Merge tag 'tc2-pm' of git://git.linaro.org/people/pawelmoll/linux into next/soc

From Pawel Moll and Nicolas Pitre:
- Fixes to the existing Vexpress DCSCB backend.

- Lorenzo's minimal SPC driver required by the TC2 MCPM backend.

- The MCPM backend enabling SMP secondary boot and CPU hotplug
  on the VExpress TC2 big.LITTLE platform.

- MCPM suspend method to the TC2 backend allowing basic CPU
  idle/suspend.  The cpuidle driver that hooks into this will be
  submitted separately.

* tag 'tc2-pm' of git://git.linaro.org/people/pawelmoll/linux:
  ARM: vexpress/TC2: implement PM suspend method
  ARM: vexpress/TC2: basic PM support
  ARM: vexpress: Add SCC to V2P-CA15_A7's device tree
  ARM: vexpress/TC2: add Serial Power Controller (SPC) support
  ARM: vexpress/dcscb: fix cache disabling sequences

Signed-off-by: Olof Johansson <olof@lixom.net>

8 files changed, 633 insertions, 21 deletions

diff --git a/Documentation/devicetree/bindings/arm/vexpress-scc.txt b/Documentation/devicetree/bindings/arm/vexpress-scc.txt
new file mode 100644
index 000000000000..ae5043e42e5d
--- /dev/null
+++ b/Documentation/devicetree/bindings/arm/vexpress-scc.txt
@@ -0,0 +1,33 @@
+ARM Versatile Express Serial Configuration Controller
+-----------------------------------------------------
+
+Test chips for ARM Versatile Express platform implement SCC (Serial
+Configuration Controller) interface, used to set initial conditions
+for the test chip.
+
+In some cases its registers are also mapped in normal address space
+and can be used to obtain runtime information about the chip internals
+(like silicon temperature sensors) and as interface to other subsystems
+like platform configuration control and power management.
+
+Required properties:
+
+- compatible value: "arm,vexpress-scc,<model>", "arm,vexpress-scc";
+                    where <model> is the full tile model name (as used
+                    in the tile's Technical Reference Manual),
+                    eg. for Coretile Express A15x2 A7x3 (V2P-CA15_A7):
+        compatible = "arm,vexpress-scc,v2p-ca15_a7", "arm,vexpress-scc";
+
+Optional properties:
+
+- reg: when the SCC is memory mapped, physical address and size of the
+       registers window
+- interrupts: when the SCC can generate a system-level interrupt
+
+Example:
+
+        scc@7fff0000 {
+                compatible = "arm,vexpress-scc,v2p-ca15_a7", "arm,vexpress-scc";
+                reg = <0 0x7fff0000 0 0x1000>;
+                interrupts = <0 95 4>;
+        };
diff --git a/arch/arm/boot/dts/vexpress-v2p-ca15_a7.dts b/arch/arm/boot/dts/vexpress-v2p-ca15_a7.dts
index d2803be4e1a8..759b0cd20013 100644
--- a/arch/arm/boot/dts/vexpress-v2p-ca15_a7.dts
+++ b/arch/arm/boot/dts/vexpress-v2p-ca15_a7.dts
@@ -125,6 +125,12 @@
                 clock-names = "apb_pclk";
         };
 
+        scc@7fff0000 {
+                compatible = "arm,vexpress-scc,v2p-ca15_a7", "arm,vexpress-scc";
+                reg = <0 0x7fff0000 0 0x1000>;
+                interrupts = <0 95 4>;
+        };
+
         timer {
                 compatible = "arm,armv7-timer";
                 interrupts = <1 13 0xf08>,
diff --git a/arch/arm/mach-vexpress/Kconfig b/arch/arm/mach-vexpress/Kconfig
index b8bbabec6310..c700e623f9d8 100644
--- a/arch/arm/mach-vexpress/Kconfig
+++ b/arch/arm/mach-vexpress/Kconfig
@@ -66,4 +66,12 @@ config ARCH_VEXPRESS_DCSCB
           This is needed to provide CPU and cluster power management
           on RTSM implementing big.LITTLE.
 
