Merge branch 'arm64/psci-rework' of git://git.kernel.org/pub/scm/linux/kernel/git/mark/linux

* 'arm64/psci-rework' of git://git.kernel.org/pub/scm/linux/kernel/git/mark/linux:
  arm64: psci: remove ACPI coupling
  arm64: psci: kill psci_power_state
  arm64: psci: account for Trusted OS instances
  arm64: psci: support unsigned return values
  arm64: psci: remove unnecessary id indirection
  arm64: smp: consistently use error codes
  arm64: smp_plat: add get_logical_index
  arm/arm64: kvm: add missing PSCI include

Conflicts:
	arch/arm64/kernel/smp.c

arch/arm/kvm/psci.c

@@ -24,6 +24,8 @@
 #include <asm/kvm_psci.h>
 #include <asm/kvm_host.h>
 
+#include <uapi/linux/psci.h>
+
 /*
  * This is an implementation of the Power State Coordination Interface
  * as described in ARM document number ARM DEN 0022A.

arch/arm64/include/asm/acpi.h

@@ -16,6 +16,7 @@
 #include <linux/irqchip/arm-gic-acpi.h>
 
 #include <asm/cputype.h>
+#include <asm/psci.h>
 #include <asm/smp_plat.h>
 
 /* Basic configuration for ACPI */
@@ -39,18 +40,6 @@ extern int acpi_disabled;
 extern int acpi_noirq;
 extern int acpi_pci_disabled;
 
-/* 1 to indicate PSCI 0.2+ is implemented */
-static inline bool acpi_psci_present(void)
-{
-	return acpi_gbl_FADT.arm_boot_flags & ACPI_FADT_PSCI_COMPLIANT;
-}
-
-/* 1 to indicate HVC must be used instead of SMC as the PSCI conduit */
-static inline bool acpi_psci_use_hvc(void)
-{
-	return acpi_gbl_FADT.arm_boot_flags & ACPI_FADT_PSCI_USE_HVC;
-}
-
 static inline void disable_acpi(void)
 {
 	acpi_disabled = 1;
@@ -88,8 +77,6 @@ static inline void arch_fix_phys_package_id(int num, u32 slot) { }
 void __init acpi_init_cpus(void);
 
 #else
-static inline bool acpi_psci_present(void) { return false; }
-static inline bool acpi_psci_use_hvc(void) { return false; }
 static inline void acpi_init_cpus(void) { }
 #endif /* CONFIG_ACPI */
 

arch/arm64/include/asm/psci.h

@@ -14,7 +14,15 @@
 #ifndef __ASM_PSCI_H
 #define __ASM_PSCI_H
 
-int psci_dt_init(void);
+int __init psci_dt_init(void);
-int psci_acpi_init(void);
+
+#ifdef CONFIG_ACPI
+int __init psci_acpi_init(void);
+bool __init acpi_psci_present(void);
+bool __init acpi_psci_use_hvc(void);
+#else
+static inline int psci_acpi_init(void) { return 0; }
+static inline bool acpi_psci_present(void) { return false; }
+#endif
 
 #endif /* __ASM_PSCI_H */

arch/arm64/include/asm/smp_plat.h

@@ -19,6 +19,8 @@
 #ifndef __ASM_SMP_PLAT_H
 #define __ASM_SMP_PLAT_H
 
+#include <linux/cpumask.h>
+
 #include <asm/types.h>
 
 struct mpidr_hash {
@@ -39,6 +41,20 @@ static inline u32 mpidr_hash_size(void)
  */
 extern u64 __cpu_logical_map[NR_CPUS];
 #define cpu_logical_map(cpu)    __cpu_logical_map[cpu]
+/*
+ * Retrieve logical cpu index corresponding to a given MPIDR.Aff*
+ *  - mpidr: MPIDR.Aff* bits to be used for the look-up
+ *
+ * Returns the cpu logical index or -EINVAL on look-up error
+ */
+static inline int get_logical_index(u64 mpidr)
+{
+	int cpu;
+	for (cpu = 0; cpu < nr_cpu_ids; cpu++)
+		if (cpu_logical_map(cpu) == mpidr)
+			return cpu;
+	return -EINVAL;
+}
 
 void __init do_post_cpus_up_work(void);
 

arch/arm64/kernel/acpi.c

@@ -89,6 +89,17 @@ void __init __acpi_unmap_table(char *map, unsigned long size)
 	early_memunmap(map, size);
 }
 
