[linux.git] / drivers / hv / hv.c

/*
 * Copyright (c) 2009, Microsoft Corporation.
 *
 * 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.
 *
 * You should have received a copy of the GNU General Public License along with
 * this program; if not, write to the Free Software Foundation, Inc., 59 Temple
 * Place - Suite 330, Boston, MA 02111-1307 USA.
 *
 * Authors:
 *   Haiyang Zhang <[email protected]>
 *   Hank Janssen  <[email protected]>
 *
 */
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt

#include <linux/kernel.h>
#include <linux/mm.h>
#include <linux/slab.h>
#include <linux/vmalloc.h>
#include <linux/hyperv.h>
#include <linux/version.h>
#include <linux/random.h>
#include <linux/clockchips.h>
#include <asm/mshyperv.h>
#include "hyperv_vmbus.h"

/* The one and only */
struct hv_context hv_context = {
	.synic_initialized	= false,
};

/*
 * If false, we're using the old mechanism for stimer0 interrupts
 * where it sends a VMbus message when it expires. The old
 * mechanism is used when running on older versions of Hyper-V
 * that don't support Direct Mode. While Hyper-V provides
 * four stimer's per CPU, Linux uses only stimer0.
 */
static bool direct_mode_enabled;
static int stimer0_irq;
static int stimer0_vector;

#define HV_TIMER_FREQUENCY (10 * 1000 * 1000) /* 100ns period */
#define HV_MAX_MAX_DELTA_TICKS 0xffffffff
#define HV_MIN_DELTA_TICKS 1

/*
 * hv_init - Main initialization routine.
 *
 * This routine must be called before any other routines in here are called
 */
int hv_init(void)
{
	hv_context.cpu_context = alloc_percpu(struct hv_per_cpu_context);
	if (!hv_context.cpu_context)
		return -ENOMEM;

	direct_mode_enabled = ms_hyperv.misc_features &
			HV_X64_STIMER_DIRECT_MODE_AVAILABLE;
	return 0;
}

/*
 * hv_post_message - Post a message using the hypervisor message IPC.
 *
 * This involves a hypercall.
 */
int hv_post_message(union hv_connection_id connection_id,
		  enum hv_message_type message_type,
		  void *payload, size_t payload_size)
{
	struct hv_input_post_message *aligned_msg;
	struct hv_per_cpu_context *hv_cpu;
	u64 status;

	if (payload_size > HV_MESSAGE_PAYLOAD_BYTE_COUNT)
		return -EMSGSIZE;

	hv_cpu = get_cpu_ptr(hv_context.cpu_context);
	aligned_msg = hv_cpu->post_msg_page;
	aligned_msg->connectionid = connection_id;
	aligned_msg->reserved = 0;
	aligned_msg->message_type = message_type;
	aligned_msg->payload_size = payload_size;
	memcpy((void *)aligned_msg->payload, payload, payload_size);

	status = hv_do_hypercall(HVCALL_POST_MESSAGE, aligned_msg, NULL);

	/* Preemption must remain disabled until after the hypercall
	 * so some other thread can't get scheduled onto this cpu and
	 * corrupt the per-cpu post_msg_page
	 */
	put_cpu_ptr(hv_cpu);

	return status & 0xFFFF;
}

/*
 * ISR for when stimer0 is operating in Direct Mode.  Direct Mode
 * does not use VMbus or any VMbus messages, so process here and not
 * in the VMbus driver code.
 */

static void hv_stimer0_isr(void)
{
	struct hv_per_cpu_context *hv_cpu;

	hv_cpu = this_cpu_ptr(hv_context.cpu_context);
	hv_cpu->clk_evt->event_handler(hv_cpu->clk_evt);
	add_interrupt_randomness(stimer0_vector, 0);
}

static int hv_ce_set_next_event(unsigned long delta,
				struct clock_event_device *evt)
{
	u64 current_tick;

