xen/acpi: ACPI PAD driver

PAD is acpi Processor Aggregator Device which provides a control point
that enables the platform to perform specific processor configuration
and control that applies to all processors in the platform.

This patch is to implement Xen acpi pad logic. When running under Xen
virt platform, native pad driver would not work. Instead Xen pad driver,
a self-contained and thin logic level, would take over acpi pad logic.

When acpi pad notify OSPM, xen pad logic intercept and parse _PUR object
to get the expected idle cpu number, and then hypercall to hypervisor.
Xen hypervisor would then do the rest work, say, core parking, to idle
specific number of cpus on its own policy.

Signed-off-by: Jan Beulich <JBeulich@suse.com>
Signed-off-by: Liu Jinsong <jinsong.liu@intel.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>

4 files changed, 215 insertions, 1 deletions

diff --git a/drivers/xen/Makefile b/drivers/xen/Makefile
index 0e8637035457..3c397170f39c 100644
--- a/drivers/xen/Makefile
+++ b/drivers/xen/Makefile
@@ -10,7 +10,8 @@ CFLAGS_features.o			:= $(nostackp)
 
 dom0-$(CONFIG_PCI) += pci.o
 dom0-$(CONFIG_USB_SUPPORT) += dbgp.o
-dom0-$(CONFIG_ACPI) += acpi.o
+dom0-$(CONFIG_ACPI) += acpi.o $(xen-pad-y)
+xen-pad-$(CONFIG_X86) += xen-acpi-pad.o
 dom0-$(CONFIG_X86) += pcpu.o
 obj-$(CONFIG_XEN_DOM0)			+= $(dom0-y)
 obj-$(CONFIG_BLOCK)			+= biomerge.o
diff --git a/drivers/xen/xen-acpi-pad.c b/drivers/xen/xen-acpi-pad.c
new file mode 100644
index 000000000000..f23ecf380088
--- /dev/null
+++ b/drivers/xen/xen-acpi-pad.c
@@ -0,0 +1,181 @@
+/*
+ * xen-acpi-pad.c - Xen pad interface
+ *
+ * Copyright (c) 2012, Intel Corporation.
+ *    Author: Liu, Jinsong <jinsong.liu@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope 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/kernel.h>
+#include <linux/types.h>
+#include <acpi/acpi_bus.h>
+#include <acpi/acpi_drivers.h>
+#include <asm/xen/hypercall.h>
+#include <xen/interface/version.h>
+
+#define ACPI_PROCESSOR_AGGREGATOR_CLASS	"acpi_pad"
+#define ACPI_PROCESSOR_AGGREGATOR_DEVICE_NAME "Processor Aggregator"
+#define ACPI_PROCESSOR_AGGREGATOR_NOTIFY 0x80
+static DEFINE_MUTEX(xen_cpu_lock);
+
+static int xen_acpi_pad_idle_cpus(unsigned int idle_nums)
+{
+	struct xen_platform_op op;
+
+	op.cmd = XENPF_core_parking;
+	op.u.core_parking.type = XEN_CORE_PARKING_SET;
+	op.u.core_parking.idle_nums = idle_nums;
+
+	return HYPERVISOR_dom0_op(&op);
+}
+
+static int xen_acpi_pad_idle_cpus_num(void)
+{
+	struct xen_platform_op op;
+
+	op.cmd = XENPF_core_parking;
+	op.u.core_parking.type = XEN_CORE_PARKING_GET;
+
+	return HYPERVISOR_dom0_op(&op)
+	       ?: op.u.core_parking.idle_nums;
+}
+
+/*
+ * Query firmware how many CPUs should be idle
+ * return -1 on failure
+ */
+static int acpi_pad_pur(acpi_handle handle)
+{
+	struct acpi_buffer buffer = {ACPI_ALLOCATE_BUFFER, NULL};
+	union acpi_object *package;
+	int num = -1;
+
+	if (ACPI_FAILURE(acpi_evaluate_object(handle, "_PUR", NULL, &buffer)))
+		return num;
+
+	if (!buffer.length || !buffer.pointer)
+		return num;
+
+	package = buffer.pointer;
+
+	if (package->type == ACPI_TYPE_PACKAGE &&
+		package->package.count == 2 &&
+		package->package.elements[0].integer.value == 1) /* rev 1 */
+		num = package->package.elements[1].integer.value;
+
+	kfree(buffer.pointer);
+	return num;
+}
+
+/* Notify firmware how many CPUs are idle */
+static void acpi_pad_ost(acpi_handle handle, int stat,
+	uint32_t idle_nums)
+{
+	union acpi_object params[3] = {
+		{.type = ACPI_TYPE_INTEGER,},
+		{.type = ACPI_TYPE_INTEGER,},
+		{.type = ACPI_TYPE_BUFFER,},
+	};
+	struct acpi_object_list arg_list = {3, params};
+
+	params[0].integer.value = ACPI_PROCESSOR_AGGREGATOR_NOTIFY;
+	params[1].integer.value =  stat;
+	params[2].buffer.length = 4;
+	params[2].buffer.pointer = (void *)&idle_nums;
+	acpi_evaluate_object(handle, "_OST", &arg_list, NULL);
+}
+
+static void acpi_pad_handle_notify(acpi_handle handle)
+{
+	int idle_nums;
+
+	mutex_lock(&xen_cpu_lock);
+	idle_nums = acpi_pad_pur(handle);
+	if (idle_nums < 0) {
+		mutex_unlock(&xen_cpu_lock);
+		return;
+	}
+
+	idle_nums = xen_acpi_pad_idle_cpus(idle_nums)
+		    ?: xen_acpi_pad_idle_cpus_num();
+	if (idle_nums >= 0)
+		acpi_pad_ost(handle, 0, idle_nums);
+	mutex_unlock(&xen_cpu_lock);
+}
+
+static void acpi_pad_notify(acpi_handle handle, u32 event,
+	void *data)
+{
+	switch (event) {
+	case ACPI_PROCESSOR_AGGREGATOR_NOTIFY:
+		acpi_pad_handle_notify(handle);
+		break;
+	default:
+		pr_warn("Unsupported event [0x%x]\n", event);
+		break;
+	}
+}
+
+static int acpi_pad_add(struct acpi_device *device)
+{
+	acpi_status status;
+
+	strcpy(acpi_device_name(device), ACPI_PROCESSOR_AGGREGATOR_DEVICE_NAME);
+	strcpy(acpi_device_class(device), ACPI_PROCESSOR_AGGREGATOR_CLASS);
+
+	status = acpi_install_notify_handler(device->handle,
+		ACPI_DEVICE_NOTIFY, acpi_pad_notify, device);
+	if (ACPI_FAILURE(status))
+		return -ENODEV;
+
+	return 0;
+}
+
+static int acpi_pad_remove(struct acpi_device *device,
+	int type)
+{
+	mutex_lock(&xen_cpu_lock);
+	xen_acpi_pad_idle_cpus(0);
+	mutex_unlock(&xen_cpu_lock);
+
+	acpi_remove_notify_handler(device->handle,
+		ACPI_DEVICE_NOTIFY, acpi_pad_notify);
+	return 0;
+}
+
+static const struct acpi_device_id pad_device_ids[] = {
+	{"ACPI000C", 0},
+	{"", 0},
+};
+
+static struct acpi_driver acpi_pad_driver = {
+	.name = "processor_aggregator",
+	.class = ACPI_PROCESSOR_AGGREGATOR_CLASS,
+	.ids = pad_device_ids,
+	.ops = {
+		.add = acpi_pad_add,
+		.remove = acpi_pad_remove,
+	},
+};
+
+static int __init xen_acpi_pad_init(void)
+{
+	/* Only DOM0 is responsible for Xen acpi pad */
+	if (!xen_initial_domain())
+		return -ENODEV;
+
+	/* Only Xen4.2 or later support Xen acpi pad */
+	if (!xen_running_on_version_or_later(4, 2))
+		return -ENODEV;
+
+	return acpi_bus_register_driver(&acpi_pad_driver);
+}
+subsys_initcall(xen_acpi_pad_init);
diff --git a/include/xen/interface/platform.h b/include/xen/interface/platform.h
index 4755b5fac9c7..5e36932ab407 100644
--- a/include/xen/interface/platform.h
+++ b/include/xen/interface/platform.h
@@ -324,6 +324,22 @@ struct xenpf_cpu_ol {
 };
 DEFINE_GUEST_HANDLE_STRUCT(xenpf_cpu_ol);
 