+bool __init acpi_psci_present(void)
+{
+	return acpi_gbl_FADT.arm_boot_flags & ACPI_FADT_PSCI_COMPLIANT;
+}
+
+/* Whether HVC must be used instead of SMC as the PSCI conduit */
+bool __init acpi_psci_use_hvc(void)
+{
+	return acpi_gbl_FADT.arm_boot_flags & ACPI_FADT_PSCI_USE_HVC;
+}
+
 /*
  * acpi_fadt_sanity_check() - Check FADT presence and carry out sanity
  *			      checks on it

arch/arm64/kernel/psci.c

@@ -15,7 +15,6 @@
 
 #define pr_fmt(fmt) "psci: " fmt
 
-#include <linux/acpi.h>
 #include <linux/init.h>
 #include <linux/of.h>
 #include <linux/smp.h>
@@ -25,8 +24,8 @@
 #include <linux/slab.h>
 #include <uapi/linux/psci.h>
 
-#include <asm/acpi.h>
 #include <asm/compiler.h>
+#include <asm/cputype.h>
 #include <asm/cpu_ops.h>
 #include <asm/errno.h>
 #include <asm/psci.h>
@@ -37,16 +36,36 @@
 #define PSCI_POWER_STATE_TYPE_STANDBY		0
 #define PSCI_POWER_STATE_TYPE_POWER_DOWN	1
 
-struct psci_power_state {
+static bool psci_power_state_loses_context(u32 state)
-	u16	id;
+{
-	u8	type;
+	return state & PSCI_0_2_POWER_STATE_TYPE_MASK;
-	u8	affinity_level;
+}
-};
+
+static bool psci_power_state_is_valid(u32 state)
+{
+	const u32 valid_mask = PSCI_0_2_POWER_STATE_ID_MASK |
+			       PSCI_0_2_POWER_STATE_TYPE_MASK |
+			       PSCI_0_2_POWER_STATE_AFFL_MASK;
+
+	return !(state & ~valid_mask);
+}
+
+/*
+ * The CPU any Trusted OS is resident on. The trusted OS may reject CPU_OFF
+ * calls to its resident CPU, so we must avoid issuing those. We never migrate
+ * a Trusted OS even if it claims to be capable of migration -- doing so will
+ * require cooperation with a Trusted OS driver.
+ */
+static int resident_cpu = -1;
+
+static bool psci_tos_resident_on(int cpu)
+{
+	return cpu == resident_cpu;
+}
 
 struct psci_operations {
-	int (*cpu_suspend)(struct psci_power_state state,
+	int (*cpu_suspend)(u32 state, unsigned long entry_point);
-			   unsigned long entry_point);
+	int (*cpu_off)(u32 state);
-	int (*cpu_off)(struct psci_power_state state);
 	int (*cpu_on)(unsigned long cpuid, unsigned long entry_point);
 	int (*migrate)(unsigned long cpuid);
 	int (*affinity_info)(unsigned long target_affinity,
@@ -56,23 +75,21 @@ struct psci_operations {
 
 static struct psci_operations psci_ops;
 