	WARN_ON(!clockevent_state_oneshot(evt));

	current_tick = hyperv_cs->read(NULL);
	current_tick += delta;
	hv_init_timer(0, current_tick);
	return 0;
}

static int hv_ce_shutdown(struct clock_event_device *evt)
{
	hv_init_timer(0, 0);
	hv_init_timer_config(0, 0);
	if (direct_mode_enabled)
		hv_disable_stimer0_percpu_irq(stimer0_irq);

	return 0;
}

static int hv_ce_set_oneshot(struct clock_event_device *evt)
{
	union hv_timer_config timer_cfg;

	timer_cfg.as_uint64 = 0;
	timer_cfg.enable = 1;
	timer_cfg.auto_enable = 1;
	if (direct_mode_enabled) {
		/*
		 * When it expires, the timer will directly interrupt
		 * on the specified hardware vector/IRQ.
		 */
		timer_cfg.direct_mode = 1;
		timer_cfg.apic_vector = stimer0_vector;
		hv_enable_stimer0_percpu_irq(stimer0_irq);
	} else {
		/*
		 * When it expires, the timer will generate a VMbus message,
		 * to be handled by the normal VMbus interrupt handler.
		 */
		timer_cfg.direct_mode = 0;
		timer_cfg.sintx = VMBUS_MESSAGE_SINT;
	}
	hv_init_timer_config(0, timer_cfg.as_uint64);
	return 0;
}

static void hv_init_clockevent_device(struct clock_event_device *dev, int cpu)
{
	dev->name = "Hyper-V clockevent";
	dev->features = CLOCK_EVT_FEAT_ONESHOT;
	dev->cpumask = cpumask_of(cpu);
	dev->rating = 1000;
	/*
	 * Avoid settint dev->owner = THIS_MODULE deliberately as doing so will
	 * result in clockevents_config_and_register() taking additional
	 * references to the hv_vmbus module making it impossible to unload.
	 */

	dev->set_state_shutdown = hv_ce_shutdown;
	dev->set_state_oneshot = hv_ce_set_oneshot;
	dev->set_next_event = hv_ce_set_next_event;
}


int hv_synic_alloc(void)
{
	int cpu;

	hv_context.hv_numa_map = kcalloc(nr_node_ids, sizeof(struct cpumask),
					 GFP_KERNEL);
	if (hv_context.hv_numa_map == NULL) {
		pr_err("Unable to allocate NUMA map\n");
		goto err;
	}

	for_each_present_cpu(cpu) {
		struct hv_per_cpu_context *hv_cpu
			= per_cpu_ptr(hv_context.cpu_context, cpu);

		memset(hv_cpu, 0, sizeof(*hv_cpu));
		tasklet_init(&hv_cpu->msg_dpc,
			     vmbus_on_msg_dpc, (unsigned long) hv_cpu);

		hv_cpu->clk_evt = kzalloc(sizeof(struct clock_event_device),
					  GFP_KERNEL);
		if (hv_cpu->clk_evt == NULL) {
			pr_err("Unable to allocate clock event device\n");
			goto err;
		}
		hv_init_clockevent_device(hv_cpu->clk_evt, cpu);

		hv_cpu->synic_message_page =
			(void *)get_zeroed_page(GFP_ATOMIC);
		if (hv_cpu->synic_message_page == NULL) {
			pr_err("Unable to allocate SYNIC message page\n");
			goto err;
		}

		hv_cpu->synic_event_page = (void *)get_zeroed_page(GFP_ATOMIC);
		if (hv_cpu->synic_event_page == NULL) {
			pr_err("Unable to allocate SYNIC event page\n");
			goto err;
		}

		hv_cpu->post_msg_page = (void *)get_zeroed_page(GFP_ATOMIC);
		if (hv_cpu->post_msg_page == NULL) {
			pr_err("Unable to allocate post msg page\n");
			goto err;
		}

		INIT_LIST_HEAD(&hv_cpu->chan_list);
	}

	if (direct_mode_enabled &&
	    hv_setup_stimer0_irq(&stimer0_irq, &stimer0_vector,
				hv_stimer0_isr))
		goto err;

	return 0;
err:
	return -ENOMEM;
}


void hv_synic_free(void)
{
	int cpu;

	for_each_present_cpu(cpu) {
		struct hv_per_cpu_context *hv_cpu
			= per_cpu_ptr(hv_context.cpu_context, cpu);

		if (hv_cpu->synic_event_page)
			free_page((unsigned long)hv_cpu->synic_event_page);
		if (hv_cpu->synic_message_page)
			free_page((unsigned long)hv_cpu->synic_message_page);
		if (hv_cpu->post_msg_page)
			free_page((unsigned long)hv_cpu->post_msg_page);
	}

	kfree(hv_context.hv_numa_map);
}

/*
 * hv_synic_init - Initialize the Synthetic Interrupt Controller.
 *
 * If it is already initialized by another entity (ie x2v shim), we need to
 * retrieve the initialized message and event pages.  Otherwise, we create and
 * initialize the message and event pages.
 */
int hv_synic_init(unsigned int cpu)
{
	struct hv_per_cpu_context *hv_cpu
		= per_cpu_ptr(hv_context.cpu_context, cpu);
	union hv_synic_simp simp;
	union hv_synic_siefp siefp;
	union hv_synic_sint shared_sint;
	union hv_synic_scontrol sctrl;

