arm: zynq: Add smp support

Zynq is dual core Cortex A9 which starts always
at zero. Using simple trampoline ensure long jump
to secondary_startup code.

Signed-off-by: Michal Simek <michal.simek@xilinx.com>
Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
diff --git a/arch/arm/mach-zynq/Kconfig b/arch/arm/mach-zynq/Kconfig
index d70651e..f4a7e63 100644
--- a/arch/arm/mach-zynq/Kconfig
+++ b/arch/arm/mach-zynq/Kconfig
@@ -8,6 +8,7 @@
 	select ICST
 	select MIGHT_HAVE_CACHE_L2X0
 	select USE_OF
+	select HAVE_SMP
 	select SPARSE_IRQ
 	select CADENCE_TTC_TIMER
 	help
diff --git a/arch/arm/mach-zynq/Makefile b/arch/arm/mach-zynq/Makefile
index 13ee09b..b595d22 100644
--- a/arch/arm/mach-zynq/Makefile
+++ b/arch/arm/mach-zynq/Makefile
@@ -4,3 +4,4 @@
 
 # Common support
 obj-y				:= common.o slcr.o
+obj-$(CONFIG_SMP)		+= headsmp.o platsmp.o
diff --git a/arch/arm/mach-zynq/common.c b/arch/arm/mach-zynq/common.c
index e1b6131..5bfe703 100644
--- a/arch/arm/mach-zynq/common.c
+++ b/arch/arm/mach-zynq/common.c
@@ -104,6 +104,7 @@
 };
 
 MACHINE_START(XILINX_EP107, "Xilinx Zynq Platform")
+	.smp		= smp_ops(zynq_smp_ops),
 	.map_io		= zynq_map_io,
 	.init_irq	= irqchip_init,
 	.init_machine	= zynq_init_machine,
diff --git a/arch/arm/mach-zynq/common.h b/arch/arm/mach-zynq/common.h
index d7ec3ca..fd308f8 100644
--- a/arch/arm/mach-zynq/common.h
+++ b/arch/arm/mach-zynq/common.h
@@ -19,6 +19,17 @@
 
 extern int zynq_slcr_init(void);
 extern void zynq_slcr_system_reset(void);
+extern void zynq_slcr_cpu_stop(int cpu);
+extern void zynq_slcr_cpu_start(int cpu);
+
+#ifdef CONFIG_SMP
+extern void secondary_startup(void);
+extern char zynq_secondary_trampoline;
+extern char zynq_secondary_trampoline_jump;
+extern char zynq_secondary_trampoline_end;
+extern int __cpuinit zynq_cpun_start(u32 address, int cpu);
+extern struct smp_operations zynq_smp_ops __initdata;
+#endif
 
