arm-trusted-firmware/services/std_svc/psci/psci_afflvl_off.c

289 lines
9.2 KiB
C
Raw Normal View History

2013-10-25 09:08:21 +01:00
/*
* Copyright (c) 2013-2014, ARM Limited and Contributors. All rights reserved.
2013-10-25 09:08:21 +01:00
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
*
* Redistributions of source code must retain the above copyright notice, this
* list of conditions and the following disclaimer.
*
* Redistributions in binary form must reproduce the above copyright notice,
* this list of conditions and the following disclaimer in the documentation
* and/or other materials provided with the distribution.
*
* Neither the name of ARM nor the names of its contributors may be used
* to endorse or promote products derived from this software without specific
* prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
* AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE
* LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.
*/
#include <stdio.h>
#include <string.h>
#include <assert.h>
#include <debug.h>
2013-10-25 09:08:21 +01:00
#include <arch_helpers.h>
#include <console.h>
#include <platform.h>
#include <psci.h>
#include "psci_private.h"
2013-10-25 09:08:21 +01:00
typedef int (*afflvl_off_handler)(unsigned long, aff_map_node *);
/*******************************************************************************
* The next three functions implement a handler for each supported affinity
* level which is called when that affinity level is turned off.
******************************************************************************/
static int psci_afflvl0_off(unsigned long mpidr, aff_map_node *cpu_node)
{
unsigned int index, plat_state;
int rc = PSCI_E_SUCCESS;
unsigned long sctlr;
2013-10-25 09:08:21 +01:00
assert(cpu_node->level == MPIDR_AFFLVL0);
psci: fix affinity level upgrade issue The psci implementation does not track target affinity level requests specified during cpu_suspend calls correctly as per the following example. 1. cpu0.cluster0 calls cpu_suspend with the target affinity level as 0 2. Only the cpu0.cluster0 is powered down while cluster0 remains powered up 3. cpu1.cluster0 calls cpu_off to power itself down to highest possible affinity level 4. cluster0 will be powered off even though cpu0.cluster0 does not allow cluster shutdown This patch introduces reference counts at affinity levels > 0 to track the number of cpus which want an affinity instance at level X to remain powered up. This instance can be turned off only if its reference count is 0. Cpus still undergo the normal state transitions (ON, OFF, ON_PENDING, SUSPEND) but the higher levels can only be either ON or OFF depending upon their reference count. The above issue is thus fixed as follows: 1. cluster0's reference count is incremented by two when cpu0 and cpu1 are initially powered on. 2. cpu0.cluster0 calls cpu_suspend with the target affinity level as 0. This does not affect the cluster0 reference count. 3. Only the cpu0.cluster0 is powered down while cluster0 remains powered up as it has a non-zero reference count. 4. cpu1.cluster0 call cpu_off to power itself down to highest possible affinity level. This decrements the cluster0 reference count. 5. cluster0 is still not powered off since its reference count will at least be 1 due to the restriction placed by cpu0. Change-Id: I433dfe82b946f5f6985b1602c2de87800504f7a9
2013-12-05 16:33:10 +00:00
/* State management: mark this cpu as turned off */
psci_set_state(cpu_node, PSCI_STATE_OFF);
2013-10-25 09:08:21 +01:00
/*
* Generic management: Get the index for clearing any lingering re-entry
* information and allow the secure world to switch itself off
2013-10-25 09:08:21 +01:00
*/
/*
* Call the cpu off handler registered by the Secure Payload Dispatcher
* to let it do any bookeeping. Assume that the SPD always reports an
* E_DENIED error if SP refuse to power down
*/
if (psci_spd_pm && psci_spd_pm->svc_off) {
rc = psci_spd_pm->svc_off(0);
if (rc)
return rc;
}
2013-10-25 09:08:21 +01:00
index = cpu_node->data;
memset(&psci_ns_entry_info[index], 0, sizeof(psci_ns_entry_info[index]));
/*
* Arch. management. Perform the necessary steps to flush all
* cpu caches.
*
* TODO: This power down sequence varies across cpus so it needs to be
* abstracted out on the basis of the MIDR like in cpu_reset_handler().
* Do the bare minimal for the time being. Fix this before porting to
* Cortex models.
*/
sctlr = read_sctlr_el3();
2013-10-25 09:08:21 +01:00
sctlr &= ~SCTLR_C_BIT;
write_sctlr_el3(sctlr);
2013-10-25 09:08:21 +01:00
/*
* CAUTION: This flush to the level of unification makes an assumption
* about the cache hierarchy at affinity level 0 (cpu) in the platform.
* Ideally the platform should tell psci which levels to flush to exit
* coherency.