+config ARCH_VEXPRESS_TC2_PM
+        bool "Versatile Express TC2 power management"
+        depends on MCPM
+        select ARM_CCI
+        help
+          Support for CPU and cluster power management on Versatile Express
+          with a TC2 (A15x2 A7x3) big.LITTLE core tile.
+
 endmenu
diff --git a/arch/arm/mach-vexpress/Makefile b/arch/arm/mach-vexpress/Makefile
index 48ba89a8149f..36ea8247123a 100644
--- a/arch/arm/mach-vexpress/Makefile
+++ b/arch/arm/mach-vexpress/Makefile
@@ -7,5 +7,6 @@ ccflags-$(CONFIG_ARCH_MULTIPLATFORM) := -I$(srctree)/$(src)/include \
 obj-y                                   := v2m.o
 obj-$(CONFIG_ARCH_VEXPRESS_CA9X4)       += ct-ca9x4.o
 obj-$(CONFIG_ARCH_VEXPRESS_DCSCB)       += dcscb.o      dcscb_setup.o
+obj-$(CONFIG_ARCH_VEXPRESS_TC2_PM)      += tc2_pm.o spc.o
 obj-$(CONFIG_SMP)                       += platsmp.o
 obj-$(CONFIG_HOTPLUG_CPU)               += hotplug.o
diff --git a/arch/arm/mach-vexpress/dcscb.c b/arch/arm/mach-vexpress/dcscb.c
index 16d57a8a9d5a..85fffa702f5b 100644
--- a/arch/arm/mach-vexpress/dcscb.c
+++ b/arch/arm/mach-vexpress/dcscb.c
@@ -136,14 +136,29 @@ static void dcscb_power_down(void)
                 /*
                  * Flush all cache levels for this cluster.
                  *
-                 * A15/A7 can hit in the cache with SCTLR.C=0, so we don't need
-                 * a preliminary flush here for those CPUs.  At least, that's
-                 * the theory -- without the extra flush, Linux explodes on
-                 * RTSM (to be investigated).
+                 * To do so we do:
+                 * - Clear the SCTLR.C bit to prevent further cache allocations
+                 * - Flush the whole cache
+                 * - Clear the ACTLR "SMP" bit to disable local coherency
+                 *
+                 * Let's do it in the safest possible way i.e. with
+                 * no memory access within the following sequence
+                 * including to the stack.
                  */
-                flush_cache_all();
-                set_cr(get_cr() & ~CR_C);
-                flush_cache_all();
+                asm volatile(
+                "mrc    p15, 0, r0, c1, c0, 0   @ get CR \n\t"
+                "bic    r0, r0, #"__stringify(CR_C)" \n\t"
+                "mcr    p15, 0, r0, c1, c0, 0   @ set CR \n\t"
+                "isb    \n\t"
+                "bl     v7_flush_dcache_all \n\t"
+                "clrex  \n\t"
+                "mrc    p15, 0, r0, c1, c0, 1   @ get AUXCR \n\t"
+                "bic    r0, r0, #(1 << 6)       @ disable local coherency \n\t"
+                "mcr    p15, 0, r0, c1, c0, 1   @ set AUXCR \n\t"
+                "isb    \n\t"
+                "dsb    "
+                : : : "r0","r1","r2","r3","r4","r5","r6","r7",
+                      "r9","r10","r11","lr","memory");
 
                 /*
                  * This is a harmless no-op.  On platforms with a real
@@ -152,9 +167,6 @@ static void dcscb_power_down(void)
                  */
                 outer_flush_all();
 
-                /* Disable local coherency by clearing the ACTLR "SMP" bit: */
-                set_auxcr(get_auxcr() & ~(1 << 6));
-
                 /*
                  * Disable cluster-level coherency by masking
                  * incoming snoops and DVM messages:
@@ -167,18 +179,22 @@ static void dcscb_power_down(void)
 
                 /*
                  * Flush the local CPU cache.
-                 *
-                 * A15/A7 can hit in the cache with SCTLR.C=0, so we don't need
-                 * a preliminary flush here for those CPUs.  At least, that's
-                 * the theory -- without the extra flush, Linux explodes on
-                 * RTSM (to be investigated).
+                 * Let's do it in the safest possible way as above.
                  */
-                flush_cache_louis();
-                set_cr(get_cr() & ~CR_C);
-                flush_cache_louis();
-
-                /* Disable local coherency by clearing the ACTLR "SMP" bit: */
-                set_auxcr(get_auxcr() & ~(1 << 6));
+                asm volatile(
+                "mrc    p15, 0, r0, c1, c0, 0   @ get CR \n\t"
+                "bic    r0, r0, #"__stringify(CR_C)" \n\t"
+                "mcr    p15, 0, r0, c1, c0, 0   @ set CR \n\t"
+                "isb    \n\t"
+                "bl     v7_flush_dcache_louis \n\t"
+                "clrex  \n\t"
+                "mrc    p15, 0, r0, c1, c0, 1   @ get AUXCR \n\t"
+                "bic    r0, r0, #(1 << 6)       @ disable local coherency \n\t"
+                "mcr    p15, 0, r0, c1, c0, 1   @ set AUXCR \n\t"
+                "isb    \n\t"
+                "dsb    "
+                : : : "r0","r1","r2","r3","r4","r5","r6","r7",
+                      "r9","r10","r11","lr","memory");
         }
 