-static int (*invoke_psci_fn)(u64, u64, u64, u64);
+typedef unsigned long (psci_fn)(unsigned long, unsigned long,
-typedef int (*psci_initcall_t)(const struct device_node *);
+				unsigned long, unsigned long);
-
+asmlinkage psci_fn __invoke_psci_fn_hvc;
-asmlinkage int __invoke_psci_fn_hvc(u64, u64, u64, u64);
+asmlinkage psci_fn __invoke_psci_fn_smc;
-asmlinkage int __invoke_psci_fn_smc(u64, u64, u64, u64);
+static psci_fn *invoke_psci_fn;
 
 enum psci_function {
 	PSCI_FN_CPU_SUSPEND,
 	PSCI_FN_CPU_ON,
 	PSCI_FN_CPU_OFF,
 	PSCI_FN_MIGRATE,
-	PSCI_FN_AFFINITY_INFO,
-	PSCI_FN_MIGRATE_INFO_TYPE,
 	PSCI_FN_MAX,
 };
 
-static DEFINE_PER_CPU_READ_MOSTLY(struct psci_power_state *, psci_power_state);
+static DEFINE_PER_CPU_READ_MOSTLY(u32 *, psci_power_state);
 
 static u32 psci_function_id[PSCI_FN_MAX];
 
@@ -92,56 +109,28 @@ static int psci_to_linux_errno(int errno)
 	return -EINVAL;
 }
 
-static u32 psci_power_state_pack(struct psci_power_state state)
+static u32 psci_get_version(void)
 {
-	return ((state.id << PSCI_0_2_POWER_STATE_ID_SHIFT)
+	return invoke_psci_fn(PSCI_0_2_FN_PSCI_VERSION, 0, 0, 0);
-			& PSCI_0_2_POWER_STATE_ID_MASK) |
-		((state.type << PSCI_0_2_POWER_STATE_TYPE_SHIFT)
-		 & PSCI_0_2_POWER_STATE_TYPE_MASK) |
-		((state.affinity_level << PSCI_0_2_POWER_STATE_AFFL_SHIFT)
-		 & PSCI_0_2_POWER_STATE_AFFL_MASK);
 }
 
-static void psci_power_state_unpack(u32 power_state,
+static int psci_cpu_suspend(u32 state, unsigned long entry_point)
-				    struct psci_power_state *state)
-{
-	state->id = (power_state & PSCI_0_2_POWER_STATE_ID_MASK) >>
-			PSCI_0_2_POWER_STATE_ID_SHIFT;
-	state->type = (power_state & PSCI_0_2_POWER_STATE_TYPE_MASK) >>
-			PSCI_0_2_POWER_STATE_TYPE_SHIFT;
-	state->affinity_level =
-			(power_state & PSCI_0_2_POWER_STATE_AFFL_MASK) >>
-			PSCI_0_2_POWER_STATE_AFFL_SHIFT;
-}
-
-static int psci_get_version(void)
 {
 	int err;
-
+	u32 fn;
-	err = invoke_psci_fn(PSCI_0_2_FN_PSCI_VERSION, 0, 0, 0);
-	return err;
-}
-
-static int psci_cpu_suspend(struct psci_power_state state,
-			    unsigned long entry_point)
-{
-	int err;
-	u32 fn, power_state;
 
 	fn = psci_function_id[PSCI_FN_CPU_SUSPEND];
-	power_state = psci_power_state_pack(state);
+	err = invoke_psci_fn(fn, state, entry_point, 0);
-	err = invoke_psci_fn(fn, power_state, entry_point, 0);
 	return psci_to_linux_errno(err);
 }
 
-static int psci_cpu_off(struct psci_power_state state)
+static int psci_cpu_off(u32 state)
 {
 	int err;
-	u32 fn, power_state;
+	u32 fn;
 
 	fn = psci_function_id[PSCI_FN_CPU_OFF];
-	power_state = psci_power_state_pack(state);
+	err = invoke_psci_fn(fn, state, 0, 0);
-	err = invoke_psci_fn(fn, power_state, 0, 0);
 	return psci_to_linux_errno(err);
 }
 