*/
dcsw_op_louis(DCCISW);
/*
* Plat. management: Perform platform specific actions to turn this
* cpu off e.g. exit cpu coherency, program the power controller etc.
*/
if (psci_plat_pm_ops->affinst_off) {
/* Get the current physical state of this cpu */
psci: fix affinity level upgrade issue The psci implementation does not track target affinity level requests specified during cpu_suspend calls correctly as per the following example. 1. cpu0.cluster0 calls cpu_suspend with the target affinity level as 0 2. Only the cpu0.cluster0 is powered down while cluster0 remains powered up 3. cpu1.cluster0 calls cpu_off to power itself down to highest possible affinity level 4. cluster0 will be powered off even though cpu0.cluster0 does not allow cluster shutdown This patch introduces reference counts at affinity levels > 0 to track the number of cpus which want an affinity instance at level X to remain powered up. This instance can be turned off only if its reference count is 0. Cpus still undergo the normal state transitions (ON, OFF, ON_PENDING, SUSPEND) but the higher levels can only be either ON or OFF depending upon their reference count. The above issue is thus fixed as follows: 1. cluster0's reference count is incremented by two when cpu0 and cpu1 are initially powered on. 2. cpu0.cluster0 calls cpu_suspend with the target affinity level as 0. This does not affect the cluster0 reference count. 3. Only the cpu0.cluster0 is powered down while cluster0 remains powered up as it has a non-zero reference count. 4. cpu1.cluster0 call cpu_off to power itself down to highest possible affinity level. This decrements the cluster0 reference count. 5. cluster0 is still not powered off since its reference count will at least be 1 due to the restriction placed by cpu0. Change-Id: I433dfe82b946f5f6985b1602c2de87800504f7a9
2013-12-05 16:33:10 +00:00
plat_state = psci_get_phys_state(cpu_node);
2013-10-25 09:08:21 +01:00
rc = psci_plat_pm_ops->affinst_off(mpidr,
cpu_node->level,
plat_state);
}
return rc;
}
static int psci_afflvl1_off(unsigned long mpidr, aff_map_node *cluster_node)
{
int rc = PSCI_E_SUCCESS;
unsigned int plat_state;
/* Sanity check the cluster level */
assert(cluster_node->level == MPIDR_AFFLVL1);
psci: fix affinity level upgrade issue The psci implementation does not track target affinity level requests specified during cpu_suspend calls correctly as per the following example. 1. cpu0.cluster0 calls cpu_suspend with the target affinity level as 0 2. Only the cpu0.cluster0 is powered down while cluster0 remains powered up 3. cpu1.cluster0 calls cpu_off to power itself down to highest possible affinity level 4. cluster0 will be powered off even though cpu0.cluster0 does not allow cluster shutdown This patch introduces reference counts at affinity levels > 0 to track the number of cpus which want an affinity instance at level X to remain powered up. This instance can be turned off only if its reference count is 0. Cpus still undergo the normal state transitions (ON, OFF, ON_PENDING, SUSPEND) but the higher levels can only be either ON or OFF depending upon their reference count. The above issue is thus fixed as follows: 1. cluster0's reference count is incremented by two when cpu0 and cpu1 are initially powered on. 2. cpu0.cluster0 calls cpu_suspend with the target affinity level as 0. This does not affect the cluster0 reference count. 3. Only the cpu0.cluster0 is powered down while cluster0 remains powered up as it has a non-zero reference count. 4. cpu1.cluster0 call cpu_off to power itself down to highest possible affinity level. This decrements the cluster0 reference count. 5. cluster0 is still not powered off since its reference count will at least be 1 due to the restriction placed by cpu0. Change-Id: I433dfe82b946f5f6985b1602c2de87800504f7a9
2013-12-05 16:33:10 +00:00
/* State management: Decrement the cluster reference count */
psci_set_state(cluster_node, PSCI_STATE_OFF);
2013-10-25 09:08:21 +01:00
/*
* Keep the physical state of this cluster handy to decide
* what action needs to be taken
*/
psci: fix affinity level upgrade issue The psci implementation does not track target affinity level requests specified during cpu_suspend calls correctly as per the following example. 1. cpu0.cluster0 calls cpu_suspend with the target affinity level as 0 2. Only the cpu0.cluster0 is powered down while cluster0 remains powered up 3. cpu1.cluster0 calls cpu_off to power itself down to highest possible affinity level 4. cluster0 will be powered off even though cpu0.cluster0 does not allow cluster shutdown This patch introduces reference counts at affinity levels > 0 to track the number of cpus which want an affinity instance at level X to remain powered up. This instance can be turned off only if its reference count is 0. Cpus still undergo the normal state transitions (ON, OFF, ON_PENDING, SUSPEND) but the higher levels can only be either ON or OFF depending upon their reference count. The above issue is thus fixed as follows: 1. cluster0's reference count is incremented by two when cpu0 and cpu1 are initially powered on. 2. cpu0.cluster0 calls cpu_suspend with the target affinity level as 0. This does not affect the cluster0 reference count. 3. Only the cpu0.cluster0 is powered down while cluster0 remains powered up as it has a non-zero reference count. 4. cpu1.cluster0 call cpu_off to power itself down to highest possible affinity level. This decrements the cluster0 reference count. 5. cluster0 is still not powered off since its reference count will at least be 1 due to the restriction placed by cpu0. Change-Id: I433dfe82b946f5f6985b1602c2de87800504f7a9
2013-12-05 16:33:10 +00:00
plat_state = psci_get_phys_state(cluster_node);
2013-10-25 09:08:21 +01:00
/*
* Arch. Management. Flush all levels of caches to PoC if
* the cluster is to be shutdown
*/
if (plat_state == PSCI_STATE_OFF)
dcsw_op_all(DCCISW);
/*
* Plat. Management. Allow the platform to do its cluster
2013-10-25 09:08:21 +01:00
* specific bookeeping e.g. turn off interconnect coherency,
* program the power controller etc.