	/* Setup the Synic's message page */
	hv_get_simp(simp.as_uint64);
	simp.simp_enabled = 1;
	simp.base_simp_gpa = virt_to_phys(hv_cpu->synic_message_page)
		>> PAGE_SHIFT;

	hv_set_simp(simp.as_uint64);

	/* Setup the Synic's event page */
	hv_get_siefp(siefp.as_uint64);
	siefp.siefp_enabled = 1;
	siefp.base_siefp_gpa = virt_to_phys(hv_cpu->synic_event_page)
		>> PAGE_SHIFT;

	hv_set_siefp(siefp.as_uint64);

	/* Setup the shared SINT. */
	hv_get_synint_state(VMBUS_MESSAGE_SINT, shared_sint.as_uint64);

	shared_sint.vector = HYPERVISOR_CALLBACK_VECTOR;
	shared_sint.masked = false;
	if (ms_hyperv.hints & HV_X64_DEPRECATING_AEOI_RECOMMENDED)
		shared_sint.auto_eoi = false;
	else
		shared_sint.auto_eoi = true;

	hv_set_synint_state(VMBUS_MESSAGE_SINT, shared_sint.as_uint64);

	/* Enable the global synic bit */
	hv_get_synic_state(sctrl.as_uint64);
	sctrl.enable = 1;

	hv_set_synic_state(sctrl.as_uint64);

	hv_context.synic_initialized = true;

	/*
	 * Register the per-cpu clockevent source.
	 */
	if (ms_hyperv.features & HV_X64_MSR_SYNTIMER_AVAILABLE)
		clockevents_config_and_register(hv_cpu->clk_evt,
						HV_TIMER_FREQUENCY,
						HV_MIN_DELTA_TICKS,
						HV_MAX_MAX_DELTA_TICKS);
	return 0;
}

/*
 * hv_synic_clockevents_cleanup - Cleanup clockevent devices
 */
void hv_synic_clockevents_cleanup(void)
{
	int cpu;

	if (!(ms_hyperv.features & HV_X64_MSR_SYNTIMER_AVAILABLE))
		return;

	if (direct_mode_enabled)
		hv_remove_stimer0_irq(stimer0_irq);

	for_each_present_cpu(cpu) {
		struct hv_per_cpu_context *hv_cpu
			= per_cpu_ptr(hv_context.cpu_context, cpu);

		clockevents_unbind_device(hv_cpu->clk_evt, cpu);
	}
}

/*
 * hv_synic_cleanup - Cleanup routine for hv_synic_init().
 */
int hv_synic_cleanup(unsigned int cpu)
{
	union hv_synic_sint shared_sint;
	union hv_synic_simp simp;
	union hv_synic_siefp siefp;
	union hv_synic_scontrol sctrl;
	struct vmbus_channel *channel, *sc;
	bool channel_found = false;
	unsigned long flags;

	if (!hv_context.synic_initialized)
		return -EFAULT;

	/*
	 * Search for channels which are bound to the CPU we're about to
	 * cleanup. In case we find one and vmbus is still connected we need to
	 * fail, this will effectively prevent CPU offlining. There is no way
	 * we can re-bind channels to different CPUs for now.
	 */
	mutex_lock(&vmbus_connection.channel_mutex);
	list_for_each_entry(channel, &vmbus_connection.chn_list, listentry) {
		if (channel->target_cpu == cpu) {
			channel_found = true;
			break;
		}
		spin_lock_irqsave(&channel->lock, flags);
		list_for_each_entry(sc, &channel->sc_list, sc_list) {
			if (sc->target_cpu == cpu) {
				channel_found = true;
				break;
			}
		}
		spin_unlock_irqrestore(&channel->lock, flags);
		if (channel_found)
			break;
	}
	mutex_unlock(&vmbus_connection.channel_mutex);

	if (channel_found && vmbus_connection.conn_state == CONNECTED)
		return -EBUSY;

