Introduce PSCI Library Interface
This patch introduces the PSCI Library interface. The major changes
introduced are as follows:
* Earlier BL31 was responsible for Architectural initialization during cold
boot via bl31_arch_setup() whereas PSCI was responsible for the same during
warm boot. This functionality is now consolidated by the PSCI library
and it does Architectural initialization via psci_arch_setup() during both
cold and warm boots.
* Earlier the warm boot entry point was always `psci_entrypoint()`. This was
not flexible enough as a library interface. Now PSCI expects the runtime
firmware to provide the entry point via `psci_setup()`. A new function
`bl31_warm_entrypoint` is introduced in BL31 and the previous
`psci_entrypoint()` is deprecated.
* The `smc_helpers.h` is reorganized to separate the SMC Calling Convention
defines from the Trusted Firmware SMC helpers. The former is now in a new
header file `smcc.h` and the SMC helpers are moved to Architecture specific
header.
* The CPU context is used by PSCI for context initialization and
restoration after power down (PSCI Context). It is also used by BL31 for SMC
handling and context management during Normal-Secure world switch (SMC
Context). The `psci_smc_handler()` interface is redefined to not use SMC
helper macros thus enabling to decouple the PSCI context from EL3 runtime
firmware SMC context. This enables PSCI to be integrated with other runtime
firmware using a different SMC context.
NOTE: With this patch the architectural setup done in `bl31_arch_setup()`
is done as part of `psci_setup()` and hence `bl31_platform_setup()` will be
invoked prior to architectural setup. It is highly unlikely that the platform
setup will depend on architectural setup and cause any failure. Please be
be aware of this change in sequence.
Change-Id: I7f497a08d33be234bbb822c28146250cb20dab73
diff --git a/lib/psci/aarch64/psci_entry.S b/lib/psci/aarch64/psci_entry.S
deleted file mode 100644
index 646ebcf..0000000
--- a/lib/psci/aarch64/psci_entry.S
+++ /dev/null
@@ -1,111 +0,0 @@
-/*
- * Copyright (c) 2013-2016, ARM Limited and Contributors. All rights reserved.
- *
- * 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 <arch.h>
-#include <asm_macros.S>
-#include <el3_common_macros.S>
-#include <psci.h>
-#include <xlat_tables.h>
-
- .globl psci_entrypoint
- .globl psci_power_down_wfi
-
- /* --------------------------------------------------------------------
- * This CPU has been physically powered up. It is either resuming from
- * suspend or has simply been turned on. In both cases, call the power
- * on finisher.
- * --------------------------------------------------------------------
- */
-func psci_entrypoint
- /*
- * On the warm boot path, most of the EL3 initialisations performed by
- * 'el3_entrypoint_common' must be skipped:
- *
- * - Only when the platform bypasses the BL1/BL31 entrypoint by
- * programming the reset address do we need to set the CPU endianness.
- * In other cases, we assume this has been taken care by the
- * entrypoint code.
- *
- * - No need to determine the type of boot, we know it is a warm boot.
- *
- * - Do not try to distinguish between primary and secondary CPUs, this
- * notion only exists for a cold boot.
- *
- * - No need to initialise the memory or the C runtime environment,
- * it has been done once and for all on the cold boot path.
- */
- el3_entrypoint_common \
- _set_endian=PROGRAMMABLE_RESET_ADDRESS \
- _warm_boot_mailbox=0 \
- _secondary_cold_boot=0 \
- _init_memory=0 \
- _init_c_runtime=0 \
- _exception_vectors=runtime_exceptions
-
- /* --------------------------------------------
- * Enable the MMU with the DCache disabled. It
- * is safe to use stacks allocated in normal
- * memory as a result. All memory accesses are
- * marked nGnRnE when the MMU is disabled. So
- * all the stack writes will make it to memory.
- * All memory accesses are marked Non-cacheable
- * when the MMU is enabled but D$ is disabled.