+/*
+ * CMD 58 and 59 are reserved for cpu hotadd and memory hotadd,
+ * which are already occupied at Xen hypervisor side.
+ */
+#define XENPF_core_parking     60
+struct xenpf_core_parking {
+	/* IN variables */
+#define XEN_CORE_PARKING_SET   1
+#define XEN_CORE_PARKING_GET   2
+	uint32_t type;
+	/* IN variables:  set cpu nums expected to be idled */
+	/* OUT variables: get cpu nums actually be idled */
+	uint32_t idle_nums;
+};
+DEFINE_GUEST_HANDLE_STRUCT(xenpf_core_parking);
+
 struct xen_platform_op {
 	uint32_t cmd;
 	uint32_t interface_version; /* XENPF_INTERFACE_VERSION */
@@ -341,6 +357,7 @@ struct xen_platform_op {
 		struct xenpf_set_processor_pminfo set_pminfo;
 		struct xenpf_pcpuinfo          pcpu_info;
 		struct xenpf_cpu_ol            cpu_ol;
+		struct xenpf_core_parking      core_parking;
 		uint8_t                        pad[128];
 	} u;
 };
diff --git a/include/xen/interface/version.h b/include/xen/interface/version.h
index 7ff6498679a3..96d8d3deea6b 100644
--- a/include/xen/interface/version.h
+++ b/include/xen/interface/version.h
@@ -63,4 +63,19 @@ struct xen_feature_info {
 /* arg == xen_domain_handle_t. */
 #define XENVER_guest_handle 8
 
+/* Check if running on Xen version (major, minor) or later */
+static inline bool
+xen_running_on_version_or_later(unsigned int major, unsigned int minor)
+{
+	unsigned int version;
+
+	if (!xen_domain())
+		return false;
+
+	version = HYPERVISOR_xen_version(XENVER_version, NULL);
+	if ((((version >> 16) == major) && ((version & 0xffff) >= minor)) ||
+		((version >> 16) > major))
+		return true;
+	return false;
+}
 #endif /* __XEN_PUBLIC_VERSION_H__ */