	/* Turn off clockevent device */
	if (ms_hyperv.features & HV_X64_MSR_SYNTIMER_AVAILABLE) {
		struct hv_per_cpu_context *hv_cpu
			= this_cpu_ptr(hv_context.cpu_context);

		clockevents_unbind_device(hv_cpu->clk_evt, cpu);
		hv_ce_shutdown(hv_cpu->clk_evt);
		put_cpu_ptr(hv_cpu);
	}

	hv_get_synint_state(VMBUS_MESSAGE_SINT, shared_sint.as_uint64);

	shared_sint.masked = 1;

	/* Need to correctly cleanup in the case of SMP!!! */
	/* Disable the interrupt */
	hv_set_synint_state(VMBUS_MESSAGE_SINT, shared_sint.as_uint64);

	hv_get_simp(simp.as_uint64);
	simp.simp_enabled = 0;
	simp.base_simp_gpa = 0;

	hv_set_simp(simp.as_uint64);

	hv_get_siefp(siefp.as_uint64);
	siefp.siefp_enabled = 0;
	siefp.base_siefp_gpa = 0;

	hv_set_siefp(siefp.as_uint64);

	/* Disable the global synic bit */
	hv_get_synic_state(sctrl.as_uint64);
	sctrl.enable = 0;
	hv_set_synic_state(sctrl.as_uint64);

