arm-trusted-firmware/lib/psci/psci_main.c
Manish Pandey ef738d19d3 feat(psci): remove cpu context init by index
Currently, the calling core (meaning the core which received the call to
CPU_ON or the powerdown path of CPU_SUSPEND on the same core) is in
charge of initialising the context for the waking core (the warmboot
entrypoint for both). This is convenient because the calling core can
write the context while in coherency and the waking core will only need
the context after its entered coherency. This avoids any cache
maintenance and makes communication simple.

However, this has 3 main problems:
a) asymmetric feature support is problematic - the calling core has no
   way of knowing the feature set of the waking core. If the two
   diverge, the architectural feature discovery via ID registers breaks
   down. We've thus far "fixed" this on a case by case basis which
   doesn't scale and introduces redundancy.

b) powerdown abandon (pabandon) introduces a contradiction - the calling
   core has to initialise the context for when the core wakes up, but
   should the core not powerdown it needs its old context intact. The only
   way to work around this is by keeping two copies of context which
   incurs a runtime and memory overhead.

c) cm_prepare_el3_exit[_ns]() doesn't have access to the entrypoint but needs
   it to make initialisation decisions. We can infer some of this from
   registers that have already been written but this is awkwardly
   limiting for what we can do. This also necessitates the split from
   the context initialisation.

We can solve all three by a making a core be in full ownership of its
own context. The calling core then only writes entrypoint information
and nothing else. The waking core then initialises its own context as it
sees fit with full knowledge of the whole picture.

The only tricky bit is cache coherency - the waking core has to be able
to coherently observe its new entrypoint. Calling cores will write to
the shared region with coherent caches on. If we make sure to read the
context only after the waking core has entered coherency, then we can
avoid cache operations and let hardware handle everything.

We can skip the spsr check for FEAT_TCR2 as it doesn't make a
difference. We can also skip enabling it twice from generic code.

Signed-off-by: Boyan Karatotev <boyan.karatotev@arm.com>
Signed-off-by: Manish Pandey <manish.pandey2@arm.com>
Change-Id: I86e7fe8b698191fc3b469e5ced1fd010f8754b0e
2025-04-11 07:27:23 +01:00

