_CTL, msr_bits);
+	rdmsrq(MSR_IA32_POWER_CTL, msr_bits);
 	msr_bits |= 0x2;
-	wrmsrl(MSR_IA32_POWER_CTL, msr_bits);
+	wrmsrq(MSR_IA32_POWER_CTL, msr_bits);
 }
 
 static void c1e_promotion_disable(void)
 {
 	unsigned long long msr_bits;
 
-	rdmsrl(MSR_IA32_POWER_CTL, msr_bits);
+	rdmsrq(MSR_IA32_POWER_CTL, msr_bits);
 	msr_bits &= ~0x2;
-	wrmsrl(MSR_IA32_POWER_CTL, msr_bits);
+	wrmsrq(MSR_IA32_POWER_CTL, msr_bits);
 }
 
 /**
@@ -2167,6 +2338,88 @@ static void __init intel_idle_cpuidle_devices_uninit(void)
 		cpuidle_unregister_device(per_cpu_ptr(intel_idle_cpuidle_devices, i));
 }
 
+static void intel_c1_demotion_toggle(void *enable)
+{
+	unsigned long long msr_val;
+
+	rdmsrl(MSR_PKG_CST_CONFIG_CONTROL, msr_val);
+	/*
+	 * Enable/disable C1 undemotion along with C1 demotion, as this is the
+	 * most sensible configuration in general.
+	 */
+	if (enable)
+		msr_val |= NHM_C1_AUTO_DEMOTE | SNB_C1_AUTO_UNDEMOTE;
+	else
+		msr_val &= ~(NHM_C1_AUTO_DEMOTE | SNB_C1_AUTO_UNDEMOTE);
+	wrmsrl(MSR_PKG_CST_CONFIG_CONTROL, msr_val);
+}
+
+static ssize_t intel_c1_demotion_store(struct device *dev,
+				       struct device_attribute *attr,
+				       const char *buf, size_t count)
+{
+	bool enable;
+	int err;
+
+	err = kstrtobool(buf, &enable);
+	if (err)
+		return err;
+
+	mutex_lock(&c1_demotion_mutex);
+	/* Enable/disable C1 demotion on all CPUs */
+	on_each_cpu(intel_c1_demotion_toggle, (void *)enable, 1);
+	mutex_unlock(&c1_demotion_mutex);
+
+	return count;
+}
+
+static ssize_t intel_c1_demotion_show(struct device *dev,
+				      struct device_attribute *attr, char *buf)
+{
+	unsigned long long msr_val;
+
+	/*
+	 * Read the MSR value for a CPU and assume it is the same for all CPUs. Any other
+	 * configuration would be a BIOS bug.
+	 */
+	rdmsrl(MSR_PKG_CST_CONFIG_CONTROL, msr_val);
+	return sysfs_emit(buf, "%d\n", !!(msr_val & NHM_C1_AUTO_DEMOTE));
+}
+static DEVICE_ATTR_RW(intel_c1_demotion);
+
+static int __init intel_idle_sysfs_init(void)
+{
+	int err;
+
+	if (!c1_demotion_supported)
+		return 0;
+
+	sysfs_root = bus_get_dev_root(&cpu_subsys);
+	if (!sysfs_root)
+		return 0;
+
+	err = sysfs_add_file_to_group(&sysfs_root->kobj,
+				      &dev_attr_intel_c1_demotion.attr,
+				      "cpuidle");
+	if (err) {
+		put_device(sysfs_root);
+		return err;
+	}
+
+	return 0;
+}
+
+static void __init intel_idle_sysfs_uninit(void)
+{
+	if (!sysfs_root)
+		return;
+
+	sysfs_remove_file_from_group(&sysfs_root->kobj,
+				     &dev_attr_intel_c1_demotion.attr,
+				     "cpuidle");
+	put_device(sysfs_root);
+}
+
 static int __init intel_idle_init(void)
 {
 	const struct x86_cpu_id *id;
@@ -2194,10 +2447,7 @@ static int __init intel_idle_init(void)
 			return -ENODEV;
 	}
 
-	if (boot_cpu_data.cpuid_level < CPUID_MWAIT_LEAF)
-		return -ENODEV;
-
-	cpuid(CPUID_MWAIT_LEAF, &eax, &ebx, &ecx, &mwait_substates);
+	cpuid(CPUID_LEAF_MWAIT, &eax, &ebx, &ecx, &mwait_substates);
 
 	if (!(ecx & CPUID5_ECX_EXTENSIONS_SUPPORTED) ||
 	    !(ecx & CPUID5_ECX_INTERRUPT_BREAK) ||
@@ -2207,11 +2457,21 @@ static int __init intel_idle_init(void)
 	pr_debug("MWAIT substates: 0x%x\n", mwait_substates);
 
 	icpu = (const struct idle_cpu *)id->driver_data;
+	if (icpu && ignore_native()) {
+		pr_debug("ignoring native CPU idle states\n");
+		icpu = NULL;
+	}
 	if (icpu) {
-		cpuidle_state_table = icpu->state_table;
+		if (icpu->state_table)
+			cpuidle_state_table = icpu->state_table;
+		else if (!intel_idle_acpi_cst_extract())
+			return -ENODEV;
+
 		auto_demotion_disable_flags = icpu->auto_demotion_disable_flags;
 		if (icpu->disable_promotion_to_c1e)
 			c1e_promotion = C1E_PROMOTION_DISABLE;
+		if (icpu->c1_demotion_supported)
+			c1_demotion_supported = true;
 		if (icpu->use_acpi || force_use_acpi)
 			intel_idle_acpi_cst_extract();
 	} else if (!intel_idle_acpi_cst_extract()) {
@@ -2225,6 +2485,10 @@ static int __init intel_idle_init(void)
 	if (!intel_idle_cpuidle_devices)
 		return -ENOMEM;
 
+	retval = intel_idle_sysfs_init();
+	if (retval)
+		pr_warn("failed to initialized sysfs");
+
 	intel_idle_cpuidle_driver_init(&intel_idle_driver);
 
 	retval = cpuidle_register_driver(&intel_idle_driver);
@@ -2249,6 +2513,7 @@ hp_setup_fail:
 	intel_idle_cpuidle_devices_uninit();
 	cpuidle_unregister_driver(&intel_idle_driver);
 init_driver_fail:
+	intel_idle_sysfs_uninit();
 	free_percpu(intel_idle_cpuidle_devices);
 	return retval;