	return 0;
}
Commit	Line	Data
3e7ee490	1	/*
3e7ee490 HJ	2	* Copyright (c) 2009, Microsoft Corporation.
	3	*
	4	* This program is free software; you can redistribute it and/or modify it
	5	* under the terms and conditions of the GNU General Public License,
	6	* version 2, as published by the Free Software Foundation.
	7	*
	8	* This program is distributed in the hope it will be useful, but WITHOUT
	9	* ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
	10	* FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
	11	* more details.
	12	*
	13	* You should have received a copy of the GNU General Public License along with
	14	* this program; if not, write to the Free Software Foundation, Inc., 59 Temple
	15	* Place - Suite 330, Boston, MA 02111-1307 USA.
	16	*
	17	* Authors:
	18	* Haiyang Zhang <[email protected]>
	19	* Hank Janssen <[email protected]>
	20	*
	21	*/
0a46618d HJ	22	#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
0a46618d HJ	23
a0086dc5 GKH	24	#include <linux/kernel.h>
a0086dc5 GKH	25	#include <linux/mm.h>
5a0e3ad6	26	#include <linux/slab.h>
b7c947f0	27	#include <linux/vmalloc.h>
46a97191	28	#include <linux/hyperv.h>
83ba0c4f	29	#include <linux/version.h>
248e742a	30	#include <linux/random.h>
4061ed9e	31	#include <linux/clockchips.h>
4061ed9e	32	#include <asm/mshyperv.h>
0f2a6619	33	#include "hyperv_vmbus.h"
3e7ee490	34
454f18a9	35	/* The one and only */
6a0aaa18 HZ	36	struct hv_context hv_context = {
6a0aaa18 HZ	37	.synic_initialized = false,
3e7ee490 HJ	38	};
3e7ee490 HJ	39
248e742a MK	40	/*
	41	* If false, we're using the old mechanism for stimer0 interrupts
	42	* where it sends a VMbus message when it expires. The old
	43	* mechanism is used when running on older versions of Hyper-V
	44	* that don't support Direct Mode. While Hyper-V provides
	45	* four stimer's per CPU, Linux uses only stimer0.
	46	*/
	47	static bool direct_mode_enabled;
	48	static int stimer0_irq;
	49	static int stimer0_vector;
	50
4061ed9e S	51	#define HV_TIMER_FREQUENCY (10 * 1000 * 1000) /* 100ns period */
	52	#define HV_MAX_MAX_DELTA_TICKS 0xffffffff
	53	#define HV_MIN_DELTA_TICKS 1
	54
3e189519	55	/*
d44890c8	56	* hv_init - Main initialization routine.
0831ad04 GKH	57	*
	58	* This routine must be called before any other routines in here are called
	59	*/
d44890c8	60	int hv_init(void)
3e7ee490	61	{
37cdd991 SH	62	hv_context.cpu_context = alloc_percpu(struct hv_per_cpu_context);
	63	if (!hv_context.cpu_context)
	64	return -ENOMEM;
	65
248e742a MK	66	direct_mode_enabled = ms_hyperv.misc_features &
248e742a MK	67	HV_X64_STIMER_DIRECT_MODE_AVAILABLE;
5433e003	68	return 0;
3e7ee490 HJ	69	}
3e7ee490 HJ	70
3e189519	71	/*
d44890c8	72	* hv_post_message - Post a message using the hypervisor message IPC.
0831ad04 GKH	73	*
	74	* This involves a hypercall.
	75	*/
415f0a02	76	int hv_post_message(union hv_connection_id connection_id,
b8dfb264 HZ	77	enum hv_message_type message_type,
b8dfb264 HZ	78	void *payload, size_t payload_size)
3e7ee490	79	{
b8dfb264	80	struct hv_input_post_message *aligned_msg;
37cdd991	81	struct hv_per_cpu_context *hv_cpu;
a108393d	82	u64 status;
3e7ee490	83
b8dfb264	84	if (payload_size > HV_MESSAGE_PAYLOAD_BYTE_COUNT)
39594abc	85	return -EMSGSIZE;
3e7ee490	86
37cdd991 SH	87	hv_cpu = get_cpu_ptr(hv_context.cpu_context);
37cdd991 SH	88	aligned_msg = hv_cpu->post_msg_page;
b8dfb264	89	aligned_msg->connectionid = connection_id;
b29ef354	90	aligned_msg->reserved = 0;
b8dfb264 HZ	91	aligned_msg->message_type = message_type;
	92	aligned_msg->payload_size = payload_size;
	93	memcpy((void *)aligned_msg->payload, payload, payload_size);
3e7ee490	94
a108393d	95	status = hv_do_hypercall(HVCALL_POST_MESSAGE, aligned_msg, NULL);
3e7ee490	96
13b9abfc MK	97	/* Preemption must remain disabled until after the hypercall
