arm/arm64: Unify PSCI client support
Subsystems (6)
Mark Rutland
maintainerv3 cover arm/arm64: Unify PSCI client support
This series unifies the 32-bit and 64-bit PSCI client code, moving the bulk of the FW invocation and probing out to a common location in drivers/firmware. As part of this, the remaining edge cases for PSCI 0.2 (Trusted OSs rejecting CPU_OFF) are accounted for, making both 32-bit adn 64-bit clients more robust.
This results in a reasonable saving in terms of lines of code, and will allow for PSCI 1.0 support to be unified form the beginning, avoiding further duplication.
The series is based on the arm64 for-next/cpu-init branch [3] due to conflicts with that series.
Since v1 [1]:
* Fix build when PSCI isn't selected
* Don't indirect migrate_info_up_cpu
* Fix ver to u32 in psci_get_version
Since v2: [2]:
* Rebase atop of Lorenzo's cpu init cleanup series [3]
* Use the correct native function IDs
* Fix cpu_disable return values
* Report invalid MIGRATE_INFO_UP_CPU return values
Many thanks to those who have reviewed and tested so far! Unfortunately, due to the volume of changes in the rebase I've dropped all Tested-by tags. I've locally tested 32-bit and 64-bit kernels with DT.
Russell, are you happy with the penultimate patch? If so, would you be happy for that to go via arm-soc (which I guess should take the series)?
Thanks,
Mark.
[1] http://lists.infradead.org/pipermail/linux-arm-kernel/2015-May/341770.html
[2] http://lists.infradead.org/pipermail/linux-arm-kernel/2015-May/344309.html
[3] git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux.git for-next/cpu-init
Mark Rutland (12):
arm/arm64: kvm: add missing PSCI include
arm64: smp_plat: add get_logical_index
arm64: smp: consistently use error codes
arm64: psci: remove unnecessary id indirection
arm64: psci: support unsigned return values
arm64: psci: account for Trusted OS instances
arm64: psci: kill psci_power_state
arm64: psci: remove ACPI coupling
arm64: psci: factor invocation code to drivers
drivers: psci: support native SMC{32,64} calls
ARM: migrate to common PSCI client code
MAINTAINERS: add PSCI entry
MAINTAINERS | 9 +
arch/arm/Kconfig | 1 +
arch/arm/include/asm/psci.h | 23 --
arch/arm/kernel/Makefile | 2 +-
arch/arm/kernel/psci.c | 299 -------------------
arch/arm/kernel/psci_smp.c | 29 +-
arch/arm/kernel/setup.c | 3 +-
arch/arm/kvm/psci.c | 2 +
arch/arm/mach-highbank/highbank.c | 2 +-
arch/arm/mach-highbank/pm.c | 8 +-
arch/arm64/Kconfig | 1 +
arch/arm64/include/asm/acpi.h | 17 +-
arch/arm64/include/asm/psci.h | 20 --
arch/arm64/include/asm/smp_plat.h | 16 +
arch/arm64/kernel/acpi.c | 11 +
arch/arm64/kernel/psci.c | 398 ++-----------------------
arch/arm64/kernel/setup.c | 2 +-
arch/arm64/kernel/smp.c | 10 +-
drivers/cpuidle/cpuidle-calxeda.c | 7 +-
drivers/firmware/Kconfig | 3 +
drivers/firmware/Makefile | 1 +
drivers/firmware/psci.c | 383 ++++++++++++++++++++++++
{arch/arm/include/asm => include/linux}/psci.h | 40 +--
23 files changed, 523 insertions(+), 764 deletions(-)
delete mode 100644 arch/arm/kernel/psci.c
delete mode 100644 arch/arm64/include/asm/psci.h
create mode 100644 drivers/firmware/psci.c
copy {arch/arm/include/asm => include/linux}/psci.h (55%)Signature · 1 line
1.9.1
Mark Rutland
maintainerPATCH v3 1/12 arm/arm64: kvm: add missing PSCI include
We make use of the PSCI function IDs, but don't explicitly include the header which defines them. Relying on transitive header includes is fragile and will be broken as headers are refactored.
This patch includes the relevant header file directly so as to avoid future breakage.
--- arch/arm/kvm/psci.c | 2 ++ 1 file changed, 2 insertions(+)
arch/arm/kvm/psci.c +2 -0
@@ -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.
Signature · 1 line
1.9.1
Mark Rutland
maintainerPATCH v3 2/12 arm64: smp_plat: add get_logical_index
In reply to Mark RutlandThe PSCI MIGRATE_INFO_UP_CPU call returns a physical ID, which we will need to map back to a Linux logical ID.
Implement a reusable get_logical_index to map from a physical ID to a logical ID.
--- arch/arm64/include/asm/smp_plat.h | 16 ++++++++++++++++ 1 file changed, 16 insertions(+)
arch/arm64/include/asm/smp_plat.h +16 -0
@@ -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 );
Signature · 1 line
1.9.1
Mark Rutland
maintainerPATCH v3 3/12 arm64: smp: consistently use error codes
In reply to Mark Rutlandcpu_kill currently returns one for success and zero for failure, which is unlike all the other cpu_operations, which return zero for success and an error code upon failure. This difference is unnecessarily confusing.
Make cpu_kill consistent with the other cpu_operations.
--- arch/arm64/kernel/psci.c | 7 +++---- arch/arm64/kernel/smp.c | 10 +++++++--- 2 files changed, 10 insertions(+), 7 deletions(-)
arch/arm64/kernel/psci.c +3 -4
@@ -501,7 +501,7 @@ 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 +512,7 @@ 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 );
@@ -521,8 +521,7 @@ 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 0 ; +return -ETIMEDOUT ; }#endif #endif
arch/arm64/kernel/smp.c +7 -3
@@ -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); }
@@ -262,6 +262,8 @@ static DECLARE_COMPLETION(cpu_died);*/ void __cpu_die (unsigned int cpu) { +int err; +if (!wait_for_completion_timeout (&cpu_died,msecs_to_jiffies (5000 ))) {pr_crit ("CPU%u: cpu didn't die\n" , cpu);return ;
@@ -274,8 +276,10 @@ 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)) -pr_warn ("CPU%d may not have shut down cleanly\n" , cpu); + err =op_cpu_kill (cpu); +if (err) +pr_warn ("CPU%d may not have shut down cleanly: %d\n" , + cpu, err); }/*
Signature · 1 line
1.9.1
Mark Rutland
maintainerPATCH v3 4/12 arm64: psci: remove unnecessary id indirection
In reply to Mark RutlandPSCI 0.1 did not define canonical IDs for CPU_ON, CPU_OFF, CPU_SUSPEND, or MIGRATE, and so these need to be provided when using firmware compliant to PSCI 0.1.
However, functions introduced in 0.2 or later have canonical IDs, and these cannot be provided via DT. There's no need to indirect the IDs via a table; they can be used directly at callsites (and already are for SYSTEM_OFF and SYSTEM_RESET).
This patch removes the unnecessary function ID indirection for AFFINITY_INFO and MIGRATE_INFO_TYPE.
--- arch/arm64/kernel/psci.c | 20 +++----------------- 1 file changed, 3 insertions(+), 17 deletions(-)
arch/arm64/kernel/psci.c +3 -17
@@ -67,8 +67,6 @@ enum psci_function {PSCI_FN_CPU_ON ,PSCI_FN_CPU_OFF ,PSCI_FN_MIGRATE , -PSCI_FN_AFFINITY_INFO , -PSCI_FN_MIGRATE_INFO_TYPE , PSCI_FN_MAX, };
@@ -168,22 +166,13 @@ static int psci_migrate(unsigned long cpuid)static int psci_affinity_info (unsigned long target_affinity,unsigned long lowest_affinity_level) { -int err; -u32 fn; - - fn = psci_function_id[PSCI_FN_AFFINITY_INFO ]; - err =invoke_psci_fn (fn, target_affinity, lowest_affinity_level,0 ); -return err; +return invoke_psci_fn (PSCI_0_2_FN64_AFFINITY_INFO , target_affinity, + lowest_affinity_level,0 ); }static int psci_migrate_info_type (void ) { -int err; -u32 fn; - - fn = psci_function_id[PSCI_FN_MIGRATE_INFO_TYPE ]; - err =invoke_psci_fn (fn,0 ,0 ,0 ); -return err; +return invoke_psci_fn (PSCI_0_2_FN_MIGRATE_INFO_TYPE ,0 ,0 ,0 ); }static int __maybe_unused cpu_psci_cpu_init_idle (unsigned int cpu)
@@ -293,11 +282,8 @@ 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;
Signature · 1 line
1.9.1
Mark Rutland
maintainerPATCH v3 5/12 arm64: psci: support unsigned return values
In reply to Mark RutlandPSCI_VERSION and MIGRATE_INFO_TYPE_UP_CPU return unsigned values, with the latter returning a 64-bit value. However, the PSCI invocation functions have prototypes returning int.
This patch upgrades the invocation functions to return unsigned long, with a new typedef to keep things legible. As PSCI_VERSION cannot return a negative value, the erroneous check against PSCI_RET_NOT_SUPPORTED is also removed. The unrelated psci_initcall_t typedef is moved closer to its first user, to avoid confusion with the invocation functions.
In preparation for sharing the code with ARM, unsigned long is used in preference of u64. In the SMC32 calling convention, the relevant fields will be 32 bits wide.