         __mcpm_cpu_down(cpu, cluster);
diff --git a/arch/arm/mach-vexpress/spc.c b/arch/arm/mach-vexpress/spc.c
new file mode 100644
index 000000000000..eefb029197ca
--- /dev/null
+++ b/arch/arm/mach-vexpress/spc.c
@@ -0,0 +1,180 @@
+/*
+ * Versatile Express Serial Power Controller (SPC) support
+ *
+ * Copyright (C) 2013 ARM Ltd.
+ *
+ * Authors: Sudeep KarkadaNagesha <sudeep.karkadanagesha@arm.com>
+ *          Achin Gupta           <achin.gupta@arm.com>
+ *          Lorenzo Pieralisi     <lorenzo.pieralisi@arm.com>
+ *
+ * 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 "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/err.h>
+#include <linux/io.h>
+#include <linux/slab.h>
+
+#include <asm/cacheflush.h>
+
+#define SPCLOG "vexpress-spc: "
+
+/* SPC wake-up IRQs status and mask */
+#define WAKE_INT_MASK           0x24
+#define WAKE_INT_RAW            0x28
+#define WAKE_INT_STAT           0x2c
+/* SPC power down registers */
+#define A15_PWRDN_EN            0x30
+#define A7_PWRDN_EN             0x34
+/* SPC per-CPU mailboxes */
+#define A15_BX_ADDR0            0x68
+#define A7_BX_ADDR0             0x78
+
+/* wake-up interrupt masks */
+#define GBL_WAKEUP_INT_MSK      (0x3 << 10)
+
+/* TC2 static dual-cluster configuration */
+#define MAX_CLUSTERS            2
+
+struct ve_spc_drvdata {
+        void __iomem *baseaddr;
+        /*
+         * A15s cluster identifier
+         * It corresponds to A15 processors MPIDR[15:8] bitfield
+         */
+        u32 a15_clusid;
+};
+
+static struct ve_spc_drvdata *info;
+
+static inline bool cluster_is_a15(u32 cluster)
+{
+        return cluster == info->a15_clusid;
+}
+
+/**
+ * ve_spc_global_wakeup_irq()
+ *
+ * Function to set/clear global wakeup IRQs. Not protected by locking since
+ * it might be used in code paths where normal cacheable locks are not
+ * working. Locking must be provided by the caller to ensure atomicity.
+ *
+ * @set: if true, global wake-up IRQs are set, if false they are cleared
+ */
+void ve_spc_global_wakeup_irq(bool set)
+{
+        u32 reg;
+
+        reg = readl_relaxed(info->baseaddr + WAKE_INT_MASK);
+
+        if (set)
+                reg |= GBL_WAKEUP_INT_MSK;
+        else
+                reg &= ~GBL_WAKEUP_INT_MSK;
+
+        writel_relaxed(reg, info->baseaddr + WAKE_INT_MASK);
+}
+
+/**
+ * ve_spc_cpu_wakeup_irq()
+ *
+ * Function to set/clear per-CPU wake-up IRQs. Not protected by locking since
+ * it might be used in code paths where normal cacheable locks are not
+ * working. Locking must be provided by the caller to ensure atomicity.
+ *
+ * @cluster: mpidr[15:8] bitfield describing cluster affinity level
+ * @cpu: mpidr[7:0] bitfield describing cpu affinity level
+ * @set: if true, wake-up IRQs are set, if false they are cleared
+ */
+void ve_spc_cpu_wakeup_irq(u32 cluster, u32 cpu, bool set)
+{
+        u32 mask, reg;
+
+        if (cluster >= MAX_CLUSTERS)
+                return;
+
+        mask = 1 << cpu;
+
+        if (!cluster_is_a15(cluster))
+                mask <<= 4;
+
+        reg = readl_relaxed(info->baseaddr + WAKE_INT_MASK);
+
+        if (set)
+                reg |= mask;
+        else
+                reg &= ~mask;
+
+        writel_relaxed(reg, info->baseaddr + WAKE_INT_MASK);
+}
+
+/**
+ * ve_spc_set_resume_addr() - set the jump address used for warm boot
+ *
+ * @cluster: mpidr[15:8] bitfield describing cluster affinity level
+ * @cpu: mpidr[7:0] bitfield describing cpu affinity level
+ * @addr: physical resume address
+ */
+void ve_spc_set_resume_addr(u32 cluster, u32 cpu, u32 addr)
+{
+        void __iomem *baseaddr;