	98	* so some other thread can't get scheduled onto this cpu and
	99	* corrupt the per-cpu post_msg_page
	100	*/
	101	put_cpu_ptr(hv_cpu);
	102
a108393d	103	return status & 0xFFFF;
3e7ee490 HJ	104	}
3e7ee490 HJ	105
248e742a MK	106	/*
	107	* ISR for when stimer0 is operating in Direct Mode. Direct Mode
	108	* does not use VMbus or any VMbus messages, so process here and not
	109	* in the VMbus driver code.
	110	*/
	111
	112	static void hv_stimer0_isr(void)
	113	{
	114	struct hv_per_cpu_context *hv_cpu;
	115
	116	hv_cpu = this_cpu_ptr(hv_context.cpu_context);
	117	hv_cpu->clk_evt->event_handler(hv_cpu->clk_evt);
	118	add_interrupt_randomness(stimer0_vector, 0);
	119	}
	120
4061ed9e S	121	static int hv_ce_set_next_event(unsigned long delta,
	122	struct clock_event_device *evt)
	123	{
a5a1d1c2	124	u64 current_tick;
4061ed9e	125
bc609cb4	126	WARN_ON(!clockevent_state_oneshot(evt));
4061ed9e	127
e546d778	128	current_tick = hyperv_cs->read(NULL);
4061ed9e	129	current_tick += delta;
619a4c8b	130	hv_init_timer(0, current_tick);
4061ed9e S	131	return 0;
	132	}
	133
bc609cb4 VK	134	static int hv_ce_shutdown(struct clock_event_device *evt)
bc609cb4 VK	135	{
619a4c8b MK	136	hv_init_timer(0, 0);
619a4c8b MK	137	hv_init_timer_config(0, 0);
248e742a MK	138	if (direct_mode_enabled)
248e742a MK	139	hv_disable_stimer0_percpu_irq(stimer0_irq);
bc609cb4 VK	140
	141	return 0;
	142	}
	143
	144	static int hv_ce_set_oneshot(struct clock_event_device *evt)
4061ed9e S	145	{
	146	union hv_timer_config timer_cfg;
	147
248e742a	148	timer_cfg.as_uint64 = 0;
bc609cb4 VK	149	timer_cfg.enable = 1;
bc609cb4 VK	150	timer_cfg.auto_enable = 1;
248e742a MK	151	if (direct_mode_enabled) {
	152	/*
	153	* When it expires, the timer will directly interrupt
	154	* on the specified hardware vector/IRQ.
	155	*/
	156	timer_cfg.direct_mode = 1;
	157	timer_cfg.apic_vector = stimer0_vector;
	158	hv_enable_stimer0_percpu_irq(stimer0_irq);
	159	} else {
	160	/*
	161	* When it expires, the timer will generate a VMbus message,
	162	* to be handled by the normal VMbus interrupt handler.
	163	*/
	164	timer_cfg.direct_mode = 0;
	165	timer_cfg.sintx = VMBUS_MESSAGE_SINT;
	166	}
619a4c8b	167	hv_init_timer_config(0, timer_cfg.as_uint64);
bc609cb4	168	return 0;
4061ed9e S	169	}
	170
	171	static void hv_init_clockevent_device(struct clock_event_device *dev, int cpu)
	172	{
	173	dev->name = "Hyper-V clockevent";
	174	dev->features = CLOCK_EVT_FEAT_ONESHOT;
	175	dev->cpumask = cpumask_of(cpu);
	176	dev->rating = 1000;
e086748c VK	177	/*
	178	* Avoid settint dev->owner = THIS_MODULE deliberately as doing so will
	179	* result in clockevents_config_and_register() taking additional
	180	* references to the hv_vmbus module making it impossible to unload.
	181	*/
4061ed9e	182
bc609cb4 VK	183	dev->set_state_shutdown = hv_ce_shutdown;
bc609cb4 VK	184	dev->set_state_oneshot = hv_ce_set_oneshot;
4061ed9e S	185	dev->set_next_event = hv_ce_set_next_event;
	186	}
	187
2608fb65 JW	188
	189	int hv_synic_alloc(void)
	190	{
2608fb65 JW	191	int cpu;
2608fb65 JW	192
6396bb22	193	hv_context.hv_numa_map = kcalloc(nr_node_ids, sizeof(struct cpumask),
597ff72f	194	GFP_KERNEL);
9f01ec53 S	195	if (hv_context.hv_numa_map == NULL) {
	196	pr_err("Unable to allocate NUMA map\n");
	197	goto err;
	198	}
	199
421b8f20	200	for_each_present_cpu(cpu) {
37cdd991 SH	201	struct hv_per_cpu_context *hv_cpu
	202	= per_cpu_ptr(hv_context.cpu_context, cpu);
	203
	204	memset(hv_cpu, 0, sizeof(*hv_cpu));
37cdd991 SH	205	tasklet_init(&hv_cpu->msg_dpc,
	206	vmbus_on_msg_dpc, (unsigned long) hv_cpu);
	207
	208	hv_cpu->clk_evt = kzalloc(sizeof(struct clock_event_device),
	209	GFP_KERNEL);
	210	if (hv_cpu->clk_evt == NULL) {
4061ed9e S	211	pr_err("Unable to allocate clock event device\n");