--- arch/arm64/kernel/psci.c | 47 ++++++++++++++++++----------------------------- 1 file changed, 18 insertions(+), 29 deletions(-)
arch/arm64/kernel/psci.c +18 -29
@@ -56,11 +56,11 @@ struct psci_operations {static struct psci_operations psci_ops; -static int (*invoke_psci_fn)(u64 ,u64 ,u64 ,u64 ); -typedef int (*psci_initcall_t )(const struct device_node *); - -asmlinkage int __invoke_psci_fn_hvc(u64, u64, u64, u64); -asmlinkage int __invoke_psci_fn_smc(u64, u64, u64, u64); +typedef unsigned long (psci_fn )(unsigned long ,unsigned long , +unsigned long ,unsigned long ); +asmlinkage psci_fn __invoke_psci_fn_hvc; +asmlinkage psci_fn __invoke_psci_fn_smc; +static psci_fn *invoke_psci_fn;enum psci_function {PSCI_FN_CPU_SUSPEND ,
@@ -112,12 +112,9 @@ static void psci_power_state_unpack(u32 power_state,PSCI_0_2_POWER_STATE_AFFL_SHIFT ; } -static int psci_get_version (void ) +static u32 psci_get_version (void ) { -int err; - - err =invoke_psci_fn (PSCI_0_2_FN_PSCI_VERSION ,0 ,0 ,0 ); -return err; +return invoke_psci_fn (PSCI_0_2_FN_PSCI_VERSION ,0 ,0 ,0 ); } static int psci_cpu_suspend(struct psci_power_state state,
@@ -296,25 +293,15 @@ static void __init psci_0_2_set_functions(void)*/ static int __init psci_probe (void ) { -int ver =psci_get_version (); - -if (ver ==PSCI_RET_NOT_SUPPORTED ) { -/* -* PSCI versions >=0.2 mandates implementation of -* 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 && -PSCI_VERSION_MINOR (ver) <2 ) { -pr_err ("Conflicting PSCI version detected.\n" ); -return -EINVAL ; - } +u32 ver =psci_get_version (); + +pr_info ("PSCIv%d.%d detected in firmware.\n" , +PSCI_VERSION_MAJOR (ver), +PSCI_VERSION_MINOR (ver)); + +if (PSCI_VERSION_MAJOR (ver) ==0 &&PSCI_VERSION_MINOR (ver) <2 ) { +pr_err ("Conflicting PSCI version detected.\n" ); +return -EINVAL ; }psci_0_2_set_functions ();
@@ -322,6 +309,8 @@ static int __init psci_probe(void)return 0 ; } +typedef int (*psci_initcall_t )(const struct device_node *); +/* * PSCI init function for PSCI versions >=0.2 *
Signature · 1 line
1.9.1
Mark Rutland
maintainerPATCH v3 6/12 arm64: psci: account for Trusted OS instances
In reply to Mark RutlandSoftware resident in the secure world (a "Trusted OS") may cause CPU_OFF calls for the CPU it is resident on to be denied. Such a denial would be fatal for the kernel, and so we must detect when this can happen before the point of no return.
This patch implements Trusted OS detection for PSCI 0.2+ systems, using MIGRATE_INFO_TYPE and MIGRATE_INFO_UP_CPU. When a trusted OS is detected as resident on a particular CPU, attempts to hot unplug that CPU will be denied early, before they can prove fatal.
Trusted OS migration is not implemented by this patch. Implementation of migratable UP trusted OSs seems unlikely, and the right policy for migration is unclear (and will likely differ across implementations). As such, it is likely that migration will require cooperation with Trusted OS drivers.
PSCI implementations prior to 0.1 do not provide the facility to detect the presence of a Trusted OS, nor the CPU any such OS is resident on, so without additional information it is not possible to handle Trusted OSs with PSCI 0.1.
--- arch/arm64/kernel/psci.c | 66 ++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 66 insertions(+)
arch/arm64/kernel/psci.c +66 -0
@@ -27,6 +27,7 @@#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>
@@ -43,6 +44,19 @@ struct psci_power_state {u8 affinity_level; }; +/* +* 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,unsigned long entry_point);
@@ -172,6 +186,11 @@ static int psci_migrate_info_type(void)return invoke_psci_fn (PSCI_0_2_FN_MIGRATE_INFO_TYPE ,0 ,0 ,0 ); } +static unsigned long psci_migrate_info_up_cpu (void ) +{ +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 ;
@@ -264,6 +283,46 @@ 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" );
@@ -306,6 +365,8 @@ static int __init psci_probe(void)psci_0_2_set_functions (); +psci_init_migrate (); +return 0 ; }
@@ -452,6 +513,11 @@ 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 ; }
Signature · 1 line
1.9.1
Mark Rutland
maintainerPATCH v3 7/12 arm64: psci: kill psci_power_state
In reply to Mark RutlandA PSCI 1.0 implementation may choose to use the new extended StateID format, the presence of which may be queried via the PSCI_FEATURES call. The layout of this new StateID format is incompatible with the existing format, and so to handle both we must abstract attempts to parse the fields.
In preparation for PSCI 1.0 support, this patch introduces psci_power_state_loses_context and psci_power_state_is_valid functions to query information from a PSCI power state, which is no longer decomposed (and hence the pack/unpack functions are removed). As it is no longer decomposed, it is now passed round as an opaque u32 token.
--- arch/arm64/kernel/psci.c | 89 ++++++++++++++++++++---------------------------- 1 file changed, 37 insertions(+), 52 deletions(-)
arch/arm64/kernel/psci.c +37 -52
@@ -38,11 +38,19 @@#define PSCI_POWER_STATE_TYPE_STANDBY 0#define PSCI_POWER_STATE_TYPE_POWER_DOWN 1 -struct psci_power_state { -u16 id; -u8 type; -u8 affinity_level; -}; +static bool psci_power_state_loses_context (u32 state) +{ +return state &PSCI_0_2_POWER_STATE_TYPE_MASK ; +} + +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
@@ -58,9 +66,8 @@ static bool psci_tos_resident_on(int cpu) }struct psci_operations { -int (*cpu_suspend)(struct psci_power_state state, -unsigned long entry_point); -int (*cpu_off)(struct psci_power_state state); +int (*cpu_suspend)(u32 state,unsigned long entry_point); +int (*cpu_off)(u32 state);int (*cpu_on)(unsigned long cpuid,unsigned long entry_point);int (*migrate)(unsigned long cpuid); int (*affinity_info)(unsigned long target_affinity,
@@ -84,7 +91,7 @@ enum psci_function { 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 ];
@@ -104,53 +111,28 @@ static int psci_to_linux_errno(int errno)return -EINVAL ; } -static u32 psci_power_state_pack (struct psci_power_state state) -{ -return ((state.id <<PSCI_0_2_POWER_STATE_ID_SHIFT ) - &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, -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 u32 psci_get_version (void ) {return invoke_psci_fn (PSCI_0_2_FN_PSCI_VERSION ,0 ,0 ,0 ); } -static int psci_cpu_suspend (struct psci_power_state state, -unsigned long entry_point) +static int psci_cpu_suspend (u32 state,unsigned long entry_point) {int err; -u32 fn, power_state; +u32 fn; fn = psci_function_id[PSCI_FN_CPU_SUSPEND ]; - power_state =psci_power_state_pack (state); - err =invoke_psci_fn (fn, power_state, entry_point,0 ); + err =invoke_psci_fn (fn, 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, power_state,0 ,0 ); + err =invoke_psci_fn (fn, state,0 ,0 );return psci_to_linux_errno (err); }
@@ -194,7 +176,7 @@ static unsigned long psci_migrate_info_up_cpu(void)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 );
@@ -223,13 +205,13 @@ 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);
@@ -238,9 +220,13 @@ 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, - i); -psci_power_state_unpack (psci_power_state, &psci_states[i]); +pr_debug ("psci-power-state %#x index %d\n" , state, i); +if (!psci_power_state_is_valid (state)) { +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;
@@ -528,9 +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 = { - .type =PSCI_POWER_STATE_TYPE_POWER_DOWN , - }; +u32 state =PSCI_POWER_STATE_TYPE_POWER_DOWN << +PSCI_0_2_POWER_STATE_TYPE_SHIFT ; ret = psci_ops.cpu_off (state);
@@ -569,7 +554,7 @@ static int cpu_psci_cpu_kill(unsigned int cpu)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));
@@ -578,7 +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
@@ -586,7 +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);
Signature · 1 line
1.9.1
Mark Rutland
maintainerPATCH v3 8/12 arm64: psci: remove ACPI coupling
In reply to Mark RutlandThe 32-bit ARM port doesn't have ACPI headers, and conditionally including them is going to look horrendous. In preparation for sharing the PSCI invocation code with 32-bit, move the acpi_psci_* function declarations and definitions such that the PSCI client code need not include ACPI headers.
While it would seem like we could simply hide the ACPI includes in psci.h, the ACPI headers have hilarious circular dependencies which make this infeasible without reorganising most of ACPICA. So rather than doing that, move the acpi_psci_* prototypes into psci.h.
The psci_acpi_init function is made dependent on CONFIG_ACPI (with a stub implementation in asm/psci.h) such that it need not be built for 32-bit ARM or kernels without ACPI support. The currently missing __init annotations are added to the prototypes in the header.
--- arch/arm64/include/asm/acpi.h | 15 +-------------- arch/arm64/include/asm/psci.h | 12 ++++++++++-- arch/arm64/kernel/acpi.c | 11 +++++++++++ arch/arm64/kernel/psci.c | 4 ++-- 4 files changed, 24 insertions(+), 18 deletions(-)
arch/arm64/include/asm/acpi.h +1 -14
@@ -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 +10 -2
@@ -14,7 +14,15 @@#ifndef __ASM_PSCI_H#define __ASM_PSCI_H -int psci_dt_init (void ); -int psci_acpi_init (void ); +int __init psci_dt_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/kernel/acpi.c +11 -0
@@ -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 +2 -2
@@ -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>
@@ -25,7 +24,6 @@#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>
@@ -446,6 +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
@@ -466,6 +465,7 @@ int __init psci_acpi_init(void) return psci_probe(); } +#endif #ifdef CONFIG_SMP
Signature · 1 line
1.9.1
PATCH v3 8/12 arm64: psci: remove ACPI coupling
Quoted 10 lines · Mark Rutland
The 32-bit ARM port doesn't have ACPI headers, and conditionally including them is going to look horrendous. In preparation for sharing the PSCI invocation code with 32-bit, move the acpi_psci_* function declarations and definitions such that the PSCI client code need not include ACPI headers.