- * So used stack memory is guaranteed to be
- * visible immediately after the MMU is enabled
- * Enabling the DCache at the same time as the
- * MMU can lead to speculatively fetched and
- * possibly stale stack memory being read from
- * other caches. This can lead to coherency
- * issues.
- * --------------------------------------------
- */
- mov x0, #DISABLE_DCACHE
- bl bl31_plat_enable_mmu
-
- bl psci_power_up_finish
-
- b el3_exit
-endfunc psci_entrypoint
-
- /* --------------------------------------------
- * This function is called to indicate to the
- * power controller that it is safe to power
- * down this cpu. It should not exit the wfi
- * and will be released from reset upon power
- * up. 'wfi_spill' is used to catch erroneous
- * exits from wfi.
- * --------------------------------------------
- */
-func psci_power_down_wfi
- dsb sy // ensure write buffer empty
- wfi
- bl plat_panic_handler
-endfunc psci_power_down_wfi
-
diff --git a/lib/psci/aarch64/psci_helpers.S b/lib/psci/aarch64/psci_helpers.S
index 87144dd..ff250a0 100644
--- a/lib/psci/aarch64/psci_helpers.S
+++ b/lib/psci/aarch64/psci_helpers.S
@@ -35,6 +35,10 @@
.globl psci_do_pwrdown_cache_maintenance
.globl psci_do_pwrup_cache_maintenance
+ .globl psci_power_down_wfi
+#if !ERROR_DEPRECATED
+ .globl psci_entrypoint
+#endif
/* -----------------------------------------------------------------------
* void psci_do_pwrdown_cache_maintenance(unsigned int power level);
@@ -152,3 +156,25 @@
ldp x29, x30, [sp], #16
ret
endfunc psci_do_pwrup_cache_maintenance
+
+/* -----------------------------------------------------------------------
+ * void psci_power_down_wfi(void);
+ * This function is called to indicate to the power controller that it
+ * is safe to power down this cpu. It should not exit the wfi and will
+ * be released from reset upon power up.
+ * -----------------------------------------------------------------------
+ */
+func psci_power_down_wfi
+ dsb sy // ensure write buffer empty
+ wfi
+ bl plat_panic_handler
+endfunc psci_power_down_wfi
+
+/* -----------------------------------------------------------------------
+ * void psci_entrypoint(void);
+ * The deprecated entry point for PSCI on warm boot for AArch64.
+ * -----------------------------------------------------------------------
+ */
+func_deprecated psci_entrypoint
+ b bl31_warm_entrypoint
+endfunc_deprecated psci_entrypoint
diff --git a/lib/psci/psci_common.c b/lib/psci/psci_common.c
index 2a0afb4..c6fea5b 100644
--- a/lib/psci/psci_common.c
+++ b/lib/psci/psci_common.c
@@ -683,7 +683,7 @@
* code to enable the gic cpu interface and for a cluster it will enable
* coherency at the interconnect level in addition to gic cpu interface.
******************************************************************************/
-void psci_power_up_finish(void)
+void psci_warmboot_entrypoint(void)
{
unsigned int end_pwrlvl, cpu_idx = plat_my_core_pos();
psci_power_state_t state_info = { {PSCI_LOCAL_STATE_RUN} };