	212	goto err;
	213	}
37cdd991	214	hv_init_clockevent_device(hv_cpu->clk_evt, cpu);
9f01ec53	215
37cdd991	216	hv_cpu->synic_message_page =
2608fb65	217	(void *)get_zeroed_page(GFP_ATOMIC);
37cdd991	218	if (hv_cpu->synic_message_page == NULL) {
2608fb65 JW	219	pr_err("Unable to allocate SYNIC message page\n");
	220	goto err;
	221	}
	222
37cdd991 SH	223	hv_cpu->synic_event_page = (void *)get_zeroed_page(GFP_ATOMIC);
37cdd991 SH	224	if (hv_cpu->synic_event_page == NULL) {
2608fb65 JW	225	pr_err("Unable to allocate SYNIC event page\n");
	226	goto err;
	227	}
b29ef354	228
37cdd991 SH	229	hv_cpu->post_msg_page = (void *)get_zeroed_page(GFP_ATOMIC);
37cdd991 SH	230	if (hv_cpu->post_msg_page == NULL) {
b29ef354 S	231	pr_err("Unable to allocate post msg page\n");
	232	goto err;
	233	}
3c7630d3	234
37cdd991	235	INIT_LIST_HEAD(&hv_cpu->chan_list);
2608fb65 JW	236	}
2608fb65 JW	237
248e742a MK	238	if (direct_mode_enabled &&
	239	hv_setup_stimer0_irq(&stimer0_irq, &stimer0_vector,
	240	hv_stimer0_isr))
	241	goto err;
	242
2608fb65 JW	243	return 0;
	244	err:
	245	return -ENOMEM;
	246	}
	247
2608fb65 JW	248
	249	void hv_synic_free(void)
	250	{
	251	int cpu;
	252
37cdd991 SH	253	for_each_present_cpu(cpu) {
	254	struct hv_per_cpu_context *hv_cpu
	255	= per_cpu_ptr(hv_context.cpu_context, cpu);
	256
	257	if (hv_cpu->synic_event_page)
	258	free_page((unsigned long)hv_cpu->synic_event_page);
	259	if (hv_cpu->synic_message_page)
	260	free_page((unsigned long)hv_cpu->synic_message_page);
	261	if (hv_cpu->post_msg_page)
	262	free_page((unsigned long)hv_cpu->post_msg_page);
	263	}
	264
9f01ec53	265	kfree(hv_context.hv_numa_map);
2608fb65 JW	266	}
2608fb65 JW	267
3e189519	268	/*
68cb8117	269	* hv_synic_init - Initialize the Synthetic Interrupt Controller.
0831ad04 GKH	270	*
	271	* If it is already initialized by another entity (ie x2v shim), we need to
	272	* retrieve the initialized message and event pages. Otherwise, we create and
	273	* initialize the message and event pages.
	274	*/
76d36ab7	275	int hv_synic_init(unsigned int cpu)
3e7ee490	276	{
37cdd991 SH	277	struct hv_per_cpu_context *hv_cpu
37cdd991 SH	278	= per_cpu_ptr(hv_context.cpu_context, cpu);
eacb1b4d GKH	279	union hv_synic_simp simp;
eacb1b4d GKH	280	union hv_synic_siefp siefp;
b8dfb264	281	union hv_synic_sint shared_sint;
eacb1b4d	282	union hv_synic_scontrol sctrl;
a73e6b7c	283
a73e6b7c	284	/* Setup the Synic's message page */
155e4a2f	285	hv_get_simp(simp.as_uint64);
f6feebe0	286	simp.simp_enabled = 1;
37cdd991	287	simp.base_simp_gpa = virt_to_phys(hv_cpu->synic_message_page)
a73e6b7c	288	>> PAGE_SHIFT;
3e7ee490	289
155e4a2f	290	hv_set_simp(simp.as_uint64);
3e7ee490	291
a73e6b7c	292	/* Setup the Synic's event page */
8e307bf8	293	hv_get_siefp(siefp.as_uint64);
f6feebe0	294	siefp.siefp_enabled = 1;
37cdd991	295	siefp.base_siefp_gpa = virt_to_phys(hv_cpu->synic_event_page)
a73e6b7c HJ	296	>> PAGE_SHIFT;
a73e6b7c HJ	297
8e307bf8	298	hv_set_siefp(siefp.as_uint64);
0831ad04	299
0831ad04	300	/* Setup the shared SINT. */
619a4c8b	301	hv_get_synint_state(VMBUS_MESSAGE_SINT, shared_sint.as_uint64);
3e7ee490	302
302a3c0f	303	shared_sint.vector = HYPERVISOR_CALLBACK_VECTOR;
b8dfb264	304	shared_sint.masked = false;
6c248aad S	305	if (ms_hyperv.hints & HV_X64_DEPRECATING_AEOI_RECOMMENDED)
	306	shared_sint.auto_eoi = false;
	307	else
	308	shared_sint.auto_eoi = true;
3e7ee490	309
619a4c8b	310	hv_set_synint_state(VMBUS_MESSAGE_SINT, shared_sint.as_uint64);
3e7ee490	311
454f18a9	312	/* Enable the global synic bit */
06d1d98a	313	hv_get_synic_state(sctrl.as_uint64);
f6feebe0	314	sctrl.enable = 1;
3e7ee490	315
06d1d98a	316	hv_set_synic_state(sctrl.as_uint64);
3e7ee490	317
6a0aaa18	318	hv_context.synic_initialized = true;