While it would seem like we could simply hide the ACPI includes in
psci.h, the ACPI headers have hilarious circular dependencies which make this infeasible without reorganising most of ACPICA. So rather than doing that, move the acpi_psci_* prototypes intopsci.h.
I really doubt "hilarious" was the first word that came to mind in trying to figure out those dependencies :-). It was not the first word I used, I assure you.
Patch LGTM: Reviewed-by: Al Stone <[email protected]>
Quoted 140 lines
The
psci_acpi_initfunction is made dependent onCONFIG_ACPI(with a stub implementation inasm/psci.h) such that it need not be built for 32-bit ARM or kernels without ACPI support. The currently missing__initannotations are added to the prototypes in the header.Signed-off-by Mark Rutland <[email protected]>Acked-by Catalin Marinas <[email protected]>Acked-by Hanjun Guo <[email protected]>Reviewed-by Ashwin Chaugule <[email protected]>Cc Al Stone <[email protected]>Cc Lorenzo Pieralisi <[email protected]>Cc Will Deacon <[email protected]>--- arch/arm64/include/asm/acpi.h | 15 +-------------- arch/arm64/include/asm/psci.h | 12 ++++++++++-- arch/arm64/kernel/acpi.c | 11 +++++++++++ arch/arm64/kernel/psci.c | 4 ++-- 4 files changed, 24 insertions(+), 18 deletions(-)arch/arm64/include/asm/acpi.h +1
-14@@ -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 +10
-2@@ -14,7 +14,15 @@#ifndef __ASM_PSCI_H#define __ASM_PSCI_H -int psci_dt_init (void ); -int psci_acpi_init (void ); +int __init psci_dt_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/kernel/acpi.c +11
-0@@ -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 +2
-2@@ -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> @@ -25,7 +24,6 @@#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> @@ -446,6 +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 @@ -466,6 +465,7 @@ int __init psci_acpi_init(void) return psci_probe(); } +#endif #ifdef CONFIG_SMP
Signature · 8 lines
ciao, al ----------------------------------- Al Stone Software Engineer Linaro Enterprise Group al.stone at linaro.org -----------------------------------
Mark Rutland
maintainerPATCH v3 9/12 arm64: psci: factor invocation code to drivers
In reply to Mark RutlandTo enable sharing with arm, move the core PSCI framework code to drivers/firmware. This results in a minor gain in lines of code, but this will quickly be amortised by the removal of code currently duplicated in arch/arm.
---
arch/arm64/Kconfig | 1 +
arch/arm64/include/asm/acpi.h | 4 +-
arch/arm64/include/asm/psci.h | 28 --
arch/arm64/kernel/psci.c | 361 +------------------------
arch/arm64/kernel/setup.c | 2 +-
drivers/firmware/Kconfig | 3 +
drivers/firmware/Makefile | 1 +
{arch/arm64/kernel => drivers/firmware}/psci.c | 251 +----------------
{arch/arm/include/asm => include/linux}/psci.h | 40 +--
9 files changed, 43 insertions(+), 648 deletions(-)
delete mode 100644 arch/arm64/include/asm/psci.h
copy {arch/arm64/kernel => drivers/firmware}/psci.c (59%)
copy {arch/arm/include/asm => include/linux}/psci.h (55%)arch/arm64/Kconfig +1 -0
@@ -19,6 +19,7 @@ config ARM64 select ARM_GIC_V2M if PCI_MSI select ARM_GIC_V3 select ARM_GIC_V3_ITS if PCI_MSI + select ARM_PSCI_FW select BUILDTIME_EXTABLE_SORT select CLONE_BACKWARDS select COMMON_CLK
arch/arm64/include/asm/acpi.h +2 -2
@@ -12,11 +12,11 @@#ifndef _ASM_ACPI_H#define _ASM_ACPI_H -#include <linux/mm.h> #include <linux/irqchip/arm-gic-acpi.h> +#include <linux/mm.h> +#include <linux/psci.h> #include <asm/cputype.h> -#include <asm/psci.h> #include <asm/smp_plat.h> /* Basic configuration for ACPI */
arch/arm64/include/asm/psci.h deleted +0 -28
@@ -1,28 +0,0 @@ -/* -* This program is free software; you can redistribute it and/or modify -* it under the terms of the GNU General Public License version 2 as -* published by the Free Software Foundation. -* -* This program is distributed in the hope that it will be useful, -* but WITHOUT ANY WARRANTY; without even the implied warranty of -* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -* GNU General Public License for more details. -* -* Copyright (C) 2013 ARM Limited -*/ - -#ifndef __ASM_PSCI_H -#define __ASM_PSCI_H - -int __init psci_dt_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/kernel/psci.c +2 -359
390 lines not on this page · raw message
arch/arm64/kernel/setup.c +1 -1
@@ -46,6 +46,7 @@#include <linux/of_platform.h> #include <linux/efi.h> #include <linux/personality.h> +#include <linux/psci.h> #include <asm/acpi.h> #include <asm/fixmap.h>
@@ -61,7 +62,6 @@#include <asm/tlbflush.h> #include <asm/traps.h> #include <asm/memblock.h> -#include <asm/psci.h> #include <asm/efi.h> #include <asm/virt.h>
drivers/firmware/Kconfig +3 -0
@@ -5,6 +5,9 @@ menu "Firmware Drivers" +config ARM_PSCI_FW + bool + config EDD tristate "BIOS Enhanced Disk Drive calls determine boot disk" depends on X86
drivers/firmware/Makefile +1 -0
@@ -1,6 +1,7 @@ # # Makefile for the linux kernel. # +obj-$(CONFIG_ARM_PSCI_FW) += psci.o obj-$(CONFIG_DMI) += dmi_scan.o obj-$(CONFIG_DMI_SYSFS) += dmi-sysfs.o obj-$(CONFIG_EDD) += edd.o
drivers/firmware/psci.c renamed from arch/arm64/kernel/psci.c +11 -240
298 lines not on this page · raw message
include/linux/psci.h renamed from arch/arm/include/asm/psci.h +22 -18
@@ -8,25 +8,23 @@* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * GNU General Public License for more details. * -* Copyright (C) 2012 ARM Limited +* Copyright (C) 2015 ARM Limited */ -#ifndef __ASM_ARM_PSCI_H -#define __ASM_ARM_PSCI_H +#ifndef __LINUX_PSCI_H +#define __LINUX_PSCI_H + +#include <linux/init.h> +#include <linux/types.h> #define PSCI_POWER_STATE_TYPE_STANDBY 0 #define PSCI_POWER_STATE_TYPE_POWER_DOWN 1 -struct psci_power_state { - u16 id; - u8 type; - u8 affinity_level; -}; +bool psci_tos_resident_on(int cpu); struct psci_operations { - int (*cpu_suspend)(struct psci_power_state state, - unsigned long entry_point); - int (*cpu_off)(struct psci_power_state state); + int (*cpu_suspend)(u32 state, unsigned long entry_point); + int (*cpu_off)(u32 state); int (*cpu_on)(unsigned long cpuid, unsigned long entry_point); int (*migrate)(unsigned long cpuid); int (*affinity_info)(unsigned long target_affinity,
@@ -35,14 +33,20 @@ struct psci_operations { };extern struct psci_operations psci_ops; -extern struct smp_operations psci_smp_ops; -#ifdef CONFIG_ARM_PSCI -int psci_init (void ); -bool psci_smp_available (void ); +#if defined(CONFIG_ARM_PSCI_FW ) +int __init psci_dt_init (void ); +#else +static inline int psci_dt_init (void ) {return 0 ; } +#endif + +#if defined(CONFIG_ARM_PSCI_FW ) && defined(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_init (void ) {return 0 ; } -static inline bool psci_smp_available (void ) {return false; } +static inline int psci_acpi_init (void ) {return 0 ; } +static inline bool acpi_psci_present (void ) {return false; }#endif -#endif /* __ASM_ARM_PSCI_H */ +#endif /* __LINUX_PSCI_H */
Signature · 1 line
1.9.1
Mark Rutland
maintainerPATCH v3 10/12 drivers: psci: support native SMC{32,64} calls
In reply to Mark RutlandA 32-bit OS cannot make calls with SMC64 IDs, while a 64-bit OS must invoke some PSCI functions with SMC64 IDs.
This patch introduces and makes use of a new macro to choose the appropriate IDs based on the register width of the OS, which will allow 32-bit callers to use the PSCI client code.