+
+        if (cluster >= MAX_CLUSTERS)
+                return;
+
+        if (cluster_is_a15(cluster))
+                baseaddr = info->baseaddr + A15_BX_ADDR0 + (cpu << 2);
+        else
+                baseaddr = info->baseaddr + A7_BX_ADDR0 + (cpu << 2);
+
+        writel_relaxed(addr, baseaddr);
+}
+
+/**
+ * ve_spc_powerdown()
+ *
+ * Function to enable/disable cluster powerdown. Not protected by locking
+ * since it might be used in code paths where normal cacheable locks are not
+ * working. Locking must be provided by the caller to ensure atomicity.
+ *
+ * @cluster: mpidr[15:8] bitfield describing cluster affinity level
+ * @enable: if true enables powerdown, if false disables it
+ */
+void ve_spc_powerdown(u32 cluster, bool enable)
+{
+        u32 pwdrn_reg;
+
+        if (cluster >= MAX_CLUSTERS)
+                return;
+
+        pwdrn_reg = cluster_is_a15(cluster) ? A15_PWRDN_EN : A7_PWRDN_EN;
+        writel_relaxed(enable, info->baseaddr + pwdrn_reg);
+}
+
+int __init ve_spc_init(void __iomem *baseaddr, u32 a15_clusid)
+{
+        info = kzalloc(sizeof(*info), GFP_KERNEL);
+        if (!info) {
+                pr_err(SPCLOG "unable to allocate mem\n");
+                return -ENOMEM;
+        }
+
+        info->baseaddr = baseaddr;
+        info->a15_clusid = a15_clusid;
+
+        /*
+         * Multi-cluster systems may need this data when non-coherent, during
+         * cluster power-up/power-down. Make sure driver info reaches main
+         * memory.
+         */
+        sync_cache_w(info);
+        sync_cache_w(&info);
+
+        return 0;
+}
diff --git a/arch/arm/mach-vexpress/spc.h b/arch/arm/mach-vexpress/spc.h
new file mode 100644
index 000000000000..5f7e4a446a17
--- /dev/null
+++ b/arch/arm/mach-vexpress/spc.h
@@ -0,0 +1,24 @@
+/*
+ * 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
+ */
+
+
+#ifndef __SPC_H_
+#define __SPC_H_
+
+int __init ve_spc_init(void __iomem *base, u32 a15_clusid);
+void ve_spc_global_wakeup_irq(bool set);
+void ve_spc_cpu_wakeup_irq(u32 cluster, u32 cpu, bool set);
+void ve_spc_set_resume_addr(u32 cluster, u32 cpu, u32 addr);
+void ve_spc_powerdown(u32 cluster, bool enable);
+
+#endif
diff --git a/arch/arm/mach-vexpress/tc2_pm.c b/arch/arm/mach-vexpress/tc2_pm.c
new file mode 100644
index 000000000000..ddd97dd4e9b7
--- /dev/null
+++ b/arch/arm/mach-vexpress/tc2_pm.c
@@ -0,0 +1,344 @@
+/*
+ * arch/arm/mach-vexpress/tc2_pm.c - TC2 power management support
+ *
+ * Created by:  Nicolas Pitre, October 2012
+ * Copyright:   (C) 2012-2013  Linaro Limited
+ *
+ * Some portions of this file were originally written by Achin Gupta
+ * Copyright:   (C) 2012  ARM Limited
+ *
+ * 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.
+ */
+
+#include <linux/init.h>
+#include <linux/io.h>
+#include <linux/kernel.h>
+#include <linux/of_address.h>
+#include <linux/spinlock.h>
+#include <linux/errno.h>
+
+#include <asm/mcpm.h>
+#include <asm/proc-fns.h>
+#include <asm/cacheflush.h>
+#include <asm/cputype.h>
+#include <asm/cp15.h>
+
+#include <linux/arm-cci.h>
+
+#include "spc.h"
+
+/* SCC conf registers */
+#define A15_CONF                0x400
+#define A7_CONF                 0x500
+#define SYS_INFO                0x700
+#define SPC_BASE                0xb00
+
+/*
+ * We can't use regular spinlocks. In the switcher case, it is possible
+ * for an outbound CPU to call power_down() after its inbound counterpart
+ * is already live using the same logical CPU number which trips lockdep
+ * debugging.
+ */
+static arch_spinlock_t tc2_pm_lock = __ARCH_SPIN_LOCK_UNLOCKED;