638 lines
16 KiB
C

/*
* Copyright (c) 2013-2025, Arm Limited and Contributors. All rights reserved.
*
* SPDX-License-Identifier: BSD-3-Clause
*/
#include <assert.h>
#include <string.h>
#include <arch.h>
#include <arch_features.h>
#include <arch_helpers.h>
#include <common/debug.h>
#include <lib/pmf/pmf.h>
#include <lib/runtime_instr.h>
#include <lib/smccc.h>
#include <plat/common/platform.h>
#include <services/arm_arch_svc.h>
#include "psci_private.h"
/*******************************************************************************
* PSCI frontend api for servicing SMCs. Described in the PSCI spec.
******************************************************************************/
int psci_cpu_on(u_register_t target_cpu,
uintptr_t entrypoint,
u_register_t context_id)
{
int rc;
entry_point_info_t *ep;
unsigned int target_idx = (unsigned int)plat_core_pos_by_mpidr(target_cpu);
/* Validate the target CPU */
if (!is_valid_mpidr(target_cpu)) {
return PSCI_E_INVALID_PARAMS;
}
ep = get_cpu_data_by_index(target_idx, warmboot_ep_info);
/* Validate the lower EL entry point and put it in the entry_point_info */
rc = psci_validate_entry_point(ep, entrypoint, context_id);
if (rc != PSCI_E_SUCCESS) {
return rc;
}
/*
* To turn this cpu on, specify which power
* levels need to be turned on
*/
return psci_cpu_on_start(target_cpu, ep);
}
unsigned int psci_version(void)
{
return PSCI_MAJOR_VER | PSCI_MINOR_VER;
}
int psci_cpu_suspend(unsigned int power_state,
uintptr_t entrypoint,
u_register_t context_id)
{
int rc;
unsigned int target_pwrlvl, is_power_down_state;
psci_power_state_t state_info = { {PSCI_LOCAL_STATE_RUN} };
plat_local_state_t cpu_pd_state;
unsigned int cpu_idx = plat_my_core_pos();
#if PSCI_OS_INIT_MODE
plat_local_state_t prev[PLAT_MAX_PWR_LVL];
#endif
#if ERRATA_SME_POWER_DOWN
/*
* If SME isn't off, attempting a real power down will only end up being
* rejected. If we got called with SME on, fall back to a normal
* suspend. We can't force SME off as in the event the power down is
* rejected for another reason (eg GIC) we'd lose the SME context.
*/
if (is_feat_sme_supported() && read_svcr() != 0) {
power_state &= ~(PSTATE_TYPE_MASK << PSTATE_TYPE_SHIFT);
power_state &= ~(PSTATE_PWR_LVL_MASK << PSTATE_PWR_LVL_SHIFT);
}
#endif /* ERRATA_SME_POWER_DOWN */
/* Validate the power_state parameter */
rc = psci_validate_power_state(power_state, &state_info);
if (rc != PSCI_E_SUCCESS) {
assert(rc == PSCI_E_INVALID_PARAMS);
return rc;
}
/*
* Get the value of the state type bit from the power state parameter.
*/
is_power_down_state = psci_get_pstate_type(power_state);
/* Sanity check the requested suspend levels */
assert(psci_validate_suspend_req(&state_info, is_power_down_state)
== PSCI_E_SUCCESS);
target_pwrlvl = psci_find_target_suspend_lvl(&state_info);
if (target_pwrlvl == PSCI_INVALID_PWR_LVL) {
ERROR("Invalid target power level for suspend operation\n");
panic();
}
/* Fast path for CPU standby.*/
if (is_cpu_standby_req(is_power_down_state, target_pwrlvl)) {
if (psci_plat_pm_ops->cpu_standby == NULL) {
return PSCI_E_INVALID_PARAMS;
}
/*
* Set the state of the CPU power domain to the platform
* specific retention state and enter the standby state.
*/
cpu_pd_state = state_info.pwr_domain_state[PSCI_CPU_PWR_LVL];
psci_set_cpu_local_state(cpu_pd_state);
#if PSCI_OS_INIT_MODE
/*
* If in OS-initiated mode, save a copy of the previous
* requested local power states and update the new requested
* local power states for this CPU.
*/
if (psci_suspend_mode == OS_INIT) {
psci_update_req_local_pwr_states(target_pwrlvl, cpu_idx,
&state_info, prev);
}
#endif
#if ENABLE_PSCI_STAT
plat_psci_stat_accounting_start(&state_info);
#endif
#if ENABLE_RUNTIME_INSTRUMENTATION
PMF_CAPTURE_TIMESTAMP(rt_instr_svc,
RT_INSTR_ENTER_HW_LOW_PWR,
PMF_NO_CACHE_MAINT);
#endif
psci_plat_pm_ops->cpu_standby(cpu_pd_state);