--- drivers/firmware/psci.c | 25 +++++++++++++++++++------ 1 file changed, 19 insertions(+), 6 deletions(-)
drivers/firmware/psci.c +19 -6
@@ -28,6 +28,18 @@#include <asm/smp_plat.h> /* +* While a 64-bit OS can make calls with SMC32 calling conventions, for some +* calls it is necessary to use SMC64 to pass or return 64-bit values. For such +* calls PSCI_0_2_FN_NATIVE(x) will choose the appropriate (native-width) +* function ID. +*/ +#ifdef CONFIG_64BIT +#define PSCI_0_2_FN_NATIVE (name) PSCI_0_2_FN64_##name +#else +#define PSCI_0_2_FN_NATIVE (name) PSCI_0_2_FN_##name +#endif + +/* * 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
@@ -122,8 +134,8 @@ static int psci_migrate(unsigned long cpuid)static int psci_affinity_info (unsigned long target_affinity,unsigned long lowest_affinity_level) { -return invoke_psci_fn (PSCI_0_2_FN64_AFFINITY_INFO , target_affinity, - lowest_affinity_level,0 ); +return invoke_psci_fn (PSCI_0_2_FN_NATIVE (AFFINITY_INFO ), + target_affinity, lowest_affinity_level,0 ); }static int psci_migrate_info_type (void )
@@ -133,7 +145,8 @@ static int psci_migrate_info_type(void)static unsigned long psci_migrate_info_up_cpu (void ) { -return invoke_psci_fn (PSCI_0_2_FN64_MIGRATE_INFO_UP_CPU ,0 ,0 ,0 ); +return invoke_psci_fn (PSCI_0_2_FN_NATIVE (MIGRATE_INFO_UP_CPU ), +0 ,0 ,0 ); }static int get_set_conduit_method (struct device_node *np)
@@ -211,16 +224,16 @@ static void __init psci_init_migrate(void)static void __init psci_0_2_set_functions (void ) {pr_info ("Using standard PSCI v0.2 function IDs\n" ); - psci_function_id[PSCI_FN_CPU_SUSPEND ] =PSCI_0_2_FN64_CPU_SUSPEND ; + psci_function_id[PSCI_FN_CPU_SUSPEND ] =PSCI_0_2_FN_NATIVE (CPU_SUSPEND ); psci_ops.cpu_suspend = psci_cpu_suspend; psci_function_id[PSCI_FN_CPU_OFF ] =PSCI_0_2_FN_CPU_OFF ; psci_ops.cpu_off = psci_cpu_off; - psci_function_id[PSCI_FN_CPU_ON ] =PSCI_0_2_FN64_CPU_ON ; + psci_function_id[PSCI_FN_CPU_ON ] =PSCI_0_2_FN_NATIVE (CPU_ON ); psci_ops.cpu_on = psci_cpu_on; - psci_function_id[PSCI_FN_MIGRATE ] =PSCI_0_2_FN64_MIGRATE ; + psci_function_id[PSCI_FN_MIGRATE ] =PSCI_0_2_FN_NATIVE (MIGRATE ); psci_ops.migrate = psci_migrate; psci_ops.affinity_info = psci_affinity_info;
Signature · 1 line
1.9.1
Mark Rutland
maintainerPATCH v3 11/12 ARM: migrate to common PSCI client code
In reply to Mark RutlandNow that the common PSCI client code has been factored out to drivers/firmware, and made safe for 32-bit use, move the 32-bit ARM code over to it. This results in a moderate reduction of duplicated lines, and will prevent further duplication as the PSCI client code is updated for PSCI 1.0 and beyond.
The two legacy platform users of the PSCI invocation code are updated to account for interface changes. In both cases the power state parameter is changed to an opaque u32 token in preparation for PSCI 1.0 power state changes.
--- arch/arm/Kconfig | 1 + arch/arm/include/asm/psci.h | 23 --- arch/arm/kernel/Makefile | 2 +- arch/arm/kernel/psci.c | 299 -------------------------------------- arch/arm/kernel/psci_smp.c | 29 +++- arch/arm/kernel/setup.c | 3 +- arch/arm/mach-highbank/highbank.c | 2 +- arch/arm/mach-highbank/pm.c | 8 +- drivers/cpuidle/cpuidle-calxeda.c | 7 +- 9 files changed, 32 insertions(+), 342 deletions(-) delete mode 100644 arch/arm/kernel/psci.c
arch/arm/Kconfig +1 -0
@@ -1465,6 +1465,7 @@ config HOTPLUG_CPU config ARM_PSCI bool "Support for the ARM Power State Coordination Interface (PSCI)" depends on CPU_V7 + select ARM_PSCI_FW help Say Y here if you want Linux to communicate with system firmware implementing the PSCI specification for CPU-centric power
arch/arm/include/asm/psci.h +0 -23
@@ -14,34 +14,11 @@#ifndef __ASM_ARM_PSCI_H#define __ASM_ARM_PSCI_H -#define PSCI_POWER_STATE_TYPE_STANDBY 0 -#define PSCI_POWER_STATE_TYPE_POWER_DOWN 1 - -struct psci_power_state { -u16 id; -u8 type; -u8 affinity_level; -}; - -struct psci_operations { -int (*cpu_suspend)(struct psci_power_state state, -unsigned long entry_point); -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, -unsigned long lowest_affinity_level); -int (*migrate_info_type)(void ); -}; - -extern struct psci_operations psci_ops;extern struct smp_operations psci_smp_ops;#ifdef CONFIG_ARM_PSCI -int psci_init (void );bool psci_smp_available (void );#else -static inline int psci_init (void ) {return 0 ; }static inline bool psci_smp_available (void ) {return false; }#endif
arch/arm/kernel/Makefile +1 -1
@@ -86,7 +86,7 @@ obj-$(CONFIG_EARLY_PRINTK) += early_printk.o obj-$(CONFIG_ARM_VIRT_EXT) += hyp-stub.o ifeq ($(CONFIG_ARM_PSCI),y) -obj-y += psci.o psci-call.o +obj-y += psci-call.o obj-$(CONFIG_SMP) += psci_smp.o endif
arch/arm/kernel/psci.c deleted +0 -299
299 lines not on this page · raw message
arch/arm/kernel/psci_smp.c +22 -7
@@ -17,6 +17,8 @@#include <linux/smp.h> #include <linux/of.h> #include <linux/delay.h> +#include <linux/psci.h> +#include <uapi/linux/psci.h> #include <asm/psci.h>
@@ -56,17 +58,29 @@ static int psci_boot_secondary(unsigned int cpu, struct task_struct *idle) }#ifdef CONFIG_HOTPLUG_CPU +int 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 ; +} +void __ref psci_cpu_die (unsigned int cpu) { -const struct psci_power_state ps = { - .type =PSCI_POWER_STATE_TYPE_POWER_DOWN , - }; +u32 state =PSCI_POWER_STATE_TYPE_POWER_DOWN << +PSCI_0_2_POWER_STATE_TYPE_SHIFT ; -if (psci_ops.cpu_off) - psci_ops.cpu_off (ps); +if (psci_ops.cpu_off) + psci_ops.cpu_off (state); -/* We should never return */ -panic ("psci: cpu %d failed to shutdown\n" , cpu); +/* We should never return */ +panic ("psci: cpu %d failed to shutdown\n" , cpu); }int __ref psci_cpu_kill (unsigned int cpu)
@@ -109,6 +123,7 @@ bool __init psci_smp_available(void)struct smp_operations __initdata psci_smp_ops = { .smp_boot_secondary = psci_boot_secondary, #ifdef CONFIG_HOTPLUG_CPU + .cpu_disable = psci_cpu_disable, .cpu_die = psci_cpu_die, .cpu_kill = psci_cpu_kill, #endif
arch/arm/kernel/setup.c +2 -1
@@ -31,6 +31,7 @@#include <linux/bug.h> #include <linux/compiler.h> #include <linux/sort.h> +#include <linux/psci.h> #include <asm/unified.h> #include <asm/cp15.h>
@@ -950,7 +951,7 @@ void __init setup_arch(char **cmdline_p)unflatten_device_tree ();arm_dt_init_cpu_maps (); -psci_init (); +psci_dt_init ();#ifdef CONFIG_SMP if (is_smp ()) {if (!mdesc->smp_init || !mdesc->smp_init ()) {
arch/arm/mach-highbank/highbank.c +1 -1
@@ -28,8 +28,8 @@#include <linux/reboot.h> #include <linux/amba/bus.h> #include <linux/platform_device.h> +#include <linux/psci.h> -#include <asm/psci.h> #include <asm/hardware/cache-l2x0.h> #include <asm/mach/arch.h> #include <asm/mach/map.h>
arch/arm/mach-highbank/pm.c +2 -6
@@ -16,18 +16,14 @@#include <linux/cpu_pm.h> #include <linux/init.h> +#include <linux/psci.h> #include <linux/suspend.h> #include <asm/suspend.h> -#include <asm/psci.h> static int highbank_suspend_finish (unsigned long val) { -const struct psci_power_state ps = { - .type =PSCI_POWER_STATE_TYPE_POWER_DOWN , - .affinity_level =1 , - }; - +u32 ps =0x01010000 ;/* Aff1 power down */ return psci_ops.cpu_suspend (ps,__pa (cpu_resume)); }
drivers/cpuidle/cpuidle-calxeda.c +3 -4
@@ -25,15 +25,14 @@#include <linux/init.h> #include <linux/mm.h> #include <linux/platform_device.h> +#include <linux/psci.h> +#include <asm/cpuidle.h> #include <asm/suspend.h> -#include <asm/psci.h> static int calxeda_idle_finish (unsigned long val) { -const struct psci_power_state ps = { - .type =PSCI_POWER_STATE_TYPE_POWER_DOWN , - }; +u32 ps =0x00010000 ;/* Aff0 power down */ return psci_ops.cpu_suspend (ps,__pa (cpu_resume)); }
Signature · 1 line
1.9.1
PATCH v3 11/12 ARM: migrate to common PSCI client code
Quoted 18 lines · Mark Rutland
Now that the common PSCI client code has been factored out to
drivers/firmware, and made safe for 32-bit use, move the 32-bit ARM code over to it. This results in a moderate reduction of duplicated lines, and will prevent further duplication as the PSCI client code is updated for PSCI 1.0 and beyond.The two legacy platform users of the PSCI invocation code are updated to account for interface changes. In both cases the power state parameter is changed to an opaque u32 token in preparation for PSCI 1.0 power state changes.