+
+#define TC2_CLUSTERS                    2
+#define TC2_MAX_CPUS_PER_CLUSTER        3
+
+static unsigned int tc2_nr_cpus[TC2_CLUSTERS];
+
+/* Keep per-cpu usage count to cope with unordered up/down requests */
+static int tc2_pm_use_count[TC2_MAX_CPUS_PER_CLUSTER][TC2_CLUSTERS];
+
+#define tc2_cluster_unused(cluster) \
+        (!tc2_pm_use_count[0][cluster] && \
+         !tc2_pm_use_count[1][cluster] && \
+         !tc2_pm_use_count[2][cluster])
+
+static int tc2_pm_power_up(unsigned int cpu, unsigned int cluster)
+{
+        pr_debug("%s: cpu %u cluster %u\n", __func__, cpu, cluster);
+        if (cluster >= TC2_CLUSTERS || cpu >= tc2_nr_cpus[cluster])
+                return -EINVAL;
+
+        /*
+         * Since this is called with IRQs enabled, and no arch_spin_lock_irq
+         * variant exists, we need to disable IRQs manually here.
+         */
+        local_irq_disable();
+        arch_spin_lock(&tc2_pm_lock);
+
+        if (tc2_cluster_unused(cluster))
+                ve_spc_powerdown(cluster, false);
+
+        tc2_pm_use_count[cpu][cluster]++;
+        if (tc2_pm_use_count[cpu][cluster] == 1) {
+                ve_spc_set_resume_addr(cluster, cpu,
+                                       virt_to_phys(mcpm_entry_point));
+                ve_spc_cpu_wakeup_irq(cluster, cpu, true);
+        } else if (tc2_pm_use_count[cpu][cluster] != 2) {
+                /*
+                 * The only possible values are:
+                 * 0 = CPU down
+                 * 1 = CPU (still) up
+                 * 2 = CPU requested to be up before it had a chance
+                 *     to actually make itself down.
+                 * Any other value is a bug.
+                 */
+                BUG();
+        }
+
+        arch_spin_unlock(&tc2_pm_lock);
+        local_irq_enable();
+
+        return 0;
+}
+
+static void tc2_pm_down(u64 residency)
+{
+        unsigned int mpidr, cpu, cluster;
+        bool last_man = false, skip_wfi = false;
+
+        mpidr = read_cpuid_mpidr();
+        cpu = MPIDR_AFFINITY_LEVEL(mpidr, 0);
+        cluster = MPIDR_AFFINITY_LEVEL(mpidr, 1);
+
+        pr_debug("%s: cpu %u cluster %u\n", __func__, cpu, cluster);
+        BUG_ON(cluster >= TC2_CLUSTERS || cpu >= TC2_MAX_CPUS_PER_CLUSTER);
+
+        __mcpm_cpu_going_down(cpu, cluster);
+
+        arch_spin_lock(&tc2_pm_lock);
+        BUG_ON(__mcpm_cluster_state(cluster) != CLUSTER_UP);
+        tc2_pm_use_count[cpu][cluster]--;
+        if (tc2_pm_use_count[cpu][cluster] == 0) {
+                ve_spc_cpu_wakeup_irq(cluster, cpu, true);
+                if (tc2_cluster_unused(cluster)) {
+                        ve_spc_powerdown(cluster, true);
+                        ve_spc_global_wakeup_irq(true);
+                        last_man = true;
+                }
+        } else if (tc2_pm_use_count[cpu][cluster] == 1) {
+                /*
+                 * A power_up request went ahead of us.
+                 * Even if we do not want to shut this CPU down,
+                 * the caller expects a certain state as if the WFI
+                 * was aborted.  So let's continue with cache cleaning.
+                 */
+                skip_wfi = true;
+        } else
+                BUG();
+
+        if (last_man && __mcpm_outbound_enter_critical(cpu, cluster)) {
+                arch_spin_unlock(&tc2_pm_lock);
+
+                if (read_cpuid_part_number() == ARM_CPU_PART_CORTEX_A15) {
+                        /*
+                         * On the Cortex-A15 we need to disable
+                         * L2 prefetching before flushing the cache.