/* Upon exit from standby, set the state back to RUN. */
psci_set_cpu_local_state(PSCI_LOCAL_STATE_RUN);
#if PSCI_OS_INIT_MODE
/*
* If in OS-initiated mode, restore the previous requested
* local power states for this CPU.
*/
if (psci_suspend_mode == OS_INIT) {
psci_restore_req_local_pwr_states(cpu_idx, prev);
}
#endif
#if ENABLE_RUNTIME_INSTRUMENTATION
PMF_CAPTURE_TIMESTAMP(rt_instr_svc,
RT_INSTR_EXIT_HW_LOW_PWR,
PMF_NO_CACHE_MAINT);
#endif
#if ENABLE_PSCI_STAT
plat_psci_stat_accounting_stop(&state_info);
/* Update PSCI stats */
psci_stats_update_pwr_up(cpu_idx, PSCI_CPU_PWR_LVL, &state_info);
#endif
return PSCI_E_SUCCESS;
}
/*
* If a power down state has been requested, we need to verify entry
* point and program entry information.
*/
if (is_power_down_state != 0U) {
entry_point_info_t *ep = get_cpu_data_by_index(cpu_idx, warmboot_ep_info);
rc = psci_validate_entry_point(ep, entrypoint, context_id);
if (rc != PSCI_E_SUCCESS) {
return rc;
}
}
/*
* Do what is needed to enter the power down state. Upon success,
* enter the final wfi which will power down this CPU. This function
* might return if the power down was abandoned for any reason, e.g.
* arrival of an interrupt
*/
rc = psci_cpu_suspend_start(cpu_idx,
target_pwrlvl,
&state_info,
is_power_down_state);
return rc;
}
int psci_system_suspend(uintptr_t entrypoint, u_register_t context_id)
{
int rc;
psci_power_state_t state_info;
unsigned int cpu_idx = plat_my_core_pos();
entry_point_info_t *ep = get_cpu_data_by_index(cpu_idx, warmboot_ep_info);
/* Check if the current CPU is the last ON CPU in the system */
if (!psci_is_last_on_cpu(cpu_idx)) {
return PSCI_E_DENIED;
}
/* Validate the entry point and get the entry_point_info */
rc = psci_validate_entry_point(ep, entrypoint, context_id);
if (rc != PSCI_E_SUCCESS) {
return rc;
}
/* Query the psci_power_state for system suspend */
psci_query_sys_suspend_pwrstate(&state_info);
/*
* Check if platform allows suspend to Highest power level
* (System level)
*/
if (psci_find_target_suspend_lvl(&state_info) < PLAT_MAX_PWR_LVL) {
return PSCI_E_DENIED;
}
/* Ensure that the psci_power_state makes sense */
assert(psci_validate_suspend_req(&state_info, PSTATE_TYPE_POWERDOWN)
== PSCI_E_SUCCESS);
assert(is_local_state_off(
state_info.pwr_domain_state[PLAT_MAX_PWR_LVL]) != 0);
/*
* Do what is needed to enter the system suspend state. This function
* might return if the power down was abandoned for any reason, e.g.
* arrival of an interrupt
*/
rc = psci_cpu_suspend_start(cpu_idx,
PLAT_MAX_PWR_LVL,
&state_info,
PSTATE_TYPE_POWERDOWN);
return rc;
}
int psci_cpu_off(void)
{
int rc;
unsigned int target_pwrlvl = PLAT_MAX_PWR_LVL;
/*
* Do what is needed to power off this CPU and possible higher power
* levels if it able to do so. Upon success, enter the final wfi
* which will power down this CPU.
*/
rc = psci_do_cpu_off(target_pwrlvl);
/*
* The only error cpu_off can return is E_DENIED. So check if that's
* indeed the case.
*/
assert(rc == PSCI_E_DENIED);
return rc;
}
int psci_affinity_info(u_register_t target_affinity,
unsigned int lowest_affinity_level)
{
unsigned int target_idx;
/* Validate the target affinity */
if (!is_valid_mpidr(target_affinity)) {
return PSCI_E_INVALID_PARAMS;
}
/* We dont support level higher than PSCI_CPU_PWR_LVL */
if (lowest_affinity_level > PSCI_CPU_PWR_LVL) {
return PSCI_E_INVALID_PARAMS;
}
/* Calculate the cpu index of the target */
target_idx = (unsigned int) plat_core_pos_by_mpidr(target_affinity);
/*
* Generic management:
* Perform cache maintanence ahead of reading the target CPU state to
* ensure that the data is not stale.
* There is a theoretical edge case where the cache may contain stale
* data for the target CPU data - this can occur under the following
* conditions:
* - the target CPU is in another cluster from the current