Signed-off-by Mark Rutland <[email protected]>Acked-by Catalin Marinas <[email protected]>Cc Ashwin Chaugule <[email protected]>Cc Lorenzo Pieralisi <[email protected]>Cc Rob Herring <[email protected]>Cc Russell King <[email protected]>Cc Will Deacon <[email protected]>
Quoted 12 lines
--- arch/arm/Kconfig | 1 + arch/arm/include/asm/psci.h | 23 --- arch/arm/kernel/Makefile | 2 +- arch/arm/kernel/psci.c | 299 -------------------------------------- arch/arm/kernel/psci_smp.c | 29 +++- arch/arm/kernel/setup.c | 3 +- arch/arm/mach-highbank/highbank.c | 2 +- arch/arm/mach-highbank/pm.c | 8 +- drivers/cpuidle/cpuidle-calxeda.c | 7 +- 9 files changed, 32 insertions(+), 342 deletions(-) delete mode 100644 arch/arm/kernel/psci.c
Mark Rutland
maintainerPATCH v3 11/12 ARM: migrate to common PSCI client code
Quoted 32 lines · Ashwin Chaugule
Quoted 18 lines · Mark Rutland
Now that the common PSCI client code has been factored out to
drivers/firmware, and made safe for 32-bit use, move the 32-bit ARM code over to it. This results in a moderate reduction of duplicated lines, and will prevent further duplication as the PSCI client code is updated for PSCI 1.0 and beyond.The two legacy platform users of the PSCI invocation code are updated to account for interface changes. In both cases the power state parameter is changed to an opaque u32 token in preparation for PSCI 1.0 power state changes.
Signed-off-by Mark Rutland <[email protected]>Acked-by Catalin Marinas <[email protected]>Cc Ashwin Chaugule <[email protected]>Cc Lorenzo Pieralisi <[email protected]>Cc Rob Herring <[email protected]>Cc Russell King <[email protected]>Cc Will Deacon <[email protected]>Quoted 12 lines
--- arch/arm/Kconfig | 1 + arch/arm/include/asm/psci.h | 23 --- arch/arm/kernel/Makefile | 2 +- arch/arm/kernel/psci.c | 299 -------------------------------------- arch/arm/kernel/psci_smp.c | 29 +++- arch/arm/kernel/setup.c | 3 +- arch/arm/mach-highbank/highbank.c | 2 +- arch/arm/mach-highbank/pm.c | 8 +- drivers/cpuidle/cpuidle-calxeda.c | 7 +- 9 files changed, 32 insertions(+), 342 deletions(-) delete mode 100644 arch/arm/kernel/psci.c
Quoted 1 line
Reviewed-by Ashwin Chaugule <[email protected]>
Cheers!
Mark.
PATCH v3 11/12 ARM: migrate to common PSCI client code
In reply to Mark RutlandQuoted 10 lines · Mark Rutland
Now that the common PSCI client code has been factored out to
drivers/firmware, and made safe for 32-bit use, move the 32-bit ARM code over to it. This results in a moderate reduction of duplicated lines, and will prevent further duplication as the PSCI client code is updated for PSCI 1.0 and beyond.The two legacy platform users of the PSCI invocation code are updated to account for interface changes. In both cases the power state parameter is changed to an opaque u32 token in preparation for PSCI 1.0 power state changes.
[...]
Quoted 26 lines
arch/arm/mach-highbank/highbank.c +1
-1@@ -28,8 +28,8 @@#include <linux/reboot.h> #include <linux/amba/bus.h> #include <linux/platform_device.h> +#include <linux/psci.h> -#include <asm/psci.h> #include <asm/hardware/cache-l2x0.h> #include <asm/mach/arch.h> #include <asm/mach/map.h> arch/arm/mach-highbank/pm.c +1
-0@@ -16,18 +16,14 @@#include <linux/cpu_pm.h> #include <linux/init.h> +#include <linux/psci.h> #include <linux/suspend.h> #include <asm/suspend.h>
arch/arm/mach-highbank/pm.c line 26
Quoted 10 lines
arch/arm/mach-highbank/pm.c +1
-6-#include <asm/psci.h> static int highbank_suspend_finish (unsigned long val) { -const struct psci_power_state ps = { - .type =PSCI_POWER_STATE_TYPE_POWER_DOWN , - .affinity_level =1 , - }; - +u32 ps =0x01010000 ;/* Aff1 power down */
Why not just pass directly instead of sticking in a variable first.
Quoted 13 lines
arch/arm/mach-highbank/pm.c +0
-0return psci_ops.cpu_suspend (ps,__pa (cpu_resume)); }drivers/cpuidle/cpuidle-calxeda.c +2
-0@@ -25,15 +25,14 @@#include <linux/init.h> #include <linux/mm.h> #include <linux/platform_device.h> +#include <linux/psci.h> +
drivers/cpuidle/cpuidle-calxeda.c line 35
Quoted 10 lines
drivers/cpuidle/cpuidle-calxeda.c +1
-4#include <asm/cpuidle.h> #include <asm/suspend.h> -#include <asm/psci.h> static int calxeda_idle_finish (unsigned long val) { -const struct psci_power_state ps = { - .type =PSCI_POWER_STATE_TYPE_POWER_DOWN , - }; +u32 ps =0x00010000 ;/* Aff0 power down */
ditto.
Quoted 6 lines
drivers/cpuidle/cpuidle-calxeda.c +0
-1return psci_ops.cpu_suspend (ps,__pa (cpu_resume)); } --1.9.1
Mark Rutland
maintainerPATCH v3 11/12 ARM: migrate to common PSCI client code
Quoted 42 lines · Rob Herring
Quoted 10 lines · Mark Rutland
Now that the common PSCI client code has been factored out to
drivers/firmware, and made safe for 32-bit use, move the 32-bit ARM code over to it. This results in a moderate reduction of duplicated lines, and will prevent further duplication as the PSCI client code is updated for PSCI 1.0 and beyond.The two legacy platform users of the PSCI invocation code are updated to account for interface changes. In both cases the power state parameter is changed to an opaque u32 token in preparation for PSCI 1.0 power state changes.
[...]
Quoted 28 lines
arch/arm/mach-highbank/highbank.c +1
-1@@ -28,8 +28,8 @@#include <linux/reboot.h> #include <linux/amba/bus.h> #include <linux/platform_device.h> +#include <linux/psci.h> -#include <asm/psci.h> #include <asm/hardware/cache-l2x0.h> #include <asm/mach/arch.h> #include <asm/mach/map.h> arch/arm/mach-highbank/pm.c +1
-1@@ -16,18 +16,14 @@#include <linux/cpu_pm.h> #include <linux/init.h> +#include <linux/psci.h> #include <linux/suspend.h> #include <asm/suspend.h> -#include <asm/psci.h>
Quoted 10 lines
arch/arm/mach-highbank/pm.c line 26
Quoted 8 lines
arch/arm/mach-highbank/pm.c +1
-5static int highbank_suspend_finish (unsigned long val) { -const struct psci_power_state ps = { - .type =PSCI_POWER_STATE_TYPE_POWER_DOWN , - .affinity_level =1 , - }; - +u32 ps =0x01010000 ;/* Aff1 power down */ Why not just pass directly instead of sticking in a variable first.
No particular reason. I'll fold it in.
Quoted 17 lines
Quoted 17 lines
arch/arm/mach-highbank/pm.c +0
-0return psci_ops.cpu_suspend (ps,__pa (cpu_resume)); }drivers/cpuidle/cpuidle-calxeda.c +2
-1@@ -25,15 +25,14 @@#include <linux/init.h> #include <linux/mm.h> #include <linux/platform_device.h> +#include <linux/psci.h> +#include <asm/cpuidle.h> #include <asm/suspend.h> -#include <asm/psci.h>
Quoted 8 lines
drivers/cpuidle/cpuidle-calxeda.c line 35
Quoted 6 lines
drivers/cpuidle/cpuidle-calxeda.c +1
-3static int calxeda_idle_finish (unsigned long val) { -const struct psci_power_state ps = { - .type =PSCI_POWER_STATE_TYPE_POWER_DOWN , - }; +u32 ps =0x00010000 ;/* Aff0 power down */ ditto.
Likewise.
Thanks,
Mark.
Quoted 8 lines
Quoted 7 lines
drivers/cpuidle/cpuidle-calxeda.c +0
-1return psci_ops.cpu_suspend (ps,__pa (cpu_resume)); } --1.9.1
Mark Rutland
maintainerPATCH v3 11/12 ARM: migrate to common PSCI client code
Hi Rob,
Quoted 8 lines · Mark Rutland
Quoted 7 lines · Rob Herring
Quoted 6 lines · Mark Rutland
Now that the common PSCI client code has been factored out to
drivers/firmware, and made safe for 32-bit use, move the 32-bit ARM code over to it. This results in a moderate reduction of duplicated lines, and will prevent further duplication as the PSCI client code is updated for PSCI 1.0 and beyond.
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
The two legacy platform users of the PSCI invocation code are updated to account for interface changes. In both cases the power state parameter is changed to an opaque u32 token in preparation for PSCI 1.0 power state changes.
Other than the changes you requested below, are you happy with the patch?
Are you happy to ack the patch with those changes folded in?
Mark.
Quoted 58 lines
Quoted 10 lines
arch/arm/mach-highbank/pm.c line 26
Quoted 8 lines
static int
highbank_suspend_finish(unsigned long val){arch/arm/mach-highbank/pm.c +1
-5-const struct psci_power_state ps = { - .type =PSCI_POWER_STATE_TYPE_POWER_DOWN , - .affinity_level =1 , - }; - +u32 ps =0x01010000 ;/* Aff1 power down */ Why not just pass directly instead of sticking in a variable first.
No particular reason. I'll fold it in.
Quoted 17 lines
Quoted 17 lines
return psci_ops.cpu_suspend(ps, __pa(cpu_resume));}
drivers/cpuidle/cpuidle-calxeda.c +2
-1@@ -25,15 +25,14 @@#include <linux/init.h> #include <linux/mm.h> #include <linux/platform_device.h> +#include <linux/psci.h> +#include <asm/cpuidle.h> #include <asm/suspend.h> -#include <asm/psci.h> Quoted 8 lines
drivers/cpuidle/cpuidle-calxeda.c line 35
Quoted 6 lines
drivers/cpuidle/cpuidle-calxeda.c +1
-3static int calxeda_idle_finish (unsigned long val) { -const struct psci_power_state ps = { - .type =PSCI_POWER_STATE_TYPE_POWER_DOWN , - }; +u32 ps =0x00010000 ;/* Aff0 power down */ ditto.