diff --git a/lib/psci/psci_lib.mk b/lib/psci/psci_lib.mk
index 93c7832..662e14a 100644
--- a/lib/psci/psci_lib.mk
+++ b/lib/psci/psci_lib.mk
@@ -28,21 +28,20 @@
# POSSIBILITY OF SUCH DAMAGE.
#
-PSCI_LIB_SOURCES := lib/el3_runtime/cpu_data_array.c \
- lib/el3_runtime/aarch64/context.S \
- lib/el3_runtime/aarch64/cpu_data.S \
- lib/el3_runtime/aarch64/context_mgmt.c \
- lib/cpus/aarch64/cpu_helpers.S \
- lib/locks/exclusive/spinlock.S \
- lib/psci/psci_off.c \
- lib/psci/psci_on.c \
- lib/psci/psci_suspend.c \
- lib/psci/psci_common.c \
- lib/psci/psci_main.c \
- lib/psci/psci_setup.c \
- lib/psci/psci_system_off.c \
- lib/psci/aarch64/psci_entry.S \
- lib/psci/aarch64/psci_helpers.S \
+PSCI_LIB_SOURCES := lib/el3_runtime/cpu_data_array.c \
+ lib/el3_runtime/aarch64/context.S \
+ lib/el3_runtime/aarch64/cpu_data.S \
+ lib/el3_runtime/aarch64/context_mgmt.c \
+ lib/cpus/aarch64/cpu_helpers.S \
+ lib/locks/exclusive/spinlock.S \
+ lib/psci/psci_off.c \
+ lib/psci/psci_on.c \
+ lib/psci/psci_suspend.c \
+ lib/psci/psci_common.c \
+ lib/psci/psci_main.c \
+ lib/psci/psci_setup.c \
+ lib/psci/psci_system_off.c \
+ lib/psci/aarch64/psci_helpers.S
ifeq (${USE_COHERENT_MEM}, 1)
PSCI_LIB_SOURCES += lib/locks/bakery/bakery_lock_coherent.c
diff --git a/lib/psci/psci_main.c b/lib/psci/psci_main.c
index 04ef10e..d412be3 100644
--- a/lib/psci/psci_main.c
+++ b/lib/psci/psci_main.c
@@ -33,8 +33,7 @@
#include <assert.h>
#include <debug.h>
#include <platform.h>
-#include <runtime_svc.h>
-#include <std_svc.h>
+#include <smcc.h>
#include <string.h>
#include "psci_private.h"
@@ -327,7 +326,7 @@
/*******************************************************************************
* PSCI top level handler for servicing SMCs.
******************************************************************************/
-uintptr_t psci_smc_handler(uint32_t smc_fid,
+u_register_t psci_smc_handler(uint32_t smc_fid,
u_register_t x1,
u_register_t x2,
u_register_t x3,
@@ -337,11 +336,11 @@
u_register_t flags)
{
if (is_caller_secure(flags))
- SMC_RET1(handle, SMC_UNK);
+ return SMC_UNK;
/* Check the fid against the capabilities */
if (!(psci_caps & define_psci_cap(smc_fid)))
- SMC_RET1(handle, SMC_UNK);
+ return SMC_UNK;
if (((smc_fid >> FUNCID_CC_SHIFT) & FUNCID_CC_MASK) == SMC_32) {
/* 32-bit PSCI function, clear top parameter bits */
@@ -352,31 +351,31 @@
switch (smc_fid) {
case PSCI_VERSION:
- SMC_RET1(handle, psci_version());
+ return psci_version();
case PSCI_CPU_OFF:
- SMC_RET1(handle, psci_cpu_off());
+ return psci_cpu_off();
case PSCI_CPU_SUSPEND_AARCH32:
- SMC_RET1(handle, psci_cpu_suspend(x1, x2, x3));
+ return psci_cpu_suspend(x1, x2, x3);
case PSCI_CPU_ON_AARCH32:
- SMC_RET1(handle, psci_cpu_on(x1, x2, x3));
+ return psci_cpu_on(x1, x2, x3);
case PSCI_AFFINITY_INFO_AARCH32:
- SMC_RET1(handle, psci_affinity_info(x1, x2));
+ return psci_affinity_info(x1, x2);
case PSCI_MIG_AARCH32:
- SMC_RET1(handle, psci_migrate(x1));
+ return psci_migrate(x1);
case PSCI_MIG_INFO_TYPE:
- SMC_RET1(handle, psci_migrate_info_type());
+ return psci_migrate_info_type();
case PSCI_MIG_INFO_UP_CPU_AARCH32:
- SMC_RET1(handle, psci_migrate_info_up_cpu());
+ return psci_migrate_info_up_cpu();
case PSCI_SYSTEM_SUSPEND_AARCH32:
- SMC_RET1(handle, psci_system_suspend(x1, x2));
+ return psci_system_suspend(x1, x2);
case PSCI_SYSTEM_OFF:
psci_system_off();
@@ -387,14 +386,14 @@
/* We should never return from psci_system_reset() */
case PSCI_FEATURES:
- SMC_RET1(handle, psci_features(x1));
+ return psci_features(x1);
#if ENABLE_PSCI_STAT
case PSCI_STAT_RESIDENCY_AARCH32:
- SMC_RET1(handle, psci_stat_residency(x1, x2));
+ return psci_stat_residency(x1, x2);
case PSCI_STAT_COUNT_AARCH32:
- SMC_RET1(handle, psci_stat_count(x1, x2));
+ return psci_stat_count(x1, x2);
#endif
default:
@@ -405,29 +404,29 @@
switch (smc_fid) {
case PSCI_CPU_SUSPEND_AARCH64:
- SMC_RET1(handle, psci_cpu_suspend(x1, x2, x3));
+ return psci_cpu_suspend(x1, x2, x3);
case PSCI_CPU_ON_AARCH64:
- SMC_RET1(handle, psci_cpu_on(x1, x2, x3));
+ return psci_cpu_on(x1, x2, x3);
case PSCI_AFFINITY_INFO_AARCH64:
- SMC_RET1(handle, psci_affinity_info(x1, x2));
+ return psci_affinity_info(x1, x2);
case PSCI_MIG_AARCH64:
- SMC_RET1(handle, psci_migrate(x1));
+ return psci_migrate(x1);
case PSCI_MIG_INFO_UP_CPU_AARCH64:
- SMC_RET1(handle, psci_migrate_info_up_cpu());
+ return psci_migrate_info_up_cpu();
case PSCI_SYSTEM_SUSPEND_AARCH64:
- SMC_RET1(handle, psci_system_suspend(x1, x2));
+ return psci_system_suspend(x1, x2);
#if ENABLE_PSCI_STAT
case PSCI_STAT_RESIDENCY_AARCH64:
- SMC_RET1(handle, psci_stat_residency(x1, x2));
+ return psci_stat_residency(x1, x2);
case PSCI_STAT_COUNT_AARCH64:
- SMC_RET1(handle, psci_stat_count(x1, x2));
+ return psci_stat_count(x1, x2);
#endif
default:
@@ -436,5 +435,5 @@
}
WARN("Unimplemented PSCI Call: 0x%x \n", smc_fid);
- SMC_RET1(handle, SMC_UNK);
+ return SMC_UNK;
}
diff --git a/lib/psci/psci_on.c b/lib/psci/psci_on.c
index d4826ed..f4bb797 100644
--- a/lib/psci/psci_on.c
+++ b/lib/psci/psci_on.c
@@ -32,11 +32,9 @@
#include <arch_helpers.h>
#include <assert.h>
#include <bl_common.h>
-#include <bl31.h>
#include <debug.h>
#include <context_mgmt.h>
#include <platform.h>
-#include <runtime_svc.h>
#include <stddef.h>
#include "psci_private.h"
@@ -177,7 +175,7 @@
* on have completed. Perform enough arch.initialization
* to run in the non-secure address space.