@@ -168,28 +157,24 @@ static int psci_migrate(unsigned long cpuid)
 static int psci_affinity_info(unsigned long target_affinity,
 		unsigned long lowest_affinity_level)
 {
-	int err;
+	return invoke_psci_fn(PSCI_0_2_FN64_AFFINITY_INFO, target_affinity,
-	u32 fn;
+			      lowest_affinity_level, 0);
-
-	fn = psci_function_id[PSCI_FN_AFFINITY_INFO];
-	err = invoke_psci_fn(fn, target_affinity, lowest_affinity_level, 0);
-	return err;
 }
 
 static int psci_migrate_info_type(void)
 {
-	int err;
+	return invoke_psci_fn(PSCI_0_2_FN_MIGRATE_INFO_TYPE, 0, 0, 0);
-	u32 fn;
+}
 
-	fn = psci_function_id[PSCI_FN_MIGRATE_INFO_TYPE];
+static unsigned long psci_migrate_info_up_cpu(void)
-	err = invoke_psci_fn(fn, 0, 0, 0);
+{
-	return err;
+	return invoke_psci_fn(PSCI_0_2_FN64_MIGRATE_INFO_UP_CPU, 0, 0, 0);
 }
 
 static int __maybe_unused cpu_psci_cpu_init_idle(unsigned int cpu)
 {
 	int i, ret, count = 0;
-	struct psci_power_state *psci_states;
+	u32 *psci_states;
 	struct device_node *state_node, *cpu_node;
 
 	cpu_node = of_get_cpu_node(cpu, NULL);
@@ -218,13 +203,13 @@ static int __maybe_unused cpu_psci_cpu_init_idle(unsigned int cpu)
 		return -ENOMEM;
 
 	for (i = 0; i < count; i++) {
-		u32 psci_power_state;
+		u32 state;
 
 		state_node = of_parse_phandle(cpu_node, "cpu-idle-states", i);
 
 		ret = of_property_read_u32(state_node,
 					   "arm,psci-suspend-param",
-					   &psci_power_state);
+					   &state);
 		if (ret) {
 			pr_warn(" * %s missing arm,psci-suspend-param property\n",
 				state_node->full_name);
@@ -233,9 +218,13 @@ static int __maybe_unused cpu_psci_cpu_init_idle(unsigned int cpu)
 		}
 
 		of_node_put(state_node);
-		pr_debug("psci-power-state %#x index %d\n", psci_power_state,
+		pr_debug("psci-power-state %#x index %d\n", state, i);
-							    i);
+		if (!psci_power_state_is_valid(state)) {
-		psci_power_state_unpack(psci_power_state, &psci_states[i]);
+			pr_warn("Invalid PSCI power state %#x\n", state);
+			ret = -EINVAL;
+			goto free_mem;
+		}
+		psci_states[i] = state;
 	}
 	/* Idle states parsed correctly, initialize per-cpu pointer */
 	per_cpu(psci_power_state, cpu) = psci_states;
@@ -278,6 +267,46 @@ static void psci_sys_poweroff(void)
 	invoke_psci_fn(PSCI_0_2_FN_SYSTEM_OFF, 0, 0, 0);
 }
 
+/*
+ * Detect the presence of a resident Trusted OS which may cause CPU_OFF to
+ * return DENIED (which would be fatal).
+ */
+static void __init psci_init_migrate(void)
+{
+	unsigned long cpuid;
+	int type, cpu;
+
+	type = psci_ops.migrate_info_type();
+
+	if (type == PSCI_0_2_TOS_MP) {
+		pr_info("Trusted OS migration not required\n");
+		return;
+	}
+
+	if (type == PSCI_RET_NOT_SUPPORTED) {
+		pr_info("MIGRATE_INFO_TYPE not supported.\n");
+		return;
+	}
+
+	if (type != PSCI_0_2_TOS_UP_MIGRATE &&
+	    type != PSCI_0_2_TOS_UP_NO_MIGRATE) {
+		pr_err("MIGRATE_INFO_TYPE returned unknown type (%d)\n", type);
+		return;
+	}
+
+	cpuid = psci_migrate_info_up_cpu();
+	if (cpuid & ~MPIDR_HWID_BITMASK) {
+		pr_warn("MIGRATE_INFO_UP_CPU reported invalid physical ID (0x%lx)\n",
+			cpuid);
+		return;
+	}
+
+	cpu = get_logical_index(cpuid);
+	resident_cpu = cpu >= 0 ? cpu : -1;
+
+	pr_info("Trusted OS resident on physical CPU 0x%lx\n", cpuid);
+}
+
 static void __init psci_0_2_set_functions(void)
 {
 	pr_info("Using standard PSCI v0.2 function IDs\n");
@@ -293,11 +322,8 @@ static void __init psci_0_2_set_functions(void)
 	psci_function_id[PSCI_FN_MIGRATE] = PSCI_0_2_FN64_MIGRATE;
 	psci_ops.migrate = psci_migrate;
 