Likewise.
Thanks,
Mark.Quoted 8 lines
Quoted 7 lines
drivers/cpuidle/cpuidle-calxeda.c +0
-1return psci_ops.cpu_suspend (ps,__pa (cpu_resume)); } --1.9.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel at lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
Mark Rutland
maintainerPATCH v3 11/12 ARM: migrate to common PSCI client code
In reply to Mark RutlandHi Russell,
Quoted 18 lines · Mark Rutland
Now that the common PSCI client code has been factored out to
drivers/firmware, and made safe for 32-bit use, move the 32-bit ARM code over to it. This results in a moderate reduction of duplicated lines, and will prevent further duplication as the PSCI client code is updated for PSCI 1.0 and beyond.The two legacy platform users of the PSCI invocation code are updated to account for interface changes. In both cases the power state parameter is changed to an opaque u32 token in preparation for PSCI 1.0 power state changes.
Signed-off-by Mark Rutland <[email protected]>Acked-by Catalin Marinas <[email protected]>Cc Ashwin Chaugule <[email protected]>Cc Lorenzo Pieralisi <[email protected]>Cc Rob Herring <[email protected]>Cc Russell King <[email protected]>Cc Will Deacon <[email protected]>
Quoted 12 lines
--- arch/arm/Kconfig | 1 + arch/arm/include/asm/psci.h | 23 --- arch/arm/kernel/Makefile | 2 +- arch/arm/kernel/psci.c | 299 -------------------------------------- arch/arm/kernel/psci_smp.c | 29 +++- arch/arm/kernel/setup.c | 3 +- arch/arm/mach-highbank/highbank.c | 2 +- arch/arm/mach-highbank/pm.c | 8 +- drivers/cpuidle/cpuidle-calxeda.c | 7 +- 9 files changed, 32 insertions(+), 342 deletions(-) delete mode 100644 arch/arm/kernel/psci.c
Do you have any objections to the arch/arm changes in this patch?
If not, could I please have your ack?
As there are arm, arm64, and drivers changes in the series I was hoping this could all go via arm-soc.
Thanks,
Mark.
Quoted 513 lines
arch/arm/Kconfig +1
-0@@ -1465,6 +1465,7 @@ config HOTPLUG_CPU config ARM_PSCI bool "Support for the ARM Power State Coordination Interface (PSCI)" depends on CPU_V7 + select ARM_PSCI_FW help Say Y here if you want Linux to communicate with system firmware implementing the PSCI specification for CPU-centric powerarch/arm/include/asm/psci.h +0
-23@@ -14,34 +14,11 @@#ifndef __ASM_ARM_PSCI_H#define __ASM_ARM_PSCI_H -#define PSCI_POWER_STATE_TYPE_STANDBY 0 -#define PSCI_POWER_STATE_TYPE_POWER_DOWN 1 - -struct psci_power_state { -u16 id; -u8 type; -u8 affinity_level; -}; - -struct psci_operations { -int (*cpu_suspend)(struct psci_power_state state, -unsigned long entry_point); -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, -unsigned long lowest_affinity_level); -int (*migrate_info_type)(void ); -}; - -extern struct psci_operations psci_ops;extern struct smp_operations psci_smp_ops;#ifdef CONFIG_ARM_PSCI -int psci_init (void );bool psci_smp_available (void );#else -static inline int psci_init (void ) {return 0 ; }static inline bool psci_smp_available (void ) {return false; }#endif arch/arm/kernel/Makefile +1
-1@@ -86,7 +86,7 @@ obj-$(CONFIG_EARLY_PRINTK) += early_printk.o obj-$(CONFIG_ARM_VIRT_EXT) += hyp-stub.o ifeq ($(CONFIG_ARM_PSCI),y) -obj-y += psci.o psci-call.o +obj-y += psci-call.o obj-$(CONFIG_SMP) += psci_smp.o endifarch/arm/kernel/psci.c deleted +0
-299@@ -1,299 +0,0 @@ -/* -* This program is free software; you can redistribute it and/or modify -* it under the terms of the GNU General Public License version 2 as -* published by the Free Software Foundation. -* -* This program is distributed in the hope that it will be useful, -* but WITHOUT ANY WARRANTY; without even the implied warranty of -* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the -* GNU General Public License for more details. -* -* Copyright (C) 2012 ARM Limited -* -* Author: Will Deacon <[email protected]> -*/ - -#define pr_fmt (fmt) "psci: " fmt - -#include <linux/init.h> -#include <linux/of.h> -#include <linux/reboot.h> -#include <linux/pm.h> -#include <uapi/linux/psci.h> - -#include <asm/compiler.h> -#include <asm/errno.h> -#include <asm/psci.h> -#include <asm/system_misc.h> - -struct psci_operations psci_ops; - -static int (*invoke_psci_fn)(u32 ,u32 ,u32 ,u32 ); -typedef int (*psci_initcall_t )(const struct device_node *); - -asmlinkage int __invoke_psci_fn_hvc(u32, u32, u32, u32); -asmlinkage int __invoke_psci_fn_smc(u32, u32, u32, u32); - -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 u32 psci_function_id[PSCI_FN_MAX ]; - -static int psci_to_linux_errno (int errno) -{ -switch (errno) { -case PSCI_RET_SUCCESS : -return 0 ; -case PSCI_RET_NOT_SUPPORTED : -return -EOPNOTSUPP ; -case PSCI_RET_INVALID_PARAMS : -return -EINVAL ; -case PSCI_RET_DENIED : -return -EPERM ; - }; - -return -EINVAL ; -} - -static u32 psci_power_state_pack (struct psci_power_state state) -{ -return ((state.id <<PSCI_0_2_POWER_STATE_ID_SHIFT ) - &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 int psci_get_version (void ) -{ -int err; - - 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, power_state, entry_point,0 ); -return psci_to_linux_errno (err); -} - -static int psci_cpu_off (struct psci_power_state state) -{ -int err; -u32 fn, power_state; - - fn = psci_function_id[PSCI_FN_CPU_OFF ]; - power_state =psci_power_state_pack (state); - err =invoke_psci_fn (fn, power_state,0 ,0 ); -return psci_to_linux_errno (err); -} - -static int psci_cpu_on (unsigned long cpuid,unsigned long entry_point) -{ -int err; -u32 fn; - - fn = psci_function_id[PSCI_FN_CPU_ON ]; - err =invoke_psci_fn (fn, cpuid, entry_point,0 ); -return psci_to_linux_errno (err); -} - -static int psci_migrate (unsigned long cpuid) -{ -int err; -u32 fn; - - fn = psci_function_id[PSCI_FN_MIGRATE ]; - err =invoke_psci_fn (fn, cpuid,0 ,0 ); -return psci_to_linux_errno (err); -} - -static int psci_affinity_info (unsigned long target_affinity, -unsigned long lowest_affinity_level) -{ -int err; -u32 fn; - - 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; -u32 fn; - - fn = psci_function_id[PSCI_FN_MIGRATE_INFO_TYPE ]; - err =invoke_psci_fn (fn,0 ,0 ,0 ); -return err; -} - -static int get_set_conduit_method (struct device_node *np) -{ -const char *method; - -pr_info ("probing for conduit method from DT.\n" ); - -if (of_property_read_string (np,"method" , &method)) { -pr_warn ("missing \"method\" property\n" ); -return -ENXIO ; - } - -if (!strcmp ("hvc" , method)) { - invoke_psci_fn = __invoke_psci_fn_hvc; - }else if (!strcmp ("smc" , method)) { - invoke_psci_fn = __invoke_psci_fn_smc; - }else { -pr_warn ("invalid \"method\" property: %s\n" , method); -return -EINVAL ; - } -return 0 ; -} - -static void psci_sys_reset (enum reboot_mode reboot_mode,const char *cmd) -{ -invoke_psci_fn (PSCI_0_2_FN_SYSTEM_RESET ,0 ,0 ,0 ); -} - -static void psci_sys_poweroff (void ) -{ -invoke_psci_fn (PSCI_0_2_FN_SYSTEM_OFF ,0 ,0 ,0 ); -} - -/* -* PSCI Function IDs for v0.2+ are well defined so use -* standard values. -*/ -static int psci_0_2_init (struct device_node *np) -{ -int err, ver; - - err =get_set_conduit_method (np); - -if (err) - goto out_put_node; - - ver =psci_get_version (); - -if (ver ==PSCI_RET_NOT_SUPPORTED ) { -/* PSCI v0.2 mandates implementation of PSCI_ID_VERSION. */ -pr_err ("PSCI firmware does not comply with the v0.2 spec.\n" ); - err = -EOPNOTSUPP ; - goto out_put_node; - }else { -pr_info ("PSCIv%d.%d detected in firmware.\n" , -PSCI_VERSION_MAJOR (ver), -PSCI_VERSION_MINOR (ver)); - -if (PSCI_VERSION_MAJOR (ver) ==0 && -PSCI_VERSION_MINOR (ver) <2 ) { - err = -EINVAL ; -pr_err ("Conflicting PSCI version detected.