+                         */
+                        asm volatile(
+                        "mcr    p15, 1, %0, c15, c0, 3 \n\t"
+                        "isb    \n\t"
+                        "dsb    "
+                        : : "r" (0x400) );
+                }
+
+                /*
+                 * We need to disable and flush the whole (L1 and L2) cache.
+                 * Let's do it in the safest possible way i.e. with
+                 * no memory access within the following sequence
+                 * including the stack.
+                 */
+                asm volatile(
+                "mrc    p15, 0, r0, c1, c0, 0   @ get CR \n\t"
+                "bic    r0, r0, #"__stringify(CR_C)" \n\t"
+                "mcr    p15, 0, r0, c1, c0, 0   @ set CR \n\t"
+                "isb    \n\t"
+                "bl     v7_flush_dcache_all \n\t"
+                "clrex  \n\t"
+                "mrc    p15, 0, r0, c1, c0, 1   @ get AUXCR \n\t"
+                "bic    r0, r0, #(1 << 6)       @ disable local coherency \n\t"
+                "mcr    p15, 0, r0, c1, c0, 1   @ set AUXCR \n\t"
+                "isb    \n\t"
+                "dsb    "
+                : : : "r0","r1","r2","r3","r4","r5","r6","r7",
+                      "r9","r10","r11","lr","memory");
+
+                cci_disable_port_by_cpu(mpidr);
+
+                __mcpm_outbound_leave_critical(cluster, CLUSTER_DOWN);
+        } else {
+                /*
+                 * If last man then undo any setup done previously.
+                 */
+                if (last_man) {
+                        ve_spc_powerdown(cluster, false);
+                        ve_spc_global_wakeup_irq(false);
+                }
+
+                arch_spin_unlock(&tc2_pm_lock);
+
+                /*
+                 * We need to disable and flush only the L1 cache.
+                 * Let's do it in the safest possible way as above.
+                 */
+                asm volatile(
+                "mrc    p15, 0, r0, c1, c0, 0   @ get CR \n\t"
+                "bic    r0, r0, #"__stringify(CR_C)" \n\t"
+                "mcr    p15, 0, r0, c1, c0, 0   @ set CR \n\t"
+                "isb    \n\t"
+                "bl     v7_flush_dcache_louis \n\t"
+                "clrex  \n\t"
+                "mrc    p15, 0, r0, c1, c0, 1   @ get AUXCR \n\t"
+                "bic    r0, r0, #(1 << 6)       @ disable local coherency \n\t"
+                "mcr    p15, 0, r0, c1, c0, 1   @ set AUXCR \n\t"
+                "isb    \n\t"
+                "dsb    "
+                : : : "r0","r1","r2","r3","r4","r5","r6","r7",
+                      "r9","r10","r11","lr","memory");
+        }
+
+        __mcpm_cpu_down(cpu, cluster);
+
+        /* Now we are prepared for power-down, do it: */
+        if (!skip_wfi)
+                wfi();
+
+        /* Not dead at this point?  Let our caller cope. */
+}
+
+static void tc2_pm_power_down(void)
+{
+        tc2_pm_down(0);
+}
+
+static void tc2_pm_suspend(u64 residency)
+{
+        unsigned int mpidr, cpu, cluster;
+
+        mpidr = read_cpuid_mpidr();
+        cpu = MPIDR_AFFINITY_LEVEL(mpidr, 0);
+        cluster = MPIDR_AFFINITY_LEVEL(mpidr, 1);
+        ve_spc_set_resume_addr(cluster, cpu, virt_to_phys(mcpm_entry_point));
+        tc2_pm_down(residency);
+}
+
+static void tc2_pm_powered_up(void)
+{
+        unsigned int mpidr, cpu, cluster;
+        unsigned long flags;
+
+        mpidr = read_cpuid_mpidr();
+        cpu = MPIDR_AFFINITY_LEVEL(mpidr, 0);
+        cluster = MPIDR_AFFINITY_LEVEL(mpidr, 1);