*/
if (psci_plat_pm_ops->affinst_off)
rc = psci_plat_pm_ops->affinst_off(mpidr,
cluster_node->level,
plat_state);
return rc;
}
static int psci_afflvl2_off(unsigned long mpidr, aff_map_node *system_node)
{
int rc = PSCI_E_SUCCESS;
unsigned int plat_state;
/* Cannot go beyond this level */
assert(system_node->level == MPIDR_AFFLVL2);
psci: fix affinity level upgrade issue The psci implementation does not track target affinity level requests specified during cpu_suspend calls correctly as per the following example. 1. cpu0.cluster0 calls cpu_suspend with the target affinity level as 0 2. Only the cpu0.cluster0 is powered down while cluster0 remains powered up 3. cpu1.cluster0 calls cpu_off to power itself down to highest possible affinity level 4. cluster0 will be powered off even though cpu0.cluster0 does not allow cluster shutdown This patch introduces reference counts at affinity levels > 0 to track the number of cpus which want an affinity instance at level X to remain powered up. This instance can be turned off only if its reference count is 0. Cpus still undergo the normal state transitions (ON, OFF, ON_PENDING, SUSPEND) but the higher levels can only be either ON or OFF depending upon their reference count. The above issue is thus fixed as follows: 1. cluster0's reference count is incremented by two when cpu0 and cpu1 are initially powered on. 2. cpu0.cluster0 calls cpu_suspend with the target affinity level as 0. This does not affect the cluster0 reference count. 3. Only the cpu0.cluster0 is powered down while cluster0 remains powered up as it has a non-zero reference count. 4. cpu1.cluster0 call cpu_off to power itself down to highest possible affinity level. This decrements the cluster0 reference count. 5. cluster0 is still not powered off since its reference count will at least be 1 due to the restriction placed by cpu0. Change-Id: I433dfe82b946f5f6985b1602c2de87800504f7a9
2013-12-05 16:33:10 +00:00
/* State management: Decrement the system reference count */
psci_set_state(system_node, PSCI_STATE_OFF);
2013-10-25 09:08:21 +01:00
/*
* Keep the physical state of the system handy to decide what
* action needs to be taken
*/
psci: fix affinity level upgrade issue The psci implementation does not track target affinity level requests specified during cpu_suspend calls correctly as per the following example. 1. cpu0.cluster0 calls cpu_suspend with the target affinity level as 0 2. Only the cpu0.cluster0 is powered down while cluster0 remains powered up 3. cpu1.cluster0 calls cpu_off to power itself down to highest possible affinity level 4. cluster0 will be powered off even though cpu0.cluster0 does not allow cluster shutdown This patch introduces reference counts at affinity levels > 0 to track the number of cpus which want an affinity instance at level X to remain powered up. This instance can be turned off only if its reference count is 0. Cpus still undergo the normal state transitions (ON, OFF, ON_PENDING, SUSPEND) but the higher levels can only be either ON or OFF depending upon their reference count. The above issue is thus fixed as follows: 1. cluster0's reference count is incremented by two when cpu0 and cpu1 are initially powered on. 2. cpu0.cluster0 calls cpu_suspend with the target affinity level as 0. This does not affect the cluster0 reference count. 3. Only the cpu0.cluster0 is powered down while cluster0 remains powered up as it has a non-zero reference count. 4. cpu1.cluster0 call cpu_off to power itself down to highest possible affinity level. This decrements the cluster0 reference count. 5. cluster0 is still not powered off since its reference count will at least be 1 due to the restriction placed by cpu0. Change-Id: I433dfe82b946f5f6985b1602c2de87800504f7a9
2013-12-05 16:33:10 +00:00
plat_state = psci_get_phys_state(system_node);
2013-10-25 09:08:21 +01:00
/* No arch. and generic bookeeping to do here currently */
/*