-	psci_function_id[PSCI_FN_AFFINITY_INFO] = PSCI_0_2_FN64_AFFINITY_INFO;
 	psci_ops.affinity_info = psci_affinity_info;
 
-	psci_function_id[PSCI_FN_MIGRATE_INFO_TYPE] =
-		PSCI_0_2_FN_MIGRATE_INFO_TYPE;
 	psci_ops.migrate_info_type = psci_migrate_info_type;
 
 	arm_pm_restart = psci_sys_reset;
@@ -310,32 +336,26 @@ static void __init psci_0_2_set_functions(void)
  */
 static int __init psci_probe(void)
 {
-	int ver = psci_get_version();
+	u32 ver = psci_get_version();
 
-	if (ver == PSCI_RET_NOT_SUPPORTED) {
+	pr_info("PSCIv%d.%d detected in firmware.\n",
-		/*
+			PSCI_VERSION_MAJOR(ver),
-		 * PSCI versions >=0.2 mandates implementation of
+			PSCI_VERSION_MINOR(ver));
-		 * PSCI_VERSION.
-		 */
-		pr_err("PSCI firmware does not comply with the v0.2 spec.\n");
-		return -EOPNOTSUPP;
-	} else {
-		pr_info("PSCIv%d.%d detected in firmware.\n",
-				PSCI_VERSION_MAJOR(ver),
-				PSCI_VERSION_MINOR(ver));
 
-		if (PSCI_VERSION_MAJOR(ver) == 0 &&
+	if (PSCI_VERSION_MAJOR(ver) == 0 && PSCI_VERSION_MINOR(ver) < 2) {
-				PSCI_VERSION_MINOR(ver) < 2) {
+		pr_err("Conflicting PSCI version detected.\n");
-			pr_err("Conflicting PSCI version detected.\n");
+		return -EINVAL;
-			return -EINVAL;
-		}
 	}
 
 	psci_0_2_set_functions();
 
+	psci_init_migrate();
+
 	return 0;
 }
 
+typedef int (*psci_initcall_t)(const struct device_node *);
+
 /*
  * PSCI init function for PSCI versions >=0.2
  *
@@ -424,6 +444,7 @@ int __init psci_dt_init(void)
 	return init_fn(np);
 }
 
+#ifdef CONFIG_ACPI
 /*
  * We use PSCI 0.2+ when ACPI is deployed on ARM64 and it's
  * explicitly clarified in SBBR
@@ -444,6 +465,7 @@ int __init psci_acpi_init(void)
 
 	return psci_probe();
 }
+#endif
 
 #ifdef CONFIG_SMP
 
@@ -477,6 +499,11 @@ static int cpu_psci_cpu_disable(unsigned int cpu)
 	/* Fail early if we don't have CPU_OFF support */
 	if (!psci_ops.cpu_off)
 		return -EOPNOTSUPP;
+
+	/* Trusted OS will deny CPU_OFF */
+	if (psci_tos_resident_on(cpu))
+		return -EPERM;
+
 	return 0;
 }
 