* - the target CPU was the last CPU to shutdown on its cluster
* - the cluster was removed from coherency as part of the CPU shutdown
*
* In this case the cache maintenace that was performed as part of the
* target CPUs shutdown was not seen by the current CPU's cluster. And
* so the cache may contain stale data for the target CPU.
*/
flush_cpu_data_by_index(target_idx,
psci_svc_cpu_data.aff_info_state);
return psci_get_aff_info_state_by_idx(target_idx);
}
int psci_migrate(u_register_t target_cpu)
{
int rc;
u_register_t resident_cpu_mpidr;
/* Validate the target cpu */
if (!is_valid_mpidr(target_cpu))
return PSCI_E_INVALID_PARAMS;
rc = psci_spd_migrate_info(&resident_cpu_mpidr);
if (rc != PSCI_TOS_UP_MIG_CAP) {
return (rc == PSCI_TOS_NOT_UP_MIG_CAP) ?
PSCI_E_DENIED : PSCI_E_NOT_SUPPORTED;
}
/*
* Migrate should only be invoked on the CPU where
* the Secure OS is resident.
*/
if (resident_cpu_mpidr != read_mpidr_el1()) {
return PSCI_E_NOT_PRESENT;
}
/* Check the validity of the specified target cpu */
if (!is_valid_mpidr(target_cpu)) {
return PSCI_E_INVALID_PARAMS;
}
assert((psci_spd_pm != NULL) && (psci_spd_pm->svc_migrate != NULL));
rc = psci_spd_pm->svc_migrate(read_mpidr_el1(), target_cpu);
assert((rc == PSCI_E_SUCCESS) || (rc == PSCI_E_INTERN_FAIL));
return rc;
}
int psci_migrate_info_type(void)
{
u_register_t resident_cpu_mpidr;
return psci_spd_migrate_info(&resident_cpu_mpidr);
}
u_register_t psci_migrate_info_up_cpu(void)
{
u_register_t resident_cpu_mpidr;
int rc;
/*
* Return value of this depends upon what
* psci_spd_migrate_info() returns.
*/
rc = psci_spd_migrate_info(&resident_cpu_mpidr);
if ((rc != PSCI_TOS_NOT_UP_MIG_CAP) && (rc != PSCI_TOS_UP_MIG_CAP))
return (u_register_t)(register_t) PSCI_E_INVALID_PARAMS;
return resident_cpu_mpidr;
}
int psci_node_hw_state(u_register_t target_cpu,
unsigned int power_level)
{
int rc;
/* Validate target_cpu */
if (!is_valid_mpidr(target_cpu))
return PSCI_E_INVALID_PARAMS;
/* Validate power_level against PLAT_MAX_PWR_LVL */
if (power_level > PLAT_MAX_PWR_LVL)
return PSCI_E_INVALID_PARAMS;
/*
* Dispatch this call to platform to query power controller, and pass on
* to the caller what it returns
*/
assert(psci_plat_pm_ops->get_node_hw_state != NULL);
rc = psci_plat_pm_ops->get_node_hw_state(target_cpu, power_level);
assert(((rc >= HW_ON) && (rc <= HW_STANDBY))
|| (rc == PSCI_E_NOT_SUPPORTED)
|| (rc == PSCI_E_INVALID_PARAMS));
return rc;
}
int psci_features(unsigned int psci_fid)
{
unsigned int local_caps = psci_caps;
if (psci_fid == SMCCC_VERSION) {
return PSCI_E_SUCCESS;
}
/* Check if it is a 64 bit function */
if (((psci_fid >> FUNCID_CC_SHIFT) & FUNCID_CC_MASK) == SMC_64) {
local_caps &= PSCI_CAP_64BIT_MASK;
}
/* Check for invalid fid */
if (!(is_std_svc_call(psci_fid) && is_valid_fast_smc(psci_fid)
&& is_psci_fid(psci_fid))) {
return PSCI_E_NOT_SUPPORTED;
}
/* Check if the psci fid is supported or not */
if ((local_caps & define_psci_cap(psci_fid)) == 0U) {
return PSCI_E_NOT_SUPPORTED;
}
/* Format the feature flags */
if ((psci_fid == PSCI_CPU_SUSPEND_AARCH32) ||
(psci_fid == PSCI_CPU_SUSPEND_AARCH64)) {
unsigned int ret = ((FF_PSTATE << FF_PSTATE_SHIFT) |
(FF_SUPPORTS_OS_INIT_MODE << FF_MODE_SUPPORT_SHIFT));
return (int)ret;
}
/* Return 0 for all other fid's */
return PSCI_E_SUCCESS;
}
#if PSCI_OS_INIT_MODE
int psci_set_suspend_mode(unsigned int mode)
{
if (psci_suspend_mode == mode) {
return PSCI_E_SUCCESS;
}
unsigned int this_core = plat_my_core_pos();
if (mode == PLAT_COORD) {
/* Check if the current CPU is the last ON CPU in the system */
if (!psci_is_last_on_cpu_safe(this_core)) {
return PSCI_E_DENIED;
}
}
if (mode == OS_INIT) {
/*
* Check if all CPUs in the system are ON or if the current
* CPU is the last ON CPU in the system.
*/
if (!(psci_are_all_cpus_on_safe(this_core) ||
psci_is_last_on_cpu_safe(this_core))) {
return PSCI_E_DENIED;
}
}