+
+        pr_debug("%s: cpu %u cluster %u\n", __func__, cpu, cluster);
+        BUG_ON(cluster >= TC2_CLUSTERS || cpu >= TC2_MAX_CPUS_PER_CLUSTER);
+
+        local_irq_save(flags);
+        arch_spin_lock(&tc2_pm_lock);
+
+        if (tc2_cluster_unused(cluster)) {
+                ve_spc_powerdown(cluster, false);
+                ve_spc_global_wakeup_irq(false);
+        }
+
+        if (!tc2_pm_use_count[cpu][cluster])
+                tc2_pm_use_count[cpu][cluster] = 1;
+
+        ve_spc_cpu_wakeup_irq(cluster, cpu, false);
+        ve_spc_set_resume_addr(cluster, cpu, 0);
+
+        arch_spin_unlock(&tc2_pm_lock);
+        local_irq_restore(flags);
+}
+
+static const struct mcpm_platform_ops tc2_pm_power_ops = {
+        .power_up       = tc2_pm_power_up,
+        .power_down     = tc2_pm_power_down,
+        .suspend        = tc2_pm_suspend,
+        .powered_up     = tc2_pm_powered_up,
+};
+
+static bool __init tc2_pm_usage_count_init(void)
+{
+        unsigned int mpidr, cpu, cluster;
+
+        mpidr = read_cpuid_mpidr();
+        cpu = MPIDR_AFFINITY_LEVEL(mpidr, 0);
+        cluster = MPIDR_AFFINITY_LEVEL(mpidr, 1);
+
+        pr_debug("%s: cpu %u cluster %u\n", __func__, cpu, cluster);
+        if (cluster >= TC2_CLUSTERS || cpu >= tc2_nr_cpus[cluster]) {
+                pr_err("%s: boot CPU is out of bound!\n", __func__);
+                return false;
+        }
+        tc2_pm_use_count[cpu][cluster] = 1;
+        return true;
+}
+
+/*
+ * Enable cluster-level coherency, in preparation for turning on the MMU.
+ */
+static void __naked tc2_pm_power_up_setup(unsigned int affinity_level)
+{
+        asm volatile (" \n"
+"       cmp     r0, #1 \n"
+"       bxne    lr \n"
+"       b       cci_enable_port_for_self ");
+}
+
+static int __init tc2_pm_init(void)
+{
+        int ret;
+        void __iomem *scc;
+        u32 a15_cluster_id, a7_cluster_id, sys_info;
+        struct device_node *np;
+
+        /*
+         * The power management-related features are hidden behind
+         * SCC registers. We need to extract runtime information like
+         * cluster ids and number of CPUs really available in clusters.
+         */
+        np = of_find_compatible_node(NULL, NULL,
+                        "arm,vexpress-scc,v2p-ca15_a7");
+        scc = of_iomap(np, 0);
+        if (!scc)
+                return -ENODEV;
+
+        a15_cluster_id = readl_relaxed(scc + A15_CONF) & 0xf;
+        a7_cluster_id = readl_relaxed(scc + A7_CONF) & 0xf;
+        if (a15_cluster_id >= TC2_CLUSTERS || a7_cluster_id >= TC2_CLUSTERS)
+                return -EINVAL;
+
+        sys_info = readl_relaxed(scc + SYS_INFO);
+        tc2_nr_cpus[a15_cluster_id] = (sys_info >> 16) & 0xf;
+        tc2_nr_cpus[a7_cluster_id] = (sys_info >> 20) & 0xf;
+
+        /*
+         * A subset of the SCC registers is also used to communicate
+         * with the SPC (power controller). We need to be able to
+         * drive it very early in the boot process to power up
+         * processors, so we initialize the SPC driver here.
+         */
+        ret = ve_spc_init(scc + SPC_BASE, a15_cluster_id);
+        if (ret)
+                return ret;
+
+        if (!cci_probed())
+                return -ENODEV;
+
+        if (!tc2_pm_usage_count_init())
+                return -EINVAL;
+
+        ret = mcpm_platform_register(&tc2_pm_power_ops);
+        if (!ret) {
+                mcpm_sync_init(tc2_pm_power_up_setup);
+                pr_info("TC2 power management initialized\n");
+        }
+        return ret;
+}
+
+early_initcall(tc2_pm_init);