\n" ); - goto out_put_node; - } - } - -pr_info ("Using standard PSCI v0.2 function IDs\n" ); - psci_function_id[PSCI_FN_CPU_SUSPEND ] =PSCI_0_2_FN_CPU_SUSPEND ; - psci_ops.cpu_suspend = psci_cpu_suspend; - - psci_function_id[PSCI_FN_CPU_OFF ] =PSCI_0_2_FN_CPU_OFF ; - psci_ops.cpu_off = psci_cpu_off; - - psci_function_id[PSCI_FN_CPU_ON ] =PSCI_0_2_FN_CPU_ON ; - psci_ops.cpu_on = psci_cpu_on; - - psci_function_id[PSCI_FN_MIGRATE ] =PSCI_0_2_FN_MIGRATE ; - psci_ops.migrate = psci_migrate; - - psci_function_id[PSCI_FN_AFFINITY_INFO ] =PSCI_0_2_FN_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; - - pm_power_off = psci_sys_poweroff; - -out_put_node: -of_node_put (np); -return err; -} - -/* -* PSCI < v0.2 get PSCI Function IDs via DT. -*/ -static int psci_0_1_init (struct device_node *np) -{ -u32 id; -int err; - - err =get_set_conduit_method (np); - -if (err) - goto out_put_node; - -pr_info ("Using PSCI v0.1 Function IDs from DT\n" ); - -if (!of_property_read_u32 (np,"cpu_suspend" , &id)) { - psci_function_id[PSCI_FN_CPU_SUSPEND ] = id; - psci_ops.cpu_suspend = psci_cpu_suspend; - } - -if (!of_property_read_u32 (np,"cpu_off" , &id)) { - psci_function_id[PSCI_FN_CPU_OFF ] = id; - psci_ops.cpu_off = psci_cpu_off; - } - -if (!of_property_read_u32 (np,"cpu_on" , &id)) { - psci_function_id[PSCI_FN_CPU_ON ] = id; - psci_ops.cpu_on = psci_cpu_on; - } - -if (!of_property_read_u32 (np,"migrate" , &id)) { - psci_function_id[PSCI_FN_MIGRATE ] = id; - psci_ops.migrate = psci_migrate; - } - -out_put_node: -of_node_put (np); -return err; -} - -static const struct of_device_id psci_of_match[]__initconst = { - { .compatible ="arm,psci" , .data = psci_0_1_init}, - { .compatible ="arm,psci-0.2" , .data = psci_0_2_init}, - {}, -}; - -int __init psci_init (void ) -{ -struct device_node *np; -const struct of_device_id *matched_np; -psci_initcall_t init_fn; - - np =of_find_matching_node_and_match (NULL , psci_of_match, &matched_np); -if (!np) -return -ENODEV ; - - init_fn = (psci_initcall_t )matched_np->data; -return init_fn (np); -}arch/arm/kernel/psci_smp.c +22
-7@@ -17,6 +17,8 @@#include <linux/smp.h> #include <linux/of.h> #include <linux/delay.h> +#include <linux/psci.h> +#include <uapi/linux/psci.h> #include <asm/psci.h> @@ -56,17 +58,29 @@ static int psci_boot_secondary(unsigned int cpu, struct task_struct *idle) }#ifdef CONFIG_HOTPLUG_CPU +int 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 ; +} +void __ref psci_cpu_die (unsigned int cpu) { -const struct psci_power_state ps = { - .type =PSCI_POWER_STATE_TYPE_POWER_DOWN , - }; +u32 state =PSCI_POWER_STATE_TYPE_POWER_DOWN << +PSCI_0_2_POWER_STATE_TYPE_SHIFT ; -if (psci_ops.cpu_off) - psci_ops.cpu_off (ps); +if (psci_ops.cpu_off) + psci_ops.cpu_off (state); -/* We should never return */ -panic ("psci: cpu %d failed to shutdown\n" , cpu); +/* We should never return */ +panic ("psci: cpu %d failed to shutdown\n" , cpu); }int __ref psci_cpu_kill (unsigned int cpu)@@ -109,6 +123,7 @@ bool __init psci_smp_available(void)struct smp_operations __initdata psci_smp_ops = { .smp_boot_secondary = psci_boot_secondary, #ifdef CONFIG_HOTPLUG_CPU + .cpu_disable = psci_cpu_disable, .cpu_die = psci_cpu_die, .cpu_kill = psci_cpu_kill, #endifarch/arm/kernel/setup.c +2
-1@@ -31,6 +31,7 @@#include <linux/bug.h> #include <linux/compiler.h> #include <linux/sort.h> +#include <linux/psci.h> #include <asm/unified.h> #include <asm/cp15.h> @@ -950,7 +951,7 @@ void __init setup_arch(char **cmdline_p)unflatten_device_tree ();arm_dt_init_cpu_maps (); -psci_init (); +psci_dt_init ();#ifdef CONFIG_SMP if (is_smp ()) {if (!mdesc->smp_init || !mdesc->smp_init ()) {arch/arm/mach-highbank/highbank.c +1
-1@@ -28,8 +28,8 @@#include <linux/reboot.h> #include <linux/amba/bus.h> #include <linux/platform_device.h> +#include <linux/psci.h> -#include <asm/psci.h> #include <asm/hardware/cache-l2x0.h> #include <asm/mach/arch.h> #include <asm/mach/map.h> arch/arm/mach-highbank/pm.c +2
-6@@ -16,18 +16,14 @@#include <linux/cpu_pm.h> #include <linux/init.h> +#include <linux/psci.h> #include <linux/suspend.h> #include <asm/suspend.h> -#include <asm/psci.h> static int highbank_suspend_finish (unsigned long val) { -const struct psci_power_state ps = { - .type =PSCI_POWER_STATE_TYPE_POWER_DOWN , - .affinity_level =1 , - }; - +u32 ps =0x01010000 ;/* Aff1 power down */ return psci_ops.cpu_suspend (ps,__pa (cpu_resume)); }drivers/cpuidle/cpuidle-calxeda.c +3
-5@@ -25,15 +25,14 @@#include <linux/init.h> #include <linux/mm.h> #include <linux/platform_device.h> +#include <linux/psci.h> +#include <asm/cpuidle.h> #include <asm/suspend.h> -#include <asm/psci.h> static int calxeda_idle_finish (unsigned long val) { -const struct psci_power_state ps = { - .type =PSCI_POWER_STATE_TYPE_POWER_DOWN , - }; +u32 ps =0x00010000 ;/* Aff0 power down */ return psci_ops.cpu_suspend (ps,__pa (cpu_resume)); } --1.9.1
Catalin Marinas
maintainerPATCH v3 11/12 ARM: migrate to common PSCI client code
Hi Mark,
Quoted 32 lines · Mark Rutland
Quoted 18 lines · Mark Rutland
Now that the common PSCI client code has been factored out to
drivers/firmware, and made safe for 32-bit use, move the 32-bit ARM code over to it. This results in a moderate reduction of duplicated lines, and will prevent further duplication as the PSCI client code is updated for PSCI 1.0 and beyond.The two legacy platform users of the PSCI invocation code are updated to account for interface changes. In both cases the power state parameter is changed to an opaque u32 token in preparation for PSCI 1.0 power state changes.
Signed-off-by Mark Rutland <[email protected]>Acked-by Catalin Marinas <[email protected]>Cc Ashwin Chaugule <[email protected]>Cc Lorenzo Pieralisi <[email protected]>Cc Rob Herring <[email protected]>Cc Russell King <[email protected]>Cc Will Deacon <[email protected]>Quoted 12 lines
--- arch/arm/Kconfig | 1 + arch/arm/include/asm/psci.h | 23 --- arch/arm/kernel/Makefile | 2 +- arch/arm/kernel/psci.c | 299 -------------------------------------- arch/arm/kernel/psci_smp.c | 29 +++- arch/arm/kernel/setup.c | 3 +- arch/arm/mach-highbank/highbank.c | 2 +- arch/arm/mach-highbank/pm.c | 8 +- drivers/cpuidle/cpuidle-calxeda.c | 7 +- 9 files changed, 32 insertions(+), 342 deletions(-) delete mode 100644 arch/arm/kernel/psci.c
Quoted 6 lines
Do you have any objections to the
arch/armchanges in this patch?If not, could I please have your ack?
As there are arm, arm64, and drivers changes in the series I was hoping this could all go via arm-soc.
If we don't hear from Russell by the end of today, I propose that we merge patches 1-8 in the arm64 tree (mostly fixes for PSCI). This series is blocking other developments like PSCI 1.0, so I really want to get parts of it merged in 4.2.
BTW, can we skip patch 11 in this series and still allow drivers/firmware/psci.c only for arm64, with the aim of moving the arm32 code later?
Thanks.
Signature · 1 line
Catalin
Mark Rutland
maintainerPATCH v3 11/12 ARM: migrate to common PSCI client code
Quoted 39 lines · Catalin Marinas
Hi Mark,
Quoted 36 lines · Mark Rutland
Quoted 18 lines · Mark Rutland
Now that the common PSCI client code has been factored out to
drivers/firmware, and made safe for 32-bit use, move the 32-bit ARM code over to it. This results in a moderate reduction of duplicated lines, and will prevent further duplication as the PSCI client code is updated for PSCI 1.0 and beyond.The two legacy platform users of the PSCI invocation code are updated to account for interface changes. In both cases the power state parameter is changed to an opaque u32 token in preparation for PSCI 1.0 power state changes.
Signed-off-by Mark Rutland <[email protected]>Acked-by Catalin Marinas <[email protected]>Cc Ashwin Chaugule <[email protected]>Cc Lorenzo Pieralisi <[email protected]>Cc Rob Herring <[email protected]>Cc Russell King <[email protected]>Cc Will Deacon <[email protected]>Quoted 12 lines
--- arch/arm/Kconfig | 1 + arch/arm/include/asm/psci.h | 23 --- arch/arm/kernel/Makefile | 2 +- arch/arm/kernel/psci.c | 299 -------------------------------------- arch/arm/kernel/psci_smp.c | 29 +++- arch/arm/kernel/setup.c | 3 +- arch/arm/mach-highbank/highbank.c | 2 +- arch/arm/mach-highbank/pm.c | 8 +- drivers/cpuidle/cpuidle-calxeda.c | 7 +- 9 files changed, 32 insertions(+), 342 deletions(-) delete mode 100644 arch/arm/kernel/psci.cDo you have any objections to the
arch/armchanges in this patch?If not, could I please have your ack?
Quoted 7 lines
Quoted 2 lines
As there are arm, arm64, and drivers changes in the series I was hoping this could all go via arm-soc.