 extern void __iomem *zynq_slcr_base;
 extern void __iomem *zynq_scu_base;
diff --git a/arch/arm/mach-zynq/headsmp.S b/arch/arm/mach-zynq/headsmp.S
new file mode 100644
index 0000000..d183cd2
--- /dev/null
+++ b/arch/arm/mach-zynq/headsmp.S
@@ -0,0 +1,24 @@
+/*
+ * Copyright (c) 2013 Steffen Trumtrar <s.trumtrar@pengutronix.de>
+ * Copyright (c) 2012-2013 Xilinx
+ *
+ * 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/linkage.h>
+#include <linux/init.h>
+
+	__CPUINIT
+
+ENTRY(zynq_secondary_trampoline)
+	ldr	r0, [pc]
+	bx	r0
+.globl zynq_secondary_trampoline_jump
+zynq_secondary_trampoline_jump:
+	/* Space for jumping address */
+	.word	/* cpu 1 */
+.globl zynq_secondary_trampoline_end
+zynq_secondary_trampoline_end:
+
+ENDPROC(zynq_secondary_trampoline)
diff --git a/arch/arm/mach-zynq/platsmp.c b/arch/arm/mach-zynq/platsmp.c
new file mode 100644
index 0000000..cdfd888
--- /dev/null
+++ b/arch/arm/mach-zynq/platsmp.c
@@ -0,0 +1,149 @@
+/*
+ * This file contains Xilinx specific SMP code, used to start up
+ * the second processor.
+ *
+ * Copyright (C) 2011-2013 Xilinx
+ *
+ * based on linux/arch/arm/mach-realview/platsmp.c
+ *
+ * Copyright (C) 2002 ARM Ltd.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * 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.
+ */
+
+#include <linux/export.h>
+#include <linux/jiffies.h>
+#include <linux/init.h>
+#include <linux/io.h>
+#include <asm/cacheflush.h>
+#include <asm/smp_scu.h>
+#include <linux/irqchip/arm-gic.h>
+#include "common.h"
+
+/*
+ * Store number of cores in the system
+ * Because of scu_get_core_count() must be in __init section and can't
+ * be called from zynq_cpun_start() because it is in __cpuinit section.
+ */
+static int ncores;
+
+/* Secondary CPU kernel startup is a 2 step process. The primary CPU
+ * starts the secondary CPU by giving it the address of the kernel and
+ * then sending it an event to wake it up. The secondary CPU then
+ * starts the kernel and tells the primary CPU it's up and running.
+ */
+static void __cpuinit zynq_secondary_init(unsigned int cpu)
+{
+	/*
+	 * if any interrupts are already enabled for the primary
+	 * core (e.g. timer irq), then they will not have been enabled
+	 * for us: do so
+	 */
+	gic_secondary_init(0);
+}
+
+int __cpuinit zynq_cpun_start(u32 address, int cpu)
+{
+	u32 trampoline_code_size = &zynq_secondary_trampoline_end -
+						&zynq_secondary_trampoline;
+
+	if (cpu > ncores) {
+		pr_warn("CPU No. is not available in the system\n");
+		return -1;
+	}
+
+	/* MS: Expectation that SLCR are directly map and accessible */
+	/* Not possible to jump to non aligned address */
+	if (!(address & 3) && (!address || (address >= trampoline_code_size))) {
+		/* Store pointer to ioremap area which points to address 0x0 */
+		static u8 __iomem *zero;
+		u32 trampoline_size = &zynq_secondary_trampoline_jump -
+						&zynq_secondary_trampoline;
+
+		zynq_slcr_cpu_stop(cpu);
+
+		if (__pa(PAGE_OFFSET)) {
+			zero = ioremap(0, trampoline_code_size);
+			if (!zero) {
+				pr_warn("BOOTUP jump vectors not accessible\n");
+				return -1;
+			}
+		} else {
+			zero = (__force u8 __iomem *)PAGE_OFFSET;
+		}
+
+		/*
+		 * This is elegant way how to jump to any address
+		 * 0x0: Load address at 0x8 to r0
+		 * 0x4: Jump by mov instruction
+		 * 0x8: Jumping address
+		 */
+		memcpy((__force void *)zero, &zynq_secondary_trampoline,
+						trampoline_size);
+		writel(address, zero + trampoline_size);
+
+		flush_cache_all();
+		outer_flush_range(0, trampoline_code_size);
+		smp_wmb();
+
+		if (__pa(PAGE_OFFSET))
+			iounmap(zero);
+
+		zynq_slcr_cpu_start(cpu);
+
+		return 0;
+	}
+
+	pr_warn("Can't start CPU%d: Wrong starting address %x\n", cpu, address);
+
+	return -1;
+}
+EXPORT_SYMBOL(zynq_cpun_start);
+
+static int __cpuinit zynq_boot_secondary(unsigned int cpu,
+						struct task_struct *idle)
+{
+	return zynq_cpun_start(virt_to_phys(secondary_startup), cpu);
+}
+
+/*
+ * Initialise the CPU possible map early - this describes the CPUs
+ * which may be present or become present in the system.
+ */
+static void __init zynq_smp_init_cpus(void)
+{
+	int i;
+
+	ncores = scu_get_core_count(zynq_scu_base);
+
+	for (i = 0; i < ncores && i < CONFIG_NR_CPUS; i++)
+		set_cpu_possible(i, true);
+}
+
+static void __init zynq_smp_prepare_cpus(unsigned int max_cpus)
+{
+	int i;
+
+	/*
+	 * Initialise the present map, which describes the set of CPUs
+	 * actually populated at the present time.
+	 */
+	for (i = 0; i < max_cpus; i++)
+		set_cpu_present(i, true);
+
+	scu_enable(zynq_scu_base);
+}
+
+struct smp_operations zynq_smp_ops __initdata = {
+	.smp_init_cpus		= zynq_smp_init_cpus,
+	.smp_prepare_cpus	= zynq_smp_prepare_cpus,
+	.smp_secondary_init	= zynq_secondary_init,
+	.smp_boot_secondary	= zynq_boot_secondary,
+};
diff --git a/arch/arm/mach-zynq/slcr.c b/arch/arm/mach-zynq/slcr.c
index d58c996..c70969b 100644
--- a/arch/arm/mach-zynq/slcr.c
+++ b/arch/arm/mach-zynq/slcr.c
@@ -33,6 +33,11 @@
 #define SLCR_UNLOCK			0x8   /* SCLR unlock register */
 
 #define SLCR_PS_RST_CTRL_OFFSET		0x200 /* PS Software Reset Control */
+
+#define SLCR_A9_CPU_CLKSTOP		0x10
+#define SLCR_A9_CPU_RST			0x1
+
+#define SLCR_A9_CPU_RST_CTRL		0x244 /* CPU Software Reset Control */
 #define SLCR_REBOOT_STATUS		0x258 /* PS Reboot Status */
 
 void __iomem *zynq_slcr_base;
@@ -62,6 +67,30 @@
 }
 
 /**
+ * zynq_slcr_cpu_start - Start cpu
+ * @cpu:	cpu number
+ */
+void zynq_slcr_cpu_start(int cpu)
+{
+	/* enable CPUn */
+	writel(SLCR_A9_CPU_CLKSTOP << cpu,
+	       zynq_slcr_base + SLCR_A9_CPU_RST_CTRL);
+	/* enable CLK for CPUn */
+	writel(0x0 << cpu, zynq_slcr_base + SLCR_A9_CPU_RST_CTRL);
+}
+
+/**
+ * zynq_slcr_cpu_stop - Stop cpu
+ * @cpu:	cpu number
+ */
+void zynq_slcr_cpu_stop(int cpu)
+{
+	/* stop CLK and reset CPUn */
+	writel((SLCR_A9_CPU_CLKSTOP | SLCR_A9_CPU_RST) << cpu,
+	       zynq_slcr_base + SLCR_A9_CPU_RST_CTRL);
+}
+
+/**
  * zynq_slcr_init
  * Returns 0 on success, negative errno otherwise.
  *