psci_suspend_mode = mode;
psci_flush_dcache_range((uintptr_t)&psci_suspend_mode,
sizeof(psci_suspend_mode));
return PSCI_E_SUCCESS;
}
#endif
/*******************************************************************************
* PSCI top level handler for servicing SMCs.
******************************************************************************/
u_register_t psci_smc_handler(uint32_t smc_fid,
u_register_t x1,
u_register_t x2,
u_register_t x3,
u_register_t x4,
void *cookie,
void *handle,
u_register_t flags)
{
u_register_t ret;
if (is_caller_secure(flags)) {
return (u_register_t)SMC_UNK;
}
/* Check the fid against the capabilities */
if ((psci_caps & define_psci_cap(smc_fid)) == 0U) {
return (u_register_t)SMC_UNK;
}
if (((smc_fid >> FUNCID_CC_SHIFT) & FUNCID_CC_MASK) == SMC_32) {
/* 32-bit PSCI function, clear top parameter bits */
uint32_t r1 = (uint32_t)x1;
uint32_t r2 = (uint32_t)x2;
uint32_t r3 = (uint32_t)x3;
switch (smc_fid) {
case PSCI_VERSION:
ret = (u_register_t)psci_version();
break;
case PSCI_CPU_OFF:
ret = (u_register_t)psci_cpu_off();
break;
case PSCI_CPU_SUSPEND_AARCH32:
ret = (u_register_t)psci_cpu_suspend(r1, r2, r3);
break;
case PSCI_CPU_ON_AARCH32:
ret = (u_register_t)psci_cpu_on(r1, r2, r3);
break;
case PSCI_AFFINITY_INFO_AARCH32:
ret = (u_register_t)psci_affinity_info(r1, r2);
break;
case PSCI_MIG_AARCH32:
ret = (u_register_t)psci_migrate(r1);
break;
case PSCI_MIG_INFO_TYPE:
ret = (u_register_t)psci_migrate_info_type();
break;
case PSCI_MIG_INFO_UP_CPU_AARCH32:
ret = psci_migrate_info_up_cpu();
break;
case PSCI_NODE_HW_STATE_AARCH32:
ret = (u_register_t)psci_node_hw_state(r1, r2);
break;
case PSCI_SYSTEM_SUSPEND_AARCH32:
ret = (u_register_t)psci_system_suspend(r1, r2);
break;
case PSCI_SYSTEM_OFF:
psci_system_off();
/* We should never return from psci_system_off() */
break;
case PSCI_SYSTEM_RESET:
psci_system_reset();
/* We should never return from psci_system_reset() */
break;
case PSCI_FEATURES:
ret = (u_register_t)psci_features(r1);
break;
#if PSCI_OS_INIT_MODE
case PSCI_SET_SUSPEND_MODE:
ret = (u_register_t)psci_set_suspend_mode(r1);
break;
#endif
#if ENABLE_PSCI_STAT
case PSCI_STAT_RESIDENCY_AARCH32:
ret = psci_stat_residency(r1, r2);
break;
case PSCI_STAT_COUNT_AARCH32:
ret = psci_stat_count(r1, r2);
break;
#endif
case PSCI_MEM_PROTECT:
ret = psci_mem_protect(r1);
break;
case PSCI_MEM_CHK_RANGE_AARCH32:
ret = psci_mem_chk_range(r1, r2);
break;
case PSCI_SYSTEM_RESET2_AARCH32:
/* We should never return from psci_system_reset2() */
ret = psci_system_reset2(r1, r2);
break;
default:
WARN("Unimplemented PSCI Call: 0x%x\n", smc_fid);
ret = (u_register_t)SMC_UNK;
break;
}
} else {
/* 64-bit PSCI function */
switch (smc_fid) {
case PSCI_CPU_SUSPEND_AARCH64:
ret = (u_register_t)
psci_cpu_suspend((unsigned int)x1, x2, x3);
break;
case PSCI_CPU_ON_AARCH64:
ret = (u_register_t)psci_cpu_on(x1, x2, x3);
break;
case PSCI_AFFINITY_INFO_AARCH64:
ret = (u_register_t)
psci_affinity_info(x1, (unsigned int)x2);
break;
case PSCI_MIG_AARCH64:
ret = (u_register_t)psci_migrate(x1);
break;
case PSCI_MIG_INFO_UP_CPU_AARCH64:
ret = psci_migrate_info_up_cpu();
break;
case PSCI_NODE_HW_STATE_AARCH64:
ret = (u_register_t)psci_node_hw_state(
x1, (unsigned int) x2);
break;
case PSCI_SYSTEM_SUSPEND_AARCH64:
ret = (u_register_t)psci_system_suspend(x1, x2);
break;
#if ENABLE_PSCI_STAT
case PSCI_STAT_RESIDENCY_AARCH64:
ret = psci_stat_residency(x1, (unsigned int) x2);
break;
case PSCI_STAT_COUNT_AARCH64:
ret = psci_stat_count(x1, (unsigned int) x2);
break;
#endif
case PSCI_MEM_CHK_RANGE_AARCH64:
ret = psci_mem_chk_range(x1, x2);
break;
case PSCI_SYSTEM_RESET2_AARCH64:
/* We should never return from psci_system_reset2() */
ret = psci_system_reset2((uint32_t) x1, x2);
break;
default:
WARN("Unimplemented PSCI Call: 0x%x\n", smc_fid);
ret = (u_register_t)SMC_UNK;
break;
}
}
return ret;
}