If we don't hear from Russell by the end of today, I propose that we merge patches 1-8 in the arm64 tree (mostly fixes for PSCI). This series is blocking other developments like PSCI 1.0, so I really want to get parts of it merged in 4.2.
I've pushed the first 8 patches to my arm64/psci-rework branch [1]. I've pushed the whole series with the tags provided today to psci/rework [2].
I tried a merge with the arm64 for-next/core branch locally, and got a trivial conflict in smp.c against the changes in "arm64: Use common outgoing-CPU-notification code". Hopefully that's easy to fix up locally.
Quoted 3 lines
BTW, can we skip patch 11 in this series and still allow
drivers/firmware/psci.conly for arm64, with the aim of moving the arm32 code later?
We could, though I would prefer to have the two unified from the beginning.
Thanks,
Mark.
[1] git://git.kernel.org/pub/scm/linux/kernel/git/mark/linux.git arm64/psci-rework
[2] git://git.kernel.org/pub/scm/linux/kernel/git/mark/linux.git psci/rework
Catalin Marinas
maintainerPATCH v3 11/12 ARM: migrate to common PSCI client code
Quoted 47 lines · Mark Rutland
Quoted 37 lines · Catalin Marinas
Quoted 36 lines · Mark Rutland
Quoted 18 lines · Mark Rutland
Now that the common PSCI client code has been factored out to
drivers/firmware, and made safe for 32-bit use, move the 32-bit ARM code over to it. This results in a moderate reduction of duplicated lines, and will prevent further duplication as the PSCI client code is updated for PSCI 1.0 and beyond.The two legacy platform users of the PSCI invocation code are updated to account for interface changes. In both cases the power state parameter is changed to an opaque u32 token in preparation for PSCI 1.0 power state changes.
Signed-off-by Mark Rutland <[email protected]>Acked-by Catalin Marinas <[email protected]>Cc Ashwin Chaugule <[email protected]>Cc Lorenzo Pieralisi <[email protected]>Cc Rob Herring <[email protected]>Cc Russell King <[email protected]>Cc Will Deacon <[email protected]>Quoted 12 lines
--- arch/arm/Kconfig | 1 + arch/arm/include/asm/psci.h | 23 --- arch/arm/kernel/Makefile | 2 +- arch/arm/kernel/psci.c | 299 -------------------------------------- arch/arm/kernel/psci_smp.c | 29 +++- arch/arm/kernel/setup.c | 3 +- arch/arm/mach-highbank/highbank.c | 2 +- arch/arm/mach-highbank/pm.c | 8 +- drivers/cpuidle/cpuidle-calxeda.c | 7 +- 9 files changed, 32 insertions(+), 342 deletions(-) delete mode 100644 arch/arm/kernel/psci.cDo you have any objections to the
arch/armchanges in this patch?If not, could I please have your ack?
Quoted 7 lines
Quoted 2 lines
As there are arm, arm64, and drivers changes in the series I was hoping this could all go via arm-soc.
If we don't hear from Russell by the end of today, I propose that we merge patches 1-8 in the arm64 tree (mostly fixes for PSCI). This series is blocking other developments like PSCI 1.0, so I really want to get parts of it merged in 4.2.
I've pushed the first 8 patches to my arm64/psci-rework branch [1].
...
Quoted 1 line
[1]
git://git.kernel.org/pub/scm/linux/kernel/git/mark/linux.gitarm64/psci-rework
Thanks. I pulled this into the arm64 for-next/core branch.
Quoted 2 lines
I've pushed the whole series with the tags provided today to psci/rework [2].
...
Quoted 1 line
[2]
git://git.kernel.org/pub/scm/linux/kernel/git/mark/linux.gitpsci/rework
As long as you don't rebase arm64/psci-rework, you can send this branch to arm-soc once Russell acks the arm part.
Thanks.
Signature · 1 line
Catalin
PATCH v3 11/12 ARM: migrate to common PSCI client code
In reply to Mark RutlandQuoted 34 lines · Mark Rutland
Hi Russell,
Quoted 18 lines · Mark Rutland
Now that the common PSCI client code has been factored out to
drivers/firmware, and made safe for 32-bit use, move the 32-bit ARM code over to it. This results in a moderate reduction of duplicated lines, and will prevent further duplication as the PSCI client code is updated for PSCI 1.0 and beyond.The two legacy platform users of the PSCI invocation code are updated to account for interface changes. In both cases the power state parameter is changed to an opaque u32 token in preparation for PSCI 1.0 power state changes.
Signed-off-by Mark Rutland <[email protected]>Acked-by Catalin Marinas <[email protected]>Cc Ashwin Chaugule <[email protected]>Cc Lorenzo Pieralisi <[email protected]>Cc Rob Herring <[email protected]>Cc Russell King <[email protected]>Cc Will Deacon <[email protected]>Quoted 12 lines
--- arch/arm/Kconfig | 1 + arch/arm/include/asm/psci.h | 23 --- arch/arm/kernel/Makefile | 2 +- arch/arm/kernel/psci.c | 299 -------------------------------------- arch/arm/kernel/psci_smp.c | 29 +++- arch/arm/kernel/setup.c | 3 +- arch/arm/mach-highbank/highbank.c | 2 +- arch/arm/mach-highbank/pm.c | 8 +- drivers/cpuidle/cpuidle-calxeda.c | 7 +- 9 files changed, 32 insertions(+), 342 deletions(-) delete mode 100644 arch/arm/kernel/psci.c
Quoted 6 lines
Do you have any objections to the
arch/armchanges in this patch?If not, could I please have your ack?
As there are arm, arm64, and drivers changes in the series I was hoping this could all go via arm-soc.
If you want me to do something, please ensure that you send stuff _To:_ me, not _Cc:_ me... I can't say when I'll be able to look at them today, but hopefully tomorrow.
Signature · 2 lines
FTTC broadband for 0.8mile line: currently at 10.5Mbps down 400kbps up according to speedtest.net.
Mark Rutland
maintainerPATCH v3 12/12 MAINTAINERS: add PSCI entry
In reply to Mark RutlandAdd myself and Lorenzo as maintainers of the PSCI client code.
--- MAINTAINERS | 9 +++++++++ 1 file changed, 9 insertions(+)
MAINTAINERS +9 -0
@@ -7789,6 +7789,15 @@ S: Maintained F: include/linux/power_supply.h F: drivers/power/ +POWER STATE COORDINATION INTERFACE (PSCI) +M: Mark Rutland <[email protected]> +M: Lorenzo Pieralisi <[email protected]> +L: linux-arm-kernel at lists.infradead.org +S: Maintained +F: drivers/firmware/psci.c +F: include/linux/psci.h +F: include/uapi/linux/psci.h + PNP SUPPORT M: Rafael J. Wysocki <[email protected]> S: Maintained
Signature · 1 line
1.9.1
v3 cover arm/arm64: Unify PSCI client support
In reply to Mark RutlandQuoted 23 lines · Mark Rutland
This series unifies the 32-bit and 64-bit PSCI client code, moving the bulk of the FW invocation and probing out to a common location in
drivers/firmware. As part of this, the remaining edge cases for PSCI 0.2 (Trusted OSs rejectingCPU_OFF) are accounted for, making both 32-bit adn 64-bit clients more robust.This results in a reasonable saving in terms of lines of code, and will allow for PSCI 1.0 support to be unified form the beginning, avoiding further duplication.
The series is based on the arm64 for-next/cpu-init branch [3] due to conflicts with that series.
Since v1 [1]:
* Fix build when PSCI isn't selected
* Don't indirectmigrate_info_up_cpu
* Fix ver to u32 inpsci_get_versionSince v2: [2]:
* Rebase atop of Lorenzo's cpu init cleanup series [3]
* Use the correct native function IDs
* Fixcpu_disablereturn values
* Report invalidMIGRATE_INFO_UP_CPUreturn values
Quoted 3 lines
Many thanks to those who have reviewed and tested so far! Unfortunately, due to the volume of changes in the rebase I've dropped all Tested-by tags. I've locally tested 32-bit and 64-bit kernels with DT.
I tested this patch set on ARM64 kernel with ACPI as I did in previous version, compile this patch set ok with CONFIG_ACPI=y/n, and CPU hotplug works with PSCI (CPU_ON/OFF),
Thanks
Hanjun
v3 cover arm/arm64: Unify PSCI client support
Quoted 28 lines · Hanjun Guo
Quoted 23 lines · Mark Rutland
This series unifies the 32-bit and 64-bit PSCI client code, moving the bulk of the FW invocation and probing out to a common location in
drivers/firmware. As part of this, the remaining edge cases for PSCI 0.2 (Trusted OSs rejectingCPU_OFF) are accounted for, making both 32-bit adn 64-bit clients more robust.This results in a reasonable saving in terms of lines of code, and will allow for PSCI 1.0 support to be unified form the beginning, avoiding further duplication.
The series is based on the arm64 for-next/cpu-init branch [3] due to conflicts with that series.
Since v1 [1]:
* Fix build when PSCI isn't selected
* Don't indirectmigrate_info_up_cpu
* Fix ver to u32 inpsci_get_versionSince v2: [2]:
* Rebase atop of Lorenzo's cpu init cleanup series [3]
* Use the correct native function IDs
* Fixcpu_disablereturn values
* Report invalidMIGRATE_INFO_UP_CPUreturn valuesQuoted 3 lines
Many thanks to those who have reviewed and tested so far! Unfortunately, due to the volume of changes in the rebase I've dropped all Tested-by tags. I've locally tested 32-bit and 64-bit kernels with DT.
Quoted 5 lines
I tested this patch set on ARM64 kernel with ACPI as I did in previous version, compile this patch set ok with
CONFIG_ACPI=y/n, and CPU hotplug works with PSCI (CPU_ON/OFF),Tested-by Hanjun Guo <[email protected]>
Thanks!
I've applied that to the first 10 patches, but not to the patch affecting 32-bit ARM nor the patch updating MAINTAINERS.
Mark.