917ea427	319
4061ed9e S	320	/*
	321	* Register the per-cpu clockevent source.
	322	*/
	323	if (ms_hyperv.features & HV_X64_MSR_SYNTIMER_AVAILABLE)
37cdd991	324	clockevents_config_and_register(hv_cpu->clk_evt,
4061ed9e S	325	HV_TIMER_FREQUENCY,
	326	HV_MIN_DELTA_TICKS,
	327	HV_MAX_MAX_DELTA_TICKS);
76d36ab7	328	return 0;
3e7ee490 HJ	329	}
3e7ee490 HJ	330
e086748c VK	331	/*
	332	* hv_synic_clockevents_cleanup - Cleanup clockevent devices
	333	*/
	334	void hv_synic_clockevents_cleanup(void)
	335	{
	336	int cpu;
	337
	338	if (!(ms_hyperv.features & HV_X64_MSR_SYNTIMER_AVAILABLE))
	339	return;
	340
248e742a MK	341	if (direct_mode_enabled)
	342	hv_remove_stimer0_irq(stimer0_irq);
	343
37cdd991 SH	344	for_each_present_cpu(cpu) {
	345	struct hv_per_cpu_context *hv_cpu
	346	= per_cpu_ptr(hv_context.cpu_context, cpu);
	347
	348	clockevents_unbind_device(hv_cpu->clk_evt, cpu);
	349	}
e086748c VK	350	}
e086748c VK	351
3e189519	352	/*
d44890c8	353	* hv_synic_cleanup - Cleanup routine for hv_synic_init().
0831ad04	354	*/
76d36ab7	355	int hv_synic_cleanup(unsigned int cpu)
3e7ee490	356	{
b8dfb264	357	union hv_synic_sint shared_sint;
eacb1b4d GKH	358	union hv_synic_simp simp;
eacb1b4d GKH	359	union hv_synic_siefp siefp;
e72e7ac5	360	union hv_synic_scontrol sctrl;
523b9408 VK	361	struct vmbus_channel channel, sc;
	362	bool channel_found = false;
	363	unsigned long flags;
3e7ee490	364
6a0aaa18	365	if (!hv_context.synic_initialized)
76d36ab7	366	return -EFAULT;
3e7ee490	367
523b9408 VK	368	/*
	369	* Search for channels which are bound to the CPU we're about to
	370	* cleanup. In case we find one and vmbus is still connected we need to
	371	* fail, this will effectively prevent CPU offlining. There is no way
	372	* we can re-bind channels to different CPUs for now.
	373	*/
	374	mutex_lock(&vmbus_connection.channel_mutex);
	375	list_for_each_entry(channel, &vmbus_connection.chn_list, listentry) {
	376	if (channel->target_cpu == cpu) {
	377	channel_found = true;
	378	break;
	379	}
	380	spin_lock_irqsave(&channel->lock, flags);
	381	list_for_each_entry(sc, &channel->sc_list, sc_list) {
	382	if (sc->target_cpu == cpu) {
	383	channel_found = true;
	384	break;
	385	}
	386	}
	387	spin_unlock_irqrestore(&channel->lock, flags);
	388	if (channel_found)
	389	break;
	390	}
	391	mutex_unlock(&vmbus_connection.channel_mutex);
	392
	393	if (channel_found && vmbus_connection.conn_state == CONNECTED)
	394	return -EBUSY;
	395
e086748c	396	/* Turn off clockevent device */
6ffc4b85	397	if (ms_hyperv.features & HV_X64_MSR_SYNTIMER_AVAILABLE) {
37cdd991 SH	398	struct hv_per_cpu_context *hv_cpu
	399	= this_cpu_ptr(hv_context.cpu_context);
	400
	401	clockevents_unbind_device(hv_cpu->clk_evt, cpu);
	402	hv_ce_shutdown(hv_cpu->clk_evt);
	403	put_cpu_ptr(hv_cpu);
6ffc4b85	404	}
e086748c	405
619a4c8b	406	hv_get_synint_state(VMBUS_MESSAGE_SINT, shared_sint.as_uint64);
3e7ee490	407
b8dfb264	408	shared_sint.masked = 1;
3e7ee490	409
7692fd4d	410	/* Need to correctly cleanup in the case of SMP!!! */
454f18a9	411	/* Disable the interrupt */
619a4c8b	412	hv_set_synint_state(VMBUS_MESSAGE_SINT, shared_sint.as_uint64);
3e7ee490	413
155e4a2f	414	hv_get_simp(simp.as_uint64);
f6feebe0 HZ	415	simp.simp_enabled = 0;
f6feebe0 HZ	416	simp.base_simp_gpa = 0;
3e7ee490	417
155e4a2f	418	hv_set_simp(simp.as_uint64);
3e7ee490	419
8e307bf8	420	hv_get_siefp(siefp.as_uint64);
f6feebe0 HZ	421	siefp.siefp_enabled = 0;
f6feebe0 HZ	422	siefp.base_siefp_gpa = 0;
3e7ee490	423
8e307bf8	424	hv_set_siefp(siefp.as_uint64);
3e7ee490	425
e72e7ac5	426	/* Disable the global synic bit */
06d1d98a	427	hv_get_synic_state(sctrl.as_uint64);
e72e7ac5	428	sctrl.enable = 0;
06d1d98a	429	hv_set_synic_state(sctrl.as_uint64);
76d36ab7 VK	430
76d36ab7 VK	431	return 0;
3e7ee490	432	}