* Plat. Management : Allow the platform to do its bookeeping
2013-10-25 09:08:21 +01:00
* at this affinity level
*/
if (psci_plat_pm_ops->affinst_off)
rc = psci_plat_pm_ops->affinst_off(mpidr,
system_node->level,
plat_state);
return rc;
}
static const afflvl_off_handler psci_afflvl_off_handlers[] = {
psci_afflvl0_off,
psci_afflvl1_off,
psci_afflvl2_off,
};
/*******************************************************************************
* This function takes an array of pointers to affinity instance nodes in the
* topology tree and calls the off handler for the corresponding affinity
* levels
******************************************************************************/
static int psci_call_off_handlers(mpidr_aff_map_nodes mpidr_nodes,
int start_afflvl,
int end_afflvl,
unsigned long mpidr)
{
int rc = PSCI_E_INVALID_PARAMS, level;
aff_map_node *node;
for (level = start_afflvl; level <= end_afflvl; level++) {
node = mpidr_nodes[level];
if (node == NULL)
continue;
/*
* TODO: In case of an error should there be a way
* of restoring what we might have torn down at
* lower affinity levels.
*/
rc = psci_afflvl_off_handlers[level](mpidr, node);
if (rc != PSCI_E_SUCCESS)
break;
}
return rc;
}
/*******************************************************************************
* Top level handler which is called when a cpu wants to power itself down.
* It's assumed that along with turning the cpu off, higher affinity levels will
* be turned off as far as possible. It traverses through all the affinity
* levels performing generic, architectural, platform setup and state management
* e.g. for a cluster that's to be powered off, it will call the platform
* specific code which will disable coherency at the interconnect level if the
* cpu is the last in the cluster. For a cpu it could mean programming the power
* the power controller etc.
*
* The state of all the relevant affinity levels is changed prior to calling the
* affinity level specific handlers as their actions would depend upon the state
* the affinity level is about to enter.
*
* The affinity level specific handlers are called in ascending order i.e. from
* the lowest to the highest affinity level implemented by the platform because
* to turn off affinity level X it is neccesary to turn off affinity level X - 1
* first.
*
* CAUTION: This function is called with coherent stacks so that coherency can
* be turned off and caches can be flushed safely.
2013-10-25 09:08:21 +01:00
******************************************************************************/
int psci_afflvl_off(unsigned long mpidr,
int start_afflvl,
int end_afflvl)
2013-10-25 09:08:21 +01:00
{
int rc = PSCI_E_SUCCESS;
mpidr_aff_map_nodes mpidr_nodes;
2013-10-25 09:08:21 +01:00
mpidr &= MPIDR_AFFINITY_MASK;;
/*
* Collect the pointers to the nodes in the topology tree for
* each affinity instance in the mpidr. If this function does
* not return successfully then either the mpidr or the affinity
* levels are incorrect. In either case, we cannot return back
* to the caller as it would not know what to do.
2013-10-25 09:08:21 +01:00
*/
rc = psci_get_aff_map_nodes(mpidr,
start_afflvl,
end_afflvl,
mpidr_nodes);
assert (rc == PSCI_E_SUCCESS);
2013-10-25 09:08:21 +01:00
/*
* This function acquires the lock corresponding to each affinity
* level so that by the time all locks are taken, the system topology
* is snapshot and state management can be done safely.
2013-10-25 09:08:21 +01:00
*/
psci_acquire_afflvl_locks(mpidr,
start_afflvl,
end_afflvl,
mpidr_nodes);
2013-10-25 09:08:21 +01:00
/* Perform generic, architecture and platform specific handling */
rc = psci_call_off_handlers(mpidr_nodes,
start_afflvl,
end_afflvl,
mpidr);
2013-10-25 09:08:21 +01:00
/*
* Release the locks corresponding to each affinity level in the
* reverse order to which they were acquired.
2013-10-25 09:08:21 +01:00
*/
psci_release_afflvl_locks(mpidr,
start_afflvl,
end_afflvl,
mpidr_nodes);
2013-10-25 09:08:21 +01:00
return rc;
}