*/
- bl31_arch_setup();
+ psci_arch_setup();
/*
* Lock the CPU spin lock to make sure that the context initialization
diff --git a/lib/psci/psci_private.h b/lib/psci/psci_private.h
index f42ce55..4935214 100644
--- a/lib/psci/psci_private.h
+++ b/lib/psci/psci_private.h
@@ -192,7 +192,6 @@
void psci_query_sys_suspend_pwrstate(psci_power_state_t *state_info);
int psci_validate_mpidr(u_register_t mpidr);
void psci_init_req_local_pwr_states(void);
-void psci_power_up_finish(void);
int psci_validate_entry_point(entry_point_info_t *ep,
uintptr_t entrypoint, u_register_t context_id);
void psci_get_parent_pwr_domain_nodes(unsigned int cpu_idx,
diff --git a/lib/psci/psci_setup.c b/lib/psci/psci_setup.c
index fac0ede..d35e000 100644
--- a/lib/psci/psci_setup.c
+++ b/lib/psci/psci_setup.c
@@ -184,15 +184,17 @@
}
/*******************************************************************************
- * This function initializes the power domain topology tree by querying the
- * platform. The power domain nodes higher than the CPU are populated in the
- * array psci_non_cpu_pd_nodes[] and the CPU power domains are populated in
- * psci_cpu_pd_nodes[]. The platform exports its static topology map through the
+ * This function does the architectural setup and takes the warm boot
+ * entry-point `mailbox_ep` as an argument. The function also initializes the
+ * power domain topology tree by querying the platform. The power domain nodes
+ * higher than the CPU are populated in the array psci_non_cpu_pd_nodes[] and
+ * the CPU power domains are populated in psci_cpu_pd_nodes[]. The platform
+ * exports its static topology map through the
* populate_power_domain_topology_tree() API. The algorithm populates the
* psci_non_cpu_pd_nodes and psci_cpu_pd_nodes iteratively by using this
- * topology map. On a platform that implements two clusters of 2 cpus each, and
- * supporting 3 domain levels, the populated psci_non_cpu_pd_nodes would look
- * like this:
+ * topology map. On a platform that implements two clusters of 2 cpus each,
+ * and supporting 3 domain levels, the populated psci_non_cpu_pd_nodes would
+ * look like this:
*
* ---------------------------------------------------
* | system node | cluster 0 node | cluster 1 node |
@@ -204,10 +206,13 @@
* | CPU 0 | CPU 1 | CPU 2 | CPU 3 |
* ------------------------------------------------
******************************************************************************/
-int psci_setup(void)
+int psci_setup(uintptr_t mailbox_ep)
{
const unsigned char *topology_tree;
+ /* Do the Architectural initialization */
+ psci_arch_setup();
+
/* Query the topology map from the platform */
topology_tree = plat_get_power_domain_tree_desc();
@@ -229,8 +234,8 @@
*/
psci_set_pwr_domains_to_run(PLAT_MAX_PWR_LVL);
- plat_setup_psci_ops((uintptr_t)psci_entrypoint,
- &psci_plat_pm_ops);
+ assert(mailbox_ep);
+ plat_setup_psci_ops(mailbox_ep, &psci_plat_pm_ops);
assert(psci_plat_pm_ops);
/* Initialize the psci capability */
@@ -259,3 +264,17 @@
return 0;
}
+
+/*******************************************************************************
+ * This duplicates what the primary cpu did after a cold boot in BL1. The same
+ * needs to be done when a cpu is hotplugged in. This function could also over-
+ * ride any EL3 setup done by BL1 as this code resides in rw memory.
+ ******************************************************************************/
+void psci_arch_setup(void)
+{
+ /* Program the counter frequency */
+ write_cntfrq_el0(plat_get_syscnt_freq2());
+
+ /* Initialize the cpu_ops pointer. */
+ init_cpu_ops();
+}
diff --git a/lib/psci/psci_suspend.c b/lib/psci/psci_suspend.c
index ee1ccef..904a4e7 100644
--- a/lib/psci/psci_suspend.c
+++ b/lib/psci/psci_suspend.c
@@ -37,7 +37,6 @@
#include <cpu_data.h>
#include <debug.h>
#include <platform.h>
-#include <runtime_svc.h>
#include <stddef.h>
#include "psci_private.h"