@@ -487,9 +514,8 @@ static void cpu_psci_cpu_die(unsigned int cpu)
 	 * There are no known implementations of PSCI actually using the
 	 * power state field, pass a sensible default for now.
 	 */
-	struct psci_power_state state = {
+	u32 state = PSCI_POWER_STATE_TYPE_POWER_DOWN <<
-		.type = PSCI_POWER_STATE_TYPE_POWER_DOWN,
+		    PSCI_0_2_POWER_STATE_TYPE_SHIFT;
-	};
 
 	ret = psci_ops.cpu_off(state);
 
@@ -501,7 +527,7 @@ static int cpu_psci_cpu_kill(unsigned int cpu)
 	int err, i;
 
 	if (!psci_ops.affinity_info)
-		return 1;
+		return 0;
 	/*
 	 * cpu_kill could race with cpu_die and we can
 	 * potentially end up declaring this cpu undead
@@ -512,7 +538,7 @@ static int cpu_psci_cpu_kill(unsigned int cpu)
 		err = psci_ops.affinity_info(cpu_logical_map(cpu), 0);
 		if (err == PSCI_0_2_AFFINITY_LEVEL_OFF) {
 			pr_info("CPU%d killed.\n", cpu);
-			return 1;
+			return 0;
 		}
 
 		msleep(10);
@@ -521,15 +547,14 @@ static int cpu_psci_cpu_kill(unsigned int cpu)
 
 	pr_warn("CPU%d may not have shut down cleanly (AFFINITY_INFO reports %d)\n",
 			cpu, err);
-	/* Make op_cpu_kill() fail. */
+	return -ETIMEDOUT;
-	return 0;
 }
 #endif
 #endif
 
 static int psci_suspend_finisher(unsigned long index)
 {
-	struct psci_power_state *state = __this_cpu_read(psci_power_state);
+	u32 *state = __this_cpu_read(psci_power_state);
 
 	return psci_ops.cpu_suspend(state[index - 1],
 				    virt_to_phys(cpu_resume));
@@ -538,7 +563,7 @@ static int psci_suspend_finisher(unsigned long index)
 static int __maybe_unused cpu_psci_cpu_suspend(unsigned long index)
 {
 	int ret;
-	struct psci_power_state *state = __this_cpu_read(psci_power_state);
+	u32 *state = __this_cpu_read(psci_power_state);
 	/*
 	 * idle state index 0 corresponds to wfi, should never be called
 	 * from the cpu_suspend operations
@@ -546,7 +571,7 @@ static int __maybe_unused cpu_psci_cpu_suspend(unsigned long index)
 	if (WARN_ON_ONCE(!index))
 		return -EINVAL;
 
-	if (state[index - 1].type == PSCI_POWER_STATE_TYPE_STANDBY)
+	if (!psci_power_state_loses_context(state[index - 1]))
 		ret = psci_ops.cpu_suspend(state[index - 1], 0);
 	else
 		ret = __cpu_suspend(index, psci_suspend_finisher);

arch/arm64/kernel/smp.c

@@ -249,7 +249,7 @@ static int op_cpu_kill(unsigned int cpu)
 	 * time and hope that it's dead, so let's skip the wait and just hope.
 	 */
 	if (!cpu_ops[cpu]->cpu_kill)
-		return 1;
+		return 0;
 
 	return cpu_ops[cpu]->cpu_kill(cpu);
 }
@@ -260,6 +260,8 @@ static int op_cpu_kill(unsigned int cpu)
  */
 void __cpu_die(unsigned int cpu)
 {
+	int err;
+
 	if (!cpu_wait_death(cpu, 5)) {
 		pr_crit("CPU%u: cpu didn't die\n", cpu);
 		return;
@@ -272,8 +274,10 @@ void __cpu_die(unsigned int cpu)
 	 * verify that it has really left the kernel before we consider
 	 * clobbering anything it might still be using.
 	 */
-	if (!op_cpu_kill(cpu))
+	err = op_cpu_kill(cpu);
-		pr_warn("CPU%d may not have shut down cleanly\n", cpu);
+	if (err)
+		pr_warn("CPU%d may not have shut down cleanly: %d\n",
+			cpu, err